diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 32dd8ccaada7bf4ce12b5d6899ea5666da53c1c6..a4e7f015f22019402b30bd2e145a4a9b0a9b72dd 100644 --- a/analysis/current/analysis-report-hybrid.json +++ b/analysis/current/analysis-report-hybrid.json @@ -1,498 +1,1305 @@ { "schema_version": "1.0", "repo": "huggingface/transformers", - "snapshot_id": "20260719T000028Z", - "generated_at": "2026-07-19T00:11:48Z", + "snapshot_id": "20260719T060027Z", + "generated_at": "2026-07-19T06:10:36Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ { - "cluster_id": "cluster-45157-14", - "summary": "Cluster of 14 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-43451-15", + "summary": "Cluster of 1 issues and 14 PRs centered on issue #47355.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, + "confidence": 0.8, + "canonical_issue_number": 47355, "canonical_pr_number": 47322, - "issue_numbers": [], + "issue_numbers": [ + 47355 + ], "pr_numbers": [ - 45157, - 46196, + 43451, 46509, 46658, + 47003, + 47096, 47143, 47163, - 47210, - 47216, - 47301, + 47298, + 47306, 47322, - 47336, - 47367, 47383, - 47387 + 47398, + 47399, + 47409 ], "evidence_types": [ + "closing_reference", + "shared_issue_target", "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 45157, - "right_pr_number": 46196, + "left_pr_number": 43451, + "right_pr_number": 46509, + "code_similarity": 0.167, + "size_similarity": 0.378, + "file_overlap": 0.097, + "area_overlap": 0.074, + "patch_similarity": 0.245, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 1375, + 1382 + ] + ], + "right_ranges": [ + [ + 795, + 802 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 1097, + 1103 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 46658, + "code_similarity": 0.276, + "size_similarity": 0.77, + "file_overlap": 0.071, + "area_overlap": 0.163, + "patch_similarity": 0.286, + "shared_filenames": [ + "src/transformers/conversion_mapping.py", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/conversion_mapping.py", + "left_ranges": [ + [ + 737, + 769 + ] + ], + "right_ranges": [ + [ + 739, + 830 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 1085, + 1091 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47003, + "code_similarity": 0.202, + "size_similarity": 0.861, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.196, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47096, + "code_similarity": 0.036, + "size_similarity": 0.099, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.108, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47143, + "code_similarity": 0.154, + "size_similarity": 0.476, + "file_overlap": 0.097, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [ + "src/transformers/models/__init__.py", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/__init__.py", + "left_ranges": [ + [ + 290, + 296 + ] + ], + "right_ranges": [ + [ + 305, + 311 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.004, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.02, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47298, "code_similarity": 0.02, - "size_similarity": 0.05, + "size_similarity": 0.055, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.07, + "patch_similarity": 0.06, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 46509, - "code_similarity": 0.032, - "size_similarity": 0.101, + "left_pr_number": 43451, + "right_pr_number": 47306, + "code_similarity": 0.044, + "size_similarity": 0.137, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.109, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47383, + "code_similarity": 0.011, + "size_similarity": 0.031, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.029, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47399, + "code_similarity": 0.003, + "size_similarity": 0.008, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.006, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47409, + "code_similarity": 0.011, + "size_similarity": 0.019, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.047, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 46658, + "code_similarity": 0.131, + "size_similarity": 0.291, + "file_overlap": 0.051, + "area_overlap": 0.081, + "patch_similarity": 0.196, + "shared_filenames": [ + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 1085, + 1091 + ] + ] + } + ] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47003, + "code_similarity": 0.115, + "size_similarity": 0.438, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.08, + "patch_similarity": 0.182, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 46658, - "code_similarity": 0.012, - "size_similarity": 0.029, + "left_pr_number": 46509, + "right_pr_number": 47096, + "code_similarity": 0.071, + "size_similarity": 0.263, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.044, + "patch_similarity": 0.121, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, + "left_pr_number": 46509, "right_pr_number": 47143, - "code_similarity": 0.022, - "size_similarity": 0.08, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.04, - "shared_filenames": [], - "shared_file_areas": [] + "code_similarity": 0.254, + "size_similarity": 0.793, + "file_overlap": 0.071, + "area_overlap": 0.143, + "patch_similarity": 0.161, + "shared_filenames": [ + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] }, { - "left_pr_number": 45157, + "left_pr_number": 46509, "right_pr_number": 47163, - "code_similarity": 0.029, - "size_similarity": 0.117, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.035, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 45157, - "right_pr_number": 47210, - "code_similarity": 0.152, - "size_similarity": 0.65, + "code_similarity": 0.006, + "size_similarity": 0.012, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.146, + "patch_similarity": 0.024, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 47216, - "code_similarity": 0.08, - "size_similarity": 0.346, + "left_pr_number": 46509, + "right_pr_number": 47298, + "code_similarity": 0.039, + "size_similarity": 0.146, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.072, + "patch_similarity": 0.064, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 47301, - "code_similarity": 0.031, - "size_similarity": 0.125, + "left_pr_number": 46509, + "right_pr_number": 47306, + "code_similarity": 0.092, + "size_similarity": 0.363, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.039, + "patch_similarity": 0.127, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, + "left_pr_number": 46509, "right_pr_number": 47322, - "code_similarity": 0.013, - "size_similarity": 0.058, + "code_similarity": 0.002, + "size_similarity": 0.006, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.009, + "patch_similarity": 0.004, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 47336, - "code_similarity": 0.016, - "size_similarity": 0.075, + "left_pr_number": 46509, + "right_pr_number": 47383, + "code_similarity": 0.022, + "size_similarity": 0.082, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.008, + "patch_similarity": 0.036, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 47367, - "code_similarity": 0.057, - "size_similarity": 0.254, + "left_pr_number": 46509, + "right_pr_number": 47398, + "code_similarity": 0.002, + "size_similarity": 0.007, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.039, + "patch_similarity": 0.005, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 47383, - "code_similarity": 0.167, - "size_similarity": 0.821, + "left_pr_number": 46509, + "right_pr_number": 47399, + "code_similarity": 0.006, + "size_similarity": 0.022, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.018, + "patch_similarity": 0.009, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 45157, - "right_pr_number": 47387, - "code_similarity": 0.176, - "size_similarity": 0.842, + "left_pr_number": 46509, + "right_pr_number": 47409, + "code_similarity": 0.021, + "size_similarity": 0.049, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.053, + "patch_similarity": 0.073, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 46509, - "code_similarity": 0.118, - "size_similarity": 0.493, + "left_pr_number": 46658, + "right_pr_number": 47003, + "code_similarity": 0.166, + "size_similarity": 0.663, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.129, + "patch_similarity": 0.221, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 46658, - "code_similarity": 0.133, - "size_similarity": 0.59, + "left_pr_number": 46658, + "right_pr_number": 47096, + "code_similarity": 0.03, + "size_similarity": 0.077, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.102, + "patch_similarity": 0.097, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, + "left_pr_number": 46658, "right_pr_number": 47143, - "code_similarity": 0.138, - "size_similarity": 0.622, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.092, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46196, - "right_pr_number": 47163, - "code_similarity": 0.006, - "size_similarity": 0.006, - "file_overlap": 0.0, + "code_similarity": 0.116, + "size_similarity": 0.367, + "file_overlap": 0.051, "area_overlap": 0.0, - "patch_similarity": 0.031, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46196, - "right_pr_number": 47210, - "code_similarity": 0.068, - "size_similarity": 0.076, - "file_overlap": 0.03, - "area_overlap": 0.069, - "patch_similarity": 0.129, + "patch_similarity": 0.183, "shared_filenames": [ - "src/transformers/integrations/sdpa_attention.py" + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" ], "shared_file_areas": [ { - "filename": "src/transformers/integrations/sdpa_attention.py", + "filename": "src/transformers/models/auto/auto_mappings.py", "left_ranges": [ [ - 78, - 89 + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 ] ], "right_ranges": [ [ - 1, - 9 + 412, + 418 ], [ - 13, - 43 - ], + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ [ - 64, - 151 - ], + 1085, + 1091 + ] + ], + "right_ranges": [ [ - 162, - 200 + 348, + 354 ], [ - 206, - 211 + 1960, + 1966 ] ] } ] }, { - "left_pr_number": 46196, - "right_pr_number": 47216, - "code_similarity": 0.047, - "size_similarity": 0.143, + "left_pr_number": 46658, + "right_pr_number": 47163, + "code_similarity": 0.003, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.015, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47298, + "code_similarity": 0.015, + "size_similarity": 0.042, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.12, + "patch_similarity": 0.045, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 47301, - "code_similarity": 0.01, - "size_similarity": 0.006, + "left_pr_number": 46658, + "right_pr_number": 47306, + "code_similarity": 0.035, + "size_similarity": 0.105, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.056, + "patch_similarity": 0.096, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, + "left_pr_number": 46658, "right_pr_number": 47322, - "code_similarity": 0.002, - "size_similarity": 0.003, + "code_similarity": 0.001, + "size_similarity": 0.002, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.009, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 47336, - "code_similarity": 0.002, - "size_similarity": 0.004, + "left_pr_number": 46658, + "right_pr_number": 47383, + "code_similarity": 0.007, + "size_similarity": 0.024, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.005, + "patch_similarity": 0.017, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 47367, - "code_similarity": 0.009, - "size_similarity": 0.013, + "left_pr_number": 46658, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.002, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.046, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 47383, - "code_similarity": 0.012, - "size_similarity": 0.041, + "left_pr_number": 46658, + "right_pr_number": 47399, + "code_similarity": 0.002, + "size_similarity": 0.006, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.027, + "patch_similarity": 0.005, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 47387, - "code_similarity": 0.026, - "size_similarity": 0.059, + "left_pr_number": 46658, + "right_pr_number": 47409, + "code_similarity": 0.008, + "size_similarity": 0.014, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.096, + "patch_similarity": 0.036, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46509, - "right_pr_number": 46658, - "code_similarity": 0.131, - "size_similarity": 0.291, - "file_overlap": 0.051, - "area_overlap": 0.081, - "patch_similarity": 0.196, + "left_pr_number": 47003, + "right_pr_number": 47096, + "code_similarity": 0.21, + "size_similarity": 0.115, + "file_overlap": 0.113, + "area_overlap": 0.352, + "patch_similarity": 0.201, "shared_filenames": [ - "src/transformers/models/auto/auto_mappings.py", - "src/transformers/models/auto/modeling_auto.py" + "src/transformers/exporters/exporter_dynamo.py", + "src/transformers/exporters/exporter_executorch.py", + "src/transformers/exporters/utils.py", + "src/transformers/models/kimi_k25/modeling_kimi_k25.py", + "src/transformers/models/kimi_k25/modular_kimi_k25.py", + "src/transformers/vision_utils.py", + "tests/exporters/test_export.py", + "tests/models/kimi_k25/test_modeling_kimi_k25.py" ], "shared_file_areas": [ { - "filename": "src/transformers/models/auto/auto_mappings.py", + "filename": "src/transformers/exporters/exporter_dynamo.py", + "left_ranges": [ + [ + 43, + 49 + ], + [ + 352, + 359 + ], + [ + 383, + 390 + ], + [ + 452, + 473 + ], + [ + 490, + 510 + ], + [ + 551, + 562 + ], + [ + 597, + 609 + ] + ], + "right_ranges": [ + [ + 350, + 357 + ], + [ + 381, + 388 + ] + ] + }, + { + "filename": "src/transformers/exporters/exporter_executorch.py", + "left_ranges": [ + [ + 70, + 88 + ], + [ + 196, + 204 + ] + ], + "right_ranges": [ + [ + 418, + 442 + ] + ] + }, + { + "filename": "src/transformers/exporters/utils.py", + "left_ranges": [ + [ + 232, + 238 + ], + [ + 432, + 441 + ], + [ + 444, + 467 + ], + [ + 472, + 498 + ], + [ + 514, + 536 + ], + [ + 662, + 706 + ], + [ + 747, + 759 + ], + [ + 808, + 815 + ] + ], + "right_ranges": [ + [ + 432, + 441 + ], + [ + 444, + 455 + ], + [ + 464, + 490 + ], + [ + 708, + 720 + ] + ] + }, + { + "filename": "src/transformers/models/kimi_k25/modeling_kimi_k25.py", + "left_ranges": [ + [ + 43, + 49 + ], + [ + 99, + 172 + ], + [ + 176, + 184 + ], + [ + 191, + 212 + ], + [ + 218, + 226 + ], + [ + 281, + 294 + ], + [ + 500, + 532 + ], + [ + 545, + 550 + ], + [ + 557, + 569 + ], + [ + 575, + 582 + ] + ], + "right_ranges": [ + [ + 43, + 49 + ], + [ + 99, + 172 + ], + [ + 176, + 184 + ], + [ + 189, + 210 + ], + [ + 216, + 224 + ], + [ + 279, + 292 + ], + [ + 498, + 530 + ], + [ + 543, + 548 + ], + [ + 555, + 567 + ], + [ + 573, + 580 + ] + ] + }, + { + "filename": "src/transformers/models/kimi_k25/modular_kimi_k25.py", + "left_ranges": [ + [ + 35, + 41 + ], + [ + 54, + 154 + ], + [ + 240, + 248 + ], + [ + 255, + 276 + ], + [ + 282, + 305 + ], + [ + 431, + 436 + ], + [ + 443, + 455 + ], + [ + 461, + 468 + ] + ], + "right_ranges": [ + [ + 35, + 41 + ], + [ + 54, + 154 + ], + [ + 240, + 248 + ], + [ + 253, + 274 + ], + [ + 280, + 303 + ], + [ + 429, + 434 + ], + [ + 441, + 453 + ], + [ + 459, + 466 + ] + ] + }, + { + "filename": "src/transformers/vision_utils.py", + "left_ranges": [ + [ + 32, + 61 + ] + ], + "right_ranges": [ + [ + 32, + 61 + ] + ] + }, + { + "filename": "tests/exporters/test_export.py", "left_ranges": [ [ - 412, - 418 + 13, + 21 + ], + [ + 26, + 32 + ], + [ + 37, + 43 + ], + [ + 49, + 132 + ], + [ + 239, + 279 ], [ - 1077, - 1083 + 307, + 341 ], [ - 1206, - 1212 - ] - ], - "right_ranges": [ + 354, + 461 + ], [ - 362, - 368 + 493, + 517 ], [ - 591, - 599 + 532, + 539 ], [ - 798, - 804 + 564, + 570 ], [ - 884, - 891 + 573, + 582 ], [ - 1084, - 1090 + 599, + 605 ], [ - 1207, - 1213 - ] - ] - }, - { - "filename": "src/transformers/models/auto/modeling_auto.py", - "left_ranges": [ + 608, + 617 + ], [ - 1097, - 1103 - ] - ], - "right_ranges": [ + 621, + 656 + ], [ - 1085, - 1091 - ] - ] - } - ] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47143, - "code_similarity": 0.254, - "size_similarity": 0.793, - "file_overlap": 0.071, - "area_overlap": 0.143, - "patch_similarity": 0.161, - "shared_filenames": [ - "src/transformers/models/auto/auto_mappings.py", - "src/transformers/models/auto/modeling_auto.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/auto/auto_mappings.py", - "left_ranges": [ + 659, + 677 + ], [ - 412, - 418 + 701, + 708 ], [ - 1077, - 1083 + 711, + 717 ], [ - 1206, - 1212 + 720, + 726 + ], + [ + 732, + 738 + ], + [ + 755, + 761 + ], + [ + 764, + 773 + ], + [ + 778, + 813 + ], + [ + 816, + 834 ] ], "right_ranges": [ [ - 412, - 418 - ], - [ - 975, - 981 + 181, + 188 ] ] }, { - "filename": "src/transformers/models/auto/modeling_auto.py", + "filename": "tests/models/kimi_k25/test_modeling_kimi_k25.py", "left_ranges": [ [ - 1097, - 1103 + 127, + 132 ] ], "right_ranges": [ [ - 348, - 354 - ], - [ - 1960, - 1966 + 127, + 132 ] ] } ] }, { - "left_pr_number": 46509, - "right_pr_number": 47163, - "code_similarity": 0.006, - "size_similarity": 0.012, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.024, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47210, - "code_similarity": 0.053, - "size_similarity": 0.155, + "left_pr_number": 47003, + "right_pr_number": 47143, + "code_similarity": 0.132, + "size_similarity": 0.553, "file_overlap": 0.0, "area_overlap": 0.0, "patch_similarity": 0.145, @@ -500,370 +1307,319 @@ "shared_file_areas": [] }, { - "left_pr_number": 46509, - "right_pr_number": 47216, - "code_similarity": 0.081, - "size_similarity": 0.291, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.152, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47301, - "code_similarity": 0.01, - "size_similarity": 0.013, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.047, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47322, - "code_similarity": 0.002, - "size_similarity": 0.006, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.004, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47336, - "code_similarity": 0.003, - "size_similarity": 0.008, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.007, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47367, - "code_similarity": 0.011, - "size_similarity": 0.026, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.04, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47383, - "code_similarity": 0.022, - "size_similarity": 0.082, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.036, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47387, - "code_similarity": 0.038, - "size_similarity": 0.119, + "left_pr_number": 47003, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.005, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.095, + "patch_similarity": 0.02, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47143, - "code_similarity": 0.116, - "size_similarity": 0.367, - "file_overlap": 0.051, + "left_pr_number": 47003, + "right_pr_number": 47298, + "code_similarity": 0.027, + "size_similarity": 0.064, + "file_overlap": 0.023, "area_overlap": 0.0, - "patch_similarity": 0.183, + "patch_similarity": 0.052, "shared_filenames": [ - "src/transformers/models/auto/auto_mappings.py", - "src/transformers/models/auto/modeling_auto.py" + "src/transformers/utils/__init__.py", + "src/transformers/utils/import_utils.py" ], "shared_file_areas": [ { - "filename": "src/transformers/models/auto/auto_mappings.py", + "filename": "src/transformers/utils/__init__.py", "left_ranges": [ [ - 362, - 368 - ], - [ - 591, - 599 - ], - [ - 798, - 804 - ], - [ - 884, - 891 - ], + 182, + 188 + ] + ], + "right_ranges": [ [ - 1084, - 1090 - ], + 253, + 259 + ] + ] + }, + { + "filename": "src/transformers/utils/import_utils.py", + "left_ranges": [ [ - 1207, - 1213 + 916, + 926 ] ], "right_ranges": [ [ - 412, - 418 - ], - [ - 975, - 981 + 756, + 785 ] ] - }, + } + ] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47306, + "code_similarity": 0.046, + "size_similarity": 0.159, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.093, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47383, + "code_similarity": 0.015, + "size_similarity": 0.036, + "file_overlap": 0.014, + "area_overlap": 0.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "tests/models/kimi_k25/test_modeling_kimi_k25.py" + ], + "shared_file_areas": [ { - "filename": "src/transformers/models/auto/modeling_auto.py", + "filename": "tests/models/kimi_k25/test_modeling_kimi_k25.py", "left_ranges": [ [ - 1085, - 1091 + 127, + 132 ] ], "right_ranges": [ [ - 348, - 354 + 18, + 24 ], [ - 1960, - 1966 + 27, + 36 + ], + [ + 216, + 401 ] ] } ] }, { - "left_pr_number": 46658, - "right_pr_number": 47163, - "code_similarity": 0.003, + "left_pr_number": 47003, + "right_pr_number": 47398, + "code_similarity": 0.001, "size_similarity": 0.003, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.015, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47210, - "code_similarity": 0.024, - "size_similarity": 0.045, + "left_pr_number": 47003, + "right_pr_number": 47399, + "code_similarity": 0.003, + "size_similarity": 0.01, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.099, + "patch_similarity": 0.007, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47216, - "code_similarity": 0.036, - "size_similarity": 0.085, + "left_pr_number": 47003, + "right_pr_number": 47409, + "code_similarity": 0.011, + "size_similarity": 0.022, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.129, + "patch_similarity": 0.042, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47301, - "code_similarity": 0.005, - "size_similarity": 0.004, + "left_pr_number": 47096, + "right_pr_number": 47143, + "code_similarity": 0.052, + "size_similarity": 0.208, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.029, + "patch_similarity": 0.069, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47322, - "code_similarity": 0.001, - "size_similarity": 0.002, + "left_pr_number": 47096, + "right_pr_number": 47163, + "code_similarity": 0.012, + "size_similarity": 0.045, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.002, + "patch_similarity": 0.021, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47336, - "code_similarity": 0.001, - "size_similarity": 0.002, + "left_pr_number": 47096, + "right_pr_number": 47298, + "code_similarity": 0.12, + "size_similarity": 0.554, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.006, + "patch_similarity": 0.06, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47367, - "code_similarity": 0.005, - "size_similarity": 0.007, + "left_pr_number": 47096, + "right_pr_number": 47306, + "code_similarity": 0.158, + "size_similarity": 0.725, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.024, + "patch_similarity": 0.089, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47383, - "code_similarity": 0.007, - "size_similarity": 0.024, + "left_pr_number": 47096, + "right_pr_number": 47322, + "code_similarity": 0.006, + "size_similarity": 0.022, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.017, + "patch_similarity": 0.011, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, - "right_pr_number": 47387, - "code_similarity": 0.035, - "size_similarity": 0.035, - "file_overlap": 0.036, - "area_overlap": 0.011, - "patch_similarity": 0.087, + "left_pr_number": 47096, + "right_pr_number": 47383, + "code_similarity": 0.103, + "size_similarity": 0.314, + "file_overlap": 0.125, + "area_overlap": 0.0, + "patch_similarity": 0.02, "shared_filenames": [ - "src/transformers/core_model_loading.py" + "tests/models/kimi_k25/test_modeling_kimi_k25.py" ], "shared_file_areas": [ { - "filename": "src/transformers/core_model_loading.py", + "filename": "tests/models/kimi_k25/test_modeling_kimi_k25.py", "left_ranges": [ [ - 954, - 978 - ], - [ - 1587, - 1597 + 127, + 132 ] ], "right_ranges": [ [ - 168, - 174 - ], - [ - 914, - 955 - ], - [ - 1480, - 1547 - ], - [ - 1696, - 1714 + 18, + 24 ], [ - 1849, - 1855 + 27, + 36 ], [ - 1888, - 1897 + 216, + 401 ] ] } ] }, { - "left_pr_number": 47143, - "right_pr_number": 47163, - "code_similarity": 0.004, - "size_similarity": 0.009, + "left_pr_number": 47096, + "right_pr_number": 47398, + "code_similarity": 0.007, + "size_similarity": 0.025, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.013, + "patch_similarity": 0.011, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47143, - "right_pr_number": 47210, - "code_similarity": 0.037, - "size_similarity": 0.123, + "left_pr_number": 47096, + "right_pr_number": 47399, + "code_similarity": 0.021, + "size_similarity": 0.083, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.084, + "patch_similarity": 0.028, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47143, - "right_pr_number": 47216, - "code_similarity": 0.061, - "size_similarity": 0.23, + "left_pr_number": 47096, + "right_pr_number": 47409, + "code_similarity": 0.045, + "size_similarity": 0.188, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.099, + "patch_similarity": 0.052, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47143, - "right_pr_number": 47301, - "code_similarity": 0.006, - "size_similarity": 0.01, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.009, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.026, + "patch_similarity": 0.013, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47143, - "right_pr_number": 47322, - "code_similarity": 0.001, - "size_similarity": 0.005, + "right_pr_number": 47298, + "code_similarity": 0.028, + "size_similarity": 0.116, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.002, + "patch_similarity": 0.033, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47143, - "right_pr_number": 47336, - "code_similarity": 0.002, - "size_similarity": 0.006, + "right_pr_number": 47306, + "code_similarity": 0.069, + "size_similarity": 0.288, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.004, + "patch_similarity": 0.079, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47143, - "right_pr_number": 47367, - "code_similarity": 0.007, - "size_similarity": 0.02, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.005, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.021, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, @@ -880,81 +1636,70 @@ }, { "left_pr_number": 47143, - "right_pr_number": 47387, - "code_similarity": 0.028, - "size_similarity": 0.095, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.005, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.06, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47163, - "right_pr_number": 47210, - "code_similarity": 0.021, - "size_similarity": 0.076, + "left_pr_number": 47143, + "right_pr_number": 47399, + "code_similarity": 0.004, + "size_similarity": 0.017, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.036, + "patch_similarity": 0.004, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47163, - "right_pr_number": 47216, - "code_similarity": 0.014, - "size_similarity": 0.04, + "left_pr_number": 47143, + "right_pr_number": 47409, + "code_similarity": 0.013, + "size_similarity": 0.039, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.039, + "patch_similarity": 0.034, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47163, - "right_pr_number": 47301, - "code_similarity": 0.192, - "size_similarity": 0.933, + "right_pr_number": 47298, + "code_similarity": 0.022, + "size_similarity": 0.08, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.033, + "patch_similarity": 0.04, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47163, - "right_pr_number": 47322, - "code_similarity": 0.101, - "size_similarity": 0.5, + "right_pr_number": 47306, + "code_similarity": 0.011, + "size_similarity": 0.032, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.008, + "patch_similarity": 0.029, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47163, - "right_pr_number": 47336, - "code_similarity": 0.13, - "size_similarity": 0.643, + "right_pr_number": 47322, + "code_similarity": 0.101, + "size_similarity": 0.5, "file_overlap": 0.0, "area_overlap": 0.0, "patch_similarity": 0.008, "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 47163, - "right_pr_number": 47367, - "code_similarity": 0.099, - "size_similarity": 0.459, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.049, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 47163, "right_pr_number": 47383, @@ -968,75 +1713,64 @@ }, { "left_pr_number": 47163, - "right_pr_number": 47387, - "code_similarity": 0.025, - "size_similarity": 0.098, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.033, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47210, - "right_pr_number": 47216, - "code_similarity": 0.127, - "size_similarity": 0.532, + "right_pr_number": 47398, + "code_similarity": 0.116, + "size_similarity": 0.571, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.14, + "patch_similarity": 0.009, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47210, - "right_pr_number": 47301, - "code_similarity": 0.025, - "size_similarity": 0.081, + "left_pr_number": 47163, + "right_pr_number": 47399, + "code_similarity": 0.113, + "size_similarity": 0.538, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.061, + "patch_similarity": 0.033, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47210, - "right_pr_number": 47322, - "code_similarity": 0.009, - "size_similarity": 0.038, + "left_pr_number": 47163, + "right_pr_number": 47409, + "code_similarity": 0.052, + "size_similarity": 0.237, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.011, + "patch_similarity": 0.032, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47210, - "right_pr_number": 47336, - "code_similarity": 0.011, - "size_similarity": 0.049, + "left_pr_number": 47298, + "right_pr_number": 47306, + "code_similarity": 0.089, + "size_similarity": 0.402, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.011, + "patch_similarity": 0.055, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47210, - "right_pr_number": 47367, - "code_similarity": 0.044, - "size_similarity": 0.165, + "left_pr_number": 47298, + "right_pr_number": 47322, + "code_similarity": 0.009, + "size_similarity": 0.04, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.071, + "patch_similarity": 0.007, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47210, + "left_pr_number": 47298, "right_pr_number": 47383, - "code_similarity": 0.112, - "size_similarity": 0.534, + "code_similarity": 0.118, + "size_similarity": 0.566, "file_overlap": 0.0, "area_overlap": 0.0, "patch_similarity": 0.033, @@ -1044,320 +1778,218 @@ "shared_file_areas": [] }, { - "left_pr_number": 47210, - "right_pr_number": 47387, - "code_similarity": 0.168, - "size_similarity": 0.772, + "left_pr_number": 47298, + "right_pr_number": 47398, + "code_similarity": 0.01, + "size_similarity": 0.046, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.088, + "patch_similarity": 0.007, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47216, - "right_pr_number": 47301, - "code_similarity": 0.02, - "size_similarity": 0.043, + "left_pr_number": 47298, + "right_pr_number": 47399, + "code_similarity": 0.034, + "size_similarity": 0.149, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.075, + "patch_similarity": 0.024, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47216, - "right_pr_number": 47322, - "code_similarity": 0.035, - "size_similarity": 0.02, - "file_overlap": 0.1, - "area_overlap": 0.0, - "patch_similarity": 0.007, - "shared_filenames": [ - "src/transformers/utils/quantization_config.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/utils/quantization_config.py", - "left_ranges": [ - [ - 1113, - 1123 - ], - [ - 1130, - 1136 - ], - [ - 1141, - 1155 - ], - [ - 1169, - 1179 - ], - [ - 1237, - 1258 - ] - ], - "right_ranges": [ - [ - 1701, - 1707 - ], - [ - 1715, - 1721 - ] - ] - } - ] - }, - { - "left_pr_number": 47216, - "right_pr_number": 47336, - "code_similarity": 0.007, - "size_similarity": 0.026, + "left_pr_number": 47298, + "right_pr_number": 47409, + "code_similarity": 0.074, + "size_similarity": 0.339, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.014, + "patch_similarity": 0.042, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47216, - "right_pr_number": 47367, - "code_similarity": 0.03, - "size_similarity": 0.088, + "left_pr_number": 47306, + "right_pr_number": 47322, + "code_similarity": 0.004, + "size_similarity": 0.016, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.086, + "patch_similarity": 0.008, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47216, + "left_pr_number": 47306, "right_pr_number": 47383, - "code_similarity": 0.061, - "size_similarity": 0.284, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.031, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47216, - "right_pr_number": 47387, - "code_similarity": 0.106, - "size_similarity": 0.411, + "code_similarity": 0.049, + "size_similarity": 0.227, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.156, + "patch_similarity": 0.026, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47301, - "right_pr_number": 47322, - "code_similarity": 0.097, - "size_similarity": 0.467, + "left_pr_number": 47306, + "right_pr_number": 47398, + "code_similarity": 0.005, + "size_similarity": 0.018, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.026, + "patch_similarity": 0.008, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47301, - "right_pr_number": 47336, - "code_similarity": 0.12, - "size_similarity": 0.6, + "left_pr_number": 47306, + "right_pr_number": 47399, + "code_similarity": 0.014, + "size_similarity": 0.06, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.0, + "patch_similarity": 0.013, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47301, - "right_pr_number": 47367, - "code_similarity": 0.11, - "size_similarity": 0.492, + "left_pr_number": 47306, + "right_pr_number": 47409, + "code_similarity": 0.045, + "size_similarity": 0.136, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.077, + "patch_similarity": 0.116, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47301, + "left_pr_number": 47322, "right_pr_number": 47383, - "code_similarity": 0.035, - "size_similarity": 0.152, + "code_similarity": 0.015, + "size_similarity": 0.071, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.031, + "patch_similarity": 0.005, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47301, - "right_pr_number": 47387, - "code_similarity": 0.03, - "size_similarity": 0.105, + "left_pr_number": 47322, + "right_pr_number": 47398, + "code_similarity": 0.195, + "size_similarity": 0.875, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.059, + "patch_similarity": 0.133, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47322, - "right_pr_number": 47336, - "code_similarity": 0.16, - "size_similarity": 0.778, + "right_pr_number": 47399, + "code_similarity": 0.054, + "size_similarity": 0.269, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.032, + "patch_similarity": 0.0, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47322, - "right_pr_number": 47367, - "code_similarity": 0.049, - "size_similarity": 0.23, + "right_pr_number": 47409, + "code_similarity": 0.029, + "size_similarity": 0.119, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.018, + "patch_similarity": 0.032, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47322, - "right_pr_number": 47383, - "code_similarity": 0.015, - "size_similarity": 0.071, + "left_pr_number": 47383, + "right_pr_number": 47398, + "code_similarity": 0.017, + "size_similarity": 0.081, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.005, + "patch_similarity": 0.006, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47322, - "right_pr_number": 47387, - "code_similarity": 0.011, - "size_similarity": 0.049, + "left_pr_number": 47383, + "right_pr_number": 47399, + "code_similarity": 0.053, + "size_similarity": 0.264, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.007, + "patch_similarity": 0.003, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47336, - "right_pr_number": 47367, - "code_similarity": 0.063, - "size_similarity": 0.295, + "left_pr_number": 47383, + "right_pr_number": 47409, + "code_similarity": 0.123, + "size_similarity": 0.599, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.024, + "patch_similarity": 0.022, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47336, - "right_pr_number": 47383, - "code_similarity": 0.019, - "size_similarity": 0.091, + "left_pr_number": 47398, + "right_pr_number": 47399, + "code_similarity": 0.062, + "size_similarity": 0.308, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.005, + "patch_similarity": 0.0, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47336, - "right_pr_number": 47387, - "code_similarity": 0.142, - "size_similarity": 0.063, - "file_overlap": 0.333, - "area_overlap": 0.06, - "patch_similarity": 0.056, + "left_pr_number": 47398, + "right_pr_number": 47409, + "code_similarity": 0.286, + "size_similarity": 0.136, + "file_overlap": 0.25, + "area_overlap": 0.5, + "patch_similarity": 0.059, "shared_filenames": [ - "src/transformers/modeling_utils.py" + "src/transformers/models/glm_moe_dsa/configuration_glm_moe_dsa.py" ], "shared_file_areas": [ { - "filename": "src/transformers/modeling_utils.py", + "filename": "src/transformers/models/glm_moe_dsa/configuration_glm_moe_dsa.py", "left_ranges": [ [ - 3559, - 3566 - ], - [ - 3664, - 3670 + 159, + 173 ] ], "right_ranges": [ [ - 62, - 68 - ], - [ - 3565, - 3578 - ], - [ - 3593, - 3600 - ], - [ - 3608, - 3620 - ], - [ - 3636, - 3666 + 158, + 166 ] ] } ] }, { - "left_pr_number": 47367, - "right_pr_number": 47383, - "code_similarity": 0.066, - "size_similarity": 0.31, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.024, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47367, - "right_pr_number": 47387, - "code_similarity": 0.053, - "size_similarity": 0.214, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.071, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47383, - "right_pr_number": 47387, - "code_similarity": 0.141, - "size_similarity": 0.691, + "left_pr_number": 47399, + "right_pr_number": 47409, + "code_similarity": 0.093, + "size_similarity": 0.441, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.018, + "patch_similarity": 0.031, "shared_filenames": [], "shared_file_areas": [] } @@ -2360,257 +2992,55 @@ 744 ], [ - 749, - 772 - ], - [ - 790, - 795 - ] - ] - } - ] - }, - { - "left_pr_number": 44129, - "right_pr_number": 44154, - "code_similarity": 0.099, - "size_similarity": 0.344, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.199, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 44129, - "right_pr_number": 44722, - "code_similarity": 0.146, - "size_similarity": 0.545, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.247, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 44154, - "right_pr_number": 44722, - "code_similarity": 0.164, - "size_similarity": 0.631, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.255, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, - { - "cluster_id": "cluster-43451-3", - "summary": "Cluster of 3 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 43451, - "issue_numbers": [], - "pr_numbers": [ - 43451, - 46836, - 47243 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 43451, - "right_pr_number": 46836, - "code_similarity": 0.278, - "size_similarity": 0.922, - "file_overlap": 0.146, - "area_overlap": 0.0, - "patch_similarity": 0.329, - "shared_filenames": [ - "docs/source/en/_toctree.yml", - "src/transformers/conversion_mapping.py", - "src/transformers/models/__init__.py", - "src/transformers/models/auto/auto_mappings.py", - "src/transformers/models/auto/modeling_auto.py", - "utils/check_repo.py" - ], - "shared_file_areas": [ - { - "filename": "docs/source/en/_toctree.yml", - "left_ranges": [ - [ - 1375, - 1382 - ] - ], - "right_ranges": [ - [ - 893, - 900 - ] - ] - }, - { - "filename": "src/transformers/conversion_mapping.py", - "left_ranges": [ - [ - 737, - 769 - ] - ], - "right_ranges": [ - [ - 127, - 133 - ], - [ - 1495, - 1545 - ] - ] - }, - { - "filename": "src/transformers/models/__init__.py", - "left_ranges": [ - [ - 290, - 296 - ] - ], - "right_ranges": [ - [ - 453, - 459 - ] - ] - }, - { - "filename": "src/transformers/models/auto/auto_mappings.py", - "left_ranges": [ - [ - 391, - 398 - ], - [ - 824, - 830 - ], - [ - 945, - 951 - ], - [ - 1068, - 1074 - ], - [ - 1202, - 1208 - ] - ], - "right_ranges": [ - [ - 637, - 647 - ], - [ - 919, - 928 - ], - [ - 1130, - 1136 - ], - [ - 1260, - 1266 - ] - ] - }, - { - "filename": "src/transformers/models/auto/modeling_auto.py", - "left_ranges": [ - [ - 331, - 338 - ], - [ - 1094, - 1100 - ], - [ - 1913, - 1919 - ] - ], - "right_ranges": [ - [ - 509, - 515 - ], - [ - 1119, - 1125 - ] - ] - }, - { - "filename": "utils/check_repo.py", - "left_ranges": [ - [ - 250, - 258 - ], - [ - 337, - 344 - ] - ], - "right_ranges": [ - [ - 267, - 274 + 749, + 772 ], [ - 494, - 501 + 790, + 795 ] ] } ] }, { - "left_pr_number": 43451, - "right_pr_number": 47243, - "code_similarity": 0.067, - "size_similarity": 0.221, + "left_pr_number": 44129, + "right_pr_number": 44154, + "code_similarity": 0.099, + "size_similarity": 0.344, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.15, + "patch_similarity": 0.199, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46836, - "right_pr_number": 47243, - "code_similarity": 0.071, - "size_similarity": 0.239, + "left_pr_number": 44129, + "right_pr_number": 44722, + "code_similarity": 0.146, + "size_similarity": 0.545, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.247, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44154, + "right_pr_number": 44722, + "code_similarity": 0.164, + "size_similarity": 0.631, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.155, + "patch_similarity": 0.255, "shared_filenames": [], "shared_file_areas": [] } ] }, { - "cluster_id": "cluster-46894-7", - "summary": "Cluster of 3 issues and 4 PRs centered on issue #46894.", + "cluster_id": "cluster-46894-9", + "summary": "Cluster of 4 issues and 5 PRs centered on issue #46894.", "status": "open", "confidence": 0.8, "canonical_issue_number": 46894, @@ -2618,13 +3048,15 @@ "issue_numbers": [ 46894, 46897, - 46901 + 46901, + 47394 ], "pr_numbers": [ 46895, 46903, 46958, - 47090 + 47090, + 47395 ], "evidence_types": [ "closing_reference", @@ -2819,6 +3251,17 @@ } ] }, + { + "left_pr_number": 46895, + "right_pr_number": 47395, + "code_similarity": 0.187, + "size_similarity": 0.683, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.333, + "shared_filenames": [], + "shared_file_areas": [] + }, { "left_pr_number": 46903, "right_pr_number": 46958, @@ -2949,6 +3392,17 @@ } ] }, + { + "left_pr_number": 46903, + "right_pr_number": 47395, + "code_similarity": 0.083, + "size_similarity": 0.276, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.186, + "shared_filenames": [], + "shared_file_areas": [] + }, { "left_pr_number": 46958, "right_pr_number": 47090, @@ -3021,6 +3475,28 @@ ] } ] + }, + { + "left_pr_number": 46958, + "right_pr_number": 47395, + "code_similarity": 0.201, + "size_similarity": 0.747, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.342, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47090, + "right_pr_number": 47395, + "code_similarity": 0.078, + "size_similarity": 0.192, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.262, + "shared_filenames": [], + "shared_file_areas": [] } ] }, @@ -3123,220 +3599,6 @@ } ] }, - { - "cluster_id": "cluster-46912-4", - "summary": "Cluster of 4 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 47298, - "issue_numbers": [], - "pr_numbers": [ - 46912, - 47175, - 47294, - 47298 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 46912, - "right_pr_number": 47175, - "code_similarity": 0.011, - "size_similarity": 0.005, - "file_overlap": 0.029, - "area_overlap": 0.0, - "patch_similarity": 0.01, - "shared_filenames": [ - "tests/models/olmo3/test_modeling_olmo3.py" - ], - "shared_file_areas": [ - { - "filename": "tests/models/olmo3/test_modeling_olmo3.py", - "left_ranges": [ - [ - 99, - 104 - ] - ], - "right_ranges": [ - [ - 213, - 227 - ], - [ - 312, - 318 - ] - ] - } - ] - }, - { - "left_pr_number": 46912, - "right_pr_number": 47294, - "code_similarity": 0.041, - "size_similarity": 0.17, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.045, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46912, - "right_pr_number": 47298, - "code_similarity": 0.052, - "size_similarity": 0.125, - "file_overlap": 0.06, - "area_overlap": 0.0, - "patch_similarity": 0.06, - "shared_filenames": [ - "tests/models/gemma3/test_modeling_gemma3.py", - "tests/models/gemma3n/test_modeling_gemma3n.py", - "tests/test_modeling_common.py" - ], - "shared_file_areas": [ - { - "filename": "tests/models/gemma3/test_modeling_gemma3.py", - "left_ranges": [ - [ - 77, - 84 - ], - [ - 163, - 168 - ], - [ - 180, - 186 - ] - ], - "right_ranges": [ - [ - 518, - 524 - ], - [ - 588, - 598 - ], - [ - 633, - 640 - ], - [ - 708, - 717 - ], - [ - 756, - 762 - ], - [ - 813, - 819 - ] - ] - }, - { - "filename": "tests/models/gemma3n/test_modeling_gemma3n.py", - "left_ranges": [ - [ - 560, - 565 - ] - ], - "right_ranges": [ - [ - 994, - 1000 - ], - [ - 1079, - 1085 - ], - [ - 1107, - 1113 - ], - [ - 1150, - 1156 - ], - [ - 1196, - 1202 - ], - [ - 1228, - 1234 - ] - ] - }, - { - "filename": "tests/test_modeling_common.py", - "left_ranges": [ - [ - 5729, - 5974 - ], - [ - 6089, - 6118 - ] - ], - "right_ranges": [ - [ - 117, - 123 - ], - [ - 3434, - 3443 - ] - ] - } - ] - }, - { - "left_pr_number": 47175, - "right_pr_number": 47294, - "code_similarity": 0.008, - "size_similarity": 0.032, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.011, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47175, - "right_pr_number": 47298, - "code_similarity": 0.019, - "size_similarity": 0.043, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.067, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47294, - "right_pr_number": 47298, - "code_similarity": 0.153, - "size_similarity": 0.734, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.044, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, { "cluster_id": "cluster-47138-4", "summary": "Cluster of 2 issues and 2 PRs centered on issue #47138.", @@ -4320,35 +4582,6 @@ } ] }, - { - "cluster_id": "cluster-47003-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "closed", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 47003, - "issue_numbers": [], - "pr_numbers": [ - 47003, - 47306 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 47003, - "right_pr_number": 47306, - "code_similarity": 0.046, - "size_similarity": 0.159, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.093, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, { "cluster_id": "cluster-47282-2", "summary": "Cluster of 2 related pull requests linked by soft_similarity.", @@ -4698,13 +4931,14 @@ ], "duplicate_issues": [ { - "cluster_id": "cluster-46894-7", + "cluster_id": "cluster-46894-9", "canonical_issue_number": 46894, "duplicate_issue_numbers": [ 46897, - 46901 + 46901, + 47394 ], - "reason": "Issues in cluster-46894-7 are treated as duplicates because they share closing_reference, shared_issue_target, soft_similarity evidence." + "reason": "Issues in cluster-46894-9 are treated as duplicates because they share closing_reference, shared_issue_target, soft_similarity evidence." }, { "cluster_id": "cluster-47138-4", @@ -4735,14 +4969,25 @@ "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence." }, { - "cluster_id": "cluster-43451-3", - "canonical_pr_number": 43451, + "cluster_id": "cluster-43451-15", + "canonical_pr_number": 47322, "duplicate_pr_numbers": [ - 46836, - 47243 + 43451, + 46509, + 46658, + 47003, + 47096, + 47143, + 47163, + 47298, + 47306, + 47383, + 47398, + 47399, + 47409 ], - "target_issue_number": null, - "reason": "PRs in cluster-43451-3 are treated as duplicates because they share soft_similarity evidence." + "target_issue_number": 47355, + "reason": "PRs in cluster-43451-15 are treated as duplicates because they converge on issue #47355 with closing_reference, shared_issue_target, soft_similarity evidence." }, { "cluster_id": "cluster-43656-2", @@ -4797,27 +5042,6 @@ "target_issue_number": 45081, "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." }, - { - "cluster_id": "cluster-45157-14", - "canonical_pr_number": 47322, - "duplicate_pr_numbers": [ - 45157, - 46196, - 46509, - 46658, - 47143, - 47163, - 47210, - 47216, - 47301, - 47336, - 47367, - 47383, - 47387 - ], - "target_issue_number": null, - "reason": "PRs in cluster-45157-14 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-45561-3", "canonical_pr_number": 45645, @@ -4864,35 +5088,16 @@ "reason": "PRs in cluster-46869-3 are treated as duplicates because they converge on issue #46869 with closing_reference, shared_issue_target evidence." }, { - "cluster_id": "cluster-46894-7", + "cluster_id": "cluster-46894-9", "canonical_pr_number": 46903, "duplicate_pr_numbers": [ 46895, 46958, - 47090 + 47090, + 47395 ], "target_issue_number": 46894, - "reason": "PRs in cluster-46894-7 are treated as duplicates because they converge on issue #46894 with closing_reference, shared_issue_target, soft_similarity evidence." - }, - { - "cluster_id": "cluster-46912-4", - "canonical_pr_number": 47298, - "duplicate_pr_numbers": [ - 46912, - 47175, - 47294 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46912-4 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-47003-2", - "canonical_pr_number": 47003, - "duplicate_pr_numbers": [ - 47306 - ], - "target_issue_number": null, - "reason": "PRs in cluster-47003-2 are treated as duplicates because they share soft_similarity evidence." + "reason": "PRs in cluster-46894-9 are treated as duplicates because they converge on issue #46894 with closing_reference, shared_issue_target, soft_similarity evidence." }, { "cluster_id": "cluster-47123-2", @@ -4923,15 +5128,15 @@ } ], "best_issue": { - "cluster_id": "cluster-46894-7", - "issue_number": 46894, - "reason": "Issue #46894 is the strongest global issue candidate because it is open, belongs to a cluster with 7 artifacts, and carries 1 discussion comments plus 10 inbound references.", - "score": 182.0 + "cluster_id": "cluster-43451-15", + "issue_number": 47355, + "reason": "Issue #47355 is the strongest global issue candidate because it is open, belongs to a cluster with 15 artifacts, and carries 5 discussion comments plus 4 inbound references.", + "score": 264.0 }, "best_pr": { - "cluster_id": "cluster-45157-14", + "cluster_id": "cluster-43451-15", "pr_number": 47322, - "reason": "PR #47322 is the strongest global PR candidate because it is open, belongs to a cluster with 14 artifacts, links to 1 issues, and carries 2 review/discussion events.", - "score": 268.986 + "reason": "PR #47322 is the strongest global PR candidate because it is open, belongs to a cluster with 15 artifacts, links to 1 issues, and carries 2 review/discussion events.", + "score": 278.986 } } diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index 9e37ac23ac44b88e18895dda0f5f3f0628291caa..55aa110cfca666d67dcbf83cfb0c03f11e1e19de 100644 --- a/analysis/current/analysis-report-hybrid.llm-reviews.json +++ b/analysis/current/analysis-report-hybrid.llm-reviews.json @@ -1,8 +1,8 @@ { "schema_version": "1.0", "repo": "huggingface/transformers", - "snapshot_id": "20260719T000028Z", - "generated_at": "2026-07-19T00:11:48Z", + "snapshot_id": "20260719T060027Z", + "generated_at": "2026-07-19T06:10:36Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -125,14 +125,14 @@ "issue:43039|issue:44918", "issue:41762|issue:43638" ], - "prepared_review_unit_hash": "874ad57478247174347a2144c06ce176d80bf0294e4f55563e37e21c248fee42", + "prepared_review_unit_hash": "f65c9956318cfbe5c0418ff2bada5621451827258d140a051fa2039ce96170f5", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, @@ -147,348 +147,354 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous issue reports with only a couple of clear duplicate pairs. The strongest duplicate matches are the DeepSpeed ZeRO-3/BERT IndexError reports and the torch 2.4.1 compatibility reports; the rest are related only at a broad subsystem level and should stay separate.", - "confidence": 0.89, - "canonical_issue_reason": "Issue 43596 is the cleanest representative of the ZeRO-3/BERT zero-sized tensor IndexError, and it has a near-identical duplicate in 43638.", + "summary": "Most items are broad, unrelated runtime/model-support issues; only the two BertModel ZeRO-3 IndexError reports look like a true duplicate pair.", + "confidence": 0.95, + "canonical_issue_reason": "issue:43596 is the broadest, clearest representative of the shared BertModel + DeepSpeed ZeRO-3 zero-sized-tensor IndexError; issue:43638 is the narrower non-pretrained Bert variant of the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "43596 is the most concrete, reproducible bug report among the true duplicates in this set, with a clearly matching counterpart in 43638.", + "best_issue_reason": "issue:43596 best captures the underlying bug and is the most general duplicate target among the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different failures: missing config attribute in Qwen3.5 vs streaming chat/completions inference error." + "reason": "Different failures: config attribute missing vs streaming chat/completions inference issue." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Unrelated model/GPU issues: GPT-OSS on AMD vs an MPS histogram error in MoE." + "reason": "Different hardware/backend bugs (AMD GPU failure vs MPS histogram op not implemented)." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different root causes and model families; dtype mismatch in Mistral-3 image processing is not the same as GPT-OSS AMD support." + "reason": "Different bug classes: dtype mismatch in Mistral-3 vs GPT-OSS running on AMD GPUs." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Both are model/runtime issues, but the concrete bug paths differ: autocast dtype mismatch vs AMD GPU failure." + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to GPT-OSS AMD GPU failure." }, { "left": "issue:45237", "right": "issue:46468", "accept": false, - "reason": "One is AMD GPU compatibility, the other is a broken official example; not the same bug." + "reason": "Different models and symptoms; one is AMD GPU runtime failure, the other a broken official example." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model-specific failures; no evidence they share the same code path." + "reason": "Different model/backend issues; not the same concrete code path." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "Deepspeed collective mismatch vs ONNX export error are unrelated." + "reason": "ZeRO-3 collective mismatch is unrelated to ModernBERT ONNX export failure." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "Different export/runtime problems in different models and code paths." + "reason": "Both are export/model errors, but the model-specific causes are different." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "ONNX export bug and auto device-map CUDA assert are not the same issue." + "reason": "ONNX export failure and Qwen3 auto device-map cuda assert are unrelated." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Table QA crash vs outdated RAG examples; unrelated." + "reason": "Table QA crash and outdated RAG examples are different docs/runtime issues." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ONNX export error is unrelated to the GenerationMixin import failure." + "reason": "ONNX export bug is unrelated to GenerationMixin import failure." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions in v5, but they refer to different removed docs content and are not clearly the same underlying change." + "reason": "Both are stale docs, but they concern different removed content and likely different edits." }, { "left": "issue:44918", "right": "issue:46165", "accept": false, - "reason": "Different training issues: Qwen3.5 embedding unpacking vs resuming local model training with trust_remote_code." + "reason": "SFT trainer embedding unpacking and training-resume/trust-remote-code are different problems." }, { "left": "issue:46165", "right": "issue:46869", "accept": false, - "reason": "Resume/trust_remote_code issue and YOLO multi-GPU device-map failure are unrelated." + "reason": "Resume-training issue is unrelated to YOLO multi-GPU device-map failure." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Both mention shape/device problems, but the model and failure mode differ substantially." + "reason": "Mask-shape IndexError and Gemma4 PLE device mismatch are not the same bug." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both are Zero-3-related errors, but one is a zero-sized index error and the other is a mask-shape mismatch." + "reason": "Both are IndexErrors, but the model/setup and failure mechanism differ." }, { "left": "issue:43638", "right": "issue:46165", "accept": false, - "reason": "Different domains: ZeRO-3/BERT initialization vs model resume/trust_remote_code." + "reason": "Bert ZeRO-3 init bug is unrelated to resume-training trust-remote-code failure." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Related in a broad sense to tensor-shape failures, but not the same concrete bug." + "reason": "Different IndexError contexts; not the same Bert ZeRO-3 initialization bug." }, { "left": "issue:44805", "right": "issue:46294", "accept": false, - "reason": "Mask-shape mismatch and video chat-template token mismatch are different problems." + "reason": "Mask/indexing error is unrelated to Qwen3.5 video token/template mismatch." }, { "left": "issue:46165", "right": "issue:46468", "accept": false, - "reason": "Resume-training issue and broken example are unrelated." + "reason": "Training resume problem and broken official example are unrelated." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Different model/runtime failures; not the same code path." + "reason": "Shape-mismatch IndexError is not the same as GPT-OSS tensor-parallel failure." }, { "left": "issue:44560", "right": "issue:46165", "accept": false, - "reason": "Video chat-template StopIteration is unrelated to training resume/trust_remote_code." + "reason": "Video input StopIteration and training-resume issue are unrelated." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice preset bug is unrelated to ONNX export." + "reason": "BarkProcessor voice preset bug is unrelated to ModernBERT ONNX export." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both involve DeepSpeed, but the concrete failures and models differ." + "reason": "Both mention distributed/ZeRO-3, but the collective mismatch and Bert init IndexError are different bugs." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same underlying DeepSpeed ZeRO-3/BERT zero-sized tensor IndexError; titles and repro context align closely." + "reason": "Same underlying BertModel + DeepSpeed ZeRO-3 zero-sized tensor IndexError; one is just the non-pretrained variant." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger Kernel cross_entropy call vs GPT-OSS MoE TP issue are unrelated." + "reason": "Liger Kernel cross-entropy regression and GPT-OSS tensor-parallel failure are unrelated." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Tokenizer detokenization bug and LoRA save/reload mismatch are different bugs." + "reason": "Different issues: detokenization regression vs LoRA merge/save/reload mismatch." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Different failures: Qwen-3 NVFP4 runtime error vs LoRA merge/save inconsistency." + "reason": "Qwen3 NVFP4 runtime failure is unrelated to LoRA merge/save behavior." }, { "left": "issue:34928", "right": "issue:35308", "accept": false, - "reason": "Both are training/distributed issues, but one is activation-checkpoint/FSDP and the other is IterableDataset length mismatch." + "reason": "Both are distributed-training crashes, but activation-checkpointing/FSDP mismatch is not the same as IterableDataset length handling." }, { "left": "issue:44841", "right": "issue:46661", "accept": false, - "reason": "Processor failure for Voxtral is unrelated to remote-code model_class config_class errors." + "reason": "Processor failure for Voxtral is unrelated to remote-code model_class config_class AttributeError." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Both are DeepSpeed-related, but the reported failures are different and not the same bug." + "reason": "DeepSpeed weight-shape error and collective mismatch are different failure modes." }, { "left": "issue:43716", "right": "issue:46869", "accept": false, - "reason": "Different models and failure modes: dtype mismatch vs multi-GPU device-map problem." + "reason": "Mistral-3 dtype mismatch and YOLO multi-GPU device-map failure are unrelated." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Both are tensor-shape failures, but they arise from different training mechanisms and error signatures." + "reason": "Different distributed/shape errors with different triggers and code paths." }, { "left": "issue:45685", "right": "issue:46869", "accept": false, - "reason": "MPS histogram error in MoE is unrelated to YOLO multi-GPU device-map auto." + "reason": "MPS histogram limitation is unrelated to YOLO multi-GPU device-map auto failure." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "Activation checkpoint/FSDP tensor recompute mismatch is unrelated to compute_loss signature changes." + "reason": "Activation-checkpointing/FSDP error is unrelated to CustomTrainer compute_loss signature breakage." }, { "left": "issue:45859", "right": "issue:45902", "accept": false, - "reason": "Both involve Qwen3.5 MoE rotary embedding, but one is CPU offload incompatibility and the other is uninitialized-memory materialization." + "reason": "Both touch Qwen3_5Moe rotary embedding, but CPU offload incompatibility and uninitialized-memory after to_empty are distinct bugs." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "ZeRO-3 BERT initialization error and GPT-OSS tensor-parallel issue are unrelated." + "reason": "Bert ZeRO-3 initialization IndexError is unrelated to GPT-OSS tensor-parallel failure." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "Different model families and different distributed-training paths." + "reason": "Different models and different distributed failure modes." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Different model-specific failures; no concrete shared bug." + "reason": "Mistral-3 dtype mismatch and GraniteMoEHybrid invalid method call are unrelated." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error and GPT-OSS AMD incompatibility are not the same underlying issue." + "reason": "Qwen-3 NVFP4 runtime error is unrelated to GPT-OSS on AMD GPUs." }, { "left": "issue:45874", "right": "issue:46294", "accept": false, - "reason": "inputs_embeds reverse-lookup performance bug is unrelated to video chat-template token mismatch." + "reason": "Reverse embedding lookup issue is unrelated to video feature/token mismatch." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Gemma3 config missing attribute under vLLM vs Mistral regex tokenizer patch crash are unrelated." + "reason": "Gemma3 vLLM config attribute error and Mistral tokenizer regex patch crash are unrelated." }, { "left": "issue:45800", "right": "issue:46515", - "accept": true, - "reason": "Same torch 2.4.1 compatibility problem; 46515 is a shorter duplicate of 45800." + "accept": false, + "reason": "Torch-version incompatibility reports are similar in theme but appear to be different compatibility cases." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Gemma4 device-map forward mismatch is unrelated to expensive reverse embedding lookup." + "reason": "Device-map forward mismatch and expensive reverse embedding lookup are different Gemma4 bugs." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Generic torch incompatibility is not the same as the Qwen-3 NVFP4 runtime failure." + "reason": "General Torch incompatibility and Qwen-3 NVFP4 runtime error are not the same issue." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "Different distributed/device issues with no shared concrete failure path." + "reason": "FSDP/checkpointing mismatch and Qwen3 auto device-map cuda assert are unrelated distributed bugs." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Autocast dtype mismatch for Phi-tiny-MoE is unrelated to GPT-OSS TP behavior." + "reason": "Autocast dtype mismatch is unrelated to GPT-OSS tensor-parallel failure." }, { "left": "issue:43541", "right": "issue:43716", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm tracing error is unrelated to Mistral-3 image-preprocessor dtype mismatch." + "reason": "Mixtral grouped_mm torch-dynamo tracing error and Mistral-3 dtype mismatch are different bugs." }, { "left": "issue:43828", "right": "issue:44292", "accept": false, - "reason": "Both are model/runtime errors, but the concrete failure modes differ." + "reason": "Different model-specific runtime errors with different root causes." }, { "left": "issue:43716", "right": "issue:45685", "accept": false, - "reason": "Mistral-3 dtype mismatch and MoE MPS histogram issue are unrelated." + "reason": "Mistral-3 dtype mismatch and MPS histogram limitation are unrelated." }, { "left": "issue:45507", "right": "issue:45685", "accept": false, - "reason": "Different model/runtime bugs; no evidence of a shared root cause." + "reason": "Invalid method call in GraniteMoEHybrid is unrelated to an MPS int histogram backend limitation." }, { "left": "issue:44292", "right": "issue:44918", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error and TRL SFT embedding unpacking issue are distinct." + "reason": "Qwen-3 NVFP4 runtime error and Qwen3.5 embedding unpacking with TRL are different failures." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Mistral-3 dtype mismatch and Qwen-3 NVFP4 error are unrelated." }, { "left": "issue:44918", "right": "issue:45161", "accept": false, - "reason": "Different model/training paths and different errors." + "reason": "TRL embedding unpacking failure and GPT-OSS tensor-parallel failure are not the same bug." }, { "left": "issue:43039", "right": "issue:44918", "accept": false, - "reason": "Liger Kernel cross-entropy call and TRL embedding unpacking are unrelated." + "reason": "Liger Kernel cross-entropy regression is unrelated to Qwen3.5 SFT trainer unpacking." }, { "left": "issue:41762", "right": "issue:43638", "accept": false, - "reason": "Both mention ZeRO-3, but Gemma3 loading failure and Bert initialization failure are different bugs." + "reason": "Gemma3 ZeRO-3 empty-tensor IndexError is similar in shape, but the model and code path differ from the BertModel bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative on the soft edges. The two accepted duplicate pairs are strongly supported by the titles and surrounding context: 43596/43638 share the same ZeRO-3/BERT zero-sized tensor IndexError pattern, and 45800/46515 both describe torch 2.4.1 incompatibility. The other pairwise rejections avoid overstating evidence and are phrased cautiously. " + "feedback": "The summary is well grounded in the packet, and the lone accepted soft-edge pair (issue:43596 vs issue:43638) is a conservative, title-supported duplicate judgment. No overstatement stands out." }, "evaluator_used": true, "retried": false, @@ -583,8 +589,8 @@ "issue:35308|issue:42617", "issue:45812|issue:45920", "issue:46901|issue:47394", - "issue:44292|issue:46015", "issue:43872|issue:46304", + "issue:44292|issue:46015", "issue:39290|issue:43531", "issue:43866|issue:44863", "issue:41093|issue:43638", @@ -592,238 +598,238 @@ "issue:43854|issue:43866", "issue:43425|issue:45070" ], - "prepared_review_unit_hash": "117baf68daa2d2eb1e32cb3341b1a904ccdce296e20513d6fc1925130b05f590", + "prepared_review_unit_hash": "1e9124a1400e676c59d1d11f6695c4dad2fcb7b4eca525c314fa73aa2104c988", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 33, - "serialized_chars": 36959, - "estimated_input_tokens": 9240, - "estimated_eval_tokens": 18736 + "serialized_chars": 36961, + "estimated_input_tokens": 9241, + "estimated_eval_tokens": 18738 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most candidate pairs are unrelated. The only clear duplicate-style match is the two DeepSpeed ZeRO-3 index-out-of-bounds reports; the rest differ by model, subsystem, or failure mode.", - "confidence": 0.82, - "canonical_issue_reason": "issue:41762 is the cleanest representative of the ZeRO-3 load-time IndexError class; issue:43596 looks like the same underlying failure on a different model, but 41762 is more broadly phrased and easier to anchor the duplicate cluster.", + "summary": "This cluster is mostly heterogeneous, with two likely duplicate subgroups: an AutoTokenizer v5 regression affecting multiple models, and a causal-LM training-loss double-shift issue. Most other soft-similarity pairs are false positives and should stay separate.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45920 is the broadest and most explicit tokenizer regression report, covering multiple affected model families and framing it as a v5-wide silent behavior change.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41762 is a strong canonical issue because it states the concrete failure mode clearly, is model-load focused, and generalizes better across the duplicate reports than the more model-specific variant.", + "best_issue_reason": "issue:45920 is the strongest anchor for the cluster\u2019s main duplicate theme: it is broad, concrete, and clearly describes the regression in a way that can subsume the narrower Granite-specific report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions after v5 removals, but they target different obsolete commands/pages and do not describe the same fix." + "reason": "Both are docs-related v5 complaints, but they refer to different removed/deprecated docs content and do not look like the same concrete change." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Different bugs: autocast dtype mismatch versus MPS histogram support for Int." + "reason": "Different failure modes: autocast dtype mismatch vs MPS histogram/int support." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Liger Kernel cross-entropy routing is unrelated to the Phi MoE autocast dtype mismatch." + "reason": "One is a Liger Kernel cross-entropy call issue; the other is an autocast dtype mismatch. No shared code path." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "Different error classes and code paths: distributed collective mismatch versus CustomTrainer API signature breakage." + "reason": "Different distributed/training errors: collective mismatch with DeepSpeed ZeRO3 vs a trainer API signature mismatch." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "Activation checkpointing/FSDP tensor-shape mismatch is unrelated to the GenerationMixin import error." + "reason": "Activation checkpointing/FSDP recompute mismatch is unrelated to a missing GenerationMixin import." }, { "left": "issue:45237", "right": "issue:46015", "accept": false, - "reason": "AMD GPU execution failure and outdated RAG docs are unrelated." + "reason": "AMD GPU compatibility and outdated RAG docs are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Cross-entropy being called is not the same as the mask/tensor shape IndexError." + "reason": "Cross-entropy/Liger Kernel behavior is unrelated to a mask-vs-tensor shape mismatch." }, { "left": "issue:45800", "right": "issue:46304", "accept": false, - "reason": "Both are version-compatibility complaints, but they concern different breakages and different release points." + "reason": "Both mention version incompatibility, but one is torch/transformers generally and the other is USE_DISTRIBUTED=0 support; not the same bug." }, { "left": "issue:43296", "right": "issue:46304", "accept": false, - "reason": "PaddleOCR-VL/vLLM model loading is unrelated to USE_DISTRIBUTED=0 support." + "reason": "vLLM model-loading failure and distributed support breakage are separate issues." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Different models and failures: missing sliding_window_pattern versus cudaErrorAssert under auto device mapping." + "reason": "Different model-specific failures: Gemma3 config attribute missing vs Qwen3 auto device-map cuda assert." }, { "left": "issue:43866", "right": "issue:46402", "accept": false, - "reason": "Checkpoint corruption and missing text encoder weights are different loading problems." + "reason": "Checkpoint corruption and missing text-encoder weight loading are not the same underlying problem." }, { "left": "issue:41762", "right": "issue:43596", - "accept": true, - "reason": "Both report the same ZeRO-3 load-time IndexError with size-0 tensors; the model differs, but the failure mode and code path match." + "accept": false, + "reason": "Both are ZeRO-3 index errors, but they involve different models and the evidence is too thin to treat as one concrete bug." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "AMD GPU support failure is unrelated to table-question-answering crashes." + "reason": "GPU compatibility for GPT-OSS is unrelated to table-question-answering crashes." }, { "left": "issue:44589", "right": "issue:46304", "accept": false, - "reason": "Missing Float8 storage support is unrelated to distributed build support." + "reason": "Float8 storage lookup failure is unrelated to distributed build/runtime support." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "Both are documentation issues, but they cover different deprecated behaviors and should not be merged." + "reason": "One is a pipeline docs mismatch; the other is a removed run command mention. Similar topic, different docs change." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger Kernel cross-entropy handling is unrelated to the Qwen3 video StopIteration." + "reason": "Cross-entropy/Liger Kernel issue is unrelated to Qwen3-vl video StopIteration." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Different model families and different failure modes: processor failure versus device-mismatch forward error." + "reason": "Different model families and different forward-path failures; not the same bug." }, { "left": "issue:45823", "right": "issue:46294", "accept": false, - "reason": "A device placement bug is not the same as chat-template video-token/feature mismatch." + "reason": "Both involve multimodal model loading/input, but one is device-map forward mismatch and the other is video token/template mismatch." }, { "left": "issue:43975", "right": "issue:46015", "accept": false, - "reason": "Incorrect detokenization and outdated RAG examples are unrelated." + "reason": "Detokenization bug and outdated RAG docs are unrelated." }, { "left": "issue:43866", "right": "issue:46772", "accept": false, - "reason": "Checkpoint corruption is unrelated to DiffusionGemma NVFP4 compatibility." + "reason": "Corrupted checkpoint and NVIDIA ModelOpt compatibility are separate problems." }, { "left": "issue:45020", "right": "issue:46304", "accept": false, - "reason": "Remote_code regressions and USE_DISTRIBUTED=0 support breakage are different issues." + "reason": "Remote-code regressions and distributed build support are different change areas." }, { "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "NemotronH checkpoint loading failure is unrelated to docs examples." + "reason": "NemotronH checkpoint loading is unrelated to RAG example documentation." }, { "left": "issue:35308", "right": "issue:42617", "accept": false, - "reason": "IterableDataset multi-GPU length mismatch and a failing 3d_parallel.py example are not the same bug." + "reason": "Both are multi-GPU/distributed failures, but the specific crash modes and entry points differ too much." }, { "left": "issue:45812", "right": "issue:45920", - "accept": false, - "reason": "Very similar tokenizer-regression reports, but they cover different model sets and are not clearly the same concrete bug from the provided evidence." + "accept": true, + "reason": "Both describe the same tokenizer-ID regression in v5; 45920 is the broader multi-model report and 45812 is a narrower Granite-specific instance." }, { "left": "issue:46901", "right": "issue:47394", - "accept": false, - "reason": "Both mention loss shifting, but the model-specific failure descriptions differ enough that they do not look like one concrete code-path bug." + "accept": true, + "reason": "Both report the same training-loss double-shift bug in causal-LM loss computation; they could plausibly be fixed by one PR." }, { - "left": "issue:44292", - "right": "issue:46015", + "left": "issue:43872", + "right": "issue:46304", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime failure is unrelated to RAG docs." + "reason": "bitsandbytes storage/type incompatibility is unrelated to distributed support changes." }, { - "left": "issue:43872", - "right": "issue:46304", + "left": "issue:44292", + "right": "issue:46015", "accept": false, - "reason": "bitsandbytes Int8Params incompatibility is unrelated to distributed build support." + "reason": "Qwen-3-8B-NVFP4 runtime error is unrelated to RAG docs." }, { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding_window, but the model and failure mode are different and not clearly the same underlying bug." + "reason": "Gemma3 config attribute error and Qwen3-MoE sliding-window issue are separate model-specific regressions." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Different loading failures: corrupted checkpoint versus implementation missing support for checkpoints." + "reason": "Different loading failures: corrupted Ovis2 checkpoint vs NemotronH checkpoint-loading implementation." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "Both are IndexErrors, but one is a mask-length mismatch and the other is ZeRO-3 empty-tensor indexing." + "reason": "Mask/tensor shape mismatch and ZeRO-3 BertModel index error are different bugs." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper model execution failure is unrelated to table-question-answering crashes." + "reason": "Whisper load failure and table QA crash are unrelated." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "Unit-test loading failure for GLM-4.7-Flash is unrelated to a corrupted Ovis2 checkpoint." + "reason": "Unit-test model-loading failure and corrupted checkpoint are not the same issue." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch 2.10 incompatibility and a pydantic PretrainedConfig regression are different issues." + "reason": "Torch/transformers incompatibility and pydantic PretrainedConfig breakage are different changes." } ] }, "evaluator_result": { "accept": true, - "feedback": "Overall grounded and conservative. The main accepted pair (41762/43596) is supported by the shared ZeRO-3 size-0 IndexError wording, and the rejected pairs generally differ by model or bug class. One minor caution: the reason for the accepted pair says the \u201ccode path match,\u201d which is stronger than what the packet directly shows, so a more cautious phrasing would be better." + "feedback": "The cluster summary is well grounded in the packet and stays conservative overall. The two accepted duplicate pairs are plausible from the issue titles, and the rejected soft matches mostly avoid overclaiming. The canonical issue choice of issue:45920 is also supported by the broader v5 tokenizer regression framing." }, "evaluator_used": true, "retried": false, @@ -895,9 +901,9 @@ ], "soft_pairs": [ "issue:43541|issue:47228", - "issue:43975|issue:45901", "issue:43792|issue:45070", "issue:43596|issue:45800", + "issue:43975|issue:45901", "issue:45362|issue:45464", "issue:43541|issue:46869", "issue:44292|issue:45901", @@ -921,8 +927,8 @@ "issue:43366|issue:45464", "issue:44291|issue:45070", "issue:43531|issue:45070", - "issue:44387|issue:45005", "issue:46869|issue:47228", + "issue:44387|issue:45005", "issue:45020|issue:45800", "issue:45507|issue:47228", "issue:43716|issue:46294", @@ -936,20 +942,20 @@ "issue:39186|issue:44918", "issue:43716|issue:44918" ], - "prepared_review_unit_hash": "c87000027e2ebd7c756e860c946e3bcdb9fed861fca1753cd24e8c7927aaf3b8", + "prepared_review_unit_hash": "f0b8d71cf51242eda3ea6adda963ab7113bcab321635a51109fb5651794af361", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 41, - "serialized_chars": 39845, + "serialized_chars": 39847, "estimated_input_tokens": 9962, "estimated_eval_tokens": 20180 }, @@ -958,265 +964,274 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the items cover many different models, runtimes, and failure modes, and none of the soft pairs look like true duplicates of the same underlying bug.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue: the candidate items are too heterogeneous to safely collapse into one duplicate target.", - "canonical_pr_reason": "No PRs were provided in this cluster.", - "best_issue_reason": "No strong representative issue exists because the soft pairs do not form a real duplicate group.", - "best_pr_reason": "No PRs were provided in this cluster.", + "summary": "Mostly a loose collection of unrelated Transformers issues, with only a couple of clear duplicate-style pairs. The strongest matches are the repeated mask-shape IndexError reports and the Qwen video input/token mismatch regression.", + "confidence": 0.57, + "canonical_issue_reason": "Issue 41093 is the best representative because it is a generic, recurring error signature and has a near-verbatim duplicate (44805), making it the clearest canonical issue among the set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 41093 is the most suitable global issue anchor here: it is concise, generic, and maps cleanly to an exact duplicate report with the same failure mode.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43541", "right": "issue:47228", "accept": false, - "reason": "Different model/regression: one is a Mixtral torch-dynamo grouped_mm error, the other a Sam3 bf16 compilation regression." - }, - { - "left": "issue:43975", - "right": "issue:45901", - "accept": false, - "reason": "Both are failures, but one is detokenization in v5 and the other a table-question-answering crash; unrelated." + "reason": "Both are regression/runtime failures, but one is a grouped_mm torch-dynamo issue and the other is a bf16 compilation regression; not the same bug." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Import failure for GenerationMixin vs a PretrainedConfig pydantic-field regression; different code paths." + "reason": "Unrelated failures: Whisper model loading vs a PretrainedConfig/pydantic field breakage." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "Both involve loading/runtime problems, but one is a ZeRO-3 empty-index error and the other is torch/transformers incompatibility." + "reason": "Different root causes: DeepSpeed ZeRO-3 empty tensor indexing vs torch/transformers version incompatibility." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Different models and symptoms: detokenization bug vs table-question-answering crash." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both mention Qwen3.5, but one is a chat crash and the other is a streaming chat/completions failure; different triggers." + "reason": "Both involve Qwen3.5, but one is a chat crash and the other is a streaming API failure; too different to treat as the same bug." }, { "left": "issue:43541", "right": "issue:46869", "accept": false, - "reason": "Mixtral torch-dynamo tracing error and YOLO multi-GPU device-map handling are different subsystems and bugs." + "reason": "Grouped_mm tracing error and YOLO multi-GPU device-map failure are different code paths." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "One is an init_empty_weights keyword/type error, the other is a table QA task crash; not the same bug." + "reason": "Different failures: NVFP4 model load error vs table QA task crash." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Gemma3 sliding_window_pattern missing vs Gemma4 device mismatch; separate model/config issues." + "reason": "Gemma3 sliding_window attribute error is unrelated to Gemma4 device mismatch." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "ImportError for GenerationMixin vs Gemma3 config attribute error; unrelated failures." + "reason": "ImportError for GenerationMixin is unrelated to Gemma3 config attribute access." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Chat/completions streaming failure on Qwen3.5 is unrelated to an MPS histogram kernel limitation." + "reason": "Chat/completions streaming failure is unrelated to the MPS histogram kernel error." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Mixtral grouped_mm tracing error and MPS histogram failure are different backends and failure modes." + "reason": "Different bugs: torch-dynamo grouped_mm tracing vs MPS int histogram support." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Mixtral grouped_mm tracing error and Phi-tiny-MoE autocast dtype mismatch are distinct bugs." + "reason": "Grouped_mm tracing failure is not the same as an autocast dtype mismatch." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Index/template failures on different models and code paths; the shared IndexError wording is not enough." + "reason": "Processor failure for Voxtral is unrelated to the template-node compilation error." }, { "left": "issue:45507", "right": "issue:46869", "accept": false, - "reason": "GraniteMoEHybrid invalid method call and YOLO device-map auto failure are distinct issues." + "reason": "GraniteMoE hybrid invalid method call is unrelated to YOLO multi-GPU device mapping." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "A tied-weight bug in Mistral3 is not the same as an Ovis2 checkpoint corruption report." + "reason": "lm_head weight tying bug is unrelated to a corrupted checkpoint report." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are index errors, but one is a mask-length mismatch and the other is a zero-sized tensor load error with Gemma3/ZeRO-3." + "reason": "Both are IndexError reports, but the shapes and loading paths differ; not enough to assume the same underlying bug." }, { "left": "issue:45399", "right": "issue:46661", "accept": false, - "reason": "FlashAttention fallback checks and remote-code model-class loading are different paths." + "reason": "Flash-Attn fallback gating and remote-code config_class lookup are different issues." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and the pydantic PretrainedConfig regression are separate config issues." + "reason": "Missing pad_token_idx and pydantic PretrainedConfig breakage are separate regressions." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "torch/transformers version incompatibility vs Gemma4 forward device mismatch; different root causes." + "reason": "Version incompatibility vs Gemma4 device mismatch are not the same bug." }, { "left": "issue:43872", "right": "issue:47228", "accept": false, - "reason": "bitsandbytes Int8Params keyword incompatibility is unrelated to Sam3 bf16 compilation regression." + "reason": "bitsandbytes constructor incompatibility is unrelated to the Sam3 bf16 compilation regression." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Mask-shape mismatch and Qwen3.5 embedding unpacking failure are different bugs." + "reason": "They both mention shape/embedding problems, but one is a mask-index mismatch and the other is a TRL SFT embedding unpacking failure." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Phi-tiny-MoE autocast dtype mismatch is unrelated to Qwen3.5 streaming chat failure." + "reason": "Autocast dtype mismatch is unrelated to streaming chat/completions failure." }, { "left": "issue:43716", "right": "issue:47228", "accept": false, - "reason": "Mistral-3 image dtype mismatch and Sam3 compilation regression are different model/runtime problems." + "reason": "Mistral-3 image preprocessor dtype mismatch is unrelated to Sam3 bf16 compilation." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM is not the same as a GGUF gpt-oss support request." + "reason": "PaddleOCR-VL loading failure and GGUF gpt-oss support are different model/loader issues." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF gpt-oss support and a Qwen3.5 chat streaming crash are unrelated." + "reason": "GGUF architecture support is unrelated to Qwen3.5 streaming inference failures." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "The _is_hf_initialized TypeError during init_empty_weights is different from the pydantic field regression." + "reason": "init_empty_weights/_is_hf_initialized TypeError is unrelated to the PretrainedConfig pydantic field issue." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding_window handling is unrelated to the PretrainedConfig pydantic breakage." + "reason": "Qwen3-MoE sliding_window behavior and pydantic config serialization are different defects." }, { - "left": "issue:44387", - "right": "issue:45005", + "left": "issue:46869", + "right": "issue:47228", "accept": false, - "reason": "int4 OOM regression and tied-weights translation-model issues are different symptoms and likely fixes." + "reason": "YOLO multi-GPU device-map failure is unrelated to Sam3 bf16 compilation regression." }, { - "left": "issue:46869", - "right": "issue:47228", + "left": "issue:44387", + "right": "issue:45005", "accept": false, - "reason": "YOLO multi-GPU device-map failure and Sam3 bf16 compilation regression are unrelated." + "reason": "Int4 memory/OOM regression and tied-weights translation-model issues are different problems." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code breakage and torch/transformers incompatibility are distinct causes." + "reason": "Remote-code model breakage is broader than, and unrelated to, the torch 2.4.1 / transformers 5.8.0 incompatibility." }, { "left": "issue:45507", "right": "issue:47228", "accept": false, - "reason": "Nemotron example breakage and Sam3 compilation regression are unrelated." + "reason": "GraniteMoE hybrid method error and Sam3 compilation regression are unrelated." }, { "left": "issue:43716", "right": "issue:46294", "accept": false, - "reason": "Mistral-3 image dtype mismatch and Qwen3.5 video token mismatch are different multimodal alignment bugs." + "reason": "Image preprocessor dtype mismatch and video token/feature mismatch are different multimodal bugs." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Two index errors, but one is a mask/tensor size mismatch and the other is an empty-dimension load error." + "reason": "Both are IndexErrors, but they occur in different contexts and don\u2019t share the same concrete failure mode." }, { "left": "issue:44479", "right": "issue:46294", - "accept": false, - "reason": "Video-input regression across Qwen variants is not the same as the video-token/feature mismatch in Qwen3.5-9B." + "accept": true, + "reason": "Both describe the same Qwen video-input regression: video features/tokens become mismatched in the chat-template path." }, { "left": "issue:43054", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2, but one reports worse text embeddings while the other reports nonsensical AutoModel/pipeline outputs; not enough to prove one bug." + "reason": "Siglip2 text-embedding quality complaint is not the same as the AutoModel/pipeline nonsensical-output report." }, { "left": "issue:46468", "right": "issue:46869", "accept": false, - "reason": "Nemotron example breakage and YOLO device-map auto failure are unrelated." + "reason": "Broken official Nemotron example and YOLO multi-GPU device-map failure are different issues." }, { "left": "issue:43596", "right": "issue:46165", "accept": false, - "reason": "Zero-3 empty-index loading error and resume-training with remote code are different failure modes." + "reason": "DeepSpeed ZeRO-3 index error is unrelated to resuming training with local models and remote code." }, { "left": "issue:46640", "right": "issue:46661", "accept": false, - "reason": "Flash Attention 2 issue and remote-code model_class/config_class error are unrelated." + "reason": "Flash Attention 2 issue and remote-code config_class error are different failure modes." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite-config text_config unwrap issue and Qwen3.5-MoE rotary embedding memory initialization bug are different bugs." + "reason": "Composite-config unwrapping bug and uninitialized rotary embedding memory are distinct Qwen3.5-MoE issues." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP 2-D weight error and Qwen3.5 embedding unpacking failure are distinct training/loading problems." + "reason": "FSDP 2-D weight requirement is unrelated to TRL SFT embedding unpacking." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Mistral-3 image dtype mismatch and Qwen3.5 embedding unpacking failure are unrelated multimodal/model-loading issues." + "reason": "Mistral-3 image preprocessor dtype mismatch is unrelated to Qwen3.5 embedding unpacking in TRL." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": true, + "reason": "Same mask-shape IndexError; these look like the same underlying indexing bug reported twice." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet and stays conservative overall. The canonical anchor choice for issue 41093 is defensible because 44805 is a near-verbatim duplicate of the same mask-shape IndexError, and the other accepted pair (44479/46294) is also a plausible same-bug match from the issue titles. The remaining soft-edge rejections are appropriately cautious and do not overclaim duplicate status." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -1288,8 +1303,8 @@ "issue:42915|issue:45399", "issue:45632|issue:45698", "issue:43404|issue:45005", - "issue:43873|issue:45005", "issue:44360|issue:44512", + "issue:43873|issue:45005", "issue:35308|issue:36331", "issue:44492|issue:44512", "issue:44918|issue:45507", @@ -1315,223 +1330,229 @@ "issue:45987|issue:46858", "issue:43950|issue:45902" ], - "prepared_review_unit_hash": "8bf3e5b67060b2810dd0af12c496b809b393356d738a11d053deeb4700f7bdfb", + "prepared_review_unit_hash": "1b474c22f9554ad15ce5d0bc75f8922d90838d276f72de4f64cea11998cea507", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37129, - "estimated_input_tokens": 9283, - "estimated_eval_tokens": 18822 + "serialized_chars": 37123, + "estimated_input_tokens": 9281, + "estimated_eval_tokens": 18818 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are not duplicates; they span unrelated model-specific regressions, docs issues, and API breakages. Clear duplicate groups appear around non-persistent buffer corruption in v5, StaticCache.get_seq_length() returning a Tensor instead of an int, quantized Gemma4 multimodal dtype casting, and disk-offloaded save failures.", - "confidence": 0.87, - "canonical_issue_reason": "Issue #43950 is the clearest representative for the non-persistent-buffer regression: it precisely states the silent corruption behavior and the transformers 5.x scope.", + "summary": "This is still a heterogeneous issue cluster. The strongest duplicate evidence is around the non-persistent-buffer regression in v5, while most other pairs are only thematically related. I removed the over-strong acceptance for the custom-module loading pair: they may be adjacent reports, but the packet does not prove a single underlying bug.", + "confidence": 0.81, + "canonical_issue_reason": "#43950 is the clearest representative regression report here: it is specific, actionable, and matches the strongest confirmed duplicate set around non-persistent buffer corruption in v5.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #43950 is the strongest overall issue in this set because it is specific, actionable, and has a clear regression description with a well-scoped fix target.", + "best_issue_reason": "#43950 is the best anchor for the cluster because it is concrete, broadly phrased, and directly covers one of the strongest same-bug pairs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Unrelated: custom-model/Jupyter initialization versus flash-attn fallback gating." + "reason": "Different problems: notebook/custom-model initialization vs flash-attn2 fallback gating." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "Different bugs in different models; perplexity regression is not GLM5 load behavior." + "reason": "Different model/perplexity reports with no clear shared code path." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "Different scope: Qwen3Moe FP8 failure versus flash-attn fallback checks." + "reason": "Qwen3Moe FP8 failure is unrelated to flash-attn fallback checks." }, { "left": "issue:45632", "right": "issue:45698", "accept": false, - "reason": "Both involve custom code loading, but one is a leaf-dir cache collision and the other is wrong module selection after save/load." + "reason": "Related custom-module loading symptoms, but the packet only shows titles and does not establish the same bug or a mergeable single fix." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Related to tied weights, but one is a specific Mistral3 lm_head issue and the other is a broader translation-model regression." + "reason": "Both mention tied weights, but they concern different model families and different failure modes." }, { - "left": "issue:43873", - "right": "issue:45005", + "left": "issue:44360", + "right": "issue:44512", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to tied-weights initialization." + "reason": "Unrelated: DSA indexer activation bug vs documentation cleanup." }, { - "left": "issue:44360", - "right": "issue:44512", + "left": "issue:43873", + "right": "issue:45005", "accept": false, - "reason": "A DSA/ReLU discussion and a docs typo are not the same bug." + "reason": "Quantization/offloading behavior is unrelated to tied-weight translation-model issues." }, { "left": "issue:35308", "right": "issue:36331", "accept": false, - "reason": "IterableDataset multi-GPU crash and a custom trainer signature error are unrelated." + "reason": "Different training bugs: iterable dataset length mismatch vs custom compute_loss signature breakage." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "Cache-strategies typo versus removed run-command docs are separate documentation issues." + "reason": "A typo in cache strategies is unrelated to a removed CLI docs reference." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different model paths and failures: TRL SFT unpacking versus invalid method call." + "reason": "Different model/trainer issues with no shared failing code path." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight regressions." + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied weights." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Different tokenizer/model failures; no shared code-path evidence." + "reason": "Different tokenizer/runtime failures on different models." }, { "left": "issue:45183", "right": "issue:46463", "accept": false, - "reason": "Both are validation/error-message bugs, but they target different validation paths and models." + "reason": "Related area, but one is broad input_ids/inputs_embeds validation and the other is a T5-specific decoder-input error message." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint loading is unrelated to tied-weight initialization." + "reason": "NemotronH checkpoint loading is unrelated to tied-weight translation-model behavior." + }, + { + "left": "issue:35308", + "right": "issue:42915", + "accept": false, + "reason": "Multi-GPU IterableDataset crash and FP8 model failure are unrelated." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "Doc/example errors versus embedding quality regression; not the same underlying bug." + "reason": "Both are SigLIP2-related, but one is docs/example mismatch and the other is an actual embedding quality issue." }, { "left": "issue:46489", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different tokenizers and different concrete round-trip behavior." + "reason": "Both are tokenizer regressions, but they affect different models and different incorrect outputs." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Same non-persistent-buffer corruption regression in transformers v5; one report is the specific manifestation and the other the broader symptom." + "reason": "Same non-persistent-buffer regression in v5, just phrased as the broader symptom versus the specific from_pretrained corruption." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "CustomTrainer kwarg mismatch is unrelated to Gemma3 ZeRO-3 loading failure." + "reason": "Different training API breakages: custom trainer loss signature vs ZeRO-3 model loading failure." }, { "left": "issue:43644", "right": "issue:43950", "accept": true, - "reason": "Same bug: v5 corrupts non-persistent buffers during from_pretrained/load materialization." + "reason": "Clear duplicate pair: both report v5 corrupting non-persistent buffers with junk values during loading/materialization." }, { "left": "issue:33666", "right": "issue:35308", "accept": false, - "reason": "Both mention multi-GPU training, but the model-specific Qwen2-VL issue is not the IterableDataset length crash." + "reason": "Both are training-related, but they describe different multi-GPU/data-loader issues." }, { "left": "issue:46032", "right": "issue:47246", - "accept": true, - "reason": "Same underlying inter-chunk recurrence bug in Mamba2-style mixer code; NemotronH is a copied variant missing the same fix." + "accept": false, + "reason": "Related Mamba-family code, but not the same concrete bug: seq_len>1 cache corruption vs NemotronH slow-path recurrence propagation." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Distinct quantization/storage incompatibilities with different failure signatures." + "reason": "Different quantization/storage incompatibilities with different error signatures." }, { "left": "issue:46899", "right": "issue:46932", "accept": true, - "reason": "Same Gemma4 quantized vision/audio embedder dtype-casting bug; #46932 is the follow-up fix report." + "reason": "Same Gemma4 quantized vision/audio embedder dtype-casting bug; #46932 is explicitly the fix for the same issue." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "Different problems: SigLIP2 embedding quality versus tokenizer/version-dependent tokenization changes." + "reason": "Text-embedding quality regression and tokenization-version change are different bugs." }, { "left": "issue:45920", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer regressions, but they concern different models and different concrete outputs." + "reason": "Broad AutoTokenizer token-id regression across models is distinct from a DeepSeek-Coder v1 encode/decode issue." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Unexpected zero probabilities is not the same as SigLIP2 text-embedding quality degradation." + "reason": "Different SigLIP2 symptoms: zero probabilities vs degraded text embeddings." }, { "left": "issue:47332", "right": "issue:47333", "accept": true, - "reason": "Same disk-offloaded save failure; only the affected Qwen3 variant differs." + "reason": "Same offloading-save failure: both report inability to save disk-offloaded Qwen3 models, differing only by model variant." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Autocast dtype mismatch and RLHF NaN explosion are different failure modes." + "reason": "Different dtype-related failures on different models and workloads." }, { "left": "issue:45987", "right": "issue:46858", "accept": true, - "reason": "Same get_seq_length contract bug: it returns a Tensor instead of an int, causing downstream cache arithmetic failures." + "reason": "Same cache contract bug: get_seq_length returns a tensor instead of a plain int, breaking downstream arithmetic." }, { "left": "issue:43950", "right": "issue:45902", "accept": false, - "reason": "Non-persistent buffer corruption is unrelated to Qwen3_5Moe rotary inv_freq uninitialized-memory reads." + "reason": "Non-persistent buffer loading corruption is unrelated to uninitialized rotary-embedding memory after materialization." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet, and the accepted soft-edge pairs are supported by strong title-level overlap without overstating shared root cause. The rejected pairs are conservative and appropriately scoped to distinct bugs/issues." + "accept": false, + "feedback": "Most of the summary is grounded, but one soft-edge acceptance is too strong: issue:45632 vs issue:45698 look related in custom-module loading, yet the packet only gives titles and does not establish that they are the same bug or a mergeable single fix. The rest of the accepted pairs are more defensible, but this one overstates the evidence, so I can\u2019t accept the cluster evaluation." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1611,8 +1632,8 @@ "issue:43643|issue:45020", "issue:44991|issue:45399", "issue:42947|issue:43856", - "issue:43824|issue:43872", "issue:45081|issue:45399", + "issue:43824|issue:43872", "issue:45685|issue:45823", "issue:41553|issue:43883", "issue:45020|issue:45823", @@ -1628,229 +1649,229 @@ "issue:45020|issue:46661", "issue:45488|issue:46489", "issue:43404|issue:43950", - "issue:43881|issue:43883", - "issue:43827|issue:44509" + "issue:43827|issue:44509", + "issue:43881|issue:43883" ], - "prepared_review_unit_hash": "6d18bf7a9f9c085cbf5da4c3144300c35896c8fde8a2943eca36dd79ee6fe083", + "prepared_review_unit_hash": "04c22ca897bb290180acee7727bb3137645cdfe6678ea1efdbf0bb2edd13c85a", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37488, - "estimated_input_tokens": 9372, - "estimated_eval_tokens": 19000 + "serialized_chars": 37489, + "estimated_input_tokens": 9373, + "estimated_eval_tokens": 19002 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: a mix of loading, tokenizer, remote-code, memory, and model-specific regressions. The only clear duplicate pairs are the v5 docs/pipeline-removal reports and the SigLIP2 implementation discrepancy reports; the rest are only theme-adjacent, not the same bug.", - "confidence": 0.77, - "canonical_issue_reason": "Issue 45020 is the broadest, most central regression report in the set and best captures the recurring remote-code / loading-breakage theme across several nearby reports.", + "summary": "Most pairs are only loosely related and should stay separate. The only clear duplicate-style match is the v5 docs cleanup around removed pipeline tasks. No pull requests were present.", + "confidence": 0.83, + "canonical_issue_reason": "Issue 44509 is the clearest and most complete statement of the v5 pipeline-docs cleanup problem; 43827 is the same issue in a narrower form.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45020 is the strongest global representative: high-signal, user-facing, and broad enough to anchor the cluster, even though the cluster overall is not a clean duplicate set.", + "best_issue_reason": "44509 is the best representative issue for the duplicate docs problem because it explicitly covers the removed pipeline task references and has the broadest scope.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43054", "right": "issue:43493", - "accept": true, - "reason": "Same underlying SigLIP2 discrepancy: HF behavior diverges from the original JAX/reference implementation." + "accept": false, + "reason": "Same model family, but one is a quality complaint and the other is a JAX/HF discrepancy report; not enough to treat as the same concrete bug." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Both are Qwen3.x-related, but one is a flash-attention crash from 3D position_ids handling and the other is a CPU-offload incompatibility in rotary embedding; different bugs." + "reason": "Both involve Qwen3.5/MoE-related runtime behavior, but the flash-attention position_ids crash and CPU offload rotary embedding issue are different code paths." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Different model families and failure modes: Gemma4 FSDP2 KeyError vs Qwen3_MoE uninitialized inv_freq after to_empty materialization." + "reason": "Gemma4 FSDP2 KeyError and Qwen3_5MoeRotaryEmbedding uninitialized-memory bug are unrelated failures." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Generic loading noise vs a specific inability to load ScandiBERT; not the same code-path bug." + "reason": "Generic loading noise vs a specific model load failure; different symptoms and likely different fixes." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Both mention training/memory, but one is about gradient checkpointing being ineffective with PEFT LoRA and the other is int4 quantization OOM; distinct issues." + "reason": "Both mention memory, but gradient checkpointing/LoRA inefficacy and int4 reserved-memory OOM are distinct regressions." }, { "left": "issue:24643", "right": "issue:28530", "accept": false, - "reason": "Training DeepSpeed weight-dimension error is unrelated to early stopping / metric_for_best_model handling." + "reason": "DeepSpeed weight-dimension training error and early-stopping metric configuration are unrelated." }, { "left": "issue:43975", "right": "issue:45237", "accept": false, - "reason": "Tokenizer detokenization regression and AMD GPU support failure are unrelated." + "reason": "Tokenizer detokenization bug and AMD GPU model support failure are different issues." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "One is a generic loading-noise report; the other is a concrete model-loading failure for cjvt/sleng-bert." + "reason": "These are both loading-related, but they concern different models and no shared underlying failure is evident." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Different model-specific errors: Gemma3 ZeRO-3 index error vs Qwen3Moe FineGrainedFP8Config failure." + "reason": "Different model families and different load-time errors; no sign of the same bug." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "Both involve remote-code, but one is a missing-fields config object issue and the other is a broader regression claim; not the same concrete bug." + "reason": "The remote-code regression theme is broad, but the specific AutoConfig missing-fields bug is not clearly the same concrete failure as the general report." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading regression and flash-attn2 fallback gating are unrelated." + "reason": "Tokenizer loading failure and flash-attn fallback gating are unrelated." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "Both are Qwen3/MoE-adjacent performance/memory reports, but they describe different symptoms and fixes." + "reason": "Both mention training/memory, but one is about gradient checkpointing with PEFT LoRA and the other about Qwen3 MoE memory usage." }, { - "left": "issue:43824", - "right": "issue:43872", + "left": "issue:45081", + "right": "issue:45399", "accept": false, - "reason": "Import error for a missing model class vs bitsandbytes Int8Params incompatibility; different subsystems and failure modes." + "reason": "Mistral tokenizer regex patch crash and flash-attn fallback blocking are separate subsystems." }, { - "left": "issue:45081", - "right": "issue:45399", + "left": "issue:43824", + "right": "issue:43872", "accept": false, - "reason": "Tokenizer regex crash and flash-attn2 fallback gating are separate bugs." + "reason": "Qwen2_5_VL import failure and bitsandbytes Int8Params incompatibility are different bugs." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "MPS histogram support error is unrelated to Gemma4 device-mismatch under device_map=auto." + "reason": "MPS histogram-int error and Gemma4 device-mismatch forward error are unrelated." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error message for Voxtral vs missing all_tied_weights_keys on MolmoForCausalLM; different models and issues." + "reason": "Voxtral tokenizer loading error and Molmo tied-weights attribute error are different model-specific bugs." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Broad remote-code regression and Gemma4 PLE device mismatch are not the same bug." + "reason": "A broad remote-code regression does not match the specific Gemma4 device-map forward bug." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL weight_only loading error and GLM5 issue are unrelated." + "reason": "Qwen3VL weight_only loading error and generic GLM5 loading failure are not the same failure mode." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "add-new-model-like failing in TOKENIZER_MAPPING_NAMES is a registration bug, not the same as remote-code loading regressions." + "reason": "add-new-model-like registry logic and remote-code loading regressions are different code paths." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to Gemma4 device placement." + "reason": "Autocast dtype mismatch and Gemma4 device mismatch are separate runtime problems." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Kimi-K2.5 tokenizer regression and remote-code breakage are different code paths." + "reason": "Remote-code breakage and Kimi-K2.5 tokenizer codec/fix_mistral_regex regression are unrelated." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Different models and failures: Voxtral tokenizer error message vs glm-4v-9b loading failure." + "reason": "Different models and different load errors; no duplicate evidence." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "LoRA hf4.x/hf5.x incompatibility and tie_word_embeddings warning are related in theme but not the same concrete bug." + "reason": "LoRA model incompatibility across HF versions and a tie_word_embeddings warning during Qwen3.5 fine-tuning are not the same bug." }, { "left": "issue:43872", "right": "issue:46869", "accept": false, - "reason": "bitsandbytes _is_hf_initialized incompatibility and YOLO multi-GPU device-map failure are unrelated." + "reason": "bitsandbytes Int8Params incompatibility and YOLO multi-GPU device-map failure are unrelated." }, { "left": "issue:45759", "right": "issue:46661", "accept": false, - "reason": "Composite config unwrapping issue and remote-code model_class.config_class AttributeError are separate code paths." + "reason": "Composite-config unwrapping for Qwen multimodal models and remote-code model_class config_class handling are different issues." }, { "left": "issue:45715", "right": "issue:46315", "accept": false, - "reason": "Tokenizer performance regressions in different conversion functions; same area, but not one bug." + "reason": "Both are tokenizer performance regressions, but convert_ids_to_tokens and convert_tokens_to_ids have different hot paths and fixes." }, { "left": "issue:45020", "right": "issue:46661", "accept": false, - "reason": "Both touch remote-code, but one is a broad regression report and the other is a specific config_class AttributeError." + "reason": "The general remote-code regression is too broad to equate with this specific config_class AttributeError without more evidence." }, { "left": "issue:45488", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer regressions affecting DeepSeek models, but they target different families/failure modes and are not clearly one underlying bug." + "reason": "Both concern tokenizer correctness, but they involve different model families and different tokenizer mechanisms." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Un-tied lm_head weights in Mistral3 and non-persistent buffer corruption are unrelated." - }, - { - "left": "issue:43881", - "right": "issue:43883", - "accept": false, - "reason": "glm-4v-9b loading failure and Molmo missing all_tied_weights_keys are different bugs." + "reason": "Tied lm_head weights and non-persistent buffer corruption are separate from one another." }, { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "Same docs regression: v5 pipeline removals are still referenced; 44509 is a clearer duplicate of 43827." + "reason": "Both report the same v5 docs problem: stale references to removed pipeline tasks/models in the summarization/translation docs." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "GLM-4v-9b loading failure and Molmo missing all_tied_weights_keys are different model-specific bugs." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative about the cluster\u2019s heterogeneity. The two accepted soft pairs are the most defensible duplicates in the set: the v5 docs/pipeline-removal issues and the SigLIP2 discrepancy reports. The remaining verdicts avoid over-collapsing theme-adjacent reports into duplicates, which matches the evidence available." + "feedback": "The evaluation is grounded in the packet. The only accepted duplicate-style pair, issue:43827 \u2194 issue:44509, is well supported by the titles and fits the v5 docs cleanup theme. The rejected soft-edge pairs are phrased conservatively and do not overclaim duplication or mergeability. No PR pairs were present." }, "evaluator_used": true, "retried": false, @@ -1895,12 +1916,12 @@ "issue:44155", "issue:44220", "issue:44479", - "issue:44488", "issue:44530", "issue:44560", "issue:44589", "issue:44805", "issue:44863", + "issue:44918", "issue:44928", "issue:44991", "issue:45305", @@ -1934,238 +1955,238 @@ "issue:43366|issue:43828", "issue:44928|issue:45588", "issue:43716|issue:46468", - "issue:44530|issue:44863", "issue:43366|issue:43531", - "issue:44863|issue:46402", + "issue:44530|issue:44863", "issue:43596|issue:46515", + "issue:44863|issue:46402", "issue:41553|issue:44991", "issue:45325|issue:45902", "issue:43366|issue:44589", "issue:43638|issue:46515", - "issue:45663|issue:45823", "issue:43716|issue:44560", + "issue:45663|issue:45823", "issue:42915|issue:43957", "issue:44155|issue:45584", "issue:43824|issue:43957", "issue:42617|issue:43366", "issue:43950|issue:45834", "issue:44560|issue:45685", - "issue:41553|issue:44488" + "issue:44560|issue:44918" ], - "prepared_review_unit_hash": "7ce77a52200e458d4f6e932c36053ecf04059d7eb7aa3ddddd8416be0564dc17", + "prepared_review_unit_hash": "b3f11ead635fc4ea842b396ce0cab22a05caa78f9db6849bff229069670c163c", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36759, - "estimated_input_tokens": 9190, - "estimated_eval_tokens": 18636 + "serialized_chars": 36762, + "estimated_input_tokens": 9191, + "estimated_eval_tokens": 18638 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are false positives across unrelated bug reports. I\u2019d only accept the exact mask-shape duplicate pair and the two training-loss double-shift reports; everything else looks like different root causes. No PRs are present.", - "confidence": 0.88, - "canonical_issue_reason": "issue:41093 is the cleanest duplicate anchor: issue:44805 repeats the same mask/tensor shape mismatch wording, so 41093 is the best representative for that exact bug.", + "summary": "This set is mostly heterogeneous and does not form one coherent duplicate cluster. The only clear duplicate-looking edge is the pair of training-loss double-shift reports; the rest are similar only at the subsystem or symptom level and should stay separate.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 46894 is the best representative because it states the underlying training-loss double-shift bug directly and is the earlier, clearer report; 46897 appears to be the same bug on another model.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46894 is the best global triage target among this set because it is open, describes the recurring training-loss double-shift regression explicitly, and generalizes to the similar Florence2 report.", + "best_issue_reason": "46894 is the strongest standalone issue in the set and the only solid anchor for deduplication within this mostly unrelated collection.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Both are audio/Whisper-related, but one is a general \u201ccan\u2019t run\u201d report and the other is a specific fbank helper issue; not the same underlying bug." + "reason": "Both are Whisper/audio-related, but one is a generic run failure and the other is a specific fbank feature extraction problem; not the same concrete bug." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Different failures in different code paths: unexpected compute_loss kwarg vs. a mask/tensor shape mismatch." + "reason": "Different failures: Trainer signature mismatch versus a tensor mask shape IndexError." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Both mention DeepSpeed, but one is weight conversion/loading and the other is gradient averaging with GAS; these are distinct bugs." + "reason": "Both mention Qwen3 and DeepSpeed, but one is about weight conversion/loading and the other is about gradient averaging; different bugs." }, { "left": "issue:46894", "right": "issue:46897", "accept": true, - "reason": "These describe the same training-loss double-shift regression pattern, just reproduced on different models; one fix could plausibly cover both." + "reason": "Same training-loss double-shift bug pattern, likely fixed by one shared change." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing/LoRA behavior is unrelated to a missing supports_gradient_checkpointing flag on a Nemotron trust_remote_code variant." + "reason": "Gradient checkpointing effectiveness and a missing supports_gradient_checkpointing flag are related only loosely; not the same bug." }, { "left": "issue:43940", "right": "issue:45313", "accept": false, - "reason": "Both are ZeRO-3 loading failures in Qwen family models, but the model targets and reported symptoms differ enough that this is not clearly the same bug." + "reason": "Both are ZeRO-3 loading issues on Qwen variants, but the models and failure modes differ enough that they do not look like one bug." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Different model families and different loading failures; only a broad \u201ccan\u2019t load checkpoint\u201d similarity." + "reason": "Different model families and different loading failures; not a duplicate." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "Training with DeepSpeed and image-processor void segmentation-map handling are unrelated." + "reason": "Unrelated issues: a DeepSpeed training runtime error versus an image processor segmentation-map failure." }, { "left": "issue:44479", "right": "issue:44560", "accept": false, - "reason": "Both are video/VL regressions, but one is a broad multi-model regression and the other is a specific StopIteration in Qwen3-vl-embedding; not enough evidence of one root cause." + "reason": "Both are video-related Qwen regressions, but one is a broad v5.3.0 regression and the other is a specific StopIteration path; not enough to merge." }, { "left": "issue:41093", "right": "issue:44805", - "accept": true, - "reason": "Near-verbatim IndexError/mask-shape mismatch reports; same underlying off-by-one shape bug." + "accept": false, + "reason": "Same generic mask-shape error wording, but no evidence they hit the same code path or model." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "GGUF gpt-oss support and an autocast dtype mismatch in Phi-tiny-MoE are unrelated." + "reason": "One is GGUF architecture support, the other is a dtype mismatch under autocast; unrelated." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "Qwen3.5 RLHF NaN explosion from position_ids/SDPA fallback is unrelated to a flash-attention crash on s_aux=None." + "reason": "Different code paths and symptoms: RLHF/SDPA fallback versus flash-attention handling of s_aux=None." }, { "left": "issue:43716", "right": "issue:46468", "accept": false, - "reason": "Different models and different problems: image-preprocessor dtype mismatch vs. an example/script loading issue." - }, - { - "left": "issue:44530", - "right": "issue:44863", - "accept": false, - "reason": "PagedAttentionCache linear_attention crash is unrelated to NemotronH checkpoint loading." + "reason": "Different models and different failures: image-preprocessor dtype mismatch versus a broken Nemotron example." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "One is GGUF architecture support, the other is a sliding_window issue in Qwen3-MoE; not the same bug." + "reason": "GGUF support for gpt-oss and a sliding_window issue in Qwen3-MoE are unrelated." }, { - "left": "issue:44863", - "right": "issue:46402", + "left": "issue:44530", + "right": "issue:44863", "accept": false, - "reason": "Both are loading-related, but NemotronH checkpoint loading and Sam3 text-encoder weights missing are different failures." + "reason": "PagedAttentionCache group-type crash and NemotronH checkpoint loading are separate problems." }, { "left": "issue:43596", "right": "issue:46515", "accept": false, - "reason": "Zero3/BertModel index error and generic torch 2.4.1 incompatibility are too different to merge." + "reason": "Zero-3/Bert index error versus a broad torch compatibility complaint; not the same bug." + }, + { + "left": "issue:44863", + "right": "issue:46402", + "accept": false, + "reason": "Different Nemotron/Sam3 loading problems with different symptoms and code paths." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Tokenizer-loading problems for different models; the specific failure modes and targets do not line up." + "reason": "Both are tokenizer/loading issues, but for different models and different error modes." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both touch position/rotary handling, but one is a still-image temporal scaling bug and the other is uninitialized inv_freq after meta-to_empty materialization." + "reason": "Different bugs: temporal position scaling in Qwen2.5-VL versus uninitialized inv_freq after meta-to-empty materialization." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "GGUF support for gpt-oss and a Float8 storage TypeError are unrelated." + "reason": "Architecture support for gpt-oss and a Float8 storage lookup error are unrelated." }, { "left": "issue:43638", "right": "issue:46515", "accept": false, - "reason": "Bert zero3 initialization failure and a generic torch incompatibility report are not the same underlying issue." + "reason": "Different root causes and symptoms; no duplicate evidence." }, { - "left": "issue:45663", - "right": "issue:45823", + "left": "issue:43716", + "right": "issue:44560", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states bug and a device mismatch in Gemma4 PLE forward are different code-path problems." + "reason": "Mistral-3 image dtype mismatch and Qwen3-vl video StopIteration are separate multimodal bugs." }, { - "left": "issue:43716", - "right": "issue:44560", + "left": "issue:45663", + "right": "issue:45823", "accept": false, - "reason": "Mistral-3 image-preprocessor dtype mismatch and a Qwen3-vl video StopIteration are unrelated." + "reason": "FSDP2 shared_kv_states logic and a PLE device-map mismatch are different forward-pass issues." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "FineGrainedFP8Config failure in Qwen3Moe and meta-device loading regressions are different problems." + "reason": "FineGrainedFP8Config failure and meta-device loading breakage are different model-loading problems." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Both are audio-generation issues, but one is embedding/token leakage between tracks and the other is empty-transcription handling after align_special_tokens." + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription handling are unrelated." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Missing Qwen2_5_VL import and meta-device loading regressions are unrelated." + "reason": "Import exposure for a missing class versus model-loading breakage on meta device; not the same issue." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "3d_parallel.py runtime failure and GGUF gpt-oss support are unrelated." + "reason": "A failed 3d_parallel.py run and GGUF gpt-oss support are unrelated." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Non-persistent buffer corruption in from_pretrained is unrelated to Kosmos2.5 long-OCR index errors." + "reason": "Non-persistent buffer corruption and Kosmos2.5 long-ocr index errors are different regressions." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Qwen3-vl video StopIteration and an MPS histogram Int limitation in moe code are unrelated." + "reason": "Qwen3-vl video StopIteration and an MPS histogram limitation on Int are unrelated." }, { - "left": "issue:41553", - "right": "issue:44488", + "left": "issue:44560", + "right": "issue:44918", "accept": false, - "reason": "A bad AutoTokenizer error message for Voxtral is not the same as failure to load cjvt/sleng-bert." + "reason": "Video embedding StopIteration and TRL SFT unpacking failure are different code paths and bugs." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded overall. The global triage and canonical-anchor choices are consistent with the packet, and the summary stays conservative. The two accepted soft pairs are plausibly the same bug based on near-identical titles, and the rejections avoid overreaching beyond the evidence. No PRs are present, as noted." + "feedback": "The summary is grounded in the packet and stays conservative. The single accepted soft edge (46894/46897) is supported by both titles describing the same training-loss double-shift pattern, while the other pairs are reasonably left as non-duplicates based on differing models, symptoms, or code paths." }, "evaluator_used": true, "retried": false, @@ -2189,6 +2210,7 @@ "issue:36010", "issue:38175", "issue:39692", + "issue:41553", "issue:41720", "issue:42491", "issue:42915", @@ -2214,7 +2236,6 @@ "issue:44863", "issue:44877", "issue:44910", - "issue:44918", "issue:44960", "issue:44977", "issue:45030", @@ -2235,7 +2256,7 @@ "issue:47254" ], "soft_pairs": [ - "issue:44560|issue:44918", + "issue:41553|issue:44488", "issue:43645|issue:43646", "issue:44683|issue:46515", "issue:36010|issue:42915", @@ -2262,224 +2283,224 @@ "issue:42491|issue:45276", "issue:44960|issue:44977", "issue:44451|issue:44488", - "issue:45313|issue:45357", + "issue:45507|issue:46468", "issue:44863|issue:46846", - "issue:45507|issue:46468" + "issue:45313|issue:45357" ], - "prepared_review_unit_hash": "52c4a14b8f287fc9f2dcd9248999604eee76b3089e3d3ed02d851f87eac1b3d1", + "prepared_review_unit_hash": "b4ab555a743f338c11d96f0bd4095eb1431176052bf2384ba7fb27992468732f", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 35971, - "estimated_input_tokens": 8993, - "estimated_eval_tokens": 18242 + "serialized_chars": 35962, + "estimated_input_tokens": 8991, + "estimated_eval_tokens": 18238 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous Transformers regression reports, with a few true duplicate pairs around the same model or v5-init bug. The strongest canonical anchor is the generic custom-model initialization regression, while most other pairs are only superficially similar.", - "confidence": 0.87, - "canonical_issue_reason": "issue:43646 is the broadest and most central Transformers v5 regression in the set: it describes the core custom-model initialization breakage without extra environment-specific constraints, making it the best anchor for the cluster.", + "summary": "This cluster is mostly a loose set of Transformers v5 regressions with only a few true duplicate pairs. The clearest duplicates are the custom-model initialization break and the non-persistent buffer junk regression; the SigLIP2 checkpoint output reports also look like the same underlying model bug.", + "confidence": 0.64, + "canonical_issue_reason": "issue:43646 is the broadest and most representative issue among the true duplicate candidates: it describes the core custom model initialization regression, while issue:43645 is just the notebook-specific manifestation.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best overall issue candidate because it captures the common underlying v5 custom-model init failure more cleanly than the more specific notebook variant and is representative of the closest duplicate-like subgroup.", + "best_issue_reason": "issue:43646 is the best cluster anchor because it is the most general, clearly stated v5 regression and likely subsumes related custom-model initialization reports better than the narrower variants.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44560", - "right": "issue:44918", + "left": "issue:41553", + "right": "issue:44488", "accept": false, - "reason": "Different failures: Qwen3-VL video StopIteration versus TRL SFT embedding unpacking." + "reason": "Both are load failures, but they concern different models and different reported failure modes; not enough evidence of the same code-path bug." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same underlying v5 custom-model initialization regression; the notebook case is just a specific manifestation." + "reason": "Same underlying custom model initialization regression in Transformers v5; one report is notebook-specific, the other is the general breakage." }, { "left": "issue:44683", "right": "issue:46515", "accept": false, - "reason": "Unrelated torch-version incompatibilities affecting different code paths." + "reason": "Different torch version compatibility problems in different contexts; no indication they are the same fixable bug." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to Qwen3MoE FP8 loading." + "reason": "Completely different failures: an import error versus a Qwen3Moe FP8 loading problem." }, { "left": "issue:44910", "right": "issue:46858", "accept": false, - "reason": "Flash-attn packed-sequence misinterpretation is a different bug from static-cache slice arithmetic on non-CUDA backends." + "reason": "Both involve attention/cache internals, but the failure mechanisms are different: illegal memory access from misread position_ids versus slice arithmetic on a device tensor." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Different models and kernels: torch.compile+SDPA versus flash-attention on Qwen3.5." + "reason": "Different models, different attention implementations, and different crash modes; too broad to be the same bug." }, { "left": "issue:45823", "right": "issue:46515", "accept": false, - "reason": "Gemma4 device-mapping mismatch is not the same as a torch 2.4.1 incompatibility." + "reason": "A Gemma4 device-mismatch issue is unrelated to a generic torch version incompatibility report." }, { "left": "issue:42947", "right": "issue:47254", "accept": false, - "reason": "Both involve gradient checkpointing, but one is ineffective LoRA checkpointing and the other is a ZeRO-3 CheckpointError." + "reason": "Both mention PEFT/gradient checkpointing, but one is a Transformers v5 regression and the other is a PEFT + DeepSpeed ZeRO-3 checkpoint error; not clearly the same issue." }, { "left": "issue:45812", "right": "issue:46489", "accept": false, - "reason": "Different tokenizer regressions affecting different model families." + "reason": "Both are tokenizer regressions, but one is a silent Granite-wide token-id bug while the other is a DeepSeek-Coder tokenizer gap; different model families and likely different fixes." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer-mapping failure is unrelated to Gemma4 PLE device mismatch." + "reason": "Tokenizer-mapping edge case and Gemma4 device placement failure are unrelated." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Different bugs: video embedding StopIteration versus mask/tensor shape mismatch." + "reason": "Both are runtime errors, but the model/task and failure mechanics are different; not the same underlying bug." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Both mention config validation, but the model targets and failure details differ too much to treat as the same bug." + "reason": "Both mention config validation, but they affect different model families and are not clearly the same validation regression." }, { "left": "issue:38175", "right": "issue:43994", "accept": true, - "reason": "Same SigLIP2-base-patch16-224 output regression; the second report is the broader user-facing manifestation." + "reason": "Same specific SigLIP2 checkpoint and same broken inference behavior; one reports zero probabilities, the other nonsensical outputs through AutoModel/pipeline." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "Zero3/Bert loading failure is unrelated to the TOKENIZER_MAPPING_NAMES add-new-model-like bug." + "reason": "A ZeRO-3/BertModel index error is unrelated to the tokenizer-mapping failure in add-new-model-like." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "Different failure modes and code paths: Zero3 Bert init versus tokenizer-mapping logic." + "reason": "Different subsystems and symptoms: DeepSpeed zero3 initialization versus tokenizer mapping logic." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "No common underlying bug; one is model-addition logic, the other is a tensor mask shape error." + "reason": "Tokenizer generation tooling and a tensor mask shape error are not the same bug." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "LoRA checkpoint portability across versions is unrelated to BitNet packed-weight unpacking during accelerate loading." + "reason": "Both involve loading/compatibility, but one is a qwen3_moe LoRA version mismatch and the other is a BitNet packed-weights accelerate-loading bug." }, { "left": "issue:45672", "right": "issue:46899", - "accept": true, - "reason": "Same Gemma4 4-bit multimodal dtype-cast problem; one report is audio-specific, the other spans audio and vision." + "accept": false, + "reason": "Related Gemma4 multimodal quantization area, but one is specifically an audio uint8 torch.finfo TypeError and the other is unsafe casting in multimodal embedders." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "_is_hf_initialized regression and Kosmos2.5 long-OCR index error are distinct issues." + "reason": "Both are v5 regressions, but the issues are unrelated: hf initialization flag breakage versus long OCR index errors." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "device_map/offload loading failure is not the same as Strix Halo mmap OOM." + "reason": "Device-map dequantized loading and Strix Halo mmap OOM are different loading problems with different fixes." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Qwen3 auto-device cuda assert and Qwen3MoE FineGrainedFP8Config failure are different problems." + "reason": "A cudaErrorAssert during auto device mapping is unrelated to a qwen3_moe FineGrainedFP8Config failure." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Same model family, but the doc-example mistakes are not the same regression as the zero-probability bug." + "reason": "Same model family, but one is a direct inference regression and the other is a documentation example with multiple distinct errors; not the same underlying bug." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "Completely different models and failure mechanisms." + "reason": "Different models and different failure classes; no shared code-path evidence." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Near-exact duplicate: both report Transformers v5 filling non-persistent buffers with junk." + "reason": "These are the same Transformers v5 non-persistent buffer regression stated with nearly identical wording." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "LoRA checkpoint incompatibility is unrelated to Gemma4 resize_token_embeddings behavior." + "reason": "LoRA version incompatibility and gemma4 resize_token_embeddings behavior are different issues." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Different models and symptoms: GLM5 versus Qwen3.5 flash-attention generation failure." + "reason": "Different models and different flash-attention failures; similar wording is not enough to merge them." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are load failures, but for different models with no clear shared concrete bug." + "reason": "Both fail to load specific BERT checkpoints, but the titles do not show the same checkpoint family or the same root cause, so this is too weak to merge." }, { - "left": "issue:45313", - "right": "issue:45357", + "left": "issue:45507", + "right": "issue:46468", "accept": false, - "reason": "Loading ZeRO-3 language_model weights and saving wrong visual encoder keys are different code paths." + "reason": "Invalid method calls in GraniteMoEHybrid and a broken official Nemotron example are unrelated." }, { "left": "issue:44863", "right": "issue:46846", "accept": false, - "reason": "NemotronH checkpoint loading and TP-plan linear_attn issues are unrelated." + "reason": "NemotronH checkpoint loading and Qwen3.5 tensor-parallel plan issues are different bugs." }, { - "left": "issue:45507", - "right": "issue:46468", + "left": "issue:45313", + "right": "issue:45357", "accept": false, - "reason": "Different Nemotron/Granite model issues with no shared concrete failure." + "reason": "Both are Qwen3.5 regressions, but one is ZeRO-3 weight loading and the other is incorrect visual encoder key saving; different code paths and fixes." } ] }, "evaluator_result": { "accept": true, - "feedback": "The canonical anchor choice and the cluster summary are grounded in the packet: 43646 is the broad custom-model v5 regression, and the listed true-duplicate/near-duplicate pairs are conservatively supported by the titles. The rejected pairs are mostly justified as distinct bugs rather than same-root-cause matches." + "feedback": "Grounded overall and conservative on the soft edges. The anchor choice is reasonable, the duplicate calls are supported by near-identical titles, and the rejections avoid over-merging on weak similarity." }, "evaluator_used": true, "retried": false, @@ -2573,8 +2594,8 @@ "issue:45663|issue:45759", "issue:39692|issue:44960", "issue:44910|issue:45588", - "issue:43404|issue:43454", "issue:46897|issue:46901", + "issue:43404|issue:43454", "issue:43873|issue:44387", "issue:42947|issue:45704", "issue:45663|issue:45859", @@ -2582,220 +2603,220 @@ "issue:44079|issue:46829", "issue:43957|issue:44756" ], - "prepared_review_unit_hash": "8190eeb422b49b83802c004722eb8cebd9c04b5bc6e2de9fa169ddad3f1c2da8", + "prepared_review_unit_hash": "4dabee40f272c1dcac34090a500c70287f69dc79c40806d6e47890ac3c19ea58", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37275, - "estimated_input_tokens": 9319, - "estimated_eval_tokens": 18894 + "serialized_chars": 37282, + "estimated_input_tokens": 9321, + "estimated_eval_tokens": 18898 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: it mixes unrelated model-loading, tokenizer, quantization/offload, training, and config bugs. I found no pair that clearly shares the same underlying code-path bug closely enough to merge as duplicates.", - "confidence": 0.84, - "canonical_issue_reason": "issue:43873 is the broadest, highest-activity issue here and best represents the recurring quantization/offloading regression theme, even though the overall cluster is not a clean duplicate set.", + "summary": "The cluster is thematically related around Transformers 5.x regressions, but the items mostly describe distinct model-specific bugs and API issues rather than one duplicate chain. I did not find any soft-edge pair that is clearly the same concrete bug/change.", + "confidence": 0.67, + "canonical_issue_reason": "issue:46459 is the best single representative: it is a concrete, high-impact composite-model loading regression with a clear repro and broad downstream impact, whereas most other issues are narrower model-specific failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 has the strongest signal for the cluster\u2019s main runtime/memory-regression theme, with substantial discussion and inbound references.", + "best_issue_reason": "issue:46459 is the strongest global candidate because it captures the cluster\u2019s main loading/dtype regression theme and has broader applicability than the more model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46459", "right": "issue:46512", "accept": false, - "reason": "Same general area (dtype during load/from_config), but one is about composite-checkpoint loading ignoring dtype and the other about stale config metadata after from_config." + "reason": "Both involve dtype handling during model construction/loading, but one is about ignored load dtype on composite checkpoints and the other about stale config.dtype after from_config; related area, different bug." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Both are loading failures, but they affect different models and different failure modes: one is cjvt/sleng-bert loading, the other is tokenizer loading for EMBEDDIA/est-roberta." + "reason": "Both are tokenizer/model load regressions on transformers>=5, but they affect different models and appear to have different failure modes." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Both involve SigLIP2-like usage, but one is a docs/example model-processor mismatch with quantization failure while the other is bad inference outputs via AutoModel/pipeline." + "reason": "Both mention SigLIP2 bad outputs, but one is a docs/example mismatch plus quantization failure while the other is an AutoModel/pipeline result issue; not clearly the same code path." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "One is about non-persistent buffers being junk-filled; the other is about custom models defined in Jupyter notebooks. Different bugs and code paths." + "reason": "One is about non-persistent buffers being junk-filled, the other about custom-model initialization in Jupyter; same release, different bugs." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Different models and different problems: Qwen3 MoE conversion under accelerate+deepspeed versus GLM-4.7-Flash loading in tests." + "reason": "Different models and symptoms: Qwen3 MoE conversion with accelerate/deepspeed versus GLM-4.7 load failures in unit tests." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both mention memory, but one is inefficient Qwen3 MoE training memory usage and the other is offloading behavior under quantization; not the same underlying bug." + "reason": "One is Qwen3 MoE training memory usage, the other is quantization/offloading behavior; broad subsystem overlap only." }, { "left": "issue:42491", "right": "issue:45335", "accept": false, - "reason": "The LoRA/qwen3_moe hf4.x compatibility issue is unrelated to T5Gemma resize_token_embeddings not updating decoder embeddings." + "reason": "LoRA checkpoint compatibility on hf5.x versus resize_token_embeddings not updating decoder embeddings; unrelated concrete failures." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "Different models and mechanisms: LoRA checkpoint incompatibility versus resize_token_embeddings not propagating to output embeddings." + "reason": "Different model-loading/compatibility issue versus a resize_token_embeddings propagation bug." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Both are model-loading issues, but they concern different model families and distinct failure signatures." + "reason": "GLM-4.7 unit-test load failure and Qwen3-VL shape mismatch are separate model-specific loading bugs." }, { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Training checkpoint-resume max_steps accounting is unrelated to optimizer argument handling for SGD." + "reason": "Trainer checkpoint resumption math and optimizer argument handling are unrelated." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both are memory-related, but one is Qwen3 MoE training memory inefficiency and the other is int4 quantization CUDA reserved memory growth/OOM." + "reason": "Training memory efficiency and int4 quantization reserved-memory OOM are different resource-management bugs." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "Serialization extra_state handling and init_empty_weights constructor-argument compatibility are separate code paths." + "reason": "save/from_pretrained extra_state handling and init_empty_weights argument compatibility are distinct serialization/loading regressions." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one is LlamaTokenizer overriding ByteLevel for DeepSeek and the other is AutoTokenizer producing wrong IDs for Granite models." + "reason": "Different tokenizer regressions affecting different model families; no shared concrete code path is evident." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "Different models and workflows: GLM5 load behavior versus sinq quantized save/load breakage." + "reason": "GLM5 loading and sinq save/load breakage are unrelated issues." }, { "left": "issue:45276", "right": "issue:45335", "accept": false, - "reason": "Similar symptom pattern, but these are different architectures and different embedding modules, so not the same concrete bug." + "reason": "Both are resize_token_embeddings propagation bugs, but they target different architectures and different nested embedding fields." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "A chat crash on Qwen3.5-35B is not the same as Gemma4Processor missing _tokenizer in transformers serve." + "reason": "Qwen3.5 chat crash and Gemma4Processor missing _tokenizer are different runtime failures." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Torch 2.10 compatibility and meta-device loading failures are different regressions with different triggers." + "reason": "Torch 2.10 compatibility and meta-device loading failures are separate regressions." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "Unexpected _is_hf_initialized support and disabling mmap on Strix Halo to avoid OOM are unrelated." + "reason": "Unexpected init_empty_weights argument and mmap-induced OOM are unrelated." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "One is a tokenizer vocab-size mismatch for a specific model; the other is a separate model-loading failure." + "reason": "Different tokenizer/vocab mismatch reports for different models; not the same underlying bug." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Qwen3 auto device mapping cudaErrorAssert and AudioFlamingo3 batched inference leakage are unrelated." + "reason": "CUDA assert on Qwen3 auto device mapping and AudioFlamingo3 batched inference leak are unrelated." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states and Qwen multimodal text_config unwrapping are different model/config bugs." + "reason": "Gemma4 FSDP2 attention state rebuild and Qwen 3.5/3.6 text_config unwrapping are different model-loading/training issues." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "SigLIP2 documentation/quantization issues and GLM5 model loading are unrelated." + "reason": "SigLIP2 docs/example failure and GLM5 loading are different models and different bug classes." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Both hit flash-attention paths, but the root causes differ: position_ids misinterpretation versus s_aux=None handling for sink-less models." + "reason": "Flash-attention illegal memory access from packed-sequence misinterpretation versus AttributeError on s_aux=None are different failures in different code paths." }, { - "left": "issue:43404", - "right": "issue:43454", + "left": "issue:46897", + "right": "issue:46901", "accept": false, - "reason": "Both involve untied lm_head weights, but they are separate model implementations and likely separate fixes." + "reason": "Both are double-shift training-loss bugs, but they affect different model families and there is no evidence of a shared implementation bug." }, { - "left": "issue:46897", - "right": "issue:46901", + "left": "issue:43404", + "right": "issue:43454", "accept": false, - "reason": "They share a double-shift pattern, but they affect different models and are not clearly the same code-path bug." + "reason": "Both mention untied lm_head weights, but they are separate multimodal model classes with no clear shared code path from the titles alone." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Quantization/offload and quantization-reserved-memory OOM are related thematically, but the reported failures point to different mechanisms." + "reason": "Quantization/offloading behavior and int4 reserved-memory OOM are related only at a high level, not the same bug." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing with PEFT LoRA is unrelated to apex.FusedRMSNorm memory leak in T5." + "reason": "Gradient checkpointing ineffectiveness in PEFT LoRA and a T5 RMSNorm memory leak are unrelated." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "FSDP2 shared_kv_states KeyError and CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding are different bugs." + "reason": "Different Gemma4/FSDP2 state bug versus Qwen3_5Moe rotary embedding CPU-offload incompatibility." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "AudioFlamingo3 token/embedding leakage and Gemma4 PLE device mismatch under device_map='auto' are distinct issues." + "reason": "AudioFlamingo3 batched inference leak and Gemma4 PLE device mismatch are distinct runtime issues." }, { "left": "issue:44079", "right": "issue:46829", "accept": false, - "reason": "ModelOutput key assignment logic and VideoPrism hidden_states output wiring are different result-shaping bugs." + "reason": "ModelOutput key assignment bug and VideoPrism hidden_states mispopulation are different output-structure problems." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading regressions and Strix Halo mmap OOM are different runtime problems." + "reason": "Meta-device loading failures and Strix Halo mmap OOM are different loading/memory issues." } ] }, @@ -2871,8 +2892,8 @@ "soft_pairs": [ "issue:43335|issue:45663", "issue:44863|issue:46468", - "issue:36010|issue:39404", "issue:44479|issue:44877", + "issue:36010|issue:39404", "issue:44387|issue:44756", "issue:45276|issue:45292", "issue:44292|issue:44912", @@ -2884,9 +2905,9 @@ "issue:43645|issue:44164", "issue:43425|issue:47228", "issue:43854|issue:45216", - "issue:44589|issue:44756", "issue:45684|issue:45698", "issue:45292|issue:45335", + "issue:44589|issue:44756", "issue:43425|issue:43828", "issue:44402|issue:44451", "issue:43646|issue:43708", @@ -2894,25 +2915,25 @@ "issue:46894|issue:46901", "issue:44683|issue:45823", "issue:43425|issue:43716", - "issue:46315|issue:46491", "issue:45812|issue:46315", + "issue:46315|issue:46491", "issue:44530|issue:45084", "issue:39404|issue:45701" ], - "prepared_review_unit_hash": "a2595a034b53d7a6bd8f2076c851f099ae3b0c0b6c1b4f179a3d3b553a10765a", + "prepared_review_unit_hash": "7c49db93f71302c5f6affd5b78db9f1c7773e1034ac35524bcb4e88e5580d069", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36279, + "serialized_chars": 36280, "estimated_input_tokens": 9070, "estimated_eval_tokens": 18396 }, @@ -2921,186 +2942,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated Transformers regression reports. The only clear duplicate-like groups are the resize_token_embeddings propagation bug and the training-loss double-shift bug; the rest differ by model, subsystem, or failure mode.", - "confidence": 0.79, - "canonical_issue_reason": "issue:45292 is the best canonical issue for the resize_token_embeddings bug because it states the underlying propagation failure in the broadest, most model-agnostic way; 45276 and 45335 look like narrower model-specific manifestations of the same problem.", + "summary": "Most pairs are unrelated despite superficial overlap. The only clear duplicate families here are the `resize_token_embeddings` regression reports and the training-loss double-shift reports; the `resize_token_embeddings` issue is the best canonical anchor.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45292 is the cleanest general statement of the shared `resize_token_embeddings` bug, and it matches the more model-specific reports in issue:45276 and issue:45335.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 is the strongest representative overall: it is concrete, broadly applicable, and has the clearest duplicate-like matches in the set.", + "best_issue_reason": "It is the most model-agnostic, reusable duplicate target among the issues shown, with clear manifestations across multiple model variants.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43335", "right": "issue:45663", "accept": false, - "reason": "Different bugs: one is a config construction issue in SwitchTransformers, the other is an FSDP2 attention state KeyError in Gemma4 training." + "reason": "Different bugs: sparse-layer construction vs Gemma4 FSDP2 shared_kv_states handling." }, { "left": "issue:44863", "right": "issue:46468", "accept": false, - "reason": "Both involve Nemotron, but one is checkpoint loading and the other is a broken example; different code paths and symptoms." + "reason": "Both are Nemotron-related, but one is checkpoint loading and the other is an example breakage; not the same concrete bug." }, { - "left": "issue:36010", - "right": "issue:39404", + "left": "issue:44479", + "right": "issue:44877", "accept": false, - "reason": "Unrelated failures: an import error for GenerationMixin versus a Whisper pipeline return_language regression." + "reason": "Different regressions in different model families and code paths." }, { - "left": "issue:44479", - "right": "issue:44877", + "left": "issue:36010", + "right": "issue:39404", "accept": false, - "reason": "Different layers of the stack: video-input regression in Qwen models versus strict config loading for granite_speech." + "reason": "ImportError for GenerationMixin is unrelated to Whisper pipeline `return_language` behavior." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Both mention OOM, but one is int4 quantization CUDA memory growth and the other is mmap on Strix Halo; not the same bug." + "reason": "Both mention OOM, but the causes differ: int4 quantization memory growth vs mmap behavior on Strix Halo." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same underlying resize_token_embeddings propagation bug; 45276 is a model-specific variant of 45292's broader embeddings/output_embeddings failure." + "reason": "Same `resize_token_embeddings` propagation bug, with 45276 being a model-specific manifestation." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization/loading problems affecting different models and formats (NVFP4 vs MXFP4 fallback)." + "reason": "Different quantization/loading problems with different symptoms and likely different fixes." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Different issues: gradient checkpointing in eval mode versus a Qwen3.5 RLHF instability from 3D position_ids and SDPA fallback." + "reason": "Gradient checkpointing eval-mode issue is unrelated to RLHF NaN from 3D position_ids / SDPA fallback." }, { "left": "issue:43957", "right": "issue:47332", "accept": false, - "reason": "Both are model-loading related, but one is meta-device loading and the other is saving a disk-offloaded model; distinct code paths." + "reason": "Different model-loading/offloading failures on different models and paths." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "Saving/from_pretrained extra_state handling is unrelated to the Strix Halo mmap/OOM issue." + "reason": "save/from_pretrained extra_state handling is unrelated to mmap-triggered OOM." }, { "left": "issue:42796", "right": "issue:46326", "accept": false, - "reason": "Both are tokenizer-related, but one is BPE marker handling and the other is continue_final_message logic for mistral_common tokenizers." + "reason": "Tokenizer BPE marker handling and `continue_final_message` are distinct tokenizer behaviors." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Different subsystems: flash_attention sink handling versus Gemma4 shared_kv_states/FSDP2 behavior." + "reason": "Flash-attention `s_aux=None` crash is a different issue from Gemma4 FSDP2 `shared_kv_states` rebuilding." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Custom model initialization in notebooks is unrelated to extra_state save/from_pretrained handling." + "reason": "Custom-model notebook init regression is unrelated to `extra_state` save/from_pretrained handling." }, { "left": "issue:43425", "right": "issue:47228", "accept": false, - "reason": "Torch version compatibility and a specific bf16 compilation regression are different problems." + "reason": "Torch compatibility breakage and Sam3 bf16 compilation regression are different code paths." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "Both involve model loading/saving, but they affect different models and different failure mechanisms." - }, - { - "left": "issue:44589", - "right": "issue:44756", - "accept": false, - "reason": "A missing Float8 storage type error is not the same as mmap-related OOM on Strix Halo." + "reason": "Unit-test model loading failure and Qwen3.5 save_pretrained regression are different problems." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Both concern save_pretrained/custom-model packaging, but one is read-only permissions propagation and the other is loading the wrong custom module." + "reason": "Permissions propagation from source files is not the same as loading the wrong custom module after save_pretrained." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Same resize_token_embeddings regression: 45292 covers the output embeddings path and 45335 is the decoder.embed_tokens manifestation." + "reason": "Both report the same `resize_token_embeddings` bug affecting model embeddings/output embeddings." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Storage-object lookup failure is unrelated to mmap/OOM behavior." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch 2.10 incompatibility is a broad environment issue, while 43828 is a specific autocast dtype mismatch in Phi-tiny-MoE." + "reason": "Torch version incompatibility is not the same as the autocast dtype-mismatch bug." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "The ScandiBERT load failure may be related to tokenizer vocab issues, but the reports do not show the same concrete bug path." + "reason": "Tokenizer vocab mismatch and model load failure are different issues." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization breakage is unrelated to the Trainer max_steps resume calculation bug." + "reason": "Custom model initialization regression is unrelated to Trainer `max_steps` miscalculation on checkpoint resume." }, { "left": "issue:43425", "right": "issue:46869", "accept": false, - "reason": "Torch incompatibility and YOLO multi-GPU device_map failures are too different to merge." + "reason": "Torch incompatibility and YOLO multi-GPU device-map failure are different concrete bugs." }, { "left": "issue:46894", "right": "issue:46901", "accept": true, - "reason": "Both describe the same training-loss double-shift bug: the model is training against shifted labels instead of the intended labels." + "reason": "Same training-loss double-shift pattern: both describe labels being shifted incorrectly during loss computation." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 device mismatch with device_map=\"auto\" are different failures." + "reason": "Compiled flex_attention regression is unrelated to Gemma4 PLE device mismatch." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "Torch 2.10 incompatibility is not the same as the Mistral-3 image preprocessor/model dtype mismatch." + "reason": "Torch version incompatibility is different from the image-preprocessor/model dtype mismatch." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Wrong token IDs and slower `convert_tokens_to_ids` are separate tokenizer regressions." }, { "left": "issue:46315", "right": "issue:46491", "accept": false, - "reason": "One is a performance regression in convert_tokens_to_ids, the other is a CodeLlama leading-space round-trip correctness bug." + "reason": "Performance regression and CodeLlama round-trip space loss are unrelated tokenizer issues." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache group-type crash is unrelated to the non-template-node compile error." + "reason": "PagedAttentionCache group-type crash is unrelated to non-template-node compilation failure." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper pipeline return_language regression and a general tokenization-version-change report do not describe the same bug." + "reason": "Whisper pipeline regression is unrelated to tokenizer-version-dependent tokenization changes." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays mostly conservative. The accepted soft edges are limited to two plausible duplicate-like clusters: the resize_token_embeddings trio and the training-loss double-shift pair. The rest of the pairwise rejections are consistent with the issue titles and do not overstate evidence. No PR-mergeability concern applies here because these are issue pairs, not pull requests." + "feedback": "The summary is grounded in the packet and stays conservative. The canonical `resize_token_embeddings` anchor at issue:45292 is well-supported by the closely related model-specific reports in issue:45276 and issue:45335, and the training-loss double-shift pair is also a reasonable duplicate callout. The unrelated pairs are correctly rejected rather than over-merged." }, "evaluator_used": true, "retried": false, @@ -3175,8 +3202,8 @@ "issue:44479|issue:44805", "issue:44589|issue:45084", "issue:39404|issue:46665", - "issue:36032|issue:43452", "issue:44164|issue:45759", + "issue:36032|issue:43452", "issue:43994|issue:44079", "issue:40990|issue:47331", "issue:44265|issue:44898", @@ -3189,8 +3216,8 @@ "issue:39404|issue:44912", "issue:45823|issue:46899", "issue:43596|issue:44683", - "issue:44756|issue:47333", "issue:44164|issue:46821", + "issue:44756|issue:47333", "issue:44488|issue:44987", "issue:43856|issue:44756", "issue:44488|issue:45081", @@ -3198,218 +3225,218 @@ "issue:43632|issue:47050", "issue:43716|issue:45084", "issue:43746|issue:47331", - "issue:45479|issue:46894", "issue:42796|issue:43404", + "issue:45479|issue:46894", "issue:43611|issue:43644" ], - "prepared_review_unit_hash": "71e8bb874e5b4b75b2e48252f98b8500b05405fab939ab6bf74462bdd2d92c43", + "prepared_review_unit_hash": "9b55eded07e48c47e6e78c96350750869e3c36bb2d3ca4c9fdab9a8d3966fdf8", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 37024, - "estimated_input_tokens": 9256, - "estimated_eval_tokens": 18768 + "serialized_chars": 37016, + "estimated_input_tokens": 9254, + "estimated_eval_tokens": 18764 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are clearly different bugs despite superficial overlap around v5 regressions and loading/tokenizer issues. No soft edge looks like a true duplicate. Issue 43611 is the best canonical representative because it is the broadest model-loading regression in the set.", - "confidence": 0.88, - "canonical_issue_reason": "issue:43611 is the broadest, most central v5 regression here: it describes a general model-loading breakage tied to base_model_prefix, which is more representative than the narrower tokenizer, multimodal, or quantization reports.", + "summary": "This cluster is mostly heterogeneous. The only clear thematic thread is a loose set of Transformers v5 loading/initialization regressions, but none of the soft pairs look like the same concrete bug or change, so all soft edges should be rejected.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43611 is the broadest and most central of the v5 regression reports here: it targets model loading breakage in Transformers 5.0.0 and best represents the shared initialization/loading theme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the best issue anchor for this cluster because it captures a general Transformers v5 loading regression and can serve as the umbrella reference for nearby model-init/load reports.", + "best_issue_reason": "issue:43611 is the best issue to anchor the cluster because it is a general, high-signal v5 model-loading regression; the other candidates are either narrower model-specific failures, tokenizer bugs, docs issues, or unrelated runtime errors.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "Both mention v5 breakage during model setup, but one is base_model_prefix loading and the other is Jupyter custom-model initialization; different failure modes and code paths." + "reason": "Both are v5 breakage reports, but one is about base_model_prefix loading and the other about custom model initialization in Jupyter; different failure modes and likely different fixes." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "One is a Qwen video-input regression; the other is a mask/tensor shape IndexError. No shared underlying bug." + "reason": "Video-input regression on Qwen VL models is unrelated to a generic mask-shape IndexError; no evidence of the same code path." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage lookup failure and template-node compile error are unrelated issues." + "reason": "One is a float8 storage lookup error, the other is a template-node compilation error; different subsystems and root causes." }, { "left": "issue:39404", "right": "issue:46665", "accept": false, - "reason": "Whisper pipeline return_language regression is unrelated to from_pretrained behavior inside Python classes." + "reason": "Whisper pipeline return_language behavior and from_pretrained-inside-classes inconsistency are separate issues." }, { - "left": "issue:36032", - "right": "issue:43452", + "left": "issue:44164", + "right": "issue:45759", "accept": false, - "reason": "Both are loading failures, but one is a T5Tokenizer method conflict and the other is gguf_file handling for AutoTokenizer/AutoModel; distinct causes." + "reason": "extra_state save/load handling and composite Qwen text_config unwrapping are different loading problems." }, { - "left": "issue:44164", - "right": "issue:45759", + "left": "issue:36032", + "right": "issue:43452", "accept": false, - "reason": "save/from_pretrained extra_state handling is a serialization bug; text_config unwrapping is a config-routing bug for composite multimodal models." + "reason": "T5 tokenizer add_special_tokens conflict and gguf_file handling for AutoTokenizer/AutoModel are distinct loader failures." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "A SigLIP2 bad-output report and a ModelOutput key assignment bug do not describe the same defect." + "reason": "Incorrect SigLIP2 outputs and ModelOutput key assignment bugs are different behaviors with different likely code paths." }, { "left": "issue:40990", "right": "issue:47331", "accept": false, - "reason": "High perplexity on gpt-oss-20b is a modeling/eval complaint; the MetalConfig report is a concrete quantized batched-generate correctness bug." + "reason": "High perplexity on gpt-oss-20b and Metal quantization batch corruption are unrelated problems." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export failure via torch_compilable_check is unrelated to Perceiver interpolation behavior on non-default resolutions." + "reason": "torch.export failures from torch_compilable_check are not the same as Perceiver non-default-resolution interpolation failures." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Both are docs issues, but they target different API removals and should not be merged." + "reason": "Both are docs issues, but they concern different pipeline/task documentation changes and not the same underlying bug." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "est-roberta tokenizer loading and Mistral regex patch crashes are separate tokenizer-specific bugs." + "reason": "Different tokenizer-loading failures on different models; not the same concrete bug." }, { "left": "issue:46050", "right": "issue:46326", "accept": false, - "reason": "Sinq save/load breakage and mistral_common continue_final_message behavior are unrelated." + "reason": "Quantized save/load breakage via sinq is unrelated to continue_final_message on mistral_common tokenizers." }, { "left": "issue:43906", "right": "issue:46326", "accept": false, - "reason": "An isolated reproduction of another issue and a mistral_common tokenizer bug are not the same concrete problem." + "reason": "An isolated reproduction of another issue and a tokenizer-specific continue_final_message bug are not the same bug." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "CamemBERT prediction regression and est-roberta tokenizer loading failure affect different model classes and paths." + "reason": "CamemBERT masked-LM prediction regression and est-roberta tokenizer loading failure are different issues." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "The reproduced issue 38071 and Qwen3.5 PagedAttentionCache crash are unrelated." + "reason": "Generic reproduction report vs Qwen3.5 PagedAttentionCache crash; different code paths." }, { "left": "issue:39404", "right": "issue:44912", "accept": false, - "reason": "Whisper return_language pipeline behavior and gpt-oss MXFP4 quantization fallback are different regressions." + "reason": "Whisper pipeline return_language and MXFP4 quantization fallback are unrelated." }, { "left": "issue:45823", "right": "issue:46899", "accept": false, - "reason": "Both are Gemma4 multimodal bugs, but one is a device-map mismatch and the other is a 4-bit dtype-cast issue; distinct defects." + "reason": "Both mention Gemma4, but one is a device-map forward mismatch and the other is a 4-bit dtype-cast bug in multimodal embedders; different fixes." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "Zero3/BertModel init failure and compiled flex_attention torch>=2.9 failure are unrelated." + "reason": "Deepspeed ZeRO-3/BertModel initialization and compiled flex_attention on torch>=2.9 are unrelated." }, { - "left": "issue:44756", - "right": "issue:47333", + "left": "issue:44164", + "right": "issue:46821", "accept": false, - "reason": "Strix Halo mmap OOM and disk-offloaded Qwen3-30B save failure are different memory/offloading problems." + "reason": "extra_state save/load handling and TimesFM window_size AttributeError are different bugs." }, { - "left": "issue:44164", - "right": "issue:46821", + "left": "issue:44756", + "right": "issue:47333", "accept": false, - "reason": "Extra_state serialization and TimesFM window_size attribute access are unrelated." + "reason": "Strix Halo mmap OOM and disk-offloaded Qwen3-30B-A3B save failure are distinct storage/offloading problems." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Both are loading complaints, but they concern different models and failure mechanisms." + "reason": "Two different model-loading failures; same broad area but not the same underlying bug." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE training memory usage is not the same bug as disabling mmap to avoid Strix Halo OOM." + "reason": "MoE training memory inefficiency and Strix Halo mmap OOM are separate memory issues with different causes." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "A generic model load failure and a Mistral regex patch crash are not the same underlying issue." + "reason": "Different tokenizer-loading failures in different model families; no shared concrete bug." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Tokenizer method conflict and Qwen3MoE FineGrainedFP8Config failure are unrelated." + "reason": "T5 tokenizer init conflict and Qwen3Moe fine-grained FP8 failure are unrelated." }, { "left": "issue:43632", "right": "issue:47050", "accept": false, - "reason": "Non-persistent buffer junk in v5 is unrelated to CanineModel fp16 ONNX export dtype mismatch." + "reason": "_is_hf_initialized flag breakage and Canine fp16 ONNX attention dtype mismatch are different problems." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Mistral-3 image-preprocessor dtype mismatch and template-node compile failure are different bugs." + "reason": "Mistral-3 image-preprocessor dtype mismatch and template-node compilation failure are unrelated." }, { "left": "issue:43746", "right": "issue:47331", "accept": false, - "reason": "GraniteSpeech PEFT checkpoint loading and MetalConfig batched quantized generation corruption are unrelated." + "reason": "PEFT local checkpoint loading for GraniteSpeech and Metal quantization batch corruption are unrelated." }, { - "left": "issue:45479", - "right": "issue:46894", + "left": "issue:42796", + "right": "issue:43404", "accept": false, - "reason": "A classification-loss edge case and a CohereASR training double-shift bug are different training bugs." + "reason": "Tokenizer BPE-marker handling and lm_head weight tying in Mistral3ForConditionalGeneration are different bugs, even if both involve Mistral-family models." }, { - "left": "issue:42796", - "right": "issue:43404", + "left": "issue:45479", + "right": "issue:46894", "accept": false, - "reason": "Ministral tokenizer BPE-marker handling and Mistral3 lm_head weight tying are separate issues." + "reason": "Zero-loss classification bug and ASR double-shift training-loss bug are different loss-shaping issues." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Both are v5 regressions, but one concerns model-loading with base_model_prefix and the other non-persistent buffer initialization; different root causes." + "reason": "Both are v5 regressions, but one is about loading with base_model_prefix and the other about non-persistent buffer initialization; not the same defect." } ] }, @@ -3483,12 +3510,12 @@ "issue:47331" ], "soft_pairs": [ - "issue:43299|issue:43931", "issue:44462|issue:46489", "issue:44661|issue:44683", + "issue:43299|issue:43931", "issue:46285|issue:46402", - "issue:45479|issue:46897", "issue:43541|issue:45084", + "issue:45479|issue:46897", "issue:44451|issue:44987", "issue:42915|issue:44530", "issue:43873|issue:47331", @@ -3497,8 +3524,8 @@ "issue:36032|issue:43645", "issue:40444|issue:46326", "issue:34567|issue:40444", - "issue:39404|issue:44671", "issue:44737|issue:45859", + "issue:39404|issue:44671", "issue:43644|issue:43646", "issue:42175|issue:46178", "issue:44987|issue:44991", @@ -3508,24 +3535,24 @@ "issue:43452|issue:43482", "issue:43482|issue:47312", "issue:43746|issue:46326", - "issue:43582|issue:47331", "issue:44811|issue:45584", + "issue:43582|issue:47331", "issue:44403|issue:45081" ], - "prepared_review_unit_hash": "d49310fec7e2c37bc23d9a9183ab7252acf98ba31bcaefa533fef1c28a96de2a", + "prepared_review_unit_hash": "3b379ffe87d35bbd0c62fc83ef513cd8fdabe2f9f01f34609e614f36bc15e36b", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36121, + "serialized_chars": 36123, "estimated_input_tokens": 9031, "estimated_eval_tokens": 18318 }, @@ -3534,184 +3561,181 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated regressions across different models/features. The only plausible duplicate pair is the GGUF loading breakage, where the specific Qwen2.5-GGUF report looks like a manifestation of the broader gguf_file/from_pretrained regression.", - "confidence": 0.81, - "canonical_issue_reason": "issue:43452 is the broadest and most representative report: it describes the generic gguf_file/from_pretrained breakage affecting both AutoTokenizer and AutoModelForCausalLM, which best captures the underlying GGUF loading regression.", + "summary": "This cluster is not a duplicate set in the strict sense; it mixes many separate Transformers regressions across tokenizers, model loading, quantization, audio, and training behavior. I found no soft edge that clearly represents the same underlying bug/change.", + "confidence": 0.88, + "canonical_issue_reason": "No single canonical issue: the items cover many distinct failures and code paths (tokenizer loading, GGUF, model init, quantization, Whisper/audio, trainer state, etc.), so collapsing them would over-merge unrelated bugs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43452 is the best overall representative because it generalizes the GGUF loading failure beyond a single model family and aligns with the only duplicate-like pair in the set.", + "best_issue_reason": "If one issue must be used as a representative anchor, `issue:43646` is the broadest user-facing Transformers v5 custom-model initialization regression, but it is still not a true canonical duplicate for the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43299", - "right": "issue:43931", - "accept": false, - "reason": "Both are Qwen3-VL-related load failures, but one is a MoE/dev-version regression and the other is a model weight-shape mismatch; different concrete bugs." - }, { "left": "issue:44462", "right": "issue:46489", "accept": false, - "reason": "Both concern tokenizers, but one is about AutoTokenizer ignoring tokenizer.json while the other is a DeepSeek-Coder-specific incorrect-output regression; not the same code-path bug." + "reason": "Both involve tokenizer behavior, but one is a repo/tokenizer.json loading regression and the other is a model-specific DeepSeek-Coder tokenizer output issue; not the same concrete bug." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Unrelated issues: tokenizer/model-mapping failure versus compiled flex_attention failing on torch>=2.9." + "reason": "Unrelated subsystems: add-new-model-like mapping failure vs compiled flex_attention on torch>=2.9." }, { - "left": "issue:46285", - "right": "issue:46402", + "left": "issue:43299", + "right": "issue:43931", "accept": false, - "reason": "Different model families and failures: CLIP structural inconsistency versus Sam3Model text-encoder weights not loading." + "reason": "Same general Qwen3-VL area, but one is a v5 dev loading regression and the other is a specific weight-shape mismatch for a concrete model; not clearly one bug." }, { - "left": "issue:45479", - "right": "issue:46897", + "left": "issue:46285", + "right": "issue:46402", "accept": false, - "reason": "Both are training-loss bugs, but they affect different models and stem from different mechanics; too broad to merge as one bug." + "reason": "Different model families and failures: CLIP text-model structure inconsistency vs Sam3Model text-encoder weights not loading." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Different failure modes and code paths: torch dynamo/grouped_mm tracing error versus template-node compilation error." + "reason": "Different failures and code paths: torch dynamo grouped_mm tracing vs template compilation error." + }, + { + "left": "issue:45479", + "right": "issue:46897", + "accept": false, + "reason": "Both are loss-related, but they are distinct model-specific training bugs with different mechanisms." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Both are model-loading regressions, but they concern different models and likely different loader issues; not the same underlying bug." + "reason": "Both are model loading regressions, but for different models and likely different causes." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "Qwen3Moe FP8 config failure is distinct from a PagedAttentionCache linear_attention crash on Qwen3.5." + "reason": "Both touch Qwen/MoE-ish areas, but one is FineGrainedFP8Config failure and the other is PagedAttentionCache linear_attention crash; not the same issue." }, { "left": "issue:43873", "right": "issue:47331", "accept": false, - "reason": "Both mention quantization/offloading, but one is a general offloading issue and the other is a specific Metal batched-generate corruption bug." + "reason": "Both mention quantization/offload, but one is offloading behavior and the other is a Metal batched-generate corruption bug in a different code path." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity regression on gpt-oss is unrelated to tokenizer post-processor updates for runtime special-token changes." + "reason": "Unrelated: perplexity regression on gpt-oss vs runtime special-token/post-processor update behavior." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "AutoConfig with trust_remote_code missing fields is unrelated to the PagedAttentionCache/Qwen3.5 crash." + "reason": "Different concerns: trust_remote_code missing fields vs cache crash on Qwen3.5 models." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "T5 tokenizer add_special_tokens conflict is a specific tokenizer-loading bug, while the other is a custom-model initialization issue in notebooks." + "reason": "Tokenizer method-name conflict is unrelated to custom model initialization in notebooks." }, { "left": "issue:40444", "right": "issue:46326", "accept": false, - "reason": "Different domains and bugs: multi-image IterableDataset finetuning versus mistral_common continue_final_message behavior." + "reason": "Different areas: multi-image IterableDataset fine-tuning failure vs mistral_common continue_final_message behavior." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "TrainerState token-count tracking is unrelated to Qwen2.5-VL multi-image dataset failures." + "reason": "TrainerState token counting bug is unrelated to Qwen2.5-VL multi-image IterableDataset failure." }, { - "left": "issue:39404", - "right": "issue:44671", + "left": "issue:44737", + "right": "issue:45859", "accept": false, - "reason": "Whisper pipeline return_language regression and CamemBERT masked-LM prediction changes are unrelated." + "reason": "Both are device-related, but one is XLNet positional encoding CPU placement and the other is Qwen3.5 MoE CPU offload incompatibility." }, { - "left": "issue:44737", - "right": "issue:45859", + "left": "issue:39404", + "right": "issue:44671", "accept": false, - "reason": "Both involve device/offload concerns, but they affect different models and different ops; not the same concrete bug." + "reason": "Whisper pipeline return_language regression is unrelated to CamemBERT masked-LM prediction changes." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Both are v5 initialization regressions, but one is about non-persistent buffers and the other about custom model initialization; similar area, not the same bug." + "reason": "Both are v5 initialization regressions, but one is non-persistent buffers being filled incorrectly and the other is custom model initialization; not the same concrete bug." }, { "left": "issue:42175", "right": "issue:46178", "accept": false, - "reason": "Backend packaging/mapping problems, but one is missing tensorflow in a torch extra and the other is an unsupported tensorflow_text backend; distinct failures." + "reason": "Both mention backend plumbing, but one is packaging/install extras and the other is a BACKENDS_MAPPING validation error for tensorflow_text." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are version-gated loading failures, but they concern different models and do not clearly share a single code-path regression." + "reason": "Both are tokenizer/model loading failures, but for different models and likely different regressions." }, { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "Related audio feature-extraction area, but one is an unsupported center arg in LasrFeatureExtractor and the other is a separate _torch_extract_fbank_features issue." + "reason": "Both are audio feature-extraction issues, but one is an unsupported `center` arg and the other is `_torch_extract_fbank_features()`; not clearly the same bug." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Different model/runtime failures: template-node compilation versus GraniteMoEHybrid calling an invalid method." + "reason": "Completely different failures: template compilation vs GraniteMoEHybrid invalid method call." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Apple Silicon allocator TypeError is unrelated to the _is_hf_initialized flag regression." + "reason": "Apple Silicon allocator warmup TypeError is unrelated to the `_is_hf_initialized` flag regression." }, { "left": "issue:43452", "right": "issue:43482", - "accept": true, - "reason": "Both are GGUF-loading regressions in transformers v5; 43482 is a model-specific manifestation of the broader gguf_file/from_pretrained breakage described in 43452." + "accept": false, + "reason": "Both involve GGUF loading, but one is a general `gguf_file`/from_pretrained break and the other is a model-specific Qwen2.5-GGUF failure; too little evidence that they are the same bug." }, { "left": "issue:43746", "right": "issue:46326", "accept": false, - "reason": "PEFT/local-checkpoint loading for GraniteSpeech is unrelated to continue_final_message behavior in mistral_common tokenizers." + "reason": "PEFT-adapter local checkpoint loading and mistral_common continue_final_message are separate problems." }, { - "left": "issue:43582", - "right": "issue:47331", + "left": "issue:44811", + "right": "issue:45584", "accept": false, - "reason": "One is an Apple Silicon allocator TypeError; the other is a Metal quantization batched-generate corruption bug." + "reason": "Both are Whisper-related, but one is `batch_decode` ignoring `skip_special_tokens` and the other is empty-transcription generation failure after `align_special_tokens`; distinct symptoms and likely distinct fixes." }, { - "left": "issue:44811", - "right": "issue:45584", + "left": "issue:43582", + "right": "issue:47331", "accept": false, - "reason": "Both are Whisper-related, but one is skip_special_tokens being ignored in batch_decode and the other is empty-transcription failure after align_special_tokens." + "reason": "Apple Silicon warmup TypeError is unrelated to Metal quantization batched-generate corruption." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Generic loading noise is unrelated to a specific Mistral tokenizer regex patch crash." + "reason": "General loading noise vs Mistral regex patch crash are different issues." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The selected canonical issue (43452) is a reasonable broad representative for the GGUF-loading reports, and the only accepted soft pair (43452/43482) is defensible as a model-specific manifestation of the broader GGUF regression. The other pair judgments are cautious and do not overstate duplication evidence." - }, - "evaluator_used": true, + "evaluator_result": null, + "evaluator_used": false, "retried": false, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -3778,9 +3802,9 @@ ], "soft_pairs": [ "issue:43262|issue:47331", + "issue:43645|issue:46661", "issue:43208|issue:47013", "issue:43792|issue:46869", - "issue:43645|issue:46661", "issue:43317|issue:43482", "issue:42915|issue:45084", "issue:44987|issue:45081", @@ -3806,202 +3830,202 @@ "issue:45923|issue:47254", "issue:44464|issue:45987" ], - "prepared_review_unit_hash": "e94a909c6bfbf38374305f05a615c57c084ca641cf882d22bc9595e4e30a3892", + "prepared_review_unit_hash": "864bc2afc3ae159e6615b72783ff24c1d50e69249fa06d77f371331b2ca3e853", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35810, - "estimated_input_tokens": 8953, - "estimated_eval_tokens": 18162 + "serialized_chars": 35814, + "estimated_input_tokens": 8954, + "estimated_eval_tokens": 18164 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose pile of unrelated closed issues with some broad overlap around Transformers v5/model-loading regressions. None of the soft pairs looks like the same concrete bug or change, so I would not merge any pair as duplicates.", - "confidence": 0.91, - "canonical_issue_reason": "Issue 43646 is the broadest representative of the largest apparent theme in the cluster: Transformers v5 regressions around custom model initialization/loading. The other v5-related reports are narrower symptom-specific variants.", + "summary": "This cluster is a loose semantic grouping of many distinct Transformers bugs, mostly around model loading, tokenizers, multimodal preprocessing, quantization, and trainer/runtime regressions. The soft pairs are only superficially related; none look like the same underlying bug or a plausible merged PR.", + "confidence": 0.96, + "canonical_issue_reason": "No true canonical issue: the items span many different code paths and failure modes, so there isn't a single duplicate to anchor the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43646 is the best single representative issue because it is broad, central, and closer to the shared v5/custom-model-init theme than the more specialized reports.", + "best_issue_reason": "If one issue must represent the cluster, issue:43646 is the broadest umbrella because it covers a generic Transformers 5.0 custom-model initialization regression and is closest to several nearby v5 loading breakage reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43262", "right": "issue:47331", "accept": false, - "reason": "Audio chat-template sampling-rate handling and MetalConfig quantization batched generation are unrelated bugs." + "reason": "Audio chat-template sampling-rate bug vs Metal quantization batch corruption; unrelated components and fixes." + }, + { + "left": "issue:43645", + "right": "issue:46661", + "accept": false, + "reason": "Both involve custom/remote-code models, but one is Jupyter initialization and the other is missing config_class on AutoModel; different bugs." }, { "left": "issue:43208", "right": "issue:47013", "accept": false, - "reason": "Both are xLSTM, but 43208 is an umbrella report with three bugs while 47013 is a specific train-mode unpacking crash." + "reason": "Both are xLSTM, but one reports multiple training bugs for smaller models while the other is a specific return_last_states unpacking crash." }, { "left": "issue:43792", "right": "issue:46869", "accept": false, - "reason": "Whisper runtime failure and YOLO multi-GPU device_map auto failure are different models and failure modes." - }, - { - "left": "issue:43645", - "right": "issue:46661", - "accept": false, - "reason": "Both concern custom/remote-code model initialization, but one is Jupyter-specific and the other is a config_class attribute error; not the same bug." + "reason": "Both are load failures, but Whisper runtime failure and YOLO multi-GPU device-map failure are different code paths." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "Dequantized model offload loading and Qwen2.5-GGUF loading in v5 are separate model-loading issues." + "reason": "GPU+CPU offload dequantized model loading vs Qwen2.5-GGUF v5 loading; different model families and failure modes." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "Qwen3Moe FP8 failure and template-compilation TypeError are unrelated." + "reason": "Qwen3Moe fine-grained FP8 failure vs template compilation error; not the same underlying bug." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Loading a specific model in v5 and a Mistral regex patch tokenizer attribute error are different problems." + "reason": "Model loading failure for physical-intelligence/fast vs Mistral tokenizer regex patch crash; separate tokenizer/model paths." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Custom model initialization breakage and non-persistent buffer junk-fill are distinct v5 regressions." + "reason": "Custom model initialization regression vs non-persistent buffer serialization junk; unrelated defects." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8Config failure and init_empty_weights _is_hf_initialized TypeError are different code paths." + "reason": "Qwen3Moe FP8 config issue vs init_empty_weights unexpected argument; different initialization/quantization problems." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Audio chat-template sampling-rate bug and MIMI padding-mask batching bug are unrelated." + "reason": "Audio sampling-rate default bug vs MIMI batch/single output mismatch from padding mask support; different modalities and logic." }, { "left": "issue:28530", "right": "issue:34567", "accept": false, - "reason": "Early stopping/eval metric handling and TrainerState num_input_tokens_seen updates are separate trainer issues." + "reason": "Early-stopping metric lookup bug vs TrainerState token counter not updating; unrelated trainer features." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "T5 tokenizer loading conflict and MobileLLM key error are different tokenizer/model-loading failures." + "reason": "T5 tokenizer method-name conflict vs MobileLLM key error; different tokenizer/model loading issues." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "AddedToken special-argument duplication and loading physical-intelligence/fast are unrelated." + "reason": "AddedToken special-argument conflict vs a model loading failure; unrelated tokenizer internals." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "MIMI batching/padding-mask mismatch and DAC from_latents vs forward STE mismatch are different model-specific bugs." + "reason": "MIMI padding-mask batch discrepancy vs DAC from_latents/forward mismatch due to missing STE; different models and bugs." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Model loading failure for ScandiBERT and Mistral regex patch backend_tokenizer crash are unrelated." + "reason": "ScandiBERT load failure vs Mistral regex patch crash; not the same bug." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Weight tying in Mistral3 and device-dependent lm_head serialization are related only at a very high level, but not the same bug." + "reason": "Both concern tied lm_head weights, but one is a missing tie in Mistral3 and the other is device-dependent serialization; related theme, different defect." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "SigLIP2 nonsensical outputs and OmDet-Turbo processor size mismatch are different vision-model issues." + "reason": "Siglip2 nonsense outputs with AutoModel/pipeline vs OmDet processor image-size mismatch; unrelated vision bugs." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Qwen3.5 hidden-state bug and Qwen2.5-VL rope index scaling bug affect different model families and code paths." + "reason": "Qwen3.5 hidden-states bug vs Qwen2.5-VL rope-index scaling bug; different model versions and code paths." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "Qwen3Moe FP8 config failure and missing Float8 storage type are separate quantization/runtime issues." + "reason": "Qwen3Moe FP8 config failure vs missing Float8 storage type; different low-level quantization/storage problems." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "T5 tokenizer load conflict and Qwen3 auto device mapping cudaErrorAssert are unrelated." + "reason": "Tokenizer attribute collision vs Qwen3 auto device-map CUDA assert; unrelated." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Qwen3-MoE sliding_window issue and granite_speech strict-config loading are different failures." + "reason": "Sliding-window issue in Qwen3-MoE vs strict config blocking granite_speech; different subsystems." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export with torch_compilable_check and OmDet-Turbo processor output-size mismatch are unrelated." + "reason": "torch.export / torch_compilable_check incompatibility vs OmDet processor size mismatch; separate issues." }, { "left": "issue:44008", "right": "issue:46899", "accept": false, - "reason": "Gemma 3n variable-name collision and Gemma4 multimodal 4-bit dtype-cast regression are different bugs." + "reason": "Gemma3n forward variable collision vs Gemma4 multimodal quantization dtype cast; different model generations and bugs." }, { "left": "issue:44479", "right": "issue:45325", "accept": false, - "reason": "Qwen video-input regression and Qwen2.5-VL still-image rope-index scaling are related only by model family, not by the same underlying defect." + "reason": "Video-input regression in Qwen multimodal models vs Qwen2.5-VL rope scaling for still images; not the same defect." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Llava Next loading segfault and Qwen3OmniMoe video-input ValueError are different multimodal failures." + "reason": "Llava Next segfault on load vs Qwen3OmniMoe video-processing ValueError; different models and failure points." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch version compatibility and loading-noise behavior are not the same code-path bug." + "reason": "Torch version incompatibility vs noisy loading output; unrelated." }, { "left": "issue:45923", "right": "issue:47254", "accept": false, - "reason": "Missing supports_gradient_checkpointing flag and PEFT/DeepSpeed checkpoint error are unrelated." + "reason": "Missing supports_gradient_checkpointing flag vs PEFT+DeepSpeed ZeRO-3 checkpoint error; both mention checkpointing but are different bugs." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Chunked generation inconsistency and StaticCache.get_seq_length return-type mismatch are distinct cache/generation bugs." + "reason": "Chunked generation inconsistency with compiled forward vs StaticCache seq-length return type bug; separate generation/cache paths." } ] }, @@ -4083,13 +4107,13 @@ "issue:43872|issue:44291", "issue:43883|issue:44488", "issue:43611|issue:44534", - "issue:45216|issue:45357", "issue:44991|issue:46315", - "issue:44403|issue:44991", "issue:43716|issue:46899", + "issue:44403|issue:44991", + "issue:45216|issue:45357", "issue:42796|issue:43746", - "issue:45812|issue:46491", "issue:44291|issue:44589", + "issue:45812|issue:46491", "issue:43381|issue:45923", "issue:43792|issue:44589", "issue:33666|issue:40444", @@ -4098,235 +4122,235 @@ "issue:42796|issue:46710", "issue:43531|issue:44291", "issue:35707|issue:40444", - "issue:43883|issue:44991", "issue:44079|issue:45663", "issue:41720|issue:45081", + "issue:43883|issue:44991", "issue:43482|issue:43525", "issue:43421|issue:44568", - "issue:44387|issue:47404", "issue:43296|issue:46821", - "issue:42175|issue:43296" + "issue:42175|issue:43296", + "issue:44387|issue:47404" ], - "prepared_review_unit_hash": "27825325d4e7914c16efe42e3beded9474cf69996229d54cd61aaad00f2d3c93", + "prepared_review_unit_hash": "d1884092157880ff10de9b7333483290619be529e5b73bbea7d9bade3aaf9b08", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37218, - "estimated_input_tokens": 9305, - "estimated_eval_tokens": 18866 + "serialized_chars": 37211, + "estimated_input_tokens": 9303, + "estimated_eval_tokens": 18862 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly noisy and heterogeneous: most items are distinct model-specific regressions, tokenizer issues, or performance bugs. I only found two clear duplicate families: the transformers v5 `_is_hf_initialized` loading regression, and the Qwen3.5 `save_pretrained` checkpoint-key regression.", - "confidence": 0.83, - "canonical_issue_reason": "issue:43632 is the broadest/root-cause report in the set: it captures the transformers v5 `_is_hf_initialized` regression that appears to underlie other loading/init failures.", + "summary": "This cluster is mostly a set of unrelated issue reports, with only a couple of close duplicate-like pairs. The strongest duplicate signal is the `_is_hf_initialized` regression, and a second likely duplicate pair is the Qwen3.5 `save_pretrained` checkpoint-key regression.", + "confidence": 0.84, + "canonical_issue_reason": "issue:44291 is the broadest and most root-cause-oriented report in the `_is_hf_initialized` regression family, so it is the best canonical issue for that duplicate pair.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43632 is the best representative issue for the cluster because it is the clearest upstream regression report and most likely to subsume related downstream loading failures.", + "best_issue_reason": "issue:44291 best represents the shared underlying bug around the new `_is_hf_initialized` argument and model loading failures in Transformers v5.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43632", "right": "issue:46821", "accept": false, - "reason": "Different bugs: `_is_hf_initialized` init regression vs `StaticCache.get_seq_length()` returning the wrong type." + "reason": "Both are v5-era regressions, but they describe different models and different failures (`_is_hf_initialized` vs `window_size` attribute access)." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Different failures: generation overhead/syncs vs a cache length contract bug." + "reason": "Generation performance overhead and `StaticCache.get_seq_length()` contract mismatch are unrelated bugs." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "Different model/code paths: Qwen2.5-VL dataset loading vs Qwen3-MoE `sliding_window` behavior." + "reason": "Different model families and different failure modes; no shared concrete code path." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Unrelated subsystems: audio fbank extraction vs compiled flex_attention on torch 2.9." + "reason": "Audio feature extraction and compiled flex attention on torch 2.9 are separate subsystems and issues." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is a training failure with multiple images and the other is a rope-index scaling regression; not the same bug." + "reason": "Both involve Qwen2.5-VL, but one is training with multiple images and the other is still-image position ID scaling; not the same bug." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same underlying regression: unexpected `_is_hf_initialized` leaking into model/bitsandbytes loading paths during transformers v5 init/loading." + "reason": "Both report the same underlying `_is_hf_initialized` regression introduced by Transformers v5, just surfacing in different loading paths." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Different bugs on different models: missing `all_tied_weights_keys` vs tokenizer load failure for `cjvt/sleng-bert`." + "reason": "Missing `all_tied_weights_keys` and inability to load `cjvt/sleng-bert` are unrelated." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "Different code paths: `base_model_prefix` loading breakage vs non-persistent buffers being filled with junk." - }, - { - "left": "issue:45216", - "right": "issue:45357", - "accept": true, - "reason": "Same regression family: Qwen3.5 `save_pretrained` is writing incorrect checkpoint keys in both reports." + "reason": "`base_model_prefix` loading failure and non-persistent buffer junk are different regressions." }, { "left": "issue:44991", "right": "issue:46315", "accept": false, - "reason": "Tokenizer load failure vs performance regression in `convert_tokens_to_ids`; not the same underlying bug." + "reason": "Tokenizer load failure for a specific model and slow `convert_tokens_to_ids` are distinct issues." + }, + { + "left": "issue:43716", + "right": "issue:46899", + "accept": false, + "reason": "Different multimodal dtype/quantization bugs in different models; not one change." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Unrelated issues: noisy loading logs vs tokenizer load failure." + "reason": "Loading noise and tokenizer load failure do not share the same concrete bug." }, { - "left": "issue:43716", - "right": "issue:46899", - "accept": false, - "reason": "Different multimodal bugs: Mistral-3 preprocessing dtype mismatch vs Gemma4 4-bit quantization cast problem." + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Both describe the same Qwen3.5 `save_pretrained` checkpoint serialization regression, with 45357 looking like a later-specific manifestation." }, { "left": "issue:42796", "right": "issue:43746", "accept": false, - "reason": "Tokenizer BPE-marker handling and PEFT local-checkpoint loading are different problems." + "reason": "Tokenizer BPE-marker handling and GraniteSpeech PEFT local-checkpoint loading are unrelated." }, { - "left": "issue:45812", - "right": "issue:46491", + "left": "issue:44291", + "right": "issue:44589", "accept": false, - "reason": "Different tokenizer regressions: Granite token IDs vs CodeLlama leading-space round-trip." + "reason": "Unexpected `_is_hf_initialized` and missing Float8 storage are different runtime failures." }, { - "left": "issue:44291", - "right": "issue:44589", + "left": "issue:45812", + "right": "issue:46491", "accept": false, - "reason": "Both are loading-time type errors, but one is about `_is_hf_initialized` and the other is about missing Float8 storage support." + "reason": "Both are tokenizer regressions, but they affect different models and different tokenization behaviors." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Different model-behavior bugs: eval-mode gradient checkpointing vs missing `supports_gradient_checkpointing` flag." + "reason": "Gradient checkpointing in eval mode and a missing `supports_gradient_checkpointing` flag are related only at a high level, not the same bug." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Different failures: Whisper model runtime/load issue vs Float8 storage lookup error." + "reason": "Whisper runtime failure and Float8 storage lookup failure are separate problems." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Different training issues: generic Qwen2-VL multi-GPU training vs Qwen2.5-VL iterable dataset with multiple images." + "reason": "Multi-GPU training and multi-image IterableDataset finetuning are different training issues." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "Different categories: perplexity regression vs Whisper `batch_decode()` special-token handling." + "reason": "High perplexity and Whisper `batch_decode` special-token handling are unrelated." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "Different bugs: `_is_hf_initialized` regression vs mmap-related OOM on Strix Halo." + "reason": "`_is_hf_initialized` regression and mmap/OOM behavior on Strix Halo are unrelated." }, { "left": "issue:42796", "right": "issue:46710", "accept": false, - "reason": "Tokenizer marker handling vs accuracy regression after upgrade; not the same concrete issue." + "reason": "Tokenizer marker handling for Ministral and DeepSeek accuracy regression are different bugs." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Different bugs: Qwen3-MoE `sliding_window` issue vs unexpected `_is_hf_initialized` kwarg during loading." + "reason": "Qwen3-MoE sliding window behavior and `_is_hf_initialized` loading errors are unrelated." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Different generation/training issues: `inputs_embeds` + `past_key_values` vs multi-image IterableDataset finetuning." - }, - { - "left": "issue:43883", - "right": "issue:44991", - "accept": false, - "reason": "Different model-loading errors on unrelated models." + "reason": "Progressive generation with embeds/past KV is not the same as multimodal IterableDataset training failure." }, { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "Different problems: `ModelOutput` key assignment vs Gemma4 FSDP2 shared KV state rebuild bug." + "reason": "`ModelOutput` key assignment and Gemma4 FSDP2 shared_kv_states rebuilding are different code paths." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Different failures: Qwen3 auto device-map cuda assert vs Mistral tokenizer regex patch crash." + "reason": "Auto device mapping CUDA assert and Mistral regex patch crash are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Missing model attribute and tokenizer loading failure do not match." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Different model/config issues: GGUF loading failure vs missing `pad_token_id` on Llama4Config." + "reason": "Qwen2.5-GGUF loading failure and missing `pad_token_id` on Llama4Config are different incompatibilities." }, { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Different tokenizer feature bugs: runtime post-processor updates vs `add_special_tokens=True` not adding BOS/EOS." - }, - { - "left": "issue:44387", - "right": "issue:47404", - "accept": false, - "reason": "Both are OOM-related, but one is int4 reserved-memory growth and the other is conversion-op transients during `device_map` loading." + "reason": "Runtime special-token post-processor updates and `add_special_tokens=True` not adding BOS/EOS are related to tokenization but not the same bug." }, { "left": "issue:43296", "right": "issue:46821", "accept": false, - "reason": "Different issues: Qwen2.5-VL finetuning failure vs TimesFM `window_size` attribute error." + "reason": "Different models and different attribute failures; no shared underlying issue." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Packaging/backend inclusion issue vs PaddleOCR-VL model loading failure; not a duplicate." + "reason": "Packaging/backends omission and model loading failure are unrelated." + }, + { + "left": "issue:44387", + "right": "issue:47404", + "accept": false, + "reason": "Both are memory-related, but one is int4 quantization OOM and the other is device_map loading transients; not the same bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative. The rejected soft pairs are mostly justified by distinct titles/code paths, and the two accepted pairs look mergeable as single bugfix families: the transformers v5 `_is_hf_initialized` loading regression and the Qwen3.5 `save_pretrained` checkpoint-key regression." + "feedback": "The summary is well grounded in the packet and stays conservative on the soft-edge calls. The two accepted duplicate-like pairs are plausible from the titles alone, and the rejected pairs are generally described with appropriate caution rather than overstated as identical bugs." }, "evaluator_used": true, "retried": false, @@ -4406,40 +4430,40 @@ "issue:44265|issue:45072", "issue:44509|issue:46909", "issue:45325|issue:45834", - "issue:43377|issue:45491", "issue:40444|issue:45381", + "issue:43377|issue:45491", "issue:43644|issue:44079", "issue:44291|issue:44877", "issue:45325|issue:45381", "issue:43994|issue:44898", "issue:40444|issue:47217", - "issue:43818|issue:46402", "issue:43454|issue:44466", + "issue:43818|issue:46402", "issue:43818|issue:45357", "issue:43873|issue:45701", "issue:43701|issue:45325", "issue:44829|issue:45479", - "issue:42796|issue:45381", "issue:43606|issue:44265", + "issue:42796|issue:45381", "issue:44479|issue:45823", "issue:43828|issue:46821", "issue:40990|issue:45584", "issue:43262|issue:43746" ], - "prepared_review_unit_hash": "3202f238bc916c6fa23202609149d731ecc0bbe4a86b843470f8f1a5cf209055", + "prepared_review_unit_hash": "ccdb9210bb0f05a707e5cfb5e59e25a2b8b79a2421c44692961f84f13fd6fb93", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36355, + "serialized_chars": 36356, "estimated_input_tokens": 9089, "estimated_eval_tokens": 18434 }, @@ -4448,168 +4472,168 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated closed issues that only share broad themes like tokenizers, multimodal models, loading/offload, and v5 regressions. The soft pairs are superficial overlaps rather than the same underlying bug or change, so all should stay separate.", - "confidence": 0.97, - "canonical_issue_reason": "No single true duplicate stands out; if one issue must represent the cluster, 43873 is the broadest/highest-activity runtime bug, but the set is too heterogeneous for a real canonical.", + "summary": "These items are mostly unrelated, model-specific regression reports rather than a single duplicate cluster. Under a strict same-code-path standard, none of the soft candidate pairs should be merged. The strongest representative issue is the broad offloading/quantization regression (#43873), which is active and well-referenced.", + "confidence": 0.84, + "canonical_issue_reason": "#43873 is the broadest and most central issue here: it has the highest discussion/inbound activity and describes a general offloading+quantization failure rather than a narrow model-specific edge case.", "canonical_pr_reason": null, - "best_issue_reason": "43873 has the strongest discussion/inbound activity and is a broad, user-facing runtime issue, so it is the most plausible representative despite the overall heterogeneity.", + "best_issue_reason": "#43873 is the best anchor for this set because it is the most general, highest-activity issue and likely to attract related offloading regressions if they truly converge.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "Both touch TensorFlow packaging/backend availability, but one is about the torch extra missing TF and the other about tensorboard-only installs causing an undefined backend error; different failure modes and fixes." + "reason": "Both involve TensorFlow backend packaging, but one is about `.[torch]` install extras omitting TF and the other is about tensorboard being present without TF; different failure modes." }, { "left": "issue:44462", "right": "issue:45812", "accept": false, - "reason": "One is about AutoTokenizer ignoring tokenizer.json; the other is a Granite token-ID regression. Same area, but not the same bug." + "reason": "One is about ignoring `tokenizer.json`, the other about wrong Granite token IDs; same area, but not the same underlying bug." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "ModelOutput key assignment and non-persistent buffer initialization are unrelated code paths and symptoms." + "reason": "`ModelOutput` key assignment and non-persistent buffer serialization are unrelated internals with different code paths." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "GGUF loading failure for Qwen2.5 vs CLIPTokenizer max-length handling are unrelated issues." + "reason": "GGUF loading in v5 and CLIPTokenizer `model_max_length` are unrelated bugs." }, { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Both are v5 loading regressions, but one is about base_model_prefix during loading and the other about custom model initialization; not the same concrete bug." + "reason": "Both are v5 loading regressions, but one is tied to `base_model_prefix` handling and the other to custom initialization; not the same concrete bug." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "Device-map/offload loading failure and MoE training memory inefficiency are different problems." + "reason": "Device-map offload/dequantization and Qwen3 MoE memory usage are distinct problems." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "torch.export + torch_compilable_check and bfloat16 dtype mismatches are separate export/inference issues." + "reason": "`torch.export` with `torch_compilable_check` and bfloat16 dtype mismatches are different inference/export failures." }, { "left": "issue:44509", "right": "issue:46909", "accept": false, - "reason": "Both are documentation-related, but they concern different docs and different content." + "reason": "Both are documentation-related, but they concern different docs/pages and different fixes." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Qwen2.5-VL rope/position-id scaling and Kosmos2.5 OCR index errors are different model-specific bugs." + "reason": "Qwen2.5-VL rope-index scaling and Kosmos2.5 long-OCR index errors are different model-specific bugs." }, { - "left": "issue:43377", - "right": "issue:45491", + "left": "issue:40444", + "right": "issue:45381", "accept": false, - "reason": "Both involve batching/padding, but the MIMI padding-mask bug and Gemma3 all-padding-window NaNs are distinct model/code paths." + "reason": "Multi-image finetuning failure in an IterableDataset is not the same as the video-position-id regression in Qwen2.5-VL inference." }, { - "left": "issue:40444", - "right": "issue:45381", + "left": "issue:43377", + "right": "issue:45491", "accept": false, - "reason": "One is a finetuning failure with multiple images per prompt; the other is a video position-id regression in Qwen2.5-VL." + "reason": "Missing padding-mask support in MIMI and Gemma3 NaNs from sliding-window all-padding cases are different batching/masking bugs." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Non-persistent buffer serialization and ModelOutput key assignment are unrelated." + "reason": "Non-persistent buffer junk and `ModelOutput` key assignment are unrelated v5 regressions." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Unexpected _is_hf_initialized init arg and strict granite_speech config loading are different initialization/config issues." + "reason": "`init_empty_weights` argument mismatch and strict config loading are different load-time failures." }, { "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Both mention Qwen2.5-VL position ids, but one is still-image rope scaling and the other is video input regression." + "reason": "They both mention Qwen2.5-VL video/vision positions, but one is still-image rope scaling and the other is video input position IDs; not clearly the same bug." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "SigLIP2 AutoModel/pipeline nonsense outputs and Perceiver interpolate_pos_encoding failure are unrelated vision-model bugs." + "reason": "Bad SigLIP2 outputs and Perceiver interpolation failures are separate vision-model issues." }, { "left": "issue:40444", "right": "issue:47217", "accept": false, - "reason": "Both are VL-related placeholder problems, but multi-image finetuning failure and user-text image-placeholder handling are not the same defect." + "reason": "Both involve VL image placeholders, but one is a finetuning/dataset issue and the other is a runtime failure when the placeholder appears in user text." }, { - "left": "issue:43818", - "right": "issue:46402", + "left": "issue:43454", + "right": "issue:44466", "accept": false, - "reason": "Video-LLaVA tower/weights issues and Sam3Model text-encoder loading are different model-loading bugs." + "reason": "Weight-tying failure in AyaVision and device-dependent `lm_head.weight` serialization inconsistency are related only at a very high level." }, { - "left": "issue:43454", - "right": "issue:44466", + "left": "issue:43818", + "right": "issue:46402", "accept": false, - "reason": "AyaVision weight tying and inconsistent lm_head serialization are about different weight-sharing/serialization paths." + "reason": "Video-LLaVA temporal-attention/weight-sharing issues are unrelated to Sam3 text-encoder loading." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Video-LLaVA architecture problems and Qwen3.5 visual-encoder save_pretrained key issues are unrelated." + "reason": "Video-LLaVA architecture bugs and Qwen3.5 visual-encoder save_pretrained keys are different subsystems and fixes." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Quantized offloading behavior and tokenizer-version-dependent tokenization are different subsystems and symptoms." + "reason": "The latter is a broad tokenization-version regression, not the same offloading/quantization bug." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint resume key mismatch and Qwen2.5-VL rope-index scaling are unrelated failures." + "reason": "`resume_from_checkpoint` key mismatch and Qwen2.5-VL rope-index scaling are unrelated." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Flash Attention 3 training degeneration and single-label/num_labels=1 zero-loss behavior are separate training bugs." + "reason": "Both can cause degenerate training, but one is tied to Flash Attention 3 behavior and the other to `problem_type`/`num_labels=1`; different root causes." }, { - "left": "issue:42796", - "right": "issue:45381", + "left": "issue:43606", + "right": "issue:44265", "accept": false, - "reason": "Tokenizer BPE marker handling and Qwen2.5-VL video position-id regression are unrelated." + "reason": "CPU-offload device mismatch and `torch.export`/compilable-check failures are distinct code paths." }, { - "left": "issue:43606", - "right": "issue:44265", + "left": "issue:42796", + "right": "issue:45381", "accept": false, - "reason": "CPU-offload device mismatch for bark-small and torch.export failures are different runtime paths." + "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL video position IDs are unrelated." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Qwen video-input regression and Gemma4 PLE device mismatch are separate model-specific bugs." + "reason": "Qwen VL video regression and Gemma4 PLE device mismatch are different model families and different failure modes." }, { "left": "issue:43828", "right": "issue:46821", "accept": false, - "reason": "Autocast dtype mismatch and TimesFM window_size AttributeError do not share the same code path." + "reason": "Autocast dtype mismatch and TimesFM `window_size` attribute errors are unrelated." }, { "left": "issue:40990", @@ -4621,7 +4645,7 @@ "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sample-rate defaulting and GraniteSpeech PEFT checkpoint loading are different speech issues." + "reason": "Audio processor sampling-rate defaulting and GraniteSpeech PEFT checkpoint loading are different bugs." } ] }, @@ -4706,20 +4730,20 @@ "issue:42796|issue:46209", "issue:43257|issue:45923", "issue:40990|issue:42947", - "issue:47404|issue:47405", "issue:44568|issue:45920", + "issue:47404|issue:47405", "issue:43454|issue:45072", "issue:43377|issue:44610", - "issue:41553|issue:43582", "issue:36331|issue:42617", + "issue:41553|issue:43582", "issue:41720|issue:43322", - "issue:36331|issue:42915", "issue:43572|issue:45759", + "issue:36331|issue:42915", + "issue:44877|issue:44991", "issue:43749|issue:44534", "issue:43454|issue:43720", - "issue:44877|issue:44991", - "issue:43377|issue:44898", "issue:43317|issue:43873", + "issue:43377|issue:44898", "issue:46305|issue:46315", "issue:41720|issue:44977", "issue:43322|issue:46050", @@ -4728,226 +4752,226 @@ "issue:44387|issue:45704", "issue:43450|issue:45701" ], - "prepared_review_unit_hash": "3168187380d694b15f2e0d8c0e0c5244108ba3a2687a611858339a1ea4bcc0d0", + "prepared_review_unit_hash": "1ea24f9cc6b50939f9fe68dfe7ce92c105e5db4b68bbb567ad792ee277f4dbd2", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37360, - "estimated_input_tokens": 9340, - "estimated_eval_tokens": 18936 + "serialized_chars": 37356, + "estimated_input_tokens": 9339, + "estimated_eval_tokens": 18934 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is highly heterogeneous: most items are unrelated bugs across tokenization, device mapping/offload, training APIs, multimodal processors, and model-specific regressions. None of the soft-similarity pairs look like the same underlying bug/change, so no duplicate merge is warranted.", - "confidence": 0.95, - "canonical_issue_reason": "If a representative issue is needed, issue 41720 is the closest hub because it is the most central concrete failure among the soft-similar items and has the strongest device-mapping/offload theme, but it still does not cover the rest of the set.", + "summary": "This cluster is mostly a loose set of Transformers bug reports with some broad thematic overlap around loading, offloading, quantization, and tokenization, but the paired items are not duplicates of the same concrete bug. I would not merge any of the soft-edge candidates.", + "confidence": 0.73, + "canonical_issue_reason": "issue:43873 is the broadest and most central issue in the set for the recurring offloading/quantization loading regressions, and it has the strongest discussion/inbound-reference signal. The other issues are either narrower model-specific failures or different code paths entirely.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 41720 is the best single anchor for triage because it is specific, reproducible, and central to the only loosely recurring subtheme (device mapping / offload failures).", + "best_issue_reason": "issue:43873 is the best representative issue for this cluster because it captures the general offloading + quantization failure theme better than the more specialized reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "Bark voice preset handling and progressive generation with inputs_embeds/past_key_values are unrelated bugs." + "reason": "Different features and failure modes: Bark voice presets vs generation with inputs_embeds/past_key_values." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Generation cache/input handling does not match a Qwen3 auto device-mapping cuda assert." + "reason": "Unrelated bugs: progressive generation state handling vs device-mapping/CUDA assert on Qwen3." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer NER batching issues and torch.export failures with torch_compilable_check are different code paths." + "reason": "Both involve batched behavior, but one is a tokenizer crash on NER padding/truncation and the other is torch.export with torch_compilable_check." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Both mention device mapping, but the models and failures differ: cudaErrorAssert vs forward device mismatch." + "reason": "Both are device-map related, but the concrete failures differ: CUDA assert on Qwen3 vs PLE forward device mismatch on Gemma4." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "CPU offload device mismatch is not the same as DAC.from_latents missing STE parity." + "reason": "Different bugs and code paths: CPU-offload device mismatch vs DAC latent reconstruction/STE mismatch." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "AyaVision lm_head tying and Bamba torch.compile/SDPA are separate model bugs." + "reason": "AyaVision lm_head tying and garbage text generation is unrelated to Bamba torch.compile + SDPA failure." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "3d_parallel script startup failure is unrelated to Qwen3Moe FP8 loading." + "reason": "No shared underlying bug; one is a 3d_parallel.py runtime issue, the other is FineGrainedFP8Config failure on Qwen3Moe." }, { "left": "issue:45823", "right": "issue:46821", "accept": false, - "reason": "Gemma4 device mismatch and TimesFM window_size AttributeError are unrelated." + "reason": "Device-mapping mismatch and TimesFM window_size AttributeError are different issues." }, { "left": "issue:42796", "right": "issue:46209", "accept": false, - "reason": "Tokenizer BPE marker handling and FP8 inference on B200 are different problems." + "reason": "Tokenizer BPE marker handling is unrelated to FP8 inference on B200." }, { "left": "issue:43257", "right": "issue:45923", "accept": false, - "reason": "Qwen3 MOE weight conversion with accelerate+deepspeed is not the same as a missing gradient-checkpointing flag." + "reason": "Accelerate+deepspeed MOE weight conversion is different from a missing supports_gradient_checkpointing flag." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity on gpt-oss and ineffective gradient checkpointing with LoRA are unrelated symptoms." + "reason": "Perplexity regression on gpt-oss is not the same bug as ineffective gradient checkpointing with LoRA." }, { - "left": "issue:47404", - "right": "issue:47405", + "left": "issue:44568", + "right": "issue:45920", "accept": false, - "reason": "Device-map loading OOM from conversion transients is different from silent wrong-Auto-class conversion skip." + "reason": "Both are tokenizer regressions, but one is missing add_special_tokens behavior for a specific tokenizer while the other is wrong token IDs across several models." }, { - "left": "issue:44568", - "right": "issue:45920", + "left": "issue:47404", + "right": "issue:47405", "accept": false, - "reason": "Missing BOS/EOS insertion for one tokenizer is not the same as broad wrong-token-id regressions across many models." + "reason": "Loading OOM from conversion-op transients is unrelated to silent conversion skipping on model_type mismatch." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "Weight tying garbage text generation and bfloat16 dtype mismatch are distinct failures." + "reason": "Weight tying/generation corruption is unrelated to dtype mismatches in bfloat16 inference." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "MIMI padding-mask batching and OmDet-Turbo image size mismatch are unrelated." + "reason": "Both mention batched inputs, but they affect different components and modalities: MIMI encoder padding masks vs OmDet processor image sizing." }, { - "left": "issue:41553", - "right": "issue:43582", + "left": "issue:36331", + "right": "issue:42617", "accept": false, - "reason": "Bad AutoTokenizer error handling for Voxtral is unrelated to Apple Silicon warmup TypeError." + "reason": "Trainer compute_loss signature breakage is unrelated to the 3d_parallel.py runtime failure." }, { - "left": "issue:36331", - "right": "issue:42617", + "left": "issue:41553", + "right": "issue:43582", "accept": false, - "reason": "CustomTrainer compute_loss API drift and 3d_parallel.py failure are separate issues." + "reason": "AutoTokenizer loading Voxtral error messaging is unrelated to an Apple Silicon TypeError in caching_allocator_warmup." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Auto device mapping cuda assert and Llava Next segmentation fault are different load-time failures." + "reason": "Different loading failures: Qwen3 auto device mapping CUDA assert vs Llava Next segmentation fault." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to unwrapping text_config for composite multimodal configs." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Trainer keyword-arg incompatibility is unrelated to Qwen3Moe FineGrainedFP8Config failure." + "reason": "Trainer loss-signature incompatibility is not the same as Qwen3Moe FP8 configuration failure." }, { - "left": "issue:43572", - "right": "issue:45759", + "left": "issue:44877", + "right": "issue:44991", "accept": false, - "reason": "StableLmConfig pad_token_idx regression and AutoModelForCausalLM text_config unwrapping are different config bugs." + "reason": "Strict config loading for granite_speech and tokenizer loading for est-roberta are different failures." }, { "left": "issue:43749", "right": "issue:44534", "accept": false, - "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk in v5 are unrelated." + "reason": "FSDP CPU RAM efficient loading breakage is not the same as non-persistent buffers being filled with junk." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "lm_head weight tying and unpacking packed weights during accelerate loading are different model-loading problems." + "reason": "AyaVision lm_head tying is unrelated to BitNet packed-weight unpacking during accelerate loading." }, { - "left": "issue:44877", - "right": "issue:44991", + "left": "issue:43317", + "right": "issue:43873", "accept": false, - "reason": "Strict granite_speech config loading and tokenizer loading for EMBEDDIA/est-roberta do not share a code path." + "reason": "Related theme, but not the same concrete bug: dequantized model offload loading failure vs broader offloading-with-quantization behavior." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "MIMI batched-output mismatch and Perceiver non-default resolution failure are unrelated vision/audio bugs." - }, - { - "left": "issue:43317", - "right": "issue:43873", - "accept": false, - "reason": "dequantized-model device_map offload failure and broad offloading-with-quantization regression are related only at a very high level, not the same concrete bug." + "reason": "MIMI batching/padding-mask mismatch is unrelated to Perceiver non-default resolution failures." }, { "left": "issue:46305", "right": "issue:46315", "accept": false, - "reason": "Sliding-window off-by-one masking and convert_tokens_to_ids slowdown are different behaviors." + "reason": "Streaming mask off-by-one and tokenizer convert_tokens_to_ids slowdown are different subsystems and symptoms." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "A800 cudaErrorAssert under auto device mapping is unrelated to Qwen3.5 flash-attention generation corruption." + "reason": "Qwen3 device-map CUDA assert is unrelated to Qwen3.5 flash-attention generation issues." }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "Llava Next load-time segmentation fault and sinq save/load quantization corruption are distinct bugs." + "reason": "Llava Next loading segfault is not the same bug as sinq quantized save/load breakage." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "Qwen3 auto device-mapping failure is unrelated to 3d_parallel.py execution failure." + "reason": "Device-mapping CUDA assert and 3d_parallel.py runtime failure are unrelated." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "High perplexity on one model and tokenization changes across versions are not the same underlying issue." + "reason": "High perplexity on a model and tokenization changes across versions are not the same underlying bug." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Int4 quantization memory growth and apex FusedRMSNorm memory leak are separate memory issues." + "reason": "Int4 CUDA reserved-memory regression is unrelated to T5 using apex.FusedRMSNorm with a memory leak." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Video processor batch-shape bug and tokenization-version changes are unrelated." + "reason": "Video processor batched-shape bug is unrelated to version-dependent tokenization changes." } ] }, @@ -5021,41 +5045,41 @@ ], "soft_pairs": [ "issue:43873|issue:44756", - "issue:44466|issue:45357", "issue:45685|issue:46821", + "issue:44466|issue:45357", "issue:26413|issue:28530", - "issue:44451|issue:47020", "issue:44186|issue:45072", + "issue:44451|issue:47020", "issue:43295|issue:45701", "issue:43065|issue:46402", "issue:46612|issue:46858", "issue:43550|issue:43720", "issue:44186|issue:44610", "issue:43646|issue:46665", - "issue:43577|issue:43708", "issue:43550|issue:43606", + "issue:43577|issue:43708", "issue:46897|issue:47317", "issue:46821|issue:46869", "issue:42915|issue:43317", "issue:41720|issue:45161", "issue:44423|issue:45406", - "issue:43404|issue:45357", - "issue:43881|issue:44488", "issue:44610|issue:44898", "issue:28530|issue:33666", + "issue:43404|issue:45357", "issue:44008|issue:45823", "issue:42915|issue:46291", + "issue:43881|issue:44488", "issue:43531|issue:43572", "issue:45704|issue:47404" ], - "prepared_review_unit_hash": "bc4bf71264780917b5377c3ff73a1d26c15fd2b0ab45889a2e41dfb39b54feb9", + "prepared_review_unit_hash": "e4ba8acee3ace3f55f96e49656f794832b0d88d3ffa4955cabe7553d5f3dfe01", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, @@ -5070,174 +5094,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No pair here looks like a true duplicate/near-duplicate; the set is a heterogeneous mix of trainer, tokenizer, model-loading, quantization, serving, and device-mapping bugs. I would not merge any soft edge candidates.", - "confidence": 0.82, - "canonical_issue_reason": "If forced to pick an anchor, issue 43646 is the broadest user-facing regression (\u2018custom model initialization\u2019), but this is only a weak umbrella and the collection is not a coherent duplicate cluster.", + "summary": "This cluster is very heterogeneous: the soft-similarity pairs mostly share broad themes (training, loading, quantization, multimodal, serving) but not the same concrete root cause. I would not merge any of the proposed soft edges.", + "confidence": 0.78, + "canonical_issue_reason": "Issue #43708 is the most concrete, well-scoped, and reproducible Trainer bug in the set, with a clear failure mode around resuming checkpoints and changing per-device batch size.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43646 is the most general representative, but it is still not a strong canonical target because the other items describe different code paths and failure modes.", + "best_issue_reason": "Issue #43708 is the strongest representative issue overall: specific symptom, clear repro path, and direct impact on training correctness.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Both involve loading/offloading, but one is quantization offload behavior and the other is mmap OOM on Strix Halo; different bugs and fixes." + "reason": "Both involve offloading/memory behavior, but one is about quantization offload while the other is Strix Halo mmap OOM; different bugs and fixes." }, { - "left": "issue:44466", - "right": "issue:45357", + "left": "issue:45685", + "right": "issue:46821", "accept": false, - "reason": "Both mention save/pretrained serialization, but one is tied lm_head serialization by device and the other is incorrect visual encoder keys in Qwen3.5." + "reason": "Completely different failures: MPS histogram support vs TimesFM window_size handling." }, { - "left": "issue:45685", - "right": "issue:46821", + "left": "issue:44466", + "right": "issue:45357", "accept": false, - "reason": "MPS histogram Int error in MoE code and TimesFM window_size AttributeError are unrelated code paths." + "reason": "Both are save/serialization regressions, but one is about tied lm_head weights and the other about incorrect visual encoder keys." }, { "left": "issue:26413", "right": "issue:28530", "accept": false, - "reason": "Checkpoint resume epoch_iterator failure and early-stopping metric lookup are separate trainer issues." + "reason": "Different Trainer issues: checkpoint resumption epoch_iterator failure vs early stopping/metric_for_best_model handling." }, { - "left": "issue:44451", - "right": "issue:47020", + "left": "issue:44186", + "right": "issue:45072", "accept": false, - "reason": "Both are model/tokenizer loading failures, but one is a specific model load regression and the other is a Unigram init type error." + "reason": "Unrelated subsystems: LayoutLMv2 tokenizer padding/NER crash vs dtype mismatches in inference." }, { - "left": "issue:44186", - "right": "issue:45072", + "left": "issue:44451", + "right": "issue:47020", "accept": false, - "reason": "LayoutLMv2 tokenizer padding/NER crash and dtype mismatch in bf16 inference are different subsystems and failure modes." + "reason": "Both affect tokenizer/model loading, but the concrete bugs are different model families and different code paths." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Processor/tokenizer API breakage is not the same as a generic tokenization change; too broad to be one bug." + "reason": "Both mention version-related tokenization behavior, but one is a processor/tokenizer API regression and the other is a broader tokenization change report." }, { "left": "issue:43065", "right": "issue:46402", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Model text encoder weights not loading are distinct Sam3 issues." + "reason": "Same model family, but one is a dummy Conv2d in Sam3PixelDecoder and the other is missing text encoder weight loading." }, { "left": "issue:46612", "right": "issue:46858", "accept": false, - "reason": "Both touch generation cache logic, but one is missing beam cache reorder support and the other is a device-tensor seq-length arithmetic bug." + "reason": "Both touch cache logic, but one is beam-search cache reorder and the other is static-cache seq length arithmetic in prefill." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Torch.compile+SDPA failure in Bamba and BitNet packed-weight unpacking during accelerate loading are unrelated." + "reason": "Different model/runtime bugs: torch.compile+SDPA for Bamba vs packed-weight unpacking during accelerate loading for BitNet." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Tokenizer crash on batched padding/truncation and OmDet-Turbo image-size mismatch are different preprocessing bugs." + "reason": "Different vision preprocessing bugs with different models and failure modes." }, { "left": "issue:43646", "right": "issue:46665", "accept": false, - "reason": "Custom model initialization regression and inconsistent from_pretrained behavior inside Python classes are related only at a very high level, not the same concrete bug." + "reason": "One is a Transformers 5.0 custom model init regression; the other is a from_pretrained behavior inconsistency inside Python classes." }, { - "left": "issue:43577", - "right": "issue:43708", + "left": "issue:43550", + "right": "issue:43606", "accept": false, - "reason": "BLIP2 dtype remaining float32 and resume_from_checkpoint max_steps miscalculation are unrelated." + "reason": "Both are runtime/device issues, but they affect different models and have different root causes." }, { - "left": "issue:43550", - "right": "issue:43606", + "left": "issue:43577", + "right": "issue:43708", "accept": false, - "reason": "Different models and failures: torch.compile/SDPA vs CPU-offload device mismatch." + "reason": "Different problems: dtype propagation in Blip2 vs Trainer max_steps recomputation on resume." }, { "left": "issue:46897", "right": "issue:47317", "accept": false, - "reason": "Both are training-loss bugs, but they concern different models and likely different loss computation paths." + "reason": "Both are training-loss bugs, but they concern different models and no shared concrete code path is established." }, { "left": "issue:46821", "right": "issue:46869", "accept": false, - "reason": "TimesFM window_size AttributeError and YOLO multi-GPU device_map failure are unrelated." + "reason": "TimesFM window_size attribute handling is unrelated to YOLO multi-GPU device_map loading." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "FineGrainedFP8Config failure and dequantized model offload loading failure both involve quantization, but not the same root cause." + "reason": "Both involve quantization/loading, but one is FineGrainedFP8Config and the other is gpu+cpu offload of a dequantized model." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Qwen3 auto device map cudaErrorAssert and GPT-OSS MoE tensor-parallel issue are different parallelism bugs." + "reason": "Different distributed/model-parallel issues: device-mapping cudaErrorAssert vs tensor-parallel support for GPT-OSS MoE." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both hit transformers serve with multimodal models, but one is a continuous-batching crash and the other is a Gemma4Processor tokenizer attribute regression." - }, - { - "left": "issue:43404", - "right": "issue:45357", - "accept": false, - "reason": "One is tied lm_head serialization in Mistral3, the other is incorrect visual encoder keys in Qwen3.5; same broad area, different bug." - }, - { - "left": "issue:43881", - "right": "issue:44488", - "accept": false, - "reason": "Two separate model-loading failures for different checkpoints; no shared code-path evidence." + "reason": "Both are serve crashes on multimodal models, but the immediate exceptions and offending code paths differ." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Both concern image resolution handling, but one is a processor producing the wrong size and the other is non-default-res interpolation failing in Perceiver." + "reason": "Different model-specific image size/preprocessing bugs." }, { "left": "issue:28530", "right": "issue:33666", "accept": false, - "reason": "Early stopping configuration bug and Qwen2-VL multi-GPU training issue are unrelated." + "reason": "Early stopping metric lookup and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "One is tied lm_head serialization in Mistral3ForConditionalGeneration; the other is incorrect visual encoder keys in Qwen3.5 save_pretrained." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Gemma3n variable-name collision in forward and Gemma4 PLE device mismatch are different model-specific bugs." + "reason": "Different Gemma-family bugs: variable name collision in forward vs device mismatch under device_map=auto." }, { "left": "issue:42915", "right": "issue:46291", "accept": false, - "reason": "FineGrainedFP8Config failure and kernels import-version requirement are unrelated quantization vs dependency issues." + "reason": "FineGrainedFP8Config loading and kernels package revision/version requirements are unrelated." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Both are model loading failures, but for different models and likely different failure mechanisms." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE and missing pad_token_idx in StableLmConfig are separate config bugs." + "reason": "Unrelated configuration bugs: Qwen3-MoE sliding window vs StableLmConfig pad_token_idx." }, { "left": "issue:45704", "right": "issue:47404", "accept": false, - "reason": "Apex RMSNorm memory leak in T5 and device_map OOM from conversion-op transients are different memory problems in different layers." + "reason": "Both relate to memory, but one is an apex RMSNorm leak and the other is device-map loading OOM from conversion transients." } ] }, @@ -5270,11 +5294,11 @@ "issue:36032", "issue:41720", "issue:42915", + "issue:43257", "issue:43316", "issue:43322", "issue:43441", "issue:43452", - "issue:43526", "issue:43531", "issue:43540", "issue:43550", @@ -5284,7 +5308,6 @@ "issue:43782", "issue:43873", "issue:43881", - "issue:43883", "issue:44186", "issue:44361", "issue:44464", @@ -5306,214 +5329,222 @@ "issue:46894", "issue:47020", "issue:47145", + "issue:47254", "issue:47317", "issue:47337" ], "soft_pairs": [ "issue:43441|issue:47337", "issue:44186|issue:44361", - "issue:44734|issue:46858", "issue:28530|issue:30064", "issue:43716|issue:46821", + "issue:44734|issue:46858", "issue:44464|issue:46858", "issue:44671|issue:44743", "issue:29159|issue:36032", "issue:43316|issue:44913", "issue:30064|issue:34634", - "issue:45479|issue:47317", "issue:43452|issue:47020", + "issue:45479|issue:47317", "issue:45632|issue:45684", - "issue:42915|issue:43322", "issue:45701|issue:47145", + "issue:42915|issue:43322", "issue:33357|issue:35707", "issue:43540|issue:43782", "issue:43873|issue:45704", - "issue:43881|issue:44991", "issue:43531|issue:45759", "issue:41720|issue:43746", + "issue:43881|issue:44991", "issue:43550|issue:43653", "issue:44534|issue:46665", "issue:46894|issue:47317", - "issue:43526|issue:43883" + "issue:43257|issue:47254", + "issue:43531|issue:46821" ], - "prepared_review_unit_hash": "e6efdb2056571f85ca5ce9426f755f714fc53f079fe8cbc5fb1b6a5b8ad4545d", + "prepared_review_unit_hash": "638da0eab1a22ebe92c843437efca133209b30c0b08ec8aab49bbe53c1d8f1c2", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 25, - "serialized_chars": 34562, - "estimated_input_tokens": 8641, - "estimated_eval_tokens": 17538 + "soft_pair_count": 26, + "serialized_chars": 34821, + "estimated_input_tokens": 8706, + "estimated_eval_tokens": 17668 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft edges are mostly topical overlaps rather than true duplicates. They span unrelated bugs across tokenizers, generation/cache handling, multimodal loading, and model-specific failures, so none should be merged as the same underlying issue.", - "confidence": 0.94, - "canonical_issue_reason": "issue:45479 is the strongest representative issue overall because it states a concrete, broadly applicable bug with clear impact; however, the full set is too heterogeneous to form one real duplicate cluster.", + "summary": "This cluster is heterogeneous and does not contain any clear duplicate PRs; the soft pairs mostly share broad subsystem keywords (tokenizers, cache, Qwen, generation) but describe different concrete failures. The best issue anchor is the most broadly discussed active bug report, not a true duplicate hub.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 43873 is the most central/open report in the set, with the strongest inbound discussion and a broad, actionable bug description. There is no strong duplicate cluster here, so it is the best available anchor rather than a true canonical duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45479 is the best issue to keep as a representative anchor: it is specific, reproducible, and broad enough to stand in for its own bug pattern.", + "best_issue_reason": "43873 has the most evidence of being a live, impactful problem and is the most suitable representative issue among otherwise unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43441", "right": "issue:47337", "accept": false, - "reason": "Different domains and symptoms: FlashAttention loading failure vs a flaky bbox-range test." + "reason": "Different domains: FlashAttention/model loading on macOS vs a flaky bbox-range test." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Both are tokenizer task crashes, but on different tokenizers and different error paths; too generic to be the same bug." - }, - { - "left": "issue:44734", - "right": "issue:46858", - "accept": false, - "reason": "Both involve cache/indexing, but 44734 is a serve KV-continuation indexing crash while 46858 is a static-cache slice-arithmetic bug on non-CUDA backends." + "reason": "Both tokenizer bugs, but different models and failure modes; not the same underlying defect." }, { "left": "issue:28530", "right": "issue:30064", "accept": false, - "reason": "Early stopping metric selection is unrelated to image processor segmentation-map handling." + "reason": "Early-stopping metric selection is unrelated to image processor void-segmentation handling." }, { "left": "issue:43716", "right": "issue:46821", "accept": false, - "reason": "Image dtype mismatch and TimesFM window_size AttributeError are unrelated failures." + "reason": "Different components and errors: multimodal dtype mismatch vs TimesFM window_size AttributeError." + }, + { + "left": "issue:44734", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache logic, but one is serve/KV-continuation indexing and the other is static-cache seq_length on non-CUDA compile paths." }, { "left": "issue:44464", "right": "issue:46858", "accept": false, - "reason": "Both mention compiled generation/cache, but the reported failures and code paths differ; not the same concrete bug." + "reason": "Chunked generation inconsistency is a different code path from static-cache slice arithmetic with torch.compile." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "Both affect generation/model behavior, but one is masked LM predictions and the other is recurrent state reset with cache." + "reason": "Different models and bugs: CamemBERT masked LM predictions vs Qwen3 recurrent-state reset with cache." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "Tokenizer inconsistency and T5 add_special_tokens name conflict are different tokenizer bugs." + "reason": "Tokenizer inconsistency is too broad; T5Tokenizer load failure is a specific special-token name collision." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Config API discrepancy vs rotary_pct persistence bug are not the same issue." + "reason": "Both are config/API-ish, but one is a Gemma3TextConfig discrepancy and the other is GPTNeoX rotary_pct persistence." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Image segmentation-map processing and BarkProcessor voice_preset are unrelated." + "reason": "Unrelated processor bugs affecting segmentation maps vs BarkProcessor voice_preset." }, { - "left": "issue:45479", - "right": "issue:47317", + "left": "issue:43452", + "right": "issue:47020", "accept": false, - "reason": "Both are training-loss-related at a high level, but the first is a specific classification zero-loss bug across models while the second is only a model-specific loss report." + "reason": "Both touch tokenizer/model loading, but they are distinct bugs with different root causes and affected classes." }, { - "left": "issue:43452", - "right": "issue:47020", + "left": "issue:45479", + "right": "issue:47317", "accept": false, - "reason": "GGUF loading/from_pretrained regression and Unigram vocab type mismatch are different loader/tokenizer problems." + "reason": "Both mention training loss, but the model families and likely failure mechanisms are different; insufficient evidence of one underlying bug." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Cache-path collision for trust_remote_code and read-only permission propagation in save_pretrained are unrelated filesystem issues." + "reason": "Cache-path collision for trust_remote_code is unrelated to file permission propagation in save_pretrained." }, { - "left": "issue:42915", - "right": "issue:43322", + "left": "issue:45701", + "right": "issue:47145", "accept": false, - "reason": "Qwen3Moe FP8 config failure and Llava Next loading segfault are different models and failure modes." + "reason": "Tokenizer-version differences and FP8 weight_scale corruption are different problems." }, { - "left": "issue:45701", - "right": "issue:47145", + "left": "issue:42915", + "right": "issue:43322", "accept": false, - "reason": "Tokenization drift across versions and FP8 weight_scale corruption are unrelated bugs." + "reason": "FineGrainedFP8Config failure and Llava Next segfault are unrelated model-loading issues." }, { "left": "issue:33357", "right": "issue:35707", "accept": false, - "reason": "MacOS CLIP bus error and progressive generation with inputs_embeds/past_key_values do not describe the same failure." + "reason": "MacOS bus error in a CLIP model is unrelated to progressive generation with inputs_embeds/past_key_values." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Both are Qwen3 multimodal issues, but one is video-input processing and the other is from_pretrained weight_only loading." + "reason": "Different multimodal bugs: video input processing vs from_pretrained weight_only loading." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantization/offloading behavior and T5 Apex RMSNorm memory leak are unrelated." - }, - { - "left": "issue:43881", - "right": "issue:44991", - "accept": false, - "reason": "GLM-4V model loading failure and tokenizer load failure for a different model are not duplicates." + "reason": "Quantization/offloading behavior is unrelated to T5's apex.FusedRMSNorm memory leak." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Sliding-window MoE behavior and from_config text_config unwrapping are different Qwen config bugs." + "reason": "Both involve Qwen3, but one is sliding_window behavior in MoE and the other is config unwrapping for multimodal models." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Auto device mapping cudaErrorAssert and GraniteSpeech PEFT local-checkpoint loading are unrelated." + "reason": "Auto device mapping CUDA assert and PEFT checkpoint loading are distinct failures." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Different model families and different loading failures; no shared code path." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "torch.compile+SDPA runtime failure and tokenizer special-token registration are different subsystems and bugs." + "reason": "torch.compile/SDPA model execution bug is unrelated to a tokenizer special-token registration issue." }, { "left": "issue:44534", "right": "issue:46665", "accept": false, - "reason": "Non-persistent buffer junk initialization and from_pretrained-in-Python-class inconsistency are not the same issue." + "reason": "Non-persistent buffers junk in v5 is not the same as inconsistent from_pretrained behavior inside Python classes." }, { "left": "issue:46894", "right": "issue:47317", "accept": false, - "reason": "Both mention training loss, but the titles do not establish the same underlying bug or code path." + "reason": "Both are training-loss complaints, but they are model-specific and the descriptions do not establish the same underlying bug." }, { - "left": "issue:43526", - "right": "issue:43883", + "left": "issue:43257", + "right": "issue:47254", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels behavior and Molmo tied-weights attribute error are unrelated." + "reason": "MoE weight conversion under accelerate/deepspeed and ZeRO-3 checkpointing errors are different distributed-training problems." + }, + { + "left": "issue:43531", + "right": "issue:46821", + "accept": false, + "reason": "Qwen3-MoE sliding-window bug is unrelated to TimesFM's window_size AttributeError." } ] }, @@ -5543,9 +5574,9 @@ "issue:39692", "issue:41720", "issue:42490", - "issue:43257", "issue:43329", "issue:43388", + "issue:43526", "issue:43531", "issue:43540", "issue:43606", @@ -5553,6 +5584,7 @@ "issue:43653", "issue:43720", "issue:43746", + "issue:43883", "issue:44062", "issue:44077", "issue:44451", @@ -5580,20 +5612,17 @@ "issue:46050", "issue:46402", "issue:46665", - "issue:46821", "issue:46858", - "issue:47254", "issue:47333", "issue:47405" ], "soft_pairs": [ - "issue:43257|issue:47254", - "issue:43531|issue:46821", + "issue:43526|issue:43883", "issue:43606|issue:43720", "issue:42490|issue:43329", - "issue:45406|issue:45464", "issue:44492|issue:44509", "issue:43531|issue:44589", + "issue:45406|issue:45464", "issue:34634|issue:41720", "issue:34567|issue:35707", "issue:43606|issue:43653", @@ -5614,190 +5643,184 @@ "issue:44625|issue:45479", "issue:43746|issue:46050" ], - "prepared_review_unit_hash": "decbdb9dbef1ebc737a5e5cbd2846de71b62badcdaf9239ce77ddd48a810194a", + "prepared_review_unit_hash": "06e0aafb4f46818d708d60c763a2b0b43a5beba3ea5193160d043e82427dd3f6", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 26, - "serialized_chars": 35768, - "estimated_input_tokens": 8942, - "estimated_eval_tokens": 18140 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 25, + "serialized_chars": 34958, + "estimated_input_tokens": 8740, + "estimated_eval_tokens": 17736 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: the items cover unrelated bugs across loading, generation, tokenization, vision, trainer metrics, docs, and offloading. I found no true duplicates among the soft-edge pairs, so all soft edges are rejected.", - "confidence": 0.97, - "canonical_issue_reason": "No clear canonical issue: the cluster does not center on one underlying bug, and the items are mostly unrelated model- or subsystem-specific regressions.", + "summary": "This cluster is a grab-bag of unrelated bug reports across tokenizers, multimodal processors, loading/offloading, training, and docs. None of the soft-edge pairs look like the same underlying bug or change, so all are rejected. I picked the broad loading-correctness regression as the canonical issue.", + "confidence": 0.96, + "canonical_issue_reason": "issue:47405 is the most broadly applicable and highest-impact bug in the set: loading a converted checkpoint through the wrong Auto class silently yields wrong weights. It is more canonical than the narrowly scoped, model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a strong global representative because the cluster spans many distinct failure modes with no shared root cause.", + "best_issue_reason": "issue:47405 is the strongest standalone issue for triage because it describes a generic, severe correctness failure in model loading rather than a model-specific edge case or documentation mismatch.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43257", - "right": "issue:47254", - "accept": false, - "reason": "Different problems: Qwen3 MOE weight conversion with accelerate+deepspeed vs PEFT + ZeRO-3 gradient checkpointing failures." - }, - { - "left": "issue:43531", - "right": "issue:46821", + "left": "issue:43526", + "right": "issue:43883", "accept": false, - "reason": "Both mention windowing, but one is Qwen3-MoE sliding_window behavior and the other is a TimesFM window_size attribute error." + "reason": "Different subsystems and failure modes: BeitImageProcessorFast label reduction vs Molmo missing all_tied_weights_keys." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "Both involve loading/offload paths, but the concrete bugs differ: Bark CPU offload device mismatch vs BitNet packed-weight unpacking." + "reason": "Both mention offload/loading, but one is a Bark CPU device mismatch and the other is BitNet packed-weight unpacking; not the same bug." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Unrelated: local-model path warning on trailing slash vs a video-branch multimodal token counting bug." - }, - { - "left": "issue:45406", - "right": "issue:45464", - "accept": false, - "reason": "Different code paths and symptoms: Gemma4Processor missing _tokenizer in serve vs Qwen3.5 streaming chat/completions failure." + "reason": "Local path import warning vs multimodal video token counting undefined variables; unrelated." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "A typo in cache strategy docs is unrelated to docs still listing removed pipeline tasks." + "reason": "A typo in cache-strategy docs is unrelated to removed pipeline tasks in docs." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Sliding-window behavior on Qwen3-MoE is unrelated to a Float8 storage lookup TypeError." + "reason": "Qwen3-MoE sliding_window behavior and Float8 storage lookup failure are different code paths." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Qwen3.5 streaming API failures." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping CUDA assert are different bugs in different models." + "reason": "BarkProcessor voice_preset bug and Qwen3 auto device mapping cuda assert are different model-specific problems." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token accounting is unrelated to progressive generation with inputs_embeds and past_key_values." + "reason": "TrainerState token counter not updating is unrelated to progressive generation with inputs_embeds/past_key_values." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Bark CPU offload device mismatch is unrelated to BigBirdTokenizer special-token registration/empty decode." + "reason": "Bark CPU offload mismatch vs BigBirdTokenizer special-token registration; no shared underlying defect." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Both are model-loading failures, but they concern different models and different underlying tokenizer/config issues." + "reason": "Both are tokenizer-load failures, but for different models and likely different causes." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "One is about unwrapping composite Qwen multimodal text_config; the other is a CPU-offload incompatibility in rotary embedding." + "reason": "Same Qwen family, but one is config unwrapping for from_config and the other is CPU-offload rotary embedding compatibility." }, { "left": "issue:43540", "right": "issue:46858", "accept": false, - "reason": "Video input ValueError in Qwen3OmniMoe is unrelated to cache.get_seq_length returning a device tensor in prefill." + "reason": "Video processing ValueError in Qwen3OmniMoe is unrelated to cache length tensor arithmetic." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "gather_for_metrics label truncation and SAM3 PCS text/box weirdness are different bugs with no shared root cause." + "reason": "Metric gathering label truncation is unrelated to SAM3 text/bounding-box behavior." }, { "left": "issue:45704", "right": "issue:47405", "accept": false, - "reason": "T5 apex.FusedRMSNorm memory leak and silent conversion skip on model_type mismatch are unrelated." + "reason": "T5 Apex RMSNorm memory leak and wrong Auto-class loading of converted checkpoints are different failures." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Loading a specific model failing and PIL image processors requiring torchvision are separate problems." + "reason": "Model loading failure for physical-intelligence/fast is unrelated to PIL backend torchvision dependency regressions." }, { "left": "issue:43644", "right": "issue:46665", "accept": false, - "reason": "Non-persistent buffers filled with junk in Transformers 5.0.0 is not the same as inconsistent from_pretrained behavior inside Python classes." + "reason": "Non-persistent buffer junk in v5.0.0 and from_pretrained behavior inside Python classes are different issues." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "patchtsmixer post_init policy and Qwen3.5 num_labels propagation are unrelated configuration issues." + "reason": "PatchTSMixer post_init policy and Qwen3.5 num_labels propagation are unrelated." }, { "left": "issue:45357", "right": "issue:46402", "accept": false, - "reason": "Both touch multimodal save/load, but one is incorrect visual encoder keys in save_pretrained and the other is missing Sam3Model text encoder weights on load." + "reason": "Both involve multimodal weights, but one is Qwen3.5 save_pretrained key naming and the other is Sam3Model text encoder loading." }, { "left": "issue:45923", "right": "issue:47333", "accept": false, - "reason": "Missing supports_gradient_checkpointing on a trust_remote_code variant is unrelated to saving a disk-offloaded Qwen3-30B-A3B." + "reason": "Missing supports_gradient_checkpointing flag is unrelated to saving a disk-offloaded model." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "AddedToken special-argument collision is a tokenizer construction bug, not the same as ScandiBERT loading failure." + "reason": "AddedToken argument duplication and ScandiBERT load failure are not the same bug." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "Both mention trainers, but one is Qwen3.5 input-embedding unpacking with TRL SFT and the other is evaluate() failing after train()." + "reason": "TRL SFT input-embedding unpacking and trainer.evaluate after trainer.train are distinct training-path issues." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver interpolation at non-default resolution and bfloat16 dtype mismatches in inference are unrelated." + "reason": "Perceiver image-size handling and dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "SigLIP2 docs example errors and OmDet-Turbo processor/model size mismatch are different documentation/processor issues." + "reason": "Docs example errors for SigLIP2 do not match OmDet-Turbo processor/model resolution mismatch." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "num_labels propagation in Qwen3.5 is not the same as the zero-loss classification bug with num_labels=1." + "reason": "num_labels propagation in Qwen3.5 is not the same as the single-label zero-loss classification bug." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "PEFT adapter loading for GraniteSpeech is unrelated to sinq quantized model save/load corruption." + "reason": "PEFT adapter loading from local checkpoints and sinq quantized save/load breakage are different save/load mechanisms." } ] }, @@ -5871,22 +5894,22 @@ ], "soft_pairs": [ "issue:44514|issue:44545", - "issue:42886|issue:44843", "issue:30064|issue:36010", + "issue:42886|issue:44843", "issue:43421|issue:43927", - "issue:46655|issue:46693", "issue:43746|issue:43873", + "issue:46655|issue:46693", "issue:43381|issue:43844", "issue:43421|issue:43653", "issue:39692|issue:43493", "issue:45081|issue:45356", - "issue:44995|issue:46858", "issue:44464|issue:47050", - "issue:44368|issue:45357", + "issue:44995|issue:46858", "issue:43673|issue:47337", "issue:43782|issue:45381", "issue:43994|issue:44671", "issue:43421|issue:44186", + "issue:44368|issue:45357", "issue:45356|issue:45907", "issue:46032|issue:46463", "issue:44857|issue:47050", @@ -5894,220 +5917,220 @@ "issue:39692|issue:47050", "issue:43596|issue:44220", "issue:43720|issue:43819", - "issue:43526|issue:43881", "issue:43540|issue:46294", "issue:43295|issue:45834", "issue:43550|issue:43927", - "issue:43550|issue:43819" + "issue:43550|issue:43819", + "issue:43526|issue:43881" ], - "prepared_review_unit_hash": "0a5fa0a46758e312a0207a68d0c9dd188abad94fb64b1e4f10cc23d871240ca1", + "prepared_review_unit_hash": "47101b6772ebee240a749a48c6759a15ee93b3f2fee12d27709bc14c9dd8a323", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36221, - "estimated_input_tokens": 9056, - "estimated_eval_tokens": 18368 + "serialized_chars": 36228, + "estimated_input_tokens": 9057, + "estimated_eval_tokens": 18370 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous; almost all suggested pairs are related only by subsystem or symptom. The only clear duplicate pair is the Qwen2_5_VLProcessor.apply_chat_template batched-input/padding=False crash.", - "confidence": 0.79, - "canonical_issue_reason": "issue:44514 is the earliest report of the exact batched-input/padding=False crash; issue:44545 is a near-identical duplicate with the same failure mode.", + "summary": "This cluster is mostly heterogeneous transformer bugs/regressions, with only one clear duplicate pair: the two Qwen2_5_VLProcessor apply_chat_template reports. The rest are related only at a broad subsystem level (tokenizers, generation, multimodal, cache/offload) and should stay separate.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43873 is the broadest and most central report in the set, with the highest discussion and multiple inbound references; the other issues are narrower, model-specific regressions or tokenizer bugs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44514 is the best canonical issue because it cleanly captures the concrete bug and is the earlier of the only true duplicate pair.", + "best_issue_reason": "issue:43873 is the best cluster anchor because it is general enough to represent the most central theme here (offloading/quantization behavior) and has the strongest engagement.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same concrete bug: identical Qwen2_5_VLProcessor.apply_chat_template crash on batched input with padding=False." + "reason": "Same concrete bug: Qwen2_5_VLProcessor.apply_chat_template crashes on batched input when padding=False; titles differ only in wording." }, { - "left": "issue:42886", - "right": "issue:44843", + "left": "issue:30064", + "right": "issue:36010", "accept": false, - "reason": "Both mention HF_HUB_OFFLINE/tokenizer loading, but 44843 is a specific Mistral-regex/model_info bug; 42886 is broader and not clearly the same root cause." + "reason": "Unrelated subsystems and failures: image processor segmentation maps vs a GenerationMixin import error." }, { - "left": "issue:30064", - "right": "issue:36010", + "left": "issue:42886", + "right": "issue:44843", "accept": false, - "reason": "Completely different areas: image segmentation map processing vs a GenerationMixin import error." + "reason": "Both involve offline/tokenizer loading, but 44843 is specifically the mistral regex patch calling model_info; 42886 is a broader cache-loading failure." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Different bugs in different code paths: tokenizer post-processor updates vs DiaConfig save/load token IDs." + "reason": "Different bugs: runtime special-token/post-processor updates vs save/load losing custom token IDs and causing generation IndexError." }, { - "left": "issue:46655", - "right": "issue:46693", + "left": "issue:43746", + "right": "issue:43873", "accept": false, - "reason": "Unrelated video padding issue vs long-context FA2 performance regression." + "reason": "Different concrete problems: PEFT local checkpoint loading vs offloading behavior with quantization." }, { - "left": "issue:43746", - "right": "issue:43873", + "left": "issue:46655", + "right": "issue:46693", "accept": false, - "reason": "Local checkpoint loading with PEFT adapters is unrelated to quantization offloading behavior." + "reason": "Different code paths: video processor padding calculation vs attention-performance regression in long-context generation." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Gradient checkpointing eval-mode restriction is unrelated to ZeRO-3 gradient growth." + "reason": "Different training/runtime issues: eval-mode gradient checkpointing vs DeepSpeed ZeRO-3 gradient growth." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Both touch tokenization, but one is runtime post-processor updates and the other is BigBird special-token registration." + "reason": "Both tokenizer-related, but one is post-processor updates after runtime special-token changes and the other is BigBird mask-token registration; not the same bug." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Docs example errors are not the same as a model implementation discrepancy against JAX." + "reason": "Same model family, but one is a broken documentation example and the other is an implementation discrepancy with JAX; not one underlying issue." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both involve Mistral regex handling, but one is an AttributeError and the other is a codec/regression report; not clearly the same bug." - }, - { - "left": "issue:44995", - "right": "issue:46858", - "accept": false, - "reason": "Different cache bugs with different triggers and symptoms." + "reason": "Both mention fix_mistral_regex, but 45081 is a tokenizer attribute crash while 45356 is a codec-handling regression; close, but not the same concrete bug." }, { "left": "issue:44464", "right": "issue:47050", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is unrelated to ONNX fp16 attention dtype mismatch." + "reason": "Different failures: compiled-forward chunked generation inconsistency vs ONNX fp16 attention dtype mismatch." }, { - "left": "issue:44368", - "right": "issue:45357", + "left": "issue:44995", + "right": "issue:46858", "accept": false, - "reason": "A warning about tie_word_embeddings is unrelated to incorrect visual-encoder keys on save_pretrained." + "reason": "Both mention cache behavior, but one is a stale indexer cache in GlmMoeDsa and the other is device-tensor seq length arithmetic in static cache/torch.compile." }, { "left": "issue:43673", "right": "issue:47337", "accept": false, - "reason": "GenerationMixin cache/chunked_prefill is unrelated to the bbox coordinate flaky test." + "reason": "GenerationMixin chunked_prefill cache issue is unrelated to a flaky bbox-range test." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Qwen3VL weight_only load failure is a different issue from Qwen2.5-VL video position IDs." + "reason": "Both are Qwen VL issues, but one is from_pretrained weight_only loading and the other is video vision_position_ids; different bugs." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different model families and different failure modes: SigLIP2 bad outputs vs CamemBERT masked LM regression." + "reason": "Completely different models and symptoms: SigLIP2 inference drift vs CamemBERT masked LM predictions." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Special-token post-processing is unrelated to LayoutLMv2Tokenizer crashes on NER/batching." + "reason": "Both tokenizer-adjacent, but one is about special-token mutation at runtime and the other about LayoutLMv2 NER/batching crashes." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "The Qwen3.5 warning about tie_word_embeddings is unrelated to a save_pretrained visual-encoder-key regression." }, { "left": "issue:45356", "right": "issue:45907", "accept": false, - "reason": "Tokenizer regex regression is unrelated to list_repo_files missing the revision argument." + "reason": "Mistral/Kimi tokenizer regression is unrelated to list_repo_files forgetting to forward revision." }, { "left": "issue:46032", "right": "issue:46463", "accept": false, - "reason": "Mamba2Mixer use_cache corruption is unrelated to a T5 decoder-input validation error message." + "reason": "Mamba2Mixer use_cache correctness bug is unrelated to a T5 decoder-input validation error message." }, { "left": "issue:44857", "right": "issue:47050", "accept": false, - "reason": "LwDetrImageLoss AMP crash is unrelated to CanineModel fp16 ONNX attention dtype mismatch." + "reason": "Different model/tooling paths: LwDetr float16 AMP crash vs CanineModel ONNX export dtype mismatch." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "SigLIP2 documentation/model example issues are unrelated to quantization offloading." + "reason": "Same broad ML area, but SigLIP2 docs/example problems are not the same bug as quantization offloading behavior." }, { "left": "issue:39692", "right": "issue:47050", "accept": false, - "reason": "Different models and unrelated failure modes." + "reason": "Unrelated model-specific issues: SigLIP2 example/model mismatch vs CanineModel ONNX dtype mismatch." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "BertModel DeepSpeed zero3 init index error is unrelated to _torch_extract_fbank_features()." + "reason": "Different failure surfaces: DeepSpeed ZeRO-3/BertModel index error vs audio feature extraction issue." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "Packed-weights loading failure and DAC from_latents mismatch are unrelated bugs." - }, - { - "left": "issue:43526", - "right": "issue:43881", - "accept": false, - "reason": "BeitImageProcessorFast reduce_labels bug is unrelated to glm-4v-9b loading failure." + "reason": "Different model internals and bugs: BitNet packed-weight loading vs DAC from_latents missing STE consistency." }, { "left": "issue:43540", "right": "issue:46294", "accept": false, - "reason": "Both are Qwen video-input issues, but they affect different models and different concrete code paths." + "reason": "Both involve Qwen video input, but one is a ValueError in video processing for Qwen3OmniMoe and the other is chat-template token/feature mismatch in Qwen3.5-9B." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Processor/tokenizer regression is unrelated to Kosmos2.5 long-OCR index errors." + "reason": "Different bugs: processor.tokenizer regression with images vs Kosmos2.5 OCR long-input index error." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "Bamba torch.compile/SDPA failure is unrelated to DiaConfig save/load token-ID loss." + "reason": "Unrelated problems: torch.compile/SDPA failure in Bamba vs DiaConfig losing custom token IDs after save/load." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Different models and different failure modes; not mergeable as one fix." + "reason": "Different model and failure mode: Bamba torch.compile/SDPA bug vs DAC forward/from_latents mismatch." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "Different issues: BeitImageProcessorFast reduce_labels returns one label vs glm-4v-9b loading failure." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative overall. The only accepted duplicate pair, issue:44514 vs issue:44545, is well supported by the packet and the summary does not overstate other relationships." + "feedback": "The summary is grounded in the packet and stays conservative. The canonical-anchor choice is reasonable given 43873\u2019s higher discussion/inbound activity, and the only accepted soft edge (44514/44545) is a clear duplicate by title and bug description. The other pairwise rejects are appropriately cautious and do not overstate evidence." }, "evaluator_used": true, "retried": false, @@ -6179,23 +6202,23 @@ "soft_pairs": [ "issue:44488|issue:47020", "issue:44062|issue:44488", - "issue:43653|issue:43720", "issue:43531|issue:43643", "issue:44220|issue:44661", + "issue:43653|issue:43720", "issue:45381|issue:46693", - "issue:30064|issue:36331", - "issue:41720|issue:44991", "issue:45072|issue:47050", + "issue:41720|issue:44991", + "issue:30064|issue:36331", "issue:44821|issue:45084", "issue:43611|issue:46665", - "issue:46650|issue:47405", "issue:46829|issue:46832", "issue:44186|issue:44898", - "issue:44991|issue:47020", + "issue:46650|issue:47405", "issue:43540|issue:44008", + "issue:44991|issue:47020", "issue:43653|issue:43819", - "issue:44843|issue:45081", "issue:34634|issue:36010", + "issue:44843|issue:45081", "issue:43643|issue:44877", "issue:43335|issue:43756", "issue:45987|issue:46829", @@ -6206,20 +6229,20 @@ "issue:44448|issue:44671", "issue:45987|issue:46932" ], - "prepared_review_unit_hash": "15ad989403be68f87189861e00d899ad79f4309f78ff476326120982bb2589b9", + "prepared_review_unit_hash": "98eeae7dfff603f22719d20dd5515c670fa1e6a8061f90b389517952176ac5b0", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35883, + "serialized_chars": 35882, "estimated_input_tokens": 8971, "estimated_eval_tokens": 18198 }, @@ -6228,180 +6251,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of model-loading, tokenizer, and model-specific runtime bugs rather than true duplicates. A few items are thematically related around remote_code / loading regressions, but the soft-edge pairs generally describe different concrete failure modes.", - "confidence": 0.89, - "canonical_issue_reason": "issue:45020 is the broadest representative of the recurring model-loading / remote_code regression theme, while the other issues are mostly narrower, model-specific variants.", + "summary": "This cluster is mostly a collection of distinct Transformers bugs, many around loading/tokenizer/modeling regressions, but they do not collapse into a single underlying issue. The soft-linked pairs share broad theme similarity only, not the same concrete code path, so none should be merged as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45020 is the broadest and most central-sounding report in the set, covering recent-version regressions with remote_code models and overlapping with several other loading/config/tokenizer complaints. It is still not an exact duplicate of the others, but it is the best umbrella candidate among these issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 best captures the shared high-level cluster theme and is the most suitable umbrella issue among the candidates.", + "best_issue_reason": "#45020 is the most suitable representative because it describes a high-level regression category rather than a narrowly model-specific failure, making it the closest thing to a cluster anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44488", "right": "issue:47020", "accept": false, - "reason": "Both are tokenizer/loading failures, but for different models and different root causes: sleng-bert load failure vs Unigram vocab type mismatch." + "reason": "Both are tokenizer-loading failures, but one is a specific model load failure and the other is a constructor/API mismatch in XLMRobertaTokenizer; different root causes and code paths." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "Different tokenizer bugs: AddedToken special-arg duplication vs a model/tokenizer load failure for a specific checkpoint." - }, - { - "left": "issue:43653", - "right": "issue:43720", - "accept": false, - "reason": "Both touch loading paths, but one is BigBirdTokenizer special-token decode behavior and the other is BitNet packed-weight unpacking in accelerate loading." + "reason": "These are unrelated tokenizer bugs: one is an AddedToken keyword conflict, the other is a model/tokenizer load failure for a specific checkpoint." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Different problems: Qwen3-MoE sliding_window handling vs AutoConfig trust_remote_code missing fields." + "reason": "One is a sliding_window handling bug in Qwen3-MoE; the other is missing fields from trust_remote_code AutoConfig loading. Similar category, different defect." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Audio feature extraction failure and tokenizer mapping/add-new-model-like failure are unrelated code paths." + "reason": "Different failure modes: one is an audio feature extraction issue, the other is an add-new-model-like/tokenizer-mapping problem." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Both are CI/loading related, but one is BigBirdTokenizer special-token decoding and the other is BitNet packed-weight unpacking during accelerate loading; not the same bug." }, { "left": "issue:45381", "right": "issue:46693", "accept": false, - "reason": "Qwen2.5-VL video position-id bug is unrelated to the FA2 long-context performance regression." + "reason": "Video input position-id mismatch for Qwen2.5-VL is unrelated to the FA2 long-context performance regression." }, { - "left": "issue:30064", - "right": "issue:36331", + "left": "issue:45072", + "right": "issue:47050", "accept": false, - "reason": "Image processor segmentation-map handling and CustomTrainer compute_loss kwargs are unrelated." + "reason": "Both mention dtype mismatches, but they occur in different models and code paths: SwitchTransformers/TimmWrapperModel versus Canine attention ONNX export." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Both are regressions, but one is a Qwen3 auto-device-map CUDA assert and the other is a tokenizer-loading failure for a different model." + "reason": "Both involve model loading/version regressions, but one is a CUDA assert with auto device mapping and the other is tokenizer loading for a specific checkpoint." }, { - "left": "issue:45072", - "right": "issue:47050", + "left": "issue:30064", + "right": "issue:36331", "accept": false, - "reason": "Both mention dtype issues, but they affect different subsystems and failure modes: bfloat16 inference vs fp16 ONNX export attention mismatch." + "reason": "Image processor segmentation-map handling and Trainer.compute_loss signature mismatch are completely separate issues." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "AutoImageProcessor URL loading and template-compilation errors are unrelated." + "reason": "One is loading an image processor from URL; the other is a template compilation error. No shared root cause." }, { "left": "issue:43611", "right": "issue:46665", "accept": false, - "reason": "Both involve loading APIs, but one is a Transformers 5.0 base_model_prefix regression and the other is inconsistent from_pretrained behavior inside Python classes." - }, - { - "left": "issue:46650", - "right": "issue:47405", - "accept": false, - "reason": "Both concern loading/saving model conversion, but one is non-idempotent reverse renaming on save_pretrained and the other is wrong Auto-class loading causing silent conversion skips." + "reason": "Both concern loading behavior, but one is base_model_prefix handling in Transformers 5.0 and the other is inconsistent from_pretrained behavior inside Python classes." }, { "left": "issue:46829", "right": "issue:46832", "accept": false, - "reason": "Different vision model bugs: output field misassignment vs RT-DETR feature-level mismatch." + "reason": "Different model families and failures: wrong hidden_states output versus RTDetr feature-level mismatch." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Tokenizer crash for LayoutLMv2 NER/padding is unrelated to Perceiver image-classification failures." + "reason": "Both are image/LayoutLM-family bugs, but one is tokenizer NER/padding truncation crashes and the other is Perceiver resolution/interpolation failure." }, { - "left": "issue:44991", - "right": "issue:47020", + "left": "issue:46650", + "right": "issue:47405", "accept": false, - "reason": "Both are tokenizer load issues, but they involve different tokenizer classes and different root causes." + "reason": "Both involve loading/conversion correctness, but one is save_pretrained key renaming corruption while the other is silent model_type mismatch during Auto class loading." }, { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Different multimodal model bugs: video-input processing in Qwen3OmniMoe vs a variable-name collision in Gemma3nModel.forward()." + "reason": "Video-input processing in Qwen3OmniMoe and Gemma3n forward variable collision are unrelated model-specific bugs." }, { - "left": "issue:43653", - "right": "issue:43819", + "left": "issue:44991", + "right": "issue:47020", "accept": false, - "reason": "BigBirdTokenizer special-token decode behavior is unrelated to DAC.from_latents forward mismatch." + "reason": "Both are tokenizer loading issues, but one is a checkpoint-specific tokenizer breakage and the other is a constructor bug in XLMRobertaTokenizer." }, { - "left": "issue:44843", - "right": "issue:45081", + "left": "issue:43653", + "right": "issue:43819", "accept": false, - "reason": "Both hit _patch_mistral_regex, but one is an unconditional model_info/offline issue and the other is an AttributeError on backend_tokenizer; different concrete bugs." + "reason": "Special-token decoding in BigBirdTokenizer and DAC.from_latents STE mismatch are unrelated defects." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "BarkProcessor voice_preset and GenerationMixin import failure are unrelated." + "reason": "BarkProcessor voice_preset and GenerationMixin import failure are different processor/import issues with no shared implementation bug." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Offline HF_hub model_info calls and mistral_regex backend_tokenizer attribute access are different failures in different helper paths." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "Both involve config loading, but one is missing fields from trust_remote_code AutoConfig and the other is strict config rejection for granite_speech." + "reason": "Both touch config loading, but one is missing fields under trust_remote_code and the other is strict config rejecting granite_speech; not the same bug." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are different architecture/configuration bugs." + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are unrelated architectural/configuration bugs." }, { "left": "issue:45987", "right": "issue:46829", "accept": false, - "reason": "StaticCache sequence-length return type and VideoPrism hidden_states assignment are unrelated." + "reason": "StaticCache sequence length return type and VideoPrism hidden_states output mismatch are different API-contract issues." }, { "left": "issue:44008", "right": "issue:46858", "accept": false, - "reason": "Gemma3n variable collision and StaticCache device-tensor return behavior are separate issues." + "reason": "Gemma3n variable collision in forward is unrelated to static cache device tensor arithmetic under torch.compile." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "SigLIP2 documentation/example errors and Kosmos2.5 OCR long-input index errors are unrelated." + "reason": "SigLIP2 documentation/model-processor mismatch and Kosmos2.5 long OCR index error are distinct issues in different examples/models." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Both are remote_code-related regressions, but they describe different breakages and should not be merged as one bug." + "reason": "Both are broad v5/remote_code regressions, but one is specifically about removal of is_torch_fx_available and the other is a general remote_code breakage report; not the same concrete defect." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are different tokenizer issues." + "reason": "LayoutLMv2 tokenizer crashes and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are unrelated tokenizer problems." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Different model-specific output regressions: Pegasus v4/v5 behavior vs CamemBERT masked LM predictions." + "reason": "Pegasus v4/v5 output differences and CamemBERT masked LM prediction errors are model-specific behavior regressions, not one bug." }, { "left": "issue:45987", "right": "issue:46932", "accept": false, - "reason": "StaticCache seq-length contract bug is unrelated to Gemma4 dtype casting for quantized embedders." + "reason": "StaticCache sequence length contract issue and Gemma4 quantized embedder dtype casting are unrelated." } ] }, @@ -6475,8 +6498,8 @@ "issue:47337" ], "soft_pairs": [ - "issue:43618|issue:43697", "issue:44805|issue:44936", + "issue:43618|issue:43697", "issue:44991|issue:46291", "issue:43322|issue:45685", "issue:45584|issue:45834", @@ -6494,208 +6517,208 @@ "issue:43644|issue:43749", "issue:45563|issue:47276", "issue:43582|issue:46512", - "issue:43931|issue:46846", "issue:43540|issue:45381", "issue:43825|issue:43994", "issue:44610|issue:45072", - "issue:34634|issue:36331", "issue:43645|issue:46665", + "issue:43931|issue:46846", + "issue:34634|issue:36331", "issue:43749|issue:44291" ], - "prepared_review_unit_hash": "a9123b0b63c4d21e6b2634555df5774b3bdb806ca01ccbbc79a46244fa3f3c48", + "prepared_review_unit_hash": "75c2c8909e383a84065343baad5af71b10c2f08d95e18686169a8f13b9b397bd", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35711, - "estimated_input_tokens": 8928, - "estimated_eval_tokens": 18112 + "serialized_chars": 35719, + "estimated_input_tokens": 8930, + "estimated_eval_tokens": 18116 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Only one pair looks like a likely duplicate: the two tokenizer issues about added-token spacing/slow-vs-fast inconsistency. The rest are distinct bugs or regressions in different models/code paths.", - "confidence": 0.86, - "canonical_issue_reason": "issue:28218 is the best canonical issue because it describes the concrete tokenizer symptom directly; issue:29159 reads as a broader wrapper/consistency report of the same behavior.", + "summary": "No pairs are strong enough to confirm as duplicates from titles alone. The tokenizer pair (28218, 29159) is the closest match, but it should be treated as related tokenizer behavior rather than the same bug. No pull requests are present.", + "confidence": 0.56, + "canonical_issue_reason": "Issue 29159 is the broadest tokenizer report and the best anchor for this sub-area, but the evidence is insufficient to collapse 28218 into it as a duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "issue:28218 is the clearest, most specific statement of the underlying bug and is the best representative of the tokenizer-spacing problem among the provided issues.", + "best_issue_reason": "29159 is the best representative of the tokenizer-related issues because it frames the behavior at the broader slow-vs-fast inconsistency level, even though the duplicate link is not strong enough to accept.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43618", - "right": "issue:43697", + "left": "issue:44805", + "right": "issue:44936", "accept": false, - "reason": "Different failures: missing CLIP attentions vs. RTDetrV2 output divergence." + "reason": "Different failure modes: mask/index shape mismatch versus trainer train/evaluate state behavior." }, { - "left": "issue:44805", - "right": "issue:44936", + "left": "issue:43618", + "right": "issue:43697", "accept": false, - "reason": "One is a tensor-mask shape mismatch; the other is trainer.evaluate() failing after training." + "reason": "CLIP attentions being dropped and RTDetr output drift are unrelated model behaviors." }, { "left": "issue:44991", "right": "issue:46291", "accept": false, - "reason": "Tokenizer loading failure is unrelated to the kernels import/version requirement error." + "reason": "Tokenizer loading failure and pipeline import/version requirement are different code paths." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Segfault while loading Llava Next models is unrelated to an MPS histogram TypeError." + "reason": "Llava Next segfault and MPS histogram TypeError are unrelated runtime failures." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation bug and Kosmos2.5 OCR long-input index error are different code paths." + "reason": "Whisper empty-transcription handling and Kosmos2.5 OCR index error are different bugs." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Runtime post-processor update bug is not the same as Whisper batch_decode ignoring skip_special_tokens." + "reason": "Special-token post-processor updates and Whisper batch_decode skipping are separate tokenizer behaviors." }, { "left": "issue:46294", "right": "issue:46463", "accept": false, - "reason": "Qwen3.5 video chat-template mismatch is unrelated to T5 decoder-input validation messaging." + "reason": "Qwen3.5 video token/feature mismatch and T5 decoder-input validation messaging are unrelated." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "CPU offload device mismatch in Bark is a different bug from bfloat16 dtype mismatches in other models." + "reason": "CPU offload device mismatch and dtype mismatch are not the same defect." }, { "left": "issue:46752", "right": "issue:47337", "accept": false, - "reason": "Empty apply_chat_template crash is unrelated to a flaky bbox-range test." + "reason": "Empty chat-template input crash and bbox-range flaky test are different problems." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Both mention inference/precision, but Bamba torch.compile SDPA failure is not the same bug as dtype mismatches." + "reason": "Torch.compile/SDPA failure is unrelated to generic dtype mismatch reports." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Qwen video-input regression and Kosmos2.5 OCR index error are distinct model/input bugs." + "reason": "Qwen video regression and Kosmos OCR index error are not the same underlying issue." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Image-preprocessor dtype mismatch and Gemma3n variable-name collision are unrelated." + "reason": "Image-preprocessor dtype mismatch and Gemma3n variable collision are unrelated." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Weight_only loading error is not the same as flash-attention generation failure." + "reason": "Weight-only load error and flash-attention generation issue differ in both symptom and path." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Input validation XOR logic and zero-loss classification behavior are different problems." + "reason": "Input validation messaging and zero-loss classification are unrelated." }, { "left": "issue:43577", "right": "issue:43818", "accept": false, - "reason": "Blip2 dtype propagation and Video-LLaVA missing temporal attention/weight sharing are unrelated." + "reason": "BLIP2 dtype retention and Video-LLaVA temporal-attention issues are different model problems." }, { "left": "issue:28218", "right": "issue:29159", - "accept": true, - "reason": "Both describe the same tokenizer spacing inconsistency around added tokens; one is the concrete symptom and the other the broader slow-vs-fast tokenizer mismatch." + "accept": false, + "reason": "Both are tokenizer-related, but the packet only shows a broad inconsistency report versus an added-token spacing symptom; not enough evidence that they are the same bug." }, { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Non-persistent buffer junk fill and FSDP CPU RAM efficient loading are separate loading regressions." + "reason": "Non-persistent buffer junk and FSDP CPU-RAM-efficient loading are different loading regressions." }, { "left": "issue:45563", "right": "issue:47276", "accept": false, - "reason": "Both are warning-message issues, but they concern different warnings in different generation modes." + "reason": "Both are warnings, but they refer to different generation features and code paths." }, { "left": "issue:43582", "right": "issue:46512", "accept": false, - "reason": "Apple Silicon warmup TypeError is unrelated to stale config.dtype after from_config(dtype=...)." - }, - { - "left": "issue:43931", - "right": "issue:46846", - "accept": false, - "reason": "Weight-shape mismatch on one Qwen3-VL model is not the same as missing linear_attn entries in TP plan." + "reason": "Apple Silicon warmup TypeError and stale config dtype after from_config are distinct dtype/loading bugs." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Video-input ValueError in Qwen3OmniMoe and qwen2.5-vl vision_position_ids regression are different bugs." + "reason": "These are different Qwen video regressions with no clear shared root cause from titles alone." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "Pipeline translation-error messaging and SigLIP2 nonsensical outputs are unrelated." + "reason": "Pipeline translation warning text and SigLIP2 output quality are unrelated." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Processor/model image-size mismatch is not the same as dtype mismatch failures." + "reason": "Processor image-size mismatch and dtype mismatch are unrelated issues." }, { - "left": "issue:34634", - "right": "issue:36331", + "left": "issue:43645", + "right": "issue:46665", "accept": false, - "reason": "BarkProcessor voice_preset failure and CustomTrainer compute_loss signature breakage are unrelated." + "reason": "Notebook custom-model initialization and from_pretrained-in-class behavior are different lifecycle bugs." }, { - "left": "issue:43645", - "right": "issue:46665", + "left": "issue:43931", + "right": "issue:46846", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 TP-plan omissions are separate loading/parallelism problems." + }, + { + "left": "issue:34634", + "right": "issue:36331", "accept": false, - "reason": "Notebook custom-model initialization breakage is not the same as from_pretrained behavior inside Python classes." + "reason": "BarkProcessor voice_preset failure and CustomTrainer signature mismatch are unrelated." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM efficient loading and init_empty_weights unexpected argument errors are different initialization/load bugs." + "reason": "FSDP CPU RAM efficient loading and init_empty_weights unexpected argument are both loading-related but not the same defect." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays mostly conservative. The one accepted soft pair, issue:28218 vs issue:29159, is reasonably supported by the titles as two descriptions of the same tokenizer spacing/slow-vs-fast inconsistency. The rest of the verdicts are appropriately negative and do not overclaim overlap." + "accept": false, + "feedback": "The overall summary is directionally fine, but the accepted soft edge overstates the evidence. From the packet we only have titles and soft similarity, so saying 28218 and 29159 are the \"same tokenizer added-token spacing inconsistency\" is stronger than the data supports. A more conservative verdict would be that they are plausibly related tokenizer issues, not clearly duplicates. The rest of the soft-edge rejections look appropriately cautious, and the no-PR note is grounded." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -6769,9 +6792,9 @@ "issue:44671|issue:45701", "issue:43742|issue:45020", "issue:47254|issue:47332", + "issue:43577|issue:43582", "issue:46697|issue:46728", "issue:42915|issue:47333", - "issue:43577|issue:43582", "issue:44589|issue:45042", "issue:43782|issue:45325", "issue:43262|issue:47217", @@ -6783,197 +6806,209 @@ "issue:45003|issue:46285", "issue:43582|issue:44857", "issue:36010|issue:36331", - "issue:45372|issue:46326", "issue:43638|issue:44936", "issue:43531|issue:45042", + "issue:45372|issue:46326", "issue:43550|issue:44912", "issue:43825|issue:43827", "issue:42175|issue:46291" ], - "prepared_review_unit_hash": "d6dab95a294c341f6c63bbdec7622bdd1324a587620ec25333d2611ba221e472", + "prepared_review_unit_hash": "5f764323ad4e141caa1d97fa4b2236a274334bcd28c62bf472bd246184ccd19a", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35855, - "estimated_input_tokens": 8964, - "estimated_eval_tokens": 18184 + "serialized_chars": 35862, + "estimated_input_tokens": 8966, + "estimated_eval_tokens": 18188 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the proposed soft pairs are mostly superficial topic overlaps (versioning, model names, training/runtime errors) rather than the same underlying bug/change. I would not merge any of the soft edges.", - "confidence": 0.98, - "canonical_issue_reason": "No single canonical issue cleanly represents this cluster; the items span unrelated failures across tokenization, model loading, training, vision/audio, and docs.", + "summary": "Mostly unrelated issue reports grouped by loose lexical similarity. I reject every soft edge: they describe different bugs, models, or subsystems. There is no real PR candidate here; among the issues, #45020 is the broadest regression-style report and the least bad canonical representative, but the cluster is not a true duplicate set.", + "confidence": 0.72, + "canonical_issue_reason": "#45020 is the broadest, most cross-cutting report in the set: a recent-version regression affecting model loading with `remote_code`, which is closer to a cluster-level umbrella than the many model-specific or subsystem-specific issues here.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the broadest and most central of the set, but it is still only a loose representative because the cluster is not actually one bug family.", + "best_issue_reason": "Best overall issue representative for this noisy cluster because it has the widest applicability and the most plausible overlap with other loading/regression reports, even though the set is not genuinely duplicate-heavy.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "Different model/backend problems: packed-weight loading vs dtype mismatches during bfloat16 inference." + "reason": "Different failures: BitNet packed-weight unpacking during accelerate loading vs dtype mismatch in bfloat16 inference for other models." }, { "left": "issue:47276", "right": "issue:47337", "accept": false, - "reason": "A warning-message issue and a flaky bbox-range test are unrelated bugs." + "reason": "A misleading warning about continuous batching is unrelated to a flaky bbox-range test." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Torch backend packaging vs Mistral tokenizer regex crash; no shared code-path bug." + "reason": "Backend installation/package-dependency issue vs a Mistral tokenizer regex crash; different code paths and symptoms." }, { "left": "issue:47254", "right": "issue:47333", "accept": false, - "reason": "Different offloading/training failures affecting different workflows and model families." + "reason": "PEFT+ZeRO-3 gradient checkpointing error vs disk-offloaded Qwen3 model save failure; not the same bug." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Metric-gathering label truncation is unrelated to Mamba2 cache correctness." + "reason": "Metric gathering tuple-label truncation and Mamba2 cache correctness are unrelated." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "Tokenizer/version behavior is too broad; one is CamemBERT MLM predictions, the other global tokenization drift." + "reason": "CamemBERT masked-LM prediction regression vs general tokenization changes; same area loosely, but not the same underlying defect." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "Generic load-time key error is not the same concrete remote_code regression." + "reason": "MobileLLM key error on load and remote_code breakage are both loading-related, but they are different concrete failure modes." }, { "left": "issue:47254", "right": "issue:47332", "accept": false, - "reason": "Checkpointing with PEFT/ZeRO-3 is unrelated to saving disk-offloaded Qwen3-VL weights." + "reason": "Gradient checkpointing/ZeRO-3 training error vs inability to save a disk-offloaded VL model." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "BLIP2 dtype handling vs an Apple Silicon allocator warmup TypeError; unrelated subsystems." }, { "left": "issue:46697", "right": "issue:46728", "accept": false, - "reason": "Both are image-size related, but the padding bug and the patch-count bug are different calculations." + "reason": "Both are image-processor bugs, but one over-pads and the other miscounts patches; different algorithms and fixes." }, { "left": "issue:42915", "right": "issue:47333", "accept": false, - "reason": "FineGrainedFP8 quantization failure is not the same as disk-offload save failure." - }, - { - "left": "issue:43577", - "right": "issue:43582", - "accept": false, - "reason": "Blip2 dtype loading and Apple Silicon allocator TypeError are unrelated." + "reason": "Qwen3MoE FineGrainedFP8 loading failure vs a disk-offloading save problem for a different Qwen3 model." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Float8 storage lookup and torchvision requirement are completely different failures." + "reason": "Float8 storage-object lookup error is unrelated to PIL backend unexpectedly requiring torchvision." }, { "left": "issue:43782", "right": "issue:45325", "accept": false, - "reason": "Qwen3-VL weight_only loading and Qwen2.5-VL rope-index scaling are unrelated." + "reason": "Qwen3-VL `weight_only=True` load error and Qwen2.5-VL rope index scaling are distinct model-specific bugs." }, { "left": "issue:43262", "right": "issue:47217", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and VL image-placeholder handling are different subsystems." + "reason": "Audio chat-template sampling-rate defaulting and VL image-placeholder handling are different template/processor issues." }, { "left": "issue:46851", "right": "issue:47337", "accept": false, - "reason": "Variable-length position_ids bug is unrelated to a bbox-coordinate flaky test." + "reason": "Variable-length position_id behavior in Qwen3.5 is unrelated to a flaky bbox-coordinate test." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "Importing GenerationMixin and Jupyter custom-model initialization are different regressions." + "reason": "GenerationMixin import failure and Jupyter custom-model initialization breakage are different compatibility issues." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Audio processor template defaults and DAC latent reconstruction mismatch do not share a code path." + "reason": "Audio processor chat-template defaulting vs DAC latent conversion mismatch; no shared bug." }, { "left": "issue:46175", "right": "issue:46932", "accept": false, - "reason": "Gemma4 one-hot patch embedding explosion vs quantized-embedder dtype casting are distinct issues." + "reason": "Gemma4 vision patch-embedder one-hot memory blowup vs quantized embedders dtype casting; same model family but not the same defect." }, { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "Tokenizer special-token registration and MXFP4 quantization loading are unrelated." + "reason": "BigBirdTokenizer special-token registration bug vs MXFP4 quantization fallback on a different model." }, { "left": "issue:45003", "right": "issue:46285", "accept": false, - "reason": "Unsafe sys.modules access is not the same as CLIP class structure inconsistency." + "reason": "Unsafe `sys.modules` access in modeling_utils and CLIP text-model structure changes are unrelated." }, { "left": "issue:43582", "right": "issue:44857", "accept": false, - "reason": "Apple Silicon allocator TypeError and CUDA AMP loss crash are different runtime failures." + "reason": "Apple Silicon M4 TypeError vs a CUDA float16 AMP crash in a loss module; different environments and causes." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "ImportError for GenerationMixin and CustomTrainer.compute_loss signature mismatch are unrelated." + "reason": "Importing `GenerationMixin` and a custom trainer kwarg mismatch are unrelated API breakages." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Zero-3/Bert index error during training and `trainer.evaluate()` failing after `train()` do not describe the same bug." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior is unrelated to PIL backend dependency requirements." }, { "left": "issue:45372", "right": "issue:46326", "accept": false, - "reason": "Both involve mistral_common, but one is a missing import and the other is continue_final_message behavior." + "reason": "Both mention mistral_common, but one is a missing import during Gemma4 processor loading and the other is a `continue_final_message` tokenizer bug." }, { - "left": "issue:43638", - "right": "issue:44936", + "left": "issue:43550", + "right": "issue:44912", "accept": false, - "reason": "ZeRO-3 index error with a non-pretrained Bert model is not the same as train-then-evaluate failure." + "reason": "Torch.compile SDPA failure in Bamba-9B-v2 and MXFP4 fallback in git-oss-20b are different model/runtime problems." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Related to v5 pipeline removals, but one is an error-message bug and the other is stale docs." + "reason": "One is an incorrect runtime error message about translation tasks; the other is docs still referencing removed pipeline APIs. Related theme, not the same issue." }, { "left": "issue:42175", "right": "issue:46291", "accept": false, - "reason": "Torch extra packaging and kernels dependency requirements are different installation/import problems." + "reason": "Missing TensorFlow backend in a torch extra install and kernels-v0.15 import/version requirements are separate dependency issues." } ] }, @@ -7060,31 +7095,31 @@ "issue:43257|issue:43696", "issue:45084|issue:45538", "issue:38175|issue:44610", - "issue:43643|issue:45759", "issue:43262|issue:44811", + "issue:43643|issue:45759", "issue:42490|issue:43257", "issue:45859|issue:47050", - "issue:43329|issue:45600", "issue:42915|issue:43742", + "issue:43329|issue:45600", "issue:43329|issue:44479", "issue:44291|issue:45042", "issue:43526|issue:44554", "issue:43329|issue:43450", - "issue:39692|issue:44912", "issue:34567|issue:40990", - "issue:44387|issue:44617", + "issue:39692|issue:44912", "issue:29159|issue:34567", + "issue:44387|issue:44617", "issue:44589|issue:47333", "issue:45081|issue:47020" ], - "prepared_review_unit_hash": "706bbdb089be613d5a9970eff2b9cf6d718e842e04ebf11f4274e676651bb377", + "prepared_review_unit_hash": "06e69fae6510e7370f35037df03e3c850fcb6129b25c9ce107700e24713e83e1", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, @@ -7099,186 +7134,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose semantic bucket of mostly unrelated bugs across tokenizer/loading/config paths, multimodal/audio/video processors, and quantization/offload regressions. None of the soft pairs look like the same underlying fix, so no duplicate edges should be accepted.", - "confidence": 0.86, - "canonical_issue_reason": "No single canonical issue; the cluster is heterogeneous and grouped by broad similarity rather than one root cause.", + "summary": "No soft duplicate pairs should be merged here. The cluster is a grab-bag of unrelated tokenizer, multimodal, quantization, offloading, and model-loading bugs; several pairs share broad subsystem language but not the same concrete failure mode.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 43329 is the best representative anchor because it is a concrete shared-helper bug affecting multiple VLM processors, but it still only covers one subset of the cluster.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer/model-loading failures, but they involve different code paths and different root causes (gguf file handling vs EST-RoBERTa tokenizer loading)." + "reason": "Both involve tokenizer loading, but they are different failure modes: gguf_file handling for AutoTokenizer/AutoModel vs a specific EMBEDDIA/est-roberta tokenizer regression." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Mistral regex patch crash and Gemma4Processor missing _tokenizer are separate loading bugs in different components." + "reason": "Different code paths and symptoms: Mistral regex patching crash in tokenizer setup vs Gemma4Processor missing _tokenizer in transformers serve." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Qwen3 MOE weight conversion with accelerate/deepspeed is unrelated to the multimodal video token-counting helper bug." + "reason": "Qwen3 MOE weight conversion with accelerate/deepspeed is unrelated to a VLM video-token counting bug in _get_num_multimodal_tokens." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "DAC latent reconstruction mismatch and DiaConfig save/load token-ID loss are different model/config problems." + "reason": "DAC latent reconstruction mismatch is unrelated to DiaConfig losing custom token IDs after save/load." }, { "left": "issue:43906", "right": "issue:46209", "accept": false, - "reason": "The reproduction of issue 38071 is not the same as the B200 FP8 inference failure." + "reason": "The first is an isolated reproduction of another issue; the second is a B200 FP8 inference problem. Not the same bug." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BEIT Fast reduce_labels behavior and RTDetrV2 output drift in v5 are distinct bugs." + "reason": "BeitImageProcessorFast reduce_labels truncation and RTDetrV2 output differences are separate model/processor issues." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "These are both tied-weights warnings, but they come from different models and different warning logic." + "reason": "One is an incorrect tied-weights warning in Qwen3-Next internals; the other is a user-facing warning during ms-swift LoRA fine-tuning. Same family, different problem." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "GLM-4.7-Flash test loading and Qwen3.5 flash-attention generation are unrelated model-specific issues." + "reason": "GLM-4.7-Flash load failure in tests is not the same as Qwen3.5 flash-attention generation issues." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Qwen3Moe FP8 loading failure and AutoImageProcessor URL loading are different subsystems." + "reason": "FineGrainedFP8Config failure for Qwen3Moe is unrelated to AutoImageProcessor URL loading." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "GraniteSpeech PEFT local checkpoint loading and Whisper batch_decode skipping specials are separate behaviors." + "reason": "PEFT adapter checkpoint loading and Whisper batch_decode skip_special_tokens are different processor/model-loading bugs." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "Accelerate/deepspeed Qwen3 MOE conversion is not the same bug as GPT-oss-20b CUDA OOM." + "reason": "Both mention Qwen3 and loading/runtime issues, but one is MOE weight conversion with accelerate/deepspeed and the other is an outright CUDA OOM on GPT-oss-20b." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template compilation error and CLIPTokenizer max-length bug are unrelated." + "reason": "Template compilation error and CLIPTokenizer model_max_length misuse are unrelated." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "SigLIP2 zero probabilities and OmDet-Turbo processor size mismatch are different model/processor regressions." + "reason": "SigLIP2 zero-probabilities and OmDet-Turbo 640px-vs-224px preprocessing mismatch are different bugs." }, { - "left": "issue:43643", - "right": "issue:45759", + "left": "issue:43262", + "right": "issue:44811", "accept": false, - "reason": "Missing fields from trust_remote_code AutoConfig and text_config unwrapping in AutoModelForCausalLM are distinct config-loading bugs." + "reason": "Audio chat-template sampling-rate defaulting and Whisper batch_decode token skipping are separate audio-processing issues." }, { - "left": "issue:43262", - "right": "issue:44811", + "left": "issue:43643", + "right": "issue:45759", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and Whisper batch_decode skip_special_tokens are unrelated processor behaviors." + "reason": "Missing fields from trust_remote_code AutoConfig and not unwrapping text_config in composite multimodal configs are distinct config-loading problems." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Trailing-slash local model path warning and Qwen3 MOE weight conversion are unrelated." + "reason": "Trailing-slash path import warning is unrelated to Qwen3 MOE weight conversion." }, { "left": "issue:45859", "right": "issue:47050", "accept": false, - "reason": "CPU offload incompatibility in Qwen3_5Moe rotary embedding and Canine ONNX dtype mismatch are different failures." + "reason": "CPU offload incompatibility in Qwen3_5MoeTextRotaryEmbedding is unrelated to a CanineModel fp16 ONNX dtype mismatch." }, { - "left": "issue:43329", - "right": "issue:45600", + "left": "issue:42915", + "right": "issue:43742", "accept": false, - "reason": "Undefined video-branch helpers and stale auto_mappings config references are not the same root cause." + "reason": "FineGrainedFP8Config failure and MobileLLM-125M key error do not describe the same defect." }, { - "left": "issue:42915", - "right": "issue:43742", + "left": "issue:43329", + "right": "issue:45600", "accept": false, - "reason": "Qwen3Moe FP8 failure and MobileLLM key error do not share a concrete code path." + "reason": "Video-token counting bug in multimodal processors is unrelated to auto_mappings.py referencing removed Sam3LiteText configs." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "Both touch video-related paths, but one is an undefined-helper bug and the other is a model-family regression." + "reason": "The first is a missing/undefined helper in video token counting; the second is a broader v5.3.0 video-input regression across several Qwen models." }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "Unexpected _is_hf_initialized argument and PIL backend requiring torchvision are separate loading/backend issues." + "reason": "init_empty_weights/_is_hf_initialized TypeError and PIL backend requiring torchvision are separate compatibility issues." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "BEIT label reduction and MPS attention correctness are unrelated." + "reason": "reduce_labels returning one label and an MPS attention correctness bug with differing head dims are unrelated." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Shared video theme is too broad; undefined helper variables and batched-shape errors are different bugs." - }, - { - "left": "issue:39692", - "right": "issue:44912", - "accept": false, - "reason": "SigLIP2 doc example errors and MXFP4 quantization fallback are unrelated." + "reason": "Video token counting undefined names and batched video processor output shape are different bugs in adjacent areas, not one concrete failure." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "Trainer token accounting and GPT-oss perplexity are not the same issue." + "reason": "TrainerState token-count tracking and GPT-oss perplexity are unrelated training/runtime issues." }, { - "left": "issue:44387", - "right": "issue:44617", + "left": "issue:39692", + "right": "issue:44912", "accept": false, - "reason": "Int4 reserved-memory regression and Sam3Video OOM are both memory symptoms, but with different causes and code paths." + "reason": "SigLIP2 documentation errors do not match a GPT-oss MXFP4 quantization fallback problem." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Tokenizer inconsistency and TrainerState token counter not updating are different components." + "reason": "Tokenizer inconsistency and TrainerState num_input_tokens_seen not updating are different subsystems and symptoms." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both mention OOM, but one is int4 quantization reserved-memory growth and the other is Sam3Video running out of memory; no shared concrete root cause." }, { "left": "issue:44589", "right": "issue:47333", "accept": false, - "reason": "Float8 storage lookup failure and disk-offload save failure are unrelated quantization/offload problems." + "reason": "Float8 storage lookup failure is not the same as save failures for disk-offloaded Qwen3-30B-A3B." }, { "left": "issue:45081", "right": "issue:47020", "accept": false, - "reason": "Mistral regex patch crash and XLMRobertaTokenizer/Unigram type mismatch are different tokenizer bugs." + "reason": "Mistral regex patching crash and XLMRobertaTokenizer Unigram initialization are distinct tokenizer bugs." } ] }, @@ -7359,209 +7394,209 @@ "issue:44912|issue:47145", "issue:28282|issue:29870", "issue:43650|issue:43906", - "issue:33357|issue:34567", "issue:28282|issue:34689", + "issue:33357|issue:34567", "issue:43540|issue:46701", "issue:44568|issue:45356", "issue:43454|issue:43577", - "issue:46421|issue:46424", "issue:44291|issue:44821", - "issue:43749|issue:44756", + "issue:46421|issue:46424", "issue:43742|issue:44291", + "issue:43749|issue:44756", "issue:44936|issue:46821", "issue:42175|issue:43824", - "issue:43404|issue:45127", - "issue:44589|issue:47332", "issue:24643|issue:29159", + "issue:44589|issue:47332", "issue:28282|issue:29127", "issue:44479|issue:46693", + "issue:43404|issue:45127", "issue:42915|issue:45538", "issue:43697|issue:45357" ], - "prepared_review_unit_hash": "6aeaa8488f783ec6c185e3b15d217525392702844c5d358c113bfb0c4827bb14", + "prepared_review_unit_hash": "19cce58d12b678c05b46c1b14eb749ae30488625a1cd980522347229ac290c68", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 34979, - "estimated_input_tokens": 8745, - "estimated_eval_tokens": 17746 + "serialized_chars": 34982, + "estimated_input_tokens": 8746, + "estimated_eval_tokens": 17748 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft pairs are almost all false positives: they share only broad themes like loading, tokenizers, cache, or quantization, but each points to a different model, code path, or failure mode. No duplicate-quality merges are justified.", - "confidence": 0.74, - "canonical_issue_reason": "issue:43749 is the strongest representative if one must be chosen: it is a concrete runtime/loading regression with the most engagement and inbound references, while many others are narrower or more model-specific.", - "canonical_pr_reason": null, - "best_issue_reason": "issue:43749 is the best overall issue candidate because it is high-signal, actionable, and broadly worded compared with the more niche or isolated reports in the set.", - "best_pr_reason": null, + "summary": "This cluster is very heterogeneous: most pairs are only loosely similar by theme (loading, tokenizers, quantization, multimodal, cache, trainer issues) but do not look like the same underlying bug. I would not merge any of the soft-edge pairs.", + "confidence": 0.63, + "canonical_issue_reason": "No single issue cleanly represents a duplicate set here; if a representative must be chosen, issue 43749 is the broadest/highest-activity infrastructure regression and most plausible cluster anchor.", + "canonical_pr_reason": "No pull requests are present in the cluster.", + "best_issue_reason": "Issue 43749 is the strongest representative candidate because it has the most discussion/inbound references and describes a broad runtime regression, making it a better cluster anchor than the highly model-specific reports.", + "best_pr_reason": "No pull requests are present, so there is no best PR candidate.", "soft_edge_verdicts": [ { "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "Both involve Qwen3 model loading, but one is a MoE weight conversion problem and the other is a vision model shape-mismatch error; different bugs." + "reason": "Different models and different failure modes: Qwen3 MOE conversion vs Qwen3-VL weight-shape loading mismatch." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "One is a loading failure for a specific model, the other is tokenizer max_length behavior; no shared underlying fix." + "reason": "Unrelated bugs: one is a model-loading failure on a specific repo, the other is tokenizer max-length handling." }, { "left": "issue:43425", "right": "issue:46291", "accept": false, - "reason": "Different dependency failures: Torch incompatibility versus kernels import/version requirements." + "reason": "Torch compatibility and pipeline import/version requirements are separate issues with different root causes." }, { "left": "issue:43525", "right": "issue:46291", "accept": false, - "reason": "Unrelated issues: missing pad_token_id on Llama4Config versus a pipelines import failure from kernels." + "reason": "Missing pad_token_id on Llama4Config is unrelated to kernels dependency validation on pipeline import." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both mention exception handling, but they are different functions and different failure modes; not the same bug." + "reason": "Both mention broad exception handling, but they are distinct code paths and distinct bugs; not the same change." }, { "left": "issue:44912", "right": "issue:47145", "accept": false, - "reason": "Different quantization formats and failures: MXFP4 fallback versus FP8 frozen checkpoint scaling omission." + "reason": "Both concern quantized checkpoint loading, but one is MXFP4 fallback behavior and the other is FP8 weight_scale omission." }, { "left": "issue:28282", "right": "issue:29870", "accept": false, - "reason": "ImportError for missing PyTorch is unrelated to a Bart generate() UnboundLocalError." + "reason": "PyTorch-not-found import error is unrelated to Bart generate() decoder_attention_mask assignment." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "A vague data request is not the same underlying issue as an isolated reproduction of another bug." + "reason": "A vague issue placeholder is not the same as a reproduction of a specific prior bug." }, { - "left": "issue:33357", - "right": "issue:34567", + "left": "issue:28282", + "right": "issue:34689", "accept": false, - "reason": "MacOS CLIP bus error and TrainerState token counting are unrelated." + "reason": "Generic PyTorch import error is not the same as a model-loading regression for Llama 3.2 Vision." }, { - "left": "issue:28282", - "right": "issue:34689", + "left": "issue:33357", + "right": "issue:34567", "accept": false, - "reason": "Both are loading-related, but one is missing PyTorch and the other is a specific model-loading regression." + "reason": "MacOS CLIP bus error and TrainerState token-count tracking are unrelated." }, { "left": "issue:43540", "right": "issue:46701", "accept": false, - "reason": "Different multimodal pipelines and different errors; no common code-path bug." + "reason": "Different multimodal failures in different models and pipelines; no shared concrete code path." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different tokenizers and different behaviors." + "reason": "Both are tokenizer regressions, but they affect different tokenizers and stem from different causes." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "Weight tying in AyaVision is unrelated to Blip2 dtype propagation." - }, - { - "left": "issue:46421", - "right": "issue:46424", - "accept": false, - "reason": "Same broad cache area, but one is lazy init recompilation and the other is max_cache_length handling; distinct fixes." + "reason": "One is broken lm_head weight tying; the other is dtype propagation for Blip2/QFormer." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "init_empty_weights TypeError and AutoImageProcessor URL loading are different code paths and errors." + "reason": "init_empty_weights argument handling and AutoImageProcessor URL loading are unrelated." }, { - "left": "issue:43749", - "right": "issue:44756", + "left": "issue:46421", + "right": "issue:46424", "accept": false, - "reason": "Both are memory/loading related, but one is FSDP RAM-efficient loading and the other is mmap OOM on Strix Halo; not the same bug." + "reason": "Both are cache-related, but one is recompilation/lazy init and the other is cache_length configuration semantics." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "Different model-loading failures with different triggers and stack traces." + "reason": "KeyError during model loading is not the same as init_empty_weights rejecting _is_hf_initialized." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU/RAM-efficient loading and Strix Halo mmap OOM are different loading/memory issues." }, { "left": "issue:44936", "right": "issue:46821", "accept": false, - "reason": "Trainer lifecycle bug versus a TimesFM-specific window_size AttributeError." + "reason": "Trainer train/evaluate sequencing bug is unrelated to TimesFM window_size AttributeError." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Backend installation/package selection issue is unrelated to a missing class import." + "reason": "Backend packaging/dependency issue and missing Qwen2.5-VL export are unrelated." }, { - "left": "issue:43404", - "right": "issue:45127", + "left": "issue:24643", + "right": "issue:29159", "accept": false, - "reason": "Both touch tied weights, but one is a missing tie in a model class and the other is a LoRA-merge regression with extended vocabulary." + "reason": "DeepSpeed weight-dimension error is unrelated to tokenizer slow/fast inconsistency." }, { "left": "issue:44589", "right": "issue:47332", "accept": false, - "reason": "Float8 storage lookup failure and disk-offloaded save failure are different bugs." - }, - { - "left": "issue:24643", - "right": "issue:29159", - "accept": false, - "reason": "DeepSpeed training weight-shape error is unrelated to slow/fast tokenizer inconsistency." + "reason": "FP8 storage lookup and disk-offloaded save failure are different bugs." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Missing PyTorch import error and LayoutLMv3 box-validation messaging are unrelated." + "reason": "Missing PyTorch import and LayoutLMv3 error messaging are unrelated." }, { "left": "issue:44479", "right": "issue:46693", "accept": false, - "reason": "Both are regressions, but one is a video-input multimodal issue and the other is a long-context performance regression." + "reason": "Both are regressions, but one is video-input behavior across Qwen VL models and the other is FA2 long-context performance." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both involve tied embeddings, but one is a specific model weight-tying bug and the other is LoRA merge behavior with extended vocab." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "Qwen3Moe FP8 config failure and CLIPTokenizer max_length bug do not share the same underlying defect." + "reason": "Qwen3Moe FP8 config failure is unrelated to CLIPTokenizer model_max_length." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RTDetrV2 output drift and Qwen3.5 save_pretrained key corruption are different regressions." + "reason": "RTDetr output differences in v5 and Qwen3.5 visual encoder key saving are unrelated regressions." } ] }, @@ -7637,9 +7672,9 @@ "soft_pairs": [ "issue:45137|issue:45405", "issue:44062|issue:44991", - "issue:44568|issue:45488", "issue:43582|issue:47254", "issue:43531|issue:44821", + "issue:44568|issue:45488", "issue:39379|issue:44554", "issue:44857|issue:46832", "issue:45538|issue:46869", @@ -7653,10 +7688,10 @@ "issue:44062|issue:45081", "issue:42175|issue:44488", "issue:44479|issue:45381", - "issue:42915|issue:47332", "issue:45042|issue:45081", - "issue:44987|issue:45907", + "issue:42915|issue:47332", "issue:43577|issue:46512", + "issue:44987|issue:45907", "issue:43493|issue:43994", "issue:38175|issue:44898", "issue:43742|issue:44589", @@ -7664,205 +7699,211 @@ "issue:44849|issue:45701", "issue:43742|issue:47333" ], - "prepared_review_unit_hash": "a24c50367fd7a975b3895245895644c90cae737215856bfc8cbed0cf5aed353f", + "prepared_review_unit_hash": "a338573a5ea32be8091588e79bbf09b766fc87c1ab1d5fe41e4323ca7936dde8", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35918, - "estimated_input_tokens": 8980, - "estimated_eval_tokens": 18216 + "serialized_chars": 35923, + "estimated_input_tokens": 8981, + "estimated_eval_tokens": 18218 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous issue set; the only plausible duplicate-like pair is the SigLIP2 pair, but the evidence is title-level and should be treated as tentative rather than definitive. No PRs are present.", - "confidence": 0.57, - "canonical_issue_reason": "issue:43493 is the best anchor for the SigLIP2 cluster because it is the more diagnostic title, but the link to issue:43994 should be treated as a plausible same-bug candidate rather than a proven root-cause/symptom pairing.", + "summary": "This cluster is mostly heterogeneous: tokenizer, dtype, multimodal, training, and device-mapping bugs are all mixed together. I only see two plausible duplicate groups: the Qwen video-input regression pair and the SigLIP2 output-discrepancy pair.", + "confidence": 0.88, + "canonical_issue_reason": "issue:44479 is the broadest and clearest representative of a real duplicate set here: it frames the Qwen video-input regression across multiple variants, while 45381 is a narrower manifestation of the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43493 is the strongest standalone issue to anchor the SigLIP2 case: it names the implementation discrepancy directly and is more specific than the downstream behavior report.", + "best_issue_reason": "issue:44479 is the best overall issue representative because it describes the widest-impact regression in the set and cleanly subsumes a more specific duplicate report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 deque error vs PEFT version pinning; unrelated failures." + "reason": "Different problems: DeepSpeed ZeRO3 deque handling vs a PEFT version bump." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "Tokenizer AddedToken argument bug vs loading est-roberta tokenizer; different code paths." - }, - { - "left": "issue:44568", - "right": "issue:45488", - "accept": false, - "reason": "Different tokenizer regressions affecting different models and mechanisms." + "reason": "Both involve tokenizers, but one is an AddedToken constructor error and the other is a tokenizer load failure for a specific model." }, { "left": "issue:43582", "right": "issue:47254", "accept": false, - "reason": "Apple Silicon allocator TypeError vs PEFT/ZeRO-3 checkpointing error; not the same bug." + "reason": "Apple Silicon allocator warmup TypeError is unrelated to PEFT + ZeRO-3 gradient-checkpointing checkpoint errors." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding-window issue vs AutoImageProcessor URL loading; unrelated." + "reason": "Qwen3-MoE sliding-window handling and AutoImageProcessor-from-URL loading are different code paths and different subsystems." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different tokenizer implementations and distinct failure modes." }, { "left": "issue:39379", "right": "issue:44554", "accept": false, - "reason": "Different attention/masking bugs on different platforms and model paths." + "reason": "mllama mask handling is not the same as the MPS attention correctness issue with mismatched value/query head dims." }, { "left": "issue:44857", "right": "issue:46832", "accept": false, - "reason": "Both are detection-model failures, but the crash conditions and code paths differ." + "reason": "RT-DETR AMP crash and RTDetrModel feature-level mismatch are separate bugs." }, { "left": "issue:45538", "right": "issue:46869", "accept": false, - "reason": "Tokenizer max length constant vs YOLO multi-GPU device-map failure; unrelated." + "reason": "CLIPTokenizer max-length behavior is unrelated to YOLO auto device-mapping failures." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Qwen3Moe FP8 config failure vs FSDP CPU RAM efficient loading; different features." + "reason": "FineGrainedFP8Config on Qwen3-MoE and FSDP CPU RAM efficient loading are different loading/training paths." }, { "left": "issue:27957", "right": "issue:28282", "accept": false, - "reason": "XLMRoberta Flash Attention 2 issue vs missing PyTorch import error; not the same underlying bug." + "reason": "XLMRoberta Flash Attention 2 support is unrelated to missing-PyTorch ImportError." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both involve video inputs, but one is Qwen3OmniMoe-specific and the other is a broader v5.3.0 regression." + "reason": "Both are Qwen video-input issues, but they involve different model families and different concrete failures, so they are not clearly the same bug." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Different model families and error symptoms; only broad video-input overlap." + "reason": "Qwen3OmniMoe video ValueError and Qwen3-vl-embedding StopIteration are different symptoms on different code paths." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Auto device mapping CUDA assert vs video-input ValueError; unrelated." + "reason": "Auto device mapping on Qwen3 and Qwen3OmniMoe video-processing errors are unrelated." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token-count tracking vs compute_loss signature mismatch; different trainer issues." + "reason": "TrainerState token-count tracking and a custom compute_loss kwarg mismatch are separate trainer bugs." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "GGUF loader breakage vs Apple Silicon allocator error; unrelated." + "reason": "gguf_file loading failures and Apple Silicon allocator TypeErrors do not share a concrete code path." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer-loading bugs, but they affect different tokenizer components and failure modes." + "reason": "AddedToken special-argument duplication is not the same as the Mistral regex patch crashing on a missing backend_tokenizer." }, { "left": "issue:42175", "right": "issue:44488", "accept": false, - "reason": "Packaging/backend dependency issue vs model loading failure; not the same bug." + "reason": "Packaging/backend dependency inclusion is unrelated to the sleng-bert loading failure." }, { "left": "issue:44479", "right": "issue:45381", - "accept": false, - "reason": "Related Qwen video-input regressions, but one is broad and the other is a specific symptom; not clearly the same fix." + "accept": true, + "reason": "Same underlying Qwen video-input regression in transformers 5.3.0; 45381 is a specific qwen2.5-vl manifestation of the broader bug in 44479." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image processor torchvision dependency vs Mistral regex patch tokenizer crash; unrelated." + "reason": "Image-processor torchvision dependency regression and Mistral tokenizer regex patch failure are unrelated." }, { - "left": "issue:44987", - "right": "issue:45907", + "left": "issue:42915", + "right": "issue:47332", "accept": false, - "reason": "Model loading failure vs hub API revision-forwarding bug; different layers." + "reason": "Qwen3-MoE FP8 configuration failure and disk-offloaded model save failure are distinct issues." }, { "left": "issue:43577", "right": "issue:46512", "accept": false, - "reason": "Both concern dtype handling, but one is about loaded module dtypes and the other about config.dtype staleness." + "reason": "Both mention dtype, but one is model/qformer weights not casting and the other is config.dtype going stale after from_config." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model-loading regression for physical-intelligence/fast is unrelated to forwarding revision in list_repo_files." }, { "left": "issue:43493", "right": "issue:43994", "accept": true, - "reason": "Likely the same SigLIP2 underlying bug: one reports an HF-vs-JAX discrepancy, the other reports bad downstream behavior on the same model family, though the evidence here is only title-level." + "reason": "Both describe the same SigLIP2 malfunction on google/siglip2-base-patch16-224: one as HF-vs-JAX discrepancy, the other as nonsensical AutoModel/pipeline outputs." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "Different model families and failure modes; only superficial image-model overlap." + "reason": "SigLIP2 zero-probability behavior and Perceiver interpolate_pos_encoding failure are different model families and bugs." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "Key loading error for MobileLLM vs missing Float8 storage type; unrelated." + "reason": "MobileLLM key error and Float8 storage lookup failure are unrelated loading/runtime errors." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Janus image-generation test failure vs Phi-4 multimodal load failure; different issues." + "reason": "Janus image-generation test failure and Phi-4 multimodal loading failure are different model-specific problems." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 hidden-state bug vs tokenization changes across versions; unrelated." + "reason": "Qwen3.5 output_hidden_states behavior is unrelated to version-dependent tokenization changes." }, { "left": "issue:43742", "right": "issue:47333", "accept": false, - "reason": "Different loading/offloading failures for different models." + "reason": "MobileLLM key error and Qwen3 offloaded-save failure are unrelated." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The overall triage is mostly conservative, but one key claim is a bit too strong for the evidence available in the packet: the SigLIP2 pair (issue:43493 vs issue:43994) is plausible, but the writeup treats 43493 as the clear root-cause report and 43994 as a downstream symptom without body-level evidence to support that distinction. Because we only have titles and similarity signals here, that should be framed more tentatively. The rest of the soft-edge rejections look reasonably cautious, and the absence of PRs is grounded." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The two accepted duplicate pairs are the strongest matches by title/body theme, and the rejected soft pairs are mostly separated by model family or bug class. No clear overstatement of evidence." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -7910,6 +7951,7 @@ "issue:45072", "issue:45325", "issue:45405", + "issue:45588", "issue:45600", "issue:45663", "issue:45672", @@ -7926,24 +7968,23 @@ "issue:46899", "issue:47276", "issue:47332", - "issue:47404", "issue:47405" ], "soft_pairs": [ - "issue:43257|issue:43756", "issue:43756|issue:45325", - "issue:43716|issue:46701", + "issue:43257|issue:43756", "issue:43756|issue:46399", + "issue:43716|issue:46701", "issue:45823|issue:46463", "issue:45672|issue:45674", "issue:46463|issue:47276", "issue:44737|issue:46463", "issue:45405|issue:45600", - "issue:43257|issue:47405", "issue:43906|issue:46693", + "issue:43257|issue:47405", "issue:24643|issue:29870", - "issue:44568|issue:46491", "issue:43577|issue:45072", + "issue:44568|issue:46491", "issue:44625|issue:44877", "issue:43540|issue:46693", "issue:43452|issue:44488", @@ -7957,218 +7998,221 @@ "issue:44913|issue:46121", "issue:43697|issue:46829", "issue:45663|issue:46846", - "issue:46846|issue:47404", - "issue:43653|issue:44361" + "issue:43653|issue:44361", + "issue:44737|issue:45588" ], - "prepared_review_unit_hash": "5f002c2980080aa1757c48e8a3c1d54aad6878faa0ca38a2eb25152c95bc408d", + "prepared_review_unit_hash": "c89520898d5d8ded27a3311c0765590f2e026b594ff11d3cff19f0ec61e2449c", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 37062, - "estimated_input_tokens": 9266, - "estimated_eval_tokens": 18788 + "serialized_chars": 37028, + "estimated_input_tokens": 9257, + "estimated_eval_tokens": 18770 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: tokenizer, config, dtype, quantization, generation, training, and loading issues with no clear duplicate core. No soft edge pair looks like the same concrete bug, so all are rejected.", - "confidence": 0.89, - "canonical_issue_reason": "No true canonical issue stands out; the issues cover unrelated failure modes across many subsystems. If one must be chosen, 44466 is the closest representative because it is a concrete v5 tied-weights regression, but it does not generalize to the rest of the cluster.", + "summary": "Mostly a heterogeneous set of unrelated issues. The only duplicate-like pair is the v5 tied-weights regression on translation models; the rest differ by model, subsystem, or failure mode and should stay separate.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 45005 is the broadest representative of the tied-weights v5 regression, and 44466 reads like a concrete device-dependent repro of that same problem.", "canonical_pr_reason": null, - "best_issue_reason": "44466 is the most specific and actionable issue in the set, but the cluster is too diverse for it to serve as a strong global representative.", + "best_issue_reason": "Issue 45005 is the best overall issue candidate because it captures the underlying tied-weights regression more broadly than the specific lm_head.weight serialization symptom in 44466.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Both touch model internals for different families, but one is Smollm3 RoPE-layer selection and the other is Qwen2.5-VL temporal position scaling; different bugs." + }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Different bugs: Qwen3 MoE conversion failure vs Smollm3 RoPE-layer mismatch." + "reason": "Qwen3 MoE weight-conversion/loading issue vs Smollm3 RoPE-layer dropping; unrelated code paths and symptoms." }, { "left": "issue:43756", - "right": "issue:45325", + "right": "issue:46399", "accept": false, - "reason": "Both are model-specific, but one is RoPE layer selection and the other is temporal position-id scaling." + "reason": "Smollm3 RoPE configuration bug and Qwen3.5 kernel-registration crash are unrelated." }, { "left": "issue:43716", "right": "issue:46701", "accept": false, - "reason": "Different code paths: image preprocessor dtype mismatch vs diffusion pipeline generation output failure." - }, - { - "left": "issue:43756", - "right": "issue:46399", - "accept": false, - "reason": "Unrelated failures: Smollm3 RoPE config bug vs Qwen3.5 kernel registration error." + "reason": "Different models and failures: Mistral-3 dtype mismatch in preprocessing vs DiffusionGemma pipeline generation breakage." }, { "left": "issue:45823", "right": "issue:46463", "accept": false, - "reason": "Different subsystems: Gemma4 device placement bug vs T5 decoder-input validation message." + "reason": "Gemma4 device-map forward mismatch and T5 decoder-input validation are not the same bug." }, { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Both involve quantized inference, but one is torch.finfo on uint8 audio embedders and the other is skip-module handling in BitsAndBytesConfig." + "reason": "Both involve 4-bit inference, but one is a uint8 finfo crash in Gemma4 audio modules and the other is a BitsAndBytesConfig exclusion regression; distinct fixes." }, { "left": "issue:46463", "right": "issue:47276", "accept": false, - "reason": "T5 validation error and synced_gpus warning text are unrelated." + "reason": "Incorrect decoder-input validation/error message vs a synced_gpus warning in continuous batching; unrelated." }, { "left": "issue:44737", "right": "issue:46463", "accept": false, - "reason": "XLNet CPU allocation bug vs T5 decoder-input validation; different models and problems." + "reason": "XLNet CPU device placement bug vs T5 validation logic; no shared underlying defect." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "PEFT version bump and stale auto_mappings refs are unrelated maintenance issues." + "reason": "PEFT version bump problem and stale auto_mappings config references are unrelated maintenance issues." }, { - "left": "issue:43257", - "right": "issue:47405", + "left": "issue:43906", + "right": "issue:46693", "accept": false, - "reason": "Both mention conversion, but one is missing MoE weight conversion and the other is silent wrong-Auto-class conversion skip." + "reason": "One is an isolated reproduction of an older issue; the other is an FA2 long-context performance regression. Not the same bug." }, { - "left": "issue:43906", - "right": "issue:46693", + "left": "issue:43257", + "right": "issue:47405", "accept": false, - "reason": "One is a repro of an old issue; the other is a FA2 long-context performance regression." + "reason": "Qwen3 MoE weight conversion problem vs silent wrong-Auto-class conversion skip; different loading failures." }, { "left": "issue:24643", "right": "issue:29870", "accept": false, - "reason": "Different bugs: DeepSpeed training weight-shape error vs Bart generate decoder mask assignment bug." + "reason": "DeepSpeed training RuntimeError about weight dimensionality and Bart generate unbound decoder_attention_mask are unrelated." }, { - "left": "issue:44568", - "right": "issue:46491", + "left": "issue:43577", + "right": "issue:45072", "accept": false, - "reason": "Both are tokenizer regressions, but one is BOS/EOS insertion and the other is leading-space round-trip preservation." + "reason": "Both mention dtype, but they affect different model families and code paths; not one concrete bug." }, { - "left": "issue:43577", - "right": "issue:45072", + "left": "issue:44568", + "right": "issue:46491", "accept": false, - "reason": "Different dtype problems in different models: Blip2 stays float32 vs SwitchTransformers/TimmWrapper bf16 mismatch." + "reason": "Different tokenizer regressions: missing BOS/EOS addition vs leading-space round-trip stripping." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation and strict Granite config loading are separate config issues." + "reason": "Qwen3.5 num_labels propagation and strict config loading for granite_speech are separate config issues." }, { "left": "issue:43540", "right": "issue:46693", "accept": false, - "reason": "Video input processing error vs FA2 long-context performance regression; unrelated failures." + "reason": "Qwen3OmniMoe video-input value error vs FA2 performance regression; unrelated." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "gguf_file loading breaks generic auto classes, while cjvt/sleng-bert is a model-specific load failure." + "reason": "Both are model/tokenizer loading complaints, but the gguf_file breakage and sleng-bert loading failure have different causes." }, { "left": "issue:43742", "right": "issue:47332", "accept": false, - "reason": "MobileLLM key error and disk-offloaded Qwen3-VL save failure are different loading/offloading problems." + "reason": "MobileLLM loading KeyError vs inability to save a disk-offloaded Qwen3-VL model; different subsystems." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "Missing PyTorch import and _gpu_broadcast_one NoneType are unrelated environment vs trainer-runtime errors." + "reason": "Missing PyTorch import error and Trainer broadcast NoneType error are unrelated." }, { "left": "issue:42796", "right": "issue:43906", "accept": false, - "reason": "Tokenizer BPE-marker handling and an isolated reproduction of another issue are not the same bug." + "reason": "Tokenizer BPE-marker handling and an isolated reproduction of another issue are not the same underlying bug." }, { "left": "issue:44466", "right": "issue:45005", - "accept": false, - "reason": "Both mention tied weights, but one is device-dependent serialization and the other is a broader translation-model complaint." + "accept": true, + "reason": "Same v5 tied-weights regression: 44466 is a concrete device-dependent serialization symptom, while 45005 describes the broader tied-weights problem on translation models." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "Gemma3 ZeRO-3 loading index error and trainer.evaluate after train are different runtime paths." + "reason": "ZeRO-3 load-time index error for Gemma3 and trainer.evaluate() failing after training are unrelated." }, { "left": "issue:46463", "right": "issue:46899", "accept": false, - "reason": "T5 decoder validation and Gemma4 multimodal 4-bit dtype cast are unrelated." + "reason": "T5 decoder validation and Gemma4 4-bit multimodal dtype casting are different bugs." }, { "left": "issue:43452", "right": "issue:44843", "accept": false, - "reason": "Generic gguf loading breakage and offline mistral regex patch behavior are different Auto* failures." + "reason": "gguf_file loading breakage and offline model_info() patching are separate from_pretrained issues with different causes." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "GPTNeoX rotary_pct reload bug and rope-validation helper type error are separate config serialization issues." + "reason": "Both involve RoPE config handling, but one is persistence of rotary_pct on reload and the other is a TypeError in validation input handling." }, { "left": "issue:43697", "right": "issue:46829", "accept": false, - "reason": "RTDetrV2 output divergence and VideoPrism hidden_states mix-up are different model-output bugs." + "reason": "Different vision-model output mismatches with different outputs being wrong; no shared code path." }, { "left": "issue:45663", "right": "issue:46846", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states KeyError vs Qwen3.5 tp-plan omissions are unrelated training/sharding bugs." + "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Qwen3.5 tensor-parallel plan omission are unrelated distributed-training bugs." }, { - "left": "issue:46846", - "right": "issue:47404", + "left": "issue:43653", + "right": "issue:44361", "accept": false, - "reason": "TP-plan missing entries and device_map loading OOM from conversion transients are different failure modes." + "reason": "Both are tokenizer-related, but one is special-token registration/empty decode and the other is an AttributeError in MLuke tasks." }, { - "left": "issue:43653", - "right": "issue:44361", + "left": "issue:44737", + "right": "issue:45588", "accept": false, - "reason": "BigBird mask-token registration and MLuke task AttributeError are unrelated tokenizer issues." + "reason": "XLNet CPU arange bug and flash_attention s_aux=None crash are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The judgment is grounded in the packet: the one accepted pair, 44466/45005, is a plausible shared v5 tied-weights regression, and the rest of the pairwise rejections stay conservative by distinguishing model, subsystem, or failure mode. The overall summary does not overclaim beyond the available titles and pair signals." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -8187,6 +8231,8 @@ "nodes": [ "issue:29159", "issue:33357", + "issue:33666", + "issue:36010", "issue:38175", "issue:40444", "issue:43295", @@ -8198,14 +8244,12 @@ "issue:43697", "issue:43720", "issue:43749", - "issue:43756", "issue:43761", "issue:43927", "issue:44361", "issue:44442", "issue:44493", "issue:44568", - "issue:44737", "issue:44756", "issue:44811", "issue:44821", @@ -8219,16 +8263,15 @@ "issue:45372", "issue:45375", "issue:45542", - "issue:45588", "issue:45701", "issue:45704", "issue:45920", "issue:46050", "issue:46170", "issue:46178", - "issue:46402", "issue:46701", "issue:46821", + "issue:46846", "issue:47000", "issue:47013", "issue:47332", @@ -8236,16 +8279,16 @@ ], "soft_pairs": [ "issue:43720|issue:44898", - "issue:44737|issue:45588", "issue:44821|issue:46821", "issue:38175|issue:43295", "issue:44493|issue:46050", + "issue:46846|issue:47404", "issue:44361|issue:44442", "issue:44756|issue:47404", "issue:43697|issue:45216", "issue:45357|issue:45920", - "issue:43688|issue:43749", "issue:29159|issue:33357", + "issue:43688|issue:43749", "issue:40444|issue:44811", "issue:44756|issue:47332", "issue:43643|issue:45375", @@ -8259,200 +8302,200 @@ "issue:43525|issue:46701", "issue:43653|issue:44568", "issue:43646|issue:43749", - "issue:43761|issue:47013", "issue:43653|issue:43927", - "issue:43756|issue:46402" + "issue:43761|issue:47013", + "issue:33666|issue:36010" ], - "prepared_review_unit_hash": "132abff23a27cd947d93a9a6394de645cabb4f82dd59baa437d7e77bd59c9a41", + "prepared_review_unit_hash": "5dcf718b4f4434fac96221ff50483b3829b1693a433c0770cdc8474cafd70267", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 36147, - "estimated_input_tokens": 9037, - "estimated_eval_tokens": 18330 + "serialized_chars": 35493, + "estimated_input_tokens": 8874, + "estimated_eval_tokens": 18004 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are unrelated model-specific regressions or load/save bugs that only weakly resemble each other by subsystem. The only broadly representative anchor is the tokenizer regression affecting multiple models; the soft edges are mostly false positives and should be rejected.", - "confidence": 0.37, - "canonical_issue_reason": "issue:45920 is the broadest and most central tokenizer regression here: it affects multiple model families, names the concrete symptom (wrong token IDs), and matches the cluster\u2019s recurring tokenizer-breakage theme better than the narrower model-specific reports.", + "summary": "These are mostly unrelated issue reports that only share broad subsystem vocabulary (tokenizer, OOM, regression, processor). None of the soft pairs look like the same underlying bug or change, so I would not merge any of them as duplicates. The most representative issue overall is the broad tokenizer regression affecting multiple models.", + "confidence": 0.62, + "canonical_issue_reason": "issue:45920 is the broadest and clearest tokenizer regression here: it describes wrong token IDs across multiple models, not a narrow model-specific symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best representative issue because it captures a cross-model v5 tokenizer failure with a clear, shared failure mode and broad impact.", + "best_issue_reason": "issue:45920 is the best single issue to keep as the canonical reference because it captures a widespread tokenizer regression with multiple affected models and a concrete failure mode.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "Different bugs in different model paths: BitNet packed-weight loading vs Perceiver image classification interpolation." - }, - { - "left": "issue:44737", - "right": "issue:45588", - "accept": false, - "reason": "XLNet CPU device placement bug and flash-attention s_aux=None crash are unrelated code paths." + "reason": "Different model families and failures: BitNet loading/unpacking vs Perceiver interpolation/output mismatch." }, { "left": "issue:44821", "right": "issue:46821", "accept": false, - "reason": "AutoImageProcessor URL loading and TimesFM window_size attribute errors are unrelated." + "reason": "Unrelated bugs: AutoImageProcessor URL loading vs TimesFM window_size AttributeError." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Different regressions: SigLIP zero probabilities vs processor.tokenizer/image-passing custom model breakage." + "reason": "Both mention processor/tokenizer, but one is SigLIP2 zero probabilities and the other is a regression in custom model code passing images to tokenizer." }, { "left": "issue:44493", "right": "issue:46050", "accept": false, - "reason": "Position-id warning/behavior and quantized-model save/load corruption are not the same bug." + "reason": "Different root causes: unexpected key warnings with position ids vs quantized model save/load breakage." + }, + { + "left": "issue:46846", + "right": "issue:47404", + "accept": false, + "reason": "Both are memory/TP-related, but one is missing TP plan entries and the other is device_map loading OOM from conversion transients." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "Both involve tokenizers, but they fail in different model-specific loader/task paths and are not the same concrete defect." + "reason": "Different tokenizer failures for different tokenizers: MLuke AttributeError vs AutoTokenizer failing on FastSpeech2ConformerTokenizer." }, { "left": "issue:44756", "right": "issue:47404", "accept": false, - "reason": "Both mention OOM, but one is mmap on Strix Halo and the other is device_map conversion-op transients; different mechanisms." + "reason": "Both involve OOM, but one is mmap on Strix Halo and the other is device_map loading/conversion-op transients." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "RTDetr output drift and Qwen3.5 checkpoint save corruption are unrelated model behaviors." + "reason": "Distinct regressions: RTDetrV2 output drift vs Qwen3.5 save_pretrained checkpoint corruption." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 visual-encoder key saving and tokenizer wrong-id regression are different subsystems and failure modes." + "reason": "Related to Qwen/Tokenizer save/load behavior, but one is visual encoder key saving and the other is wrong token IDs across many models." }, { - "left": "issue:43688", - "right": "issue:43749", + "left": "issue:29159", + "right": "issue:33357", "accept": false, - "reason": "Aux-loss normalization and FSDP CPU RAM loading are unrelated training/runtime issues." + "reason": "Tokenizer consistency bug vs MacOS bus error on CLIP; no shared underlying failure." }, { - "left": "issue:29159", - "right": "issue:33357", + "left": "issue:43688", + "right": "issue:43749", "accept": false, - "reason": "Tokenizer slow/fast inconsistency and MacOS CLIP bus error are unrelated." + "reason": "Different subsystems: auxiliary loss normalization vs FSDP CPU RAM efficient loading." }, { "left": "issue:40444", "right": "issue:44811", "accept": false, - "reason": "IterableDataset multi-image finetuning failure and Whisper batch_decode skip_special_tokens are unrelated." + "reason": "Different features: multi-image VL finetuning failure vs Whisper batch_decode skip_special_tokens bug." }, { "left": "issue:44756", "right": "issue:47332", "accept": false, - "reason": "Both are loading/offloading memory issues, but they involve different triggers and code paths." + "reason": "Both involve memory/offloading, but the causes differ: mmap OOM vs inability to save a disk-offloaded model." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both concern missing config fields, but one is trust_remote_code return-shape issues and the other is a specific Qwen vision config field dropped by strict loading." + "reason": "Both concern config fields, but one is generic trust_remote_code missing fields and the other is a specific Qwen3_5MoeVisionConfig field dropped by @strict." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Offline model_info call and tokenizer load failure for a specific repo are different tokenizer-loading regressions." + "reason": "Offline model_info patch bug vs a tokenizer load failure for EMBEDDIA/est-roberta; different code paths and symptoms." }, { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Gemma processor ImportError from mistral_common and Windows ROCm torch.distributed import/type errors are unrelated." + "reason": "Different dependency/import problems: mistral_common ReasoningEffort import vs Windows ROCm torch.distributed dependency issues." }, { "left": "issue:44857", "right": "issue:47000", "accept": false, - "reason": "LwDetr AMP/CUDA crash and RTDetrLoss infeasible cost matrix are different detection-loss failures." + "reason": "Different RTDetr issues: float16 AMP crash vs infeasible cost matrix in RTDetrLoss." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Processor/tokenizer regression and Kimi-K2.5 codec/regex warning regression are different bugs." + "reason": "Different regression targets: processor.tokenizer/images handling vs Kimi-K2.5 codec handling and mistral regex warning." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Aux-loss normalization bug and T5 apex fused RMSNorm memory leak are unrelated." + "reason": "Unrelated training/model bugs: auxiliary loss normalization vs T5 apex.FusedRMSNorm memory leak." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Aux-loss normalization and version-dependent tokenization changes are not the same defect." + "reason": "Different bug classes: auxiliary loss normalization vs tokenization changes across versions." }, { "left": "issue:45542", "right": "issue:46178", "accept": false, - "reason": "Backend-mapping tf backend error and tensorflow_text backend definition error are separate dependency/backend issues." + "reason": "Both are backend/import errors, but they affect different missing backends and code paths." }, { "left": "issue:43525", "right": "issue:46701", "accept": false, - "reason": "Llama4Config pad_token_id missing and DiffusionGemma pipeline failure are unrelated model bugs." + "reason": "Llama4Config pad_token_id attribute error is unrelated to DiffusionGemma image-text pipeline failure." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both involve special tokens, but BigBird mask-token registration and mdeberta BOS/EOS insertion are distinct tokenizer behaviors." + "reason": "Both involve special tokens, but one is BigBird mask token registration and the other is mdeberta-v3 BOS/EOS insertion behavior." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "Custom model initialization breakage and FSDP CPU RAM loading are unrelated." + "reason": "Generic custom model initialization breakage vs FSDP CPU RAM efficient loading; different failure modes." }, { - "left": "issue:43761", - "right": "issue:47013", + "left": "issue:43653", + "right": "issue:43927", "accept": false, - "reason": "CLIPVision hidden_states regression and xLSTM return_last_states crash are unrelated model execution bugs." + "reason": "Different token/config issues: BigBird special-token decode bug vs DiaConfig losing custom token IDs on save/load." }, { - "left": "issue:43653", - "right": "issue:43927", + "left": "issue:43761", + "right": "issue:47013", "accept": false, - "reason": "BigBird special-token decode bug and DiaConfig save/load token-id loss are different tokenizer/config issues." + "reason": "Different model/runtime problems: CLIPVision hidden_states regression vs xLSTM train-mode state unpacking crash." }, { - "left": "issue:43756", - "right": "issue:46402", + "left": "issue:33666", + "right": "issue:36010", "accept": false, - "reason": "Smollm3 RoPE-layer mismatch and Sam3Model text encoder weight loading are unrelated model-specific regressions." + "reason": "Multi-GPU Qwen2-VL training request vs GenerationMixin import error; not the same underlying issue." } ] }, @@ -8480,7 +8523,6 @@ "issue:32090", "issue:33357", "issue:33666", - "issue:36010", "issue:36331", "issue:41720", "issue:42898", @@ -8495,6 +8537,7 @@ "issue:43643", "issue:43688", "issue:43720", + "issue:43756", "issue:43761", "issue:43782", "issue:43927", @@ -8503,7 +8546,6 @@ "issue:44466", "issue:44479", "issue:44484", - "issue:44514", "issue:44554", "issue:44561", "issue:44610", @@ -8512,10 +8554,10 @@ "issue:44734", "issue:44743", "issue:44977", - "issue:45290", "issue:45405", "issue:45440", "issue:46050", + "issue:46402", "issue:46491", "issue:46612", "issue:46693", @@ -8526,7 +8568,7 @@ "issue:47276" ], "soft_pairs": [ - "issue:33666|issue:36010", + "issue:43756|issue:46402", "issue:44464|issue:44743", "issue:46832|issue:46932", "issue:44479|issue:44625", @@ -8537,15 +8579,15 @@ "issue:33666|issue:36331", "issue:41720|issue:44464", "issue:32090|issue:33666", - "issue:44977|issue:46693", "issue:44466|issue:46050", + "issue:44977|issue:46693", "issue:43504|issue:43550", "issue:44464|issue:44977", "issue:45405|issue:47254", - "issue:43688|issue:45440", + "issue:43278|issue:43688", "issue:43782|issue:46693", "issue:43643|issue:44561", - "issue:43278|issue:43688", + "issue:43688|issue:45440", "issue:43504|issue:43720", "issue:43720|issue:43927", "issue:43606|issue:43927", @@ -8556,229 +8598,222 @@ "issue:44484|issue:44704", "issue:43525|issue:43572", "issue:43577|issue:44554", - "issue:28218|issue:33357", - "issue:44514|issue:45290" + "issue:28218|issue:33357" ], - "prepared_review_unit_hash": "3d2ecd720e80c3454233ea51cf37250bf5a1d04b5e044e75f83f6f84a6334c0a", + "prepared_review_unit_hash": "938ca0e0d75da8d1ca4d16206152a5087f18c4b0d1df88199ea5aae7f3dd0de6", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 37452, - "estimated_input_tokens": 9363, - "estimated_eval_tokens": 18982 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36542, + "estimated_input_tokens": 9136, + "estimated_eval_tokens": 18528 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a broad mix of unrelated bugs and regressions; none of the soft candidate pairs look like the same underlying issue, so they should stay separate.", - "confidence": 0.95, - "canonical_issue_reason": "No single issue is a true duplicate anchor here; the set is too heterogeneous across tokenization, loading, generation, training, and config regressions.", + "summary": "This cluster is mostly noisy and spans unrelated bug reports. The only clearly on-theme item is the tokenizer spacing report, so I\u2019d anchor the cluster on issue #28218; no PRs are present.", + "confidence": 0.76, + "canonical_issue_reason": "#28218 is the broadest and most direct match for the tokenizer-space bug class, and its title matches the shared underlying symptom better than the more model-specific variants.", "canonical_pr_reason": null, - "best_issue_reason": "issue:28218 is the cleanest, most self-contained bug report in the set and has downstream references, but it only represents one narrow tokenizer problem.", + "best_issue_reason": "#28218 is the best representative issue because it is the clearest, most general statement of the tokenizer spacing problem and can subsume related whitespace-regression reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:33666", - "right": "issue:36010", + "left": "issue:43756", + "right": "issue:46402", "accept": false, - "reason": "Different failures: multi-GPU training vs GenerationMixin import error." + "reason": "Unrelated model/blog RoPE-layer discrepancy vs. missing text-encoder weights; only superficial transformer/model similarity." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Both involve generation internals, but one is compiled-forward inconsistency and the other is recurrent-state reset in a specific modular model." + "reason": "Both involve generation/caching, but one is chunked generation under compiled forward and the other is recurrent-state reset in modular Qwen3.5; different code paths and symptoms." }, { "left": "issue:46832", "right": "issue:46932", "accept": false, - "reason": "Unrelated model families and bugs: RT-DETR feature-level mismatch vs Gemma4 quantized embedders dtype casting." + "reason": "Both concern RT-DETR/Gemma4 dtype or model wiring issues, but they are different models and different failure modes." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Different Qwen3.x regressions: video input handling vs num_labels propagation." + "reason": "Video-input regression across Qwen VL variants vs. num_labels propagation in Qwen3.5; not the same bug." }, { "left": "issue:43761", "right": "issue:46829", "accept": false, - "reason": "Both mention hidden states, but they are distinct model/output bugs with different code paths and symptoms." + "reason": "Different outputs are wrong in different classes: CLIPVisionModel hidden_states regression vs. VideoPrism output field mismatch." }, { "left": "issue:43550", "right": "issue:44361", "accept": false, - "reason": "Completely different areas: torch.compile/SDPA model failure vs tokenizer AttributeError." + "reason": "One is a torch.compile/SDPA failure in Bamba, the other is a tokenizer AttributeError in MLukeTokenizer; unrelated." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "One is gguf loading for Auto* APIs; the other is dtype propagation on Blip2 loading." + "reason": "GGUF loading/tokenizer path breakage vs. Blip2 dtype loading bug; different subsystems and fixes." }, { "left": "issue:42898", "right": "issue:46491", "accept": false, - "reason": "Both are tokenization/spacing related, but one is a v5 cleanup behavior change and the other is a CodeLlama round-trip regression." + "reason": "Both mention tokenization spacing, but 42898 is a v5 cleanup-behavior change while 46491 is a CodeLlama round-trip regression; not clearly the same concrete bug." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Different training-time errors: multi-GPU training vs CustomTrainer compute_loss signature mismatch." + "reason": "Multi-GPU training for Qwen2-VL vs. Trainer custom compute_loss signature mismatch; unrelated." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Unrelated problems: device-mapping cuda assert vs compiled-forward generation inconsistency." + "reason": "Qwen3 auto device mapping cuda assert vs. compiled-forward generation inconsistency; different failure mechanisms." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer broadcast error is unrelated to Qwen2-VL multi-GPU training." - }, - { - "left": "issue:44977", - "right": "issue:46693", - "accept": false, - "reason": "Both involve FA2, but one is a correctness bug and the other is a performance regression." + "reason": "Trainer GPU broadcast NoneType error vs. Qwen2-VL multi-GPU training; both distributed-training themed but not the same bug." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Different save/load bugs: tied-weight serialization vs sinq quantized model persistence." + "reason": "Serialization of lm_head weights vs. sinq quantized save/load corruption; both persistence-related but distinct problems." + }, + { + "left": "issue:44977", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug vs. FA2 long-context performance regression; same model family but different behavior and symptom." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "Legacy-field loading bug vs torch.compile/SDPA model failure." + "reason": "Legacy-field model loading bug vs. torch.compile SDPA failure; different model-loading code paths." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Different generation regressions affecting different mechanisms and models." + "reason": "Compiled-forward chunked generation inconsistency is not the same concrete issue as flash-attention generation failure." }, { "left": "issue:45405", "right": "issue:47254", "accept": false, - "reason": "Version bump complaint vs PEFT/DeepSpeed gradient-checkpointing error." + "reason": "PEFT version pin complaint vs. checkpoint error with PEFT+DeepSpeed; related dependency, but not the same underlying bug." }, { - "left": "issue:43688", - "right": "issue:45440", + "left": "issue:43278", + "right": "issue:43688", "accept": false, - "reason": "Both are MoE-related, but they describe different correctness issues in different implementations." + "reason": "Both concern tensor dtypes or normalization, but one is BF16/FP32 eval drift and the other is auxiliary-loss normalization; unrelated." }, { "left": "issue:43782", "right": "issue:46693", "accept": false, - "reason": "From-pretrained weight_only load error is unrelated to FA2 long-context performance regression." + "reason": "Qwen3VL weight_only load error vs. long-context FA2 performance regression; different models and failure modes." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "Both involve trust_remote_code compatibility, but the causes differ: missing config fields vs removed helper function." + "reason": "AutoConfig missing fields with trust_remote_code vs. trust_remote_code breakage from removed is_torch_fx_available; different root causes." }, { - "left": "issue:43278", - "right": "issue:43688", + "left": "issue:43688", + "right": "issue:45440", "accept": false, - "reason": "Embedding dtype drift is unrelated to MoE auxiliary-loss normalization." + "reason": "Aux-loss normalization bug vs. DeepseekV3MoE divergence from remote implementation; both model-accuracy issues but not the same code-path problem." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Different bugs in different models: legacy preset loading vs BitNet packed-weight loading." + "reason": "Legacy field loading for BEiT vs. BitNet packed-weights unpacking during accelerate loading; unrelated loading bugs." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "Packed-weight loading bug and tokenizer AttributeError are unrelated." + "reason": "BitNet accelerate loading issue vs. tokenizer task AttributeError; entirely different components." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch vs config save/load losing custom token IDs." + "reason": "CPU offload device mismatch in bark-small vs. DiaConfig losing custom token IDs on save/load; both config/load-related but not the same bug." }, { "left": "issue:44734", "right": "issue:46612", "accept": false, - "reason": "Both are cache-related generation bugs, but one is serve-side KV indexing and the other is beam-search cache reorder skipping in specific models." + "reason": "KV-cache continuation crash in serve vs. beam-search cache reorder skipped for several models; both cache-related but different execution paths and symptoms." }, { "left": "issue:44625", "right": "issue:47276", "accept": false, - "reason": "Qwen3.5 num_labels propagation is unrelated to a synced_gpus warning message." + "reason": "Qwen3.5 num_labels propagation bug vs. misleading synced_gpus warning; unrelated." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "Legacy-field loading bug is unrelated to OmDet-Turbo processor image-size mismatch." + "reason": "Legacy preset loading bug vs. processor image-size mismatch; different model families and failures." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "Same as above: BitNet loading vs MLukeTokenizer task failure are unrelated." + "reason": "Packed-weight loading bug vs. MLukeTokenizer task AttributeError; unrelated." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "save_pretrained shard-size question is unrelated to AutoProcessor kwarg forwarding." + "reason": "Default max_shard_size question vs. AutoProcessor not passing kwargs to cached_file; not the same bug or fix." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Similar symptom class, but different configs and different missing fields; not enough to call the same bug." + "reason": "Missing pad_token_id on Llama4Config vs. missing pad_token_idx on StableLmConfig; similar naming, but different configs and likely separate omissions." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "Loading dtype propagation bug is unrelated to MPS attention correctness." + "reason": "Blip2 dtype remains float32 vs. MPS attention correctness issue; unrelated." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer spacing regression and MacOS bus error are unrelated." - }, - { - "left": "issue:44514", - "right": "issue:45290", - "accept": false, - "reason": "Both hit apply_chat_template, but on different inputs and different failure modes; not the same underlying bug." + "reason": "Tokenizer adds an extra space after an added token vs. MacOS bus error loading CLIP; completely unrelated." } ] }, @@ -8825,12 +8860,14 @@ "issue:44361", "issue:44466", "issue:44488", + "issue:44514", "issue:44554", "issue:44568", "issue:44589", "issue:44843", "issue:44964", "issue:44977", + "issue:45290", "issue:45405", "issue:45491", "issue:45520", @@ -8844,15 +8881,14 @@ "issue:46365", "issue:46489", "issue:46710", - "issue:46821", - "issue:46832", "issue:46846", "issue:47332", "issue:47404" ], "soft_pairs": [ - "issue:43611|issue:46846", + "issue:44514|issue:45290", "issue:44964|issue:45405", + "issue:43611|issue:46846", "issue:45823|issue:46050", "issue:44977|issue:46209", "issue:39692|issue:43295", @@ -8862,227 +8898,226 @@ "issue:43334|issue:43572", "issue:43994|issue:46192", "issue:42796|issue:44977", - "issue:43257|issue:47404", "issue:45405|issue:47332", + "issue:43257|issue:47404", "issue:44554|issue:46365", "issue:45491|issue:46002", - "issue:44568|issue:46489", "issue:43646|issue:44466", - "issue:44568|issue:45812", + "issue:44568|issue:46489", "issue:43278|issue:43701", "issue:43572|issue:44291", + "issue:44568|issue:45812", "issue:43927|issue:44361", "issue:43493|issue:43844", - "issue:43278|issue:43844", "issue:46209|issue:46710", + "issue:43278|issue:43844", "issue:43873|issue:47404", - "issue:44589|issue:45520", - "issue:43824|issue:44589", "issue:43504|issue:43606", - "issue:46821|issue:46832" + "issue:44589|issue:45520", + "issue:43824|issue:44589" ], - "prepared_review_unit_hash": "8f03b97ad03b96972e41bda141fc14b3b49f357047aa37e8c7529abc2a52fbc3", + "prepared_review_unit_hash": "31adae51167a9d999f69e43e3e042a23d49200340b917a839042daa7da8bf9c8", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36132, - "estimated_input_tokens": 9033, - "estimated_eval_tokens": 18322 + "serialized_chars": 36175, + "estimated_input_tokens": 9044, + "estimated_eval_tokens": 18344 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is heterogeneous: mostly unrelated bug reports across tokenizers, multimodal configs, offloading, quantization, and model-loading regressions. None of the suggested soft pairs look like the same underlying defect closely enough to merge or deduplicate.", - "confidence": 0.94, - "canonical_issue_reason": "No clear canonical issue: the items span many unrelated models and failure modes, so there isn\u2019t a single representative bug for the cluster.", + "summary": "This cluster is theme-based rather than duplicate-based: most items are distinct Transformers regressions across tokenizers, multimodal processors, loading, quantization, offloading, and v5 compatibility. None of the soft pairs look like the same underlying bug/change with enough specificity to merge.", + "confidence": 0.22, + "canonical_issue_reason": "If one issue must represent the cluster, issue 43646 is the broadest high-level regression report (custom model initialization breaks in Transformers 5.0.0) and is closest to the cluster\u2019s general v5-breakage theme.", "canonical_pr_reason": null, - "best_issue_reason": "No best issue candidate; the cluster is too diverse and no issue clearly subsumes the others.", + "best_issue_reason": "Issue 43646 is the best single representative because it is the most generic and broadly applicable of the v5 regression reports in this set, though the overall cluster is still heterogeneous.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43611", - "right": "issue:46846", + "left": "issue:44514", + "right": "issue:45290", "accept": false, - "reason": "Both are loading/model-configuration problems, but one is a generic base_model_prefix regression and the other is a Qwen3.5 tensor-parallel planning bug; different code paths and fixes." + "reason": "Both involve apply_chat_template/processor code, but they are different failure modes: batched Qwen2.5-VL padding vs tool-call assistant messages." }, { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "Completely different topics: Phi-4 multimodal loading versus an unreleased PEFT version requirement." + "reason": "A model-loading failure for Phi-4 multimodal is unrelated to a PEFT version bump issue." + }, + { + "left": "issue:43611", + "right": "issue:46846", + "accept": false, + "reason": "Different bugs: base_model_prefix loading regression vs missing linear_attn TP plan entries causing OOM/shape errors." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Both mention model loading, but one is a Gemma3/Gemma4 device-map mismatch and the other is a sinq quantization save/load breakage; not the same bug." + "reason": "Gemma4 device-mismatch with auto device_map is not the same as sinq quantization save/load breakage." }, { "left": "issue:44977", "right": "issue:46209", "accept": false, - "reason": "Different failure modes and stacks: flash-attention generation behavior for Qwen3.5 versus FP8 inference on B200." + "reason": "Qwen3.5 flash-attn generation issues and FP8/B200 inference failures are separate code paths and hardware-specific problems." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "Both involve processor/tokenizer usage, but the SigLIP2 docs-example errors are not the same underlying regression as the v4.57.5 custom-code processor.tokenizer breakage." + "reason": "Both mention processor/tokenizer usage, but one is a SigLIP2 docs example and the other is a regression in processor.tokenizer access with image passing." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "Unrelated issues: SigLIP2 implementation discrepancy versus a pipeline error message about translation tasks." + "reason": "SigLIP2 implementation divergence and a pipeline error message for translation tasks are unrelated." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Both are multimodal/config-loading issues, but they target different model families and different config plumbing; not close enough to be one fix." + "reason": "Both are multimodal Qwen config issues, but one is a missing pad_token_id attribute and the other is text_config unwrapping in from_config." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "Different tokenizer problems: a specific model failing to load versus an offline-mode bug in _patch_mistral_regex." + "reason": "Loading a specific model failing is unrelated to AutoTokenizer's offline-breakage in _patch_mistral_regex." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Superficially similar missing-pad-token attribute errors, but they affect different configs (Qwen3-VL text config vs StableLmConfig) and likely different fixes." + "reason": "Different missing config fields for different models: pad_token_id in Qwen3-VL vs pad_token_idx in StableLmConfig." }, { "left": "issue:43994", "right": "issue:46192", "accept": false, - "reason": "Different tokenizer/pipeline behaviors: SigLIP2 wrong outputs versus return_tensors being ignored when text_kwargs is passed." + "reason": "SigLIP2 wrong outputs and return_tensors being ignored are different API/behavior bugs." }, { "left": "issue:42796", "right": "issue:44977", "accept": false, - "reason": "Tokenizer BPE-marker handling for Ministral is unrelated to Qwen3.5 flash-attention generation issues." + "reason": "Tokenizer BPE marker handling for Ministral is unrelated to Qwen3.5 flash-attention generation." }, { - "left": "issue:43257", - "right": "issue:47404", + "left": "issue:45405", + "right": "issue:47332", "accept": false, - "reason": "Both touch loading/offloading, but one is MOE weight conversion under accelerate+deepspeed and the other is a device_map memory budgeting problem during conversion; distinct bugs." + "reason": "A dependency version bump and a disk-offloaded model save failure are unrelated." }, { - "left": "issue:45405", - "right": "issue:47332", + "left": "issue:43257", + "right": "issue:47404", "accept": false, - "reason": "No relation: PEFT version pinning issue versus a disk-offloaded Qwen3-VL save failure." + "reason": "Qwen3 MOE weight conversion with accelerate/deepspeed is a different loading path from device_map OOMs during transient conversion ops." }, { "left": "issue:44554", "right": "issue:46365", "accept": false, - "reason": "Unrelated: MPS attention correctness versus an import failure for Trainer caused by datasets." + "reason": "MPS attention correctness and Trainer import conflicts with datasets are unrelated." }, { "left": "issue:45491", "right": "issue:46002", "accept": false, - "reason": "Both mention NaNs, but they occur in different models and circumstances; not the same underlying defect." - }, - { - "left": "issue:44568", - "right": "issue:46489", - "accept": false, - "reason": "Separate tokenizer regressions for different models and code paths; one is special-token insertion, the other is DeepSeek-Coder v1 token output on v5+." + "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention on GPU and the other is classification head initialization/checkpoint behavior." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Both are v5-related regressions, but one is custom model initialization and the other is state_dict serialization of tied weights; different fixes." + "reason": "Custom model initialization breakage is not the same as inconsistent lm_head.weight serialization tied to device." }, { "left": "issue:44568", - "right": "issue:45812", + "right": "issue:46489", "accept": false, - "reason": "Different tokenizer regressions affecting different model families; not one shared bug." + "reason": "These are separate tokenizer regressions affecting different model families and likely different causes." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "One is an embedding dtype drift between train/eval, the other a resume_from_checkpoint key mismatch." + "reason": "Embedding dtype drift during evaluation is unrelated to resume_from_checkpoint key mismatch." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Different failure modes: missing config field versus init_empty_weights receiving an unexpected argument." + "reason": "Missing pad_token_idx in a config and init_empty_weights receiving an unexpected argument are different v5 compatibility issues." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and symptoms; not enough evidence of one underlying bug." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "Distinct tokenizer/config bugs affecting different models and code paths." + "reason": "DiaConfig save/load token-ID loss and MLukeTokenizer task AttributeError are unrelated." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation discrepancy is unrelated to the ZeRO-3 gradient growth issue." + "reason": "SigLIP2 JAX/HF discrepancy and a ZeRO-3 gradient increase issue do not share a concrete code path." }, { - "left": "issue:43278", - "right": "issue:43844", + "left": "issue:46209", + "right": "issue:46710", "accept": false, - "reason": "Both concern training/eval behavior, but one is dtype change and the other is gradient inflation under DeepSpeed ZeRO-3." + "reason": "FP8 inference on B200 and a post-upgrade accuracy regression are too different to be the same bug." }, { - "left": "issue:46209", - "right": "issue:46710", + "left": "issue:43278", + "right": "issue:43844", "accept": false, - "reason": "Different problems: FP8 inference on B200 versus an accuracy regression after a Transformers upgrade." + "reason": "Embedding dtype changes in evaluation are unrelated to ZeRO-3 gradient anomalies." }, { "left": "issue:43873", "right": "issue:47404", "accept": false, - "reason": "Both involve offloading and memory, but one is quantization offloading behavior and the other is max_memory budgeting for conversion transients; not the same concrete bug." + "reason": "Quantization offloading behavior and device_map OOM from conversion transients are related only at a broad subsystem level, not the same bug." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "A legacy-field model loading failure and a CPU-offload device mismatch are separate issues." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Unrelated runtime/environment issues: missing Float8 storage object versus flash_attn import handling on Python 3.13." + "reason": "Float8 storage lookup failure and a flash_attn import KeyError on Python 3.13 are unrelated." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "Different loading/import failures affecting different subsystems." - }, - { - "left": "issue:43504", - "right": "issue:43606", - "accept": false, - "reason": "Different model bugs: a legacy-field load failure for BEiT segmentation versus CPU-offload device mismatch for bark-small." - }, - { - "left": "issue:46821", - "right": "issue:46832", - "accept": false, - "reason": "Different model-specific forward/config errors: TimesFM window_size AttributeError versus RT-DETR feature-level mismatch." + "reason": "Qwen2.5-VL import error and Float8 storage resolution failure are different problems in different areas." } ] }, @@ -9125,7 +9160,6 @@ "issue:43701", "issue:43844", "issue:43873", - "issue:43937", "issue:43957", "issue:43994", "issue:44112", @@ -9148,20 +9182,21 @@ "issue:46050", "issue:46209", "issue:46285", - "issue:46701", "issue:46710", + "issue:46821", "issue:46832", "issue:47013", "issue:47145", "issue:47254" ], "soft_pairs": [ - "issue:44493|issue:46285", + "issue:46821|issue:46832", "issue:43504|issue:43653", + "issue:44493|issue:46285", "issue:43994|issue:44493", "issue:43606|issue:44112", - "issue:45245|issue:46710", "issue:15354|issue:38175", + "issue:45245|issue:46710", "issue:42490|issue:43696", "issue:45072|issue:46832", "issue:43873|issue:46050", @@ -9171,209 +9206,208 @@ "issue:28218|issue:34567", "issue:42947|issue:43381", "issue:29159|issue:32090", - "issue:45084|issue:45520", "issue:43844|issue:44964", + "issue:45084|issue:45520", "issue:47013|issue:47254", "issue:45229|issue:45823", - "issue:45200|issue:45229", "issue:44423|issue:44734", + "issue:45200|issue:45229", "issue:45588|issue:46050", "issue:43493|issue:43701", "issue:44849|issue:47145", "issue:37428|issue:43957", - "issue:41720|issue:46710", - "issue:43937|issue:46701" + "issue:41720|issue:46710" ], - "prepared_review_unit_hash": "509e53e0d0fc4cab4b205129338096d93d100da9a49c8c3a0e1c0a526790b58f", + "prepared_review_unit_hash": "15d64157a5b026637c75d6d3745c97fd8c2e57d4c2ea54f1aa8834775de8248a", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35954, - "estimated_input_tokens": 8989, - "estimated_eval_tokens": 18234 + "serialized_chars": 35398, + "estimated_input_tokens": 8850, + "estimated_eval_tokens": 17956 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft pair looks like the same underlying bug or change. The set is mostly a mix of unrelated tokenizer, loading, training, and serving issues with only superficial similarity in broad subsystem wording.", - "confidence": 0.92, - "canonical_issue_reason": "issue:44493 is the broadest, most centrally framed regression in the set, affecting many models and likely representing the most general issue candidate among otherwise unrelated reports.", + "summary": "The set is heterogeneous; the soft pairs mostly share broad subsystems or model families, but each describes a distinct bug or failure mode. I would reject all proposed duplicate links and avoid choosing a single canonical issue for the whole batch.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue is a clear canonical representative because the items span unrelated bugs across tokenizers, loading, training, quantization, serving, and model-specific regressions.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44493 is the best single issue candidate because it is broad, regression-like, and not overly model-specific, making it the most plausible representative if a canonical issue must be chosen.", + "best_issue_reason": "No issue is a strong global canonical choice; the list does not form one coherent duplicate cluster, so any pick would be arbitrary.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44493", - "right": "issue:46285", + "left": "issue:46821", + "right": "issue:46832", "accept": false, - "reason": "Different bug classes: unexpected position-id keys vs CLIP class structure/API inconsistency." + "reason": "Different models and failure modes: a TimesFM window_size AttributeError vs an RT-DETR forward shape/feature-level mismatch." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "Different models and failures: legacy field loading for BEiT vs BigBirdTokenizer special-token registration." + "reason": "Unrelated components: Beit semantic segmentation loading a legacy field vs BigBirdTokenizer special-token decoding." + }, + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Both involve model structure/key shape concerns, but one is a broad unexpected-key regression and the other is a CLIP text-model attribute inconsistency." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "Model-specific inference corruption in SigLIP2 is not the same as a broad position-id key regression." + "reason": "Different symptoms and models: SigLIP2 nonsensical outputs vs a generic unexpected-key issue in many models." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "Both are CI failures, but the root causes differ: CPU offload device mismatch vs a stale GraniteSpeech test override." + "reason": "Both are CI/device related, but one is a CPU offload mismatch in bark-small and the other is a stale GraniteSpeech test." }, { - "left": "issue:45245", - "right": "issue:46710", + "left": "issue:15354", + "right": "issue:38175", "accept": false, - "reason": "Unrelated symptoms: category-count limit error vs an accuracy regression after version upgrade." + "reason": "TorchScript GeneratorExp export failure is unrelated to SigLIP2 zero-probability output behavior." }, { - "left": "issue:15354", - "right": "issue:38175", + "left": "issue:45245", + "right": "issue:46710", "accept": false, - "reason": "TorchScript GeneratorExp export failure is unrelated to SigLIP2 zero-probability inference output." + "reason": "A category-cardinality runtime error is unrelated to a downstream accuracy regression on a specific Llama model." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Path/import warning on trailing slash is unrelated to CUDA OOM during GPT-oss-20b inference." + "reason": "Local-path import warning and CUDA OOM on GPT-oss are unrelated failure classes." }, { "left": "issue:45072", "right": "issue:46832", "accept": false, - "reason": "Different code paths: bfloat16 dtype mismatch in inference vs RT-DETR feature-level shape/type error." + "reason": "bfloat16 dtype mismatches in inference are not the same bug as RT-DETR having too many feature levels." }, { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Quantization offloading behavior is unrelated to save/load breakage in sinq quantized models." + "reason": "Quantization/offloading issues are adjacent, but one is runtime offloading behavior and the other is save/load corruption for sinq checkpoints." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is not the same as Whisper empty-transcription failure after token alignment." + "reason": "Compiled-forward chunked generation inconsistency is unrelated to Whisper empty-transcription generation after token alignment." }, { "left": "issue:44849", "right": "issue:46209", "accept": false, - "reason": "Qwen3.5 hidden-state bug is unrelated to FP8 inference failure on B200." + "reason": "Qwen3.5 hidden-state handling and FP8 inference failure on B200 are different model/path problems." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Different Qwen3.5 symptoms and likely different code paths: hidden states vs incorrect visual encoder keys on save." + "reason": "Hidden-state output bug is unrelated to saving incorrect visual encoder keys." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Tokenizer extra-space behavior is unrelated to TrainerState token-count tracking." + "reason": "Tokenizer spacing after added tokens does not match a TrainerState counter not updating." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Both involve gradient checkpointing, but one is a PEFT/LoRA effectiveness issue and the other is eval-mode incompatibility." + "reason": "Both mention gradient checkpointing, but one is ineffective LoRA behavior while the other is an eval-mode restriction." }, { "left": "issue:29159", "right": "issue:32090", "accept": false, - "reason": "Tokenizer slow/fast inconsistency is unrelated to Trainer NoneType broadcast failure." + "reason": "Tokenizer inconsistency and a Trainer broadcast NoneType error are unrelated." }, { - "left": "issue:45084", - "right": "issue:45520", + "left": "issue:43844", + "right": "issue:44964", "accept": false, - "reason": "Template compilation error is not the same as the flash_attn import KeyError on Python 3.13." + "reason": "ZeRO-3 gradient inflation is not the same as a model-loading regression on transformers 5.x." }, { - "left": "issue:43844", - "right": "issue:44964", + "left": "issue:45084", + "right": "issue:45520", "accept": false, - "reason": "DeepSpeed ZeRO-3 gradient explosion is unrelated to model loading failure for Phi-4 multimodal." + "reason": "A template-node compilation TypeError is unrelated to a Python 3.13 flash_attn import KeyError." }, { "left": "issue:47013", "right": "issue:47254", "accept": false, - "reason": "xLSTM train-mode state unpacking bug is unrelated to PEFT + DeepSpeed checkpointing errors." + "reason": "xLSTM return_last_states unpacking and PEFT/DeepSpeed checkpoint errors are different training-path failures." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Both mention Gemma4, but one is multi-GPU CUDA OOM and the other is a PLE device-mismatch forward bug." + "reason": "Both are Gemma4 issues, but one is multi-GPU CUDA OOM and the other is a device-mismatch forward bug." }, { - "left": "issue:45200", - "right": "issue:45229", + "left": "issue:44423", + "right": "issue:44734", "accept": false, - "reason": "Text-only mm_token_type_ids defaulting is a different problem from multi-GPU inference OOM." + "reason": "Both touch transformers serve, but continuous batching crashes and KV-cache continuation tensor indexing are distinct defects." }, { - "left": "issue:44423", - "right": "issue:44734", + "left": "issue:45200", + "right": "issue:45229", "accept": false, - "reason": "Continuous-batching multimodal crash and KV-cache continuation crash share a service context but not the same tensor/path failure." + "reason": "mm_token_type_ids defaulting and multi-GPU OOM are unrelated." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "Flash-attention sinkless-model AttributeError is unrelated to quantized save/load corruption." + "reason": "Flash-attention sink handling and sinq checkpoint save/load corruption are unrelated." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 HF-vs-JAX discrepancy is unrelated to resume_from_checkpoint key mismatch." + "reason": "SigLIP2 implementation discrepancy and resume_from_checkpoint key mismatch do not share the same underlying bug." }, { "left": "issue:44849", "right": "issue:47145", "accept": false, - "reason": "Qwen3.5 hidden-states bug is unrelated to FP8 frozen-checkpoint loading without weight_scale application." + "reason": "Qwen3.5 hidden-state output bug and FP8 weight_scale omission are unrelated." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Flash-attention import error is unrelated to meta-device model loading breakage." + "reason": "A flash attention import error on Python 3.13 is unrelated to meta-device model loading breakage." }, { "left": "issue:41720", "right": "issue:46710", "accept": false, - "reason": "Qwen3 auto device-map cudaErrorAssert is unrelated to DeepSeek accuracy regression after upgrade." - }, - { - "left": "issue:43937", - "right": "issue:46701", - "accept": false, - "reason": "Invalid GenerationConfig for GLM-5 is not the same as DiffusionGemma pipeline output incompatibility." + "reason": "A device-map cudaErrorAssert on A800 is unrelated to an accuracy regression after upgrading Transformers." } ] }, @@ -9418,9 +9452,9 @@ "issue:43824", "issue:43844", "issue:43873", + "issue:43937", "issue:43957", "issue:44186", - "issue:44521", "issue:44554", "issue:44610", "issue:44869", @@ -9428,7 +9462,6 @@ "issue:45072", "issue:45198", "issue:45200", - "issue:45290", "issue:45406", "issue:45446", "issue:45520", @@ -9447,20 +9480,21 @@ "issue:47404" ], "soft_pairs": [ + "issue:43937|issue:46701", "issue:43295|issue:43450", "issue:43065|issue:46369", "issue:44554|issue:45446", "issue:47332|issue:47404", "issue:44977|issue:46701", "issue:44869|issue:46752", - "issue:29942|issue:30333", "issue:43404|issue:46710", "issue:45072|issue:45198", + "issue:29942|issue:30333", "issue:45588|issue:45759", "issue:43295|issue:43873", "issue:43572|issue:46821", - "issue:45406|issue:46121", "issue:43540|issue:45200", + "issue:45406|issue:46121", "issue:43572|issue:43957", "issue:43295|issue:44610", "issue:43450|issue:43575", @@ -9475,217 +9509,216 @@ "issue:43334|issue:43525", "issue:43295|issue:43493", "issue:39692|issue:46192", - "issue:42915|issue:43824", - "issue:44521|issue:45290" + "issue:42915|issue:43824" ], - "prepared_review_unit_hash": "600af6a7f65072fff9953dc810ae1dc819a978c4230f6999bbf0bf68131bd98c", + "prepared_review_unit_hash": "2ebec9aa252096a70a321c3e30f78ab9b8e17575d2fe26732e5615a3c5cc8f66", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36919, - "estimated_input_tokens": 9230, - "estimated_eval_tokens": 18716 + "serialized_chars": 36298, + "estimated_input_tokens": 9075, + "estimated_eval_tokens": 18406 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose collection of unrelated Transformers issues. The soft pairs mostly share only broad subsystem keywords (multimodal, offloading, flash-attn, configs) but describe different failures, so none look like true duplicates or mergeable PR-equivalents.", - "confidence": 0.94, - "canonical_issue_reason": "issue:44521 is the best standalone representative in this set: it describes a concrete, reproducible user-facing bug, has active discussion/inbound references, and is more actionable than the broader or more speculative reports.", + "summary": "The set is a loose collection of unrelated Transformers bugs. Several are multimodal/processor or loading regressions, but none of the soft pairs look like the same concrete underlying fix, so all soft edges should be rejected.", + "confidence": 0.37, + "canonical_issue_reason": "Issue 43295 is the broadest representative of the loose processor/tokenizer regression theme, and it is the closest hub among the candidates even though it is not a true duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the strongest issue to keep as the representative item because it is specific, well-scoped, and clearly framed as a defect rather than a broad symptom.", + "best_issue_reason": "Issue 43295 is the best single issue to anchor the cluster because it is the most general and centrally connected to the processor/tokenizer breakage pattern seen in a few neighboring reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43937", + "right": "issue:46701", + "accept": false, + "reason": "Different bugs: invalid GenerationConfig for GLM-5 vs a pipeline failure in DiffusionGemma output handling." + }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Both are multimodal/processor regressions, but one is about processor.tokenizer/images handling and the other is about batched video shape; different code paths and symptoms." + "reason": "Both involve processors, but one is a regression around processor.tokenizer access while the other is batched video shape handling; not the same fix." }, { "left": "issue:43065", "right": "issue:46369", "accept": false, - "reason": "Same SAM3 family, but one concerns a dummy Conv2d in the decoder architecture and the other is a memory overflow in sam3-litetext; not the same bug." + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to sam3-litetext memory overflow." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "Unrelated problems: MPS attention correctness vs a flex_attention import/version check issue." + "reason": "MPS attention correctness with mismatched head dims is unrelated to a PyTorch version gate for AuxRequest import." }, { "left": "issue:47332", "right": "issue:47404", "accept": false, - "reason": "Both involve memory/offloading, but one is about saving a disk-offloaded model and the other about load-time OOM from conversion transients; different failure mechanisms." + "reason": "Both mention loading/offloading memory, but one is save-time disk offload for Qwen3-VL and the other is device_map conversion OOM during weight conversion; not the same bug." }, { "left": "issue:44977", "right": "issue:46701", "accept": false, - "reason": "Different models and different failures: flash-attn generation bug vs an image-text-to-text pipeline output issue." + "reason": "Flash-attention generation failure in Qwen3.5 is unrelated to DiffusionGemma pipeline output issues." }, { "left": "issue:44869", "right": "issue:46752", "accept": false, - "reason": "Both are crashes, but one is Whisper timestamp decoding and the other is apply_chat_template on an empty list; not the same underlying bug." - }, - { - "left": "issue:29942", - "right": "issue:30333", - "accept": false, - "reason": "Flash Attention 2 test failures and MLflow reporting after training failure are unrelated subsystems." + "reason": "Whisper timestamp decoding on a trailing replacement character is different from apply_chat_template crashing on an empty list." }, { "left": "issue:43404", "right": "issue:46710", "accept": false, - "reason": "Weight tying in Mistral3 and an accuracy regression in DeepSeek are distinct issues at different layers." + "reason": "Missing lm_head tying in Mistral3ForConditionalGeneration is a model-weight bug, not an accuracy regression after a Transformers upgrade." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "One is a bfloat16 dtype mismatch in inference wrappers; the other is a save_pretrained/tokenization failure in Wav2Vec2." + "reason": "CI dtype mismatch for SwitchTransformers/TimmWrapperModel is unrelated to Wav2Vec2 save_pretrained/tokenization failure." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 test failures and MLflow reporting on failed training jobs are completely different areas." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "A flash_attention integration crash on sink-less models is not the same as from_config failing to unwrap text_config for composite configs." + "reason": "Sink-less flash_attention crash on s_aux=None is unrelated to AutoModelForCausalLM not unwrapping text_config for multimodal configs." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Processor/tokenizer regression and quantization offloading failure are only loosely related at the multimodal/model-loading level." + "reason": "processor.tokenizer regression and quantized offloading behavior are different problems despite both touching multimodal loading paths." }, { "left": "issue:43572", "right": "issue:46821", "accept": false, - "reason": "Missing pad_token_idx and TimesFM window_size errors are different config attribute bugs." + "reason": "StableLmConfig missing pad_token_idx is unrelated to TimesFM window_size AttributeError." }, { - "left": "issue:45406", - "right": "issue:46121", + "left": "issue:43540", + "right": "issue:45200", "accept": false, - "reason": "Gemma4Processor lacking _tokenizer and rope-validation TypeError are unrelated code paths." + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to Gemma 4 mm_token_type_ids defaulting behavior." }, { - "left": "issue:43540", - "right": "issue:45200", + "left": "issue:45406", + "right": "issue:46121", "accept": false, - "reason": "Video processing failure in Qwen3OmniMoe is unrelated to Gemma 4 mm_token_type_ids defaulting behavior." + "reason": "Gemma4Processor missing _tokenizer and rope dict conversion on JSON-loaded lists are unrelated failures." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "A missing config attribute and a meta-device loading breakage are different regressions." + "reason": "Missing pad_token_idx in StableLmConfig is not the same as meta-device loading breaking some models." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Custom processor/tokenizer regression is not the same as OmDet's processor producing the wrong image size." + "reason": "Both are processor/model mismatches, but one is a processor.tokenizer regression and the other is a fixed input-size mismatch for OmDet-Turbo; different code paths." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Batched video shape issues and tensor-parallel OOM on model load are different problems." + "reason": "Batched video processor shape issues are unrelated to Qwen2-57B-A14B-Instruct OOM on tensor-parallel load." }, { "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "Tokenizer crashes on NER/padding are unrelated to return_tensors being ignored when text_kwargs is passed." + "reason": "LayoutLMv2Tokenizer crashes on NER padding/truncation are different from return_tensors being ignored when text_kwargs is passed." }, { "left": "issue:45823", "right": "issue:46192", "accept": false, - "reason": "Device placement mismatch in Gemma4 PLE is a different bug from return_tensors semantics." + "reason": "Gemma4 PLE device_map auto mismatch is a device-placement bug, not the return_tensors/text_kwargs API issue." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "One is an internal flash_attention import error, the other is a missing top-level Qwen2.5-VL export." + "reason": "Importing _flash_supports_window_size and importing Qwen2_5_VLForConditionalGeneration are unrelated import failures." }, { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "torch.compile/SDPA failure and ZeRO-3 gradient growth are different training/runtime bugs." + "reason": "torch.compile with SDPA in Bamba-9B-v2 is unrelated to gradient explosion under HfDeepSpeedConfig + ZeRO-3." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "The flash_attn backend KeyError on Python 3.13 is unrelated to the tensorflow_text backend mapping error." + "reason": "Python 3.13 flash_attn import KeyError is unrelated to an invalid BACKENDS_MAPPING backend value." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Text-only token type defaults and device_map forward mismatches are distinct Gemma 4 issues." + "reason": "mm_token_type_ids defaulting for fine-tuning is unrelated to Gemma4 PLE device mismatch at forward time." }, { "left": "issue:29870", "right": "issue:29942", "accept": false, - "reason": "Bart generate decoder_attention_mask bug and Flash Attention 2 test failures do not describe the same defect." + "reason": "BART generate decoder_attention_mask UnboundLocalError is unrelated to Flash Attention 2 test failures." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "A local-path import warning and a SigLIP2 implementation discrepancy are not the same underlying issue." + "reason": "Trailing-slash local model import warning is unrelated to SigLIP2 implementation discrepancy vs JAX." }, { "left": "issue:43334", "right": "issue:43525", "accept": false, - "reason": "Both mention missing pad_token_id, but they affect different models/config classes and likely different code paths." + "reason": "Qwen3-VL missing pad_token_id and Llama4 missing pad_token_id are similar in shape but refer to different model families and likely different config fixes." }, { "left": "issue:43295", "right": "issue:43493", "accept": false, - "reason": "Processor/tokenizer regression and SigLIP2 implementation mismatch are different bugs." + "reason": "processor.tokenizer regression and SigLIP2 implementation discrepancy are different bugs in different subsystems." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "Documentation example errors are unrelated to return_tensors being ignored in tokenizer kwargs." + "reason": "SigLIP2 documentation/example errors are unrelated to the return_tensors ignored API bug." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "FineGrainedFP8Config failure in Qwen3Moe is unrelated to the Qwen2.5-VL import error." - }, - { - "left": "issue:44521", - "right": "issue:45290", - "accept": false, - "reason": "Both involve apply_chat_template, but one yields all-zero assistant masks for multimodal inputs while the other crashes on tool-call assistant messages with no content; different triggers and outcomes." + "reason": "Qwen3Moe failing with FineGrainedFP8Config is unrelated to the Qwen2_5_VL import error." } ] }, @@ -9732,22 +9765,21 @@ "issue:44442", "issue:44466", "issue:44479", + "issue:44521", "issue:44561", "issue:44610", "issue:44737", - "issue:44792", "issue:44857", "issue:44898", "issue:45083", "issue:45127", "issue:45198", "issue:45200", + "issue:45290", "issue:45325", "issue:45381", "issue:45412", "issue:45520", - "issue:45563", - "issue:45593", "issue:46192", "issue:46209", "issue:46234", @@ -9759,6 +9791,7 @@ "issue:47405" ], "soft_pairs": [ + "issue:44521|issue:45290", "issue:43824|issue:43975", "issue:43278|issue:43856", "issue:28218|issue:35141", @@ -9766,231 +9799,223 @@ "issue:45381|issue:46209", "issue:43975|issue:45520", "issue:44737|issue:46894", - "issue:44466|issue:45127", "issue:43950|issue:46285", + "issue:44466|issue:45127", "issue:43653|issue:44112", - "issue:44442|issue:45198", "issue:43452|issue:46512", - "issue:43611|issue:47405", + "issue:44442|issue:45198", "issue:44561|issue:46234", - "issue:45200|issue:45325", + "issue:43611|issue:47405", "issue:44479|issue:46192", - "issue:45200|issue:46899", - "issue:43819|issue:44112", + "issue:45200|issue:45325", "issue:43335|issue:43550", + "issue:43819|issue:44112", + "issue:45200|issue:46899", "issue:44898|issue:45412", "issue:41720|issue:46209", - "issue:45200|issue:46932", "issue:44610|issue:44857", + "issue:45200|issue:46932", "issue:37428|issue:43425", "issue:43819|issue:45083", "issue:43335|issue:43653", - "issue:43606|issue:45083", - "issue:44792|issue:45563", - "issue:44857|issue:45593" + "issue:43606|issue:45083" ], - "prepared_review_unit_hash": "81bf1a5ecac31315dbd9f6e41310a043b1d8c4fd432176ab5b2000258dc97d3f", + "prepared_review_unit_hash": "61ab64d3629e490a05cb636fdd66f5ae76d50b8260770129a2f77bd433a2b96c", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36817, - "estimated_input_tokens": 9205, - "estimated_eval_tokens": 18666 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35989, + "estimated_input_tokens": 8998, + "estimated_eval_tokens": 18252 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a loose set of unrelated issue reports, not a duplicate cluster. None of the soft pairs describe the same concrete bug or change, so all soft edges should be rejected. I picked the tokenizer spacing report as the best representative issue because it is tightly scoped and clearly reproducible.", + "summary": "The items are mostly unrelated transformer bugs/regressions across tokenization, multimodal chat templates, model loading, dtype/device handling, and architecture-specific failures. No soft pair looks like the same underlying fix, so all soft edges are rejected. There are no PRs in the set.", "confidence": 0.91, - "canonical_issue_reason": "Issue 28218 is the clearest, most self-contained bug report here: a specific tokenizer behavior with direct user impact and some inbound reference activity. It is the best representative among otherwise unrelated issues.", + "canonical_issue_reason": "issue:44521 is the strongest anchor: it is open, highly discussed, and has the most inbound references, making it the best representative user-facing bug report in this mixed set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 28218 is the strongest standalone triage candidate because it states a precise symptom ('additional space after the added token') and is likely to be useful as a canonical reference for duplicates.", + "best_issue_reason": "issue:44521 is the best global issue candidate because it has the highest visible activity and references, and it describes a concrete, reproducible regression rather than an abstract compatibility note.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both involve chat templating, but one is about multimodal assistant masks and the other about tool-call messages with no content; different failure modes and fixes." + }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "Missing Qwen2.5-VL import error vs DeepSeek detokenization regression; different component and failure mode." + "reason": "Import error for a missing VL class is unrelated to a detokenization regression in DeepSeek Coder." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "BF16/FP32 embedding dtype regression vs Qwen3 MoE memory-usage complaint; not the same bug." + "reason": "BF16->FP32 dtype drift in eval is unrelated to Qwen3 MoE memory usage during training." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer extra-space behavior vs embedding reinit after resize; both tokenizer/model-adjacent but clearly different defects." + "reason": "Tokenizer spacing after added tokens is a different bug from output embedding reinitialization when resizing embeddings." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate default vs batched video shape bug; unrelated processor paths." + "reason": "Audio chat-template sampling-rate defaulting and batched video output shape are distinct processor bugs." }, { "left": "issue:45381", "right": "issue:46209", "accept": false, - "reason": "Qwen2.5-VL positional-id regression vs FP8 inference failure on B200; different code paths and hardware-specific issue." + "reason": "Qwen2.5-VL vision position IDs regression is unrelated to FP8 inference failure on B200." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Detokenization bug vs Python 3.13 flash-attn import key error; unrelated." + "reason": "A decoding regression and a Python 3.13 flash_attn import error are unrelated." }, { "left": "issue:44737", "right": "issue:46894", "accept": false, - "reason": "XLNet CPU device placement bug vs CohereASR training-loss double-shift bug; not the same underlying issue." - }, - { - "left": "issue:44466", - "right": "issue:45127", - "accept": false, - "reason": "Serialization/tied-weights inconsistency vs LoRA merge collapse after vocab extension; different symptoms and fixes." + "reason": "CPU device placement in XLNet positional encoding is not the same as the CohereASR double-shift training-loss bug." }, { "left": "issue:43950", "right": "issue:46285", "accept": false, - "reason": "Non-persistent buffer corruption on load vs CLIP text model structure inconsistency; not a duplicate." + "reason": "Non-persistent buffer corruption on load is unrelated to CLIP text-model structure inconsistency." }, { - "left": "issue:43653", - "right": "issue:44112", + "left": "issue:44466", + "right": "issue:45127", "accept": false, - "reason": "Tokenizer special-token registration bug vs GraniteSpeech CI test flake; different artifacts and root causes." + "reason": "Both mention tied weights, but one is serialization inconsistency and the other is LoRA merge collapse with extended vocabulary; not the same code path." }, { - "left": "issue:44442", - "right": "issue:45198", + "left": "issue:43653", + "right": "issue:44112", "accept": false, - "reason": "AutoTokenizer load failure for FastSpeech2Conformer vs Wav2Vec2 save/tokenization failure; same broad area but distinct bugs." + "reason": "BigBird tokenizer special-token registration is unrelated to a GraniteSpeech CI device-override test issue." }, { "left": "issue:43452", "right": "issue:46512", "accept": false, - "reason": "gguf_file/from_pretrained loading issue vs from_config dtype-staleness bug; different mechanisms." + "reason": "GGUF loading breakage is unrelated to from_config(dtype=...) leaving config.dtype stale." }, { - "left": "issue:43611", - "right": "issue:47405", + "left": "issue:44442", + "right": "issue:45198", "accept": false, - "reason": "base_model_prefix loading regression vs silent conversion skip on wrong Auto class/model_type mismatch; related to loading, but not the same bug." + "reason": "FastSpeech2ConformerTokenizer loading failure is unrelated to Wav2Vec2 save_pretrained/tokenization failures." }, { "left": "issue:44561", "right": "issue:46234", "accept": false, - "reason": "Removal of is_torch_fx_available vs missing is_falcon_mamba_ssm_available; different missing-API regressions." + "reason": "Both touch import_utils, but one is removal of an API breaking trust_remote_code models and the other is a missing availability helper; different fixes." }, { - "left": "issue:45200", - "right": "issue:45325", + "left": "issue:43611", + "right": "issue:47405", "accept": false, - "reason": "Gemma4 mm_token_type_ids defaulting issue vs Qwen2.5-VL rope-index scaling bug; unrelated multimodal bugs." + "reason": "Base-model-prefix loading breakage is different from silent conversion skipping on wrong Auto class/model_type mismatch." }, { "left": "issue:44479", "right": "issue:46192", "accept": false, - "reason": "Qwen video-input regression vs return_tensors being ignored when text_kwargs is passed; different call sites and behavior." + "reason": "Qwen video-input regression is unrelated to return_tensors being ignored when text_kwargs is passed." }, { "left": "issue:45200", - "right": "issue:46899", + "right": "issue:45325", "accept": false, - "reason": "Missing mm_token_type_ids default vs unsafe dtype casting in quantized embedders; distinct multimodal failures." + "reason": "Gemma4 text-only mm_token_type_ids defaults and Qwen2.5-VL still-image position scaling are separate multimodal bugs." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "A config bug creating a sparse layer at zero layers is unrelated to a torch.compile/SDPA failure in Bamba-9B-v2." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "DAC forward/latents mismatch vs GraniteSpeech stale-device CI test; not the same code path." + "reason": "DAC forward/latents mismatch is unrelated to a stale CI device override test." }, { - "left": "issue:43335", - "right": "issue:43550", + "left": "issue:45200", + "right": "issue:46899", "accept": false, - "reason": "SwitchTransformers sparse-layer construction bug vs Bamba torch.compile SDPA failure; unrelated model/config bugs." + "reason": "Defaulting mm_token_type_ids to zeros is not the same as unsafe dtype casting in quantized multimodal embedders." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Perceiver interpolation/resolution bug vs RT-DETR memory leak; different symptoms and fixes." + "reason": "Perceiver interpolation failure and RT-DETR memory reclamation are unrelated model behaviors." }, { "left": "issue:41720", "right": "issue:46209", "accept": false, - "reason": "Qwen3 auto device-mapping assert on A800 vs FP8 inference failure on B200; hardware-specific but not the same issue." - }, - { - "left": "issue:45200", - "right": "issue:46932", - "accept": false, - "reason": "Gemma4 token-type-id default bug vs quantized vision/audio dtype cast bug; related model family but different defect." + "reason": "Qwen3 auto device mapping cuda assert and FP8 inference on B200 are different hardware/runtime failures." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Processor/model input-size mismatch vs float16 AMP crash in LwDetrImageLoss; different problems." + "reason": "Processor output resolution mismatch is unrelated to an AMP/CUDA loss crash." }, { - "left": "issue:37428", - "right": "issue:43425", + "left": "issue:45200", + "right": "issue:46932", + "accept": false, + "reason": "Both involve Gemma4 multimodal dtype behavior, but one is token-type defaults and the other is quantized embedder casting; not the same bug." + }, + { + "left": "issue:37428", + "right": "issue:43425", "accept": false, - "reason": "Missing flash-attn import symbol vs Torch 2.10 incompatibility; one is a symbol/export issue, the other a version-compatibility issue." + "reason": "A missing flash-attention helper import is not the same as generic Torch 2.10 incompatibility." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "DAC latent mismatch vs qwen3_omni_moe feature-length helper behavior; unrelated internals." + "reason": "DAC latent reconstruction mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "Sparse-layer config bug vs tokenizer special-token registration bug; no shared underlying defect." + "reason": "SwitchTransformers sparse-layer creation and BigBird tokenizer special-token registration are different subsystems and bugs." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU offload device mismatch in Bark vs qwen3_omni_moe feature-length helper behavior; different subsystems." - }, - { - "left": "issue:44792", - "right": "issue:45563", - "accept": false, - "reason": "Janus image-generation test failure vs paged generate stale warning; one is a failing test, the other a warning regression." - }, - { - "left": "issue:44857", - "right": "issue:45593", - "accept": false, - "reason": "AMP/CUDA crash in LwDetrImageLoss vs D-FINE auxiliary-loss omission; different model behaviors and likely fixes." + "reason": "CPU offload device mismatch in Bark is unrelated to qwen3_omni_moe feature-length calculation." } ] }, @@ -10018,11 +10043,8 @@ "issue:29942", "issue:32090", "issue:33357", - "issue:34689", - "issue:35141", "issue:37428", "issue:38175", - "issue:39692", "issue:42490", "issue:43335", "issue:43388", @@ -10037,17 +10059,19 @@ "issue:44112", "issue:44206", "issue:44351", - "issue:44448", "issue:44493", "issue:44617", - "issue:44743", + "issue:44792", "issue:44855", + "issue:44857", "issue:45083", "issue:45290", "issue:45412", "issue:45520", + "issue:45563", "issue:45584", "issue:45588", + "issue:45593", "issue:45759", "issue:45859", "issue:45923", @@ -10064,16 +10088,18 @@ "issue:47405" ], "soft_pairs": [ + "issue:44792|issue:45563", + "issue:44857|issue:45593", "issue:28282|issue:33357", "issue:46512|issue:46821", - "issue:45759|issue:46285", - "issue:43827|issue:43901", "issue:38175|issue:45412", - "issue:29942|issue:32090", + "issue:43827|issue:43901", + "issue:45759|issue:46285", "issue:43335|issue:43504", + "issue:29942|issue:32090", "issue:44206|issue:45584", - "issue:45923|issue:47405", "issue:43550|issue:44112", + "issue:45923|issue:47405", "issue:44855|issue:45083", "issue:46424|issue:46858", "issue:43388|issue:43526", @@ -10084,204 +10110,195 @@ "issue:37428|issue:44351", "issue:45290|issue:46326", "issue:42490|issue:43854", - "issue:45520|issue:46170", "issue:44617|issue:46697", + "issue:45520|issue:46170", "issue:43819|issue:44493", - "issue:45588|issue:45859", - "issue:44448|issue:44743", - "issue:34689|issue:35141", - "issue:39692|issue:43526" + "issue:45588|issue:45859" ], - "prepared_review_unit_hash": "c2fcff9cafd7aced50b16c5e9b8517098ab8c807b60237cfdb8ef6ac7b81081a", + "prepared_review_unit_hash": "7594aa04d362b8c01525d9e32c4fc5d8cf6ba37e3fde19985e4a0c4c1586f186", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 36364, - "estimated_input_tokens": 9091, - "estimated_eval_tokens": 18438 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35338, + "estimated_input_tokens": 8835, + "estimated_eval_tokens": 17926 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are largely unrelated, with the soft edges reflecting broad textual similarity rather than shared root causes. I would not merge any of the candidate pairs as duplicates.", - "confidence": 0.79, - "canonical_issue_reason": "If one issue must serve as the representative, issue 47405 is the strongest fit: it describes a concrete, high-severity model-loading correctness bug where using the wrong Auto class silently yields random weights.", + "summary": "This cluster is a loose similarity bucket, not a duplicate set. The soft pairs span unrelated bugs across generation, model loading, config handling, cache behavior, docs, and CI, so none should be merged as the same underlying issue.", + "confidence": 0.19, + "canonical_issue_reason": "If one issue must anchor the cluster, issue 43335 is the best representative only because it is a concrete code bug with the broadest set of weak similarity links here; however, it is not actually a duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 47405 is the best single issue to keep as canonical because it is clearly scoped, user-facing, and has strong impact on model correctness rather than a narrow subsystem quirk.", + "best_issue_reason": "Issue 43335 is the least-bad representative for triage, but the set is too heterogeneous for a true canonical issue.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "Different problems: janus image generation test failure vs a stale paged-generate warning about num_return_sequences." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "Different code paths: AMP/CUDA loss crash vs D-FINE not using auxiliary losses when denoising is off." + }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "One is a PyTorch import error, the other is a Mac bus error in a CLIP model; different failure modes and likely different fixes." + "reason": "Both are runtime failures, but one is a PyTorch import error and the other is a MacOS bus error for a CLIP model." }, { "left": "issue:46512", "right": "issue:46821", "accept": false, - "reason": "Stale dtype metadata in from_config is unrelated to the TimesFM window_size AttributeError." + "reason": "Unrelated bugs: stale config dtype after from_config vs window_size raising AttributeError in TimesFM 2.5." }, { - "left": "issue:45759", - "right": "issue:46285", + "left": "issue:38175", + "right": "issue:45412", "accept": false, - "reason": "Composite Qwen config unwrapping and CLIP text-model structure inconsistency affect different code paths and model families." + "reason": "Different behavior and subsystem: unexpected zero probabilities in a model vs RT-DETR memory not being released on deletion." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs-related, but one is about pipeline() removal and the other about return_all_scores behavior; not the same underlying issue." + "reason": "Both are docs-related, but they cover different APIs and different stale references; not the same bug." }, { - "left": "issue:38175", - "right": "issue:45412", + "left": "issue:45759", + "right": "issue:46285", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to RT-DETR memory release behavior." + "reason": "Different model/config issues: Qwen multimodal from_config unwrap problem vs CLIP class structure inconsistency." }, { - "left": "issue:29942", - "right": "issue:32090", + "left": "issue:43335", + "right": "issue:43504", "accept": false, - "reason": "Flash Attention 2 test failures and a Trainer broadcast TypeError are distinct bugs." + "reason": "Both involve Switch/Beit config-loading bugs, but one creates sparse layers incorrectly and the other fails to load a pretrained preset due to a legacy field." }, { - "left": "issue:43335", - "right": "issue:43504", + "left": "issue:29942", + "right": "issue:32090", "accept": false, - "reason": "A sparse-layer creation bug in SwitchTransformersConfig is unrelated to Beit pretrained preset loading with a legacy field." + "reason": "Flash Attention test failures vs Trainer _gpu_broadcast_one TypeError are distinct failures in different paths." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Unsupported center arg in LasrFeatureExtractor is a different bug from Whisper empty-transcription generation failure." + "reason": "Different regressions: LaserFeatureExtractor unsupported argument crash vs Whisper empty-transcription failure after token alignment." }, { - "left": "issue:45923", - "right": "issue:47405", + "left": "issue:43550", + "right": "issue:44112", "accept": false, - "reason": "Missing gradient_checkpointing support and silent wrong Auto-class checkpoint loading are not the same problem." + "reason": "Both are CI/runtime failures, but one is a torch.compile SDPA issue and the other is a stale GraniteSpeech device override test." }, { - "left": "issue:43550", - "right": "issue:44112", + "left": "issue:45923", + "right": "issue:47405", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to a stale CI device-override test in GraniteSpeech." + "reason": "Different classes of bug: missing gradient_checkpointing flag vs silent model_type mismatch during auto-class loading." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 import parsing failure in DebertaV2Model is unrelated to qwen3_omni_moe feature-extractor helper behavior." + "reason": "Different parser/processor bugs: Python 3.13 torch.jit import failure vs helper length-calculation behavior in qwen3_omni_moe." }, { "left": "issue:46424", "right": "issue:46858", "accept": false, - "reason": "Both involve cache internals, but one is a config knob being ignored and the other is device-tensor arithmetic in _prefill; different root causes." + "reason": "Cache settings bug vs static-cache seq_length returning a device tensor; related theme only, not same issue." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Tuple label handling in gather_for_metrics is a different bug from BeitImageProcessorFast reduce_labels returning only one label." + "reason": "Both affect label handling, but one drops tuple elements in gather_for_metrics while the other returns only one label in BeitImageProcessorFast." }, { "left": "issue:46431", "right": "issue:46566", "accept": false, - "reason": "Missing chat template support and weights \u10d3\u10d0\u10e0\u10e9ing on meta device are unrelated model/processor loading issues." + "reason": "Different loading failures: missing chat template on a processor vs DiffusionGemma weights left on meta with device_map=auto." }, { "left": "issue:43335", "right": "issue:44112", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and GraniteSpeech CI device override are unrelated." + "reason": "SwitchTransformers sparse-layer construction bug is unrelated to a GraniteSpeech CI device-override test." }, { "left": "issue:43335", "right": "issue:43606", "accept": false, - "reason": "SwitchTransformers configuration and Bark CPU offload device mismatch are different bugs." + "reason": "SwitchTransformers config bug vs bark-small CPU offload device mismatch; same broad area, different root cause." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "Sparse-layer config creation is unrelated to qwen3_omni_moe feature extraction helper behavior." + "reason": "Different components: sparse layer creation in SwitchTransformers vs helper output-length behavior in qwen3_omni_moe." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Different missing imports (_flash_supports_window_size vs HybridCache) with no shared code path or fix." + "reason": "Different import errors: missing flash_attention_utils symbol vs cannot import HybridCache." }, { "left": "issue:45290", "right": "issue:46326", "accept": false, - "reason": "Both touch chat templates, but one crashes on assistant tool-call messages and the other concerns continue_final_message in mistral_common tokenizers." + "reason": "Different chat-template failures: tool-call assistant message crash vs continue_final_message bug for mistral_common tokenizers." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "A local-path trailing-slash warning is unrelated to the GLM-4.7-Flash unit-test loading failure." + "reason": "Both affect model loading, but one is a local-path trailing-slash warning and the other is a unit-test model-loading failure for GLM-4.7-Flash." }, { - "left": "issue:45520", - "right": "issue:46170", + "left": "issue:44617", + "right": "issue:46697", "accept": false, - "reason": "Python 3.13 flash_attn import issues and Windows ROCm torch.distributed problems are different platform-specific failures." + "reason": "Completely different image/video processing bugs: CUDA OOM in Sam3Video vs over-padding in Swin2SR image processor." }, { - "left": "issue:44617", - "right": "issue:46697", + "left": "issue:45520", + "right": "issue:46170", "accept": false, - "reason": "CUDA OOM in Sam3Video is unrelated to Swin2SR image over-padding." + "reason": "Both are import-time issues on unusual platforms, but one is flash_attn on Python 3.13 and the other is Windows ROCm torch.distributed dependency handling." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC.from_latents/forward mismatch is unrelated to unexpected position-id keys across models." + "reason": "Different numeric/model bugs: DAC latent reconstruction mismatch vs unexpected position-id keys across many models." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "One is a flash_attention integration crash on s_aux=None; the other is a CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding. Different concrete bugs." - }, - { - "left": "issue:44448", - "right": "issue:44743", - "accept": false, - "reason": "Pegasus v4/v5 output differences are unrelated to recurrent-state reset behavior in qwen3_5 modular code." - }, - { - "left": "issue:34689", - "right": "issue:35141", - "accept": false, - "reason": "Llama 3.2 Vision load regression and token-embedding reinitialization after resize are distinct model-loading/training issues." - }, - { - "left": "issue:39692", - "right": "issue:43526", - "accept": false, - "reason": "SigLIP2 docs example errors are unrelated to BeitImageProcessorFast's reduce_labels bug." + "reason": "Not the same bug: flash_attention integration crashes on s_aux=None, while Qwen3_5MoeTextRotaryEmbedding is incompatible with CPU offload." } ] }, @@ -10325,6 +10342,7 @@ "issue:42490", "issue:43278", "issue:43329", + "issue:43526", "issue:43550", "issue:43575", "issue:43606", @@ -10333,12 +10351,11 @@ "issue:43825", "issue:44112", "issue:44206", - "issue:44360", "issue:44448", "issue:44493", "issue:44514", "issue:44521", - "issue:44561", + "issue:44743", "issue:44779", "issue:44857", "issue:45072", @@ -10354,6 +10371,9 @@ "issue:47000" ], "soft_pairs": [ + "issue:44448|issue:44743", + "issue:34689|issue:35141", + "issue:39692|issue:43526", "issue:44206|issue:45823", "issue:39692|issue:44493", "issue:43606|issue:44857", @@ -10361,17 +10381,17 @@ "issue:44514|issue:44521", "issue:29127|issue:32090", "issue:28218|issue:32090", - "issue:43720|issue:44112", "issue:43701|issue:45083", + "issue:43720|issue:44112", "issue:44448|issue:44779", "issue:44857|issue:45072", "issue:38175|issue:44493", "issue:32090|issue:33357", - "issue:45310|issue:45478", "issue:42490|issue:43575", "issue:43278|issue:43825", "issue:43550|issue:45083", "issue:26413|issue:31501", + "issue:45310|issue:45478", "issue:37428|issue:42175", "issue:45127|issue:45245", "issue:45459|issue:46234", @@ -10379,205 +10399,216 @@ "issue:29942|issue:35141", "issue:45593|issue:47000", "issue:29870|issue:32090", - "issue:44360|issue:44561", "issue:43278|issue:43329" ], - "prepared_review_unit_hash": "1db5461fa1121f500430d8792381870c81992f38b455db7a11bb10779cef8f43", + "prepared_review_unit_hash": "0c9dbcd5419aa58e2e3efc4b8bd673f8c1362466fb3d1bef1d6f1faca913a15f", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 27, - "serialized_chars": 35292, - "estimated_input_tokens": 8823, - "estimated_eval_tokens": 17902 + "soft_pair_count": 29, + "serialized_chars": 35968, + "estimated_input_tokens": 8992, + "estimated_eval_tokens": 18240 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Only one soft pair looks like the same underlying bug: the two identical Qwen3.5 MoE from_pretrained reports. The rest are similar at the subsystem level but describe different concrete failures, so they should stay separate.", - "confidence": 0.95, - "canonical_issue_reason": "issue:45310 is the best canonical issue: it has the exact same bug title as issue:45478, was filed first, and already has an inbound reference, making it the cleanest anchor for the duplicate report.", + "summary": "This cluster is mostly a set of unrelated Transformers issues grouped by weak text similarity. I only found one clear duplicate pair: the two identical Qwen3.5 MoE `from_pretrained` errors. The rest share only broad subsystems (tokenizers, training, vision, dtype, etc.) but describe different bugs or code paths.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45310 is the earliest, clearest report of the duplicated Qwen3.5 MoE `from_pretrained` failure and can serve as the primary tracker for that bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest representative issue in this cluster because it is the earliest, most explicit duplicate of the Qwen3.5 MoE from_pretrained failure and is the most likely primary tracker.", + "best_issue_reason": "issue:45310 is the best issue to keep as the canonical representative because it is the first of the identical duplicate pair and has the same concrete failure mode as issue:45478.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Different models and failures: Pegasus output drift vs Qwen3 recurrent state reset." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "One is a Llama 3.2 model-loading regression; the other is embedding reinitialization after resizing. Not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Documentation/example errors vs BeitImageProcessorFast label reduction bug." + }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Both mention device/dtype mismatch, but they involve different models and different failure modes; this is broad thematic similarity, not the same bug." + "reason": "Different subsystems and symptoms: feature extractor arg crash vs PLE device mismatch." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 documentation errors and unexpected position-id key warnings are different problems affecting different code paths." + "reason": "SigLIP2 doc/example issues are unrelated to position-id key warnings." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "One is a CPU offload device mismatch in bark-small; the other is an AMP/CUDA loss crash in LwDetrImageLoss. Same general area, different bug." + "reason": "Both are runtime failures, but in different models and code paths (CPU offload vs AMP loss)." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "Completely different models and errors: German DistilBertModel issue versus a torch.jit.script GeneratorExp export problem." + "reason": "German DistilBert issue vs TorchScript generator expression export error; unrelated." }, { "left": "issue:44514", "right": "issue:44521", "accept": false, - "reason": "Both involve Qwen2.5 VL chat templating, but one crashes on batched input with padding=False while the other returns all-zero assistant masks; not the same concrete defect." + "reason": "Same broad processor family, but one is a batched padding crash and the other is incorrect assistant masks." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 box validation messaging and a Trainer _gpu_broadcast_one NoneType error are unrelated bugs." + "reason": "LayoutLMv3 error messaging issue vs Trainer GPU broadcast NoneType error." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer spacing after added tokens and a Trainer broadcast NoneType failure are unrelated code paths." + "reason": "Tokenizer added-space bug is unrelated to Trainer broadcast failure." }, { - "left": "issue:43720", - "right": "issue:44112", + "left": "issue:43701", + "right": "issue:45083", "accept": false, - "reason": "BitNet packed-weight loading and GraniteSpeech stale device override tests are different CI failures, not one underlying bug." + "reason": "Checkpoint key mismatch vs qwen3_omni_moe feature-extraction helper behavior; different bugs." }, { - "left": "issue:43701", - "right": "issue:45083", + "left": "issue:43720", + "right": "issue:44112", "accept": false, - "reason": "resume_from_checkpoint key mismatch and qwen3_omni_moe feature-length helper behavior are unrelated issues." + "reason": "Accelerate loading/unpacked weights bug vs GraniteSpeech CI test flake; not the same issue." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Different models and different regressions: Pegasus output drift versus DeepSeek tokenizer correctness." + "reason": "Pegasus output mismatch vs DeepSeek tokenizer regression; different components and failures." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "Both are dtype-related, but they concern different models and different failure sites; not mergeable as one fix." + "reason": "Both mention dtype, but the affected models and failure mechanisms differ." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "Zero probabilities in SigLIP2 and unexpected position-id keys across models are distinct symptoms with different likely causes." + "reason": "SigLIP2 zero probabilities vs unexpected position-id key warnings; unrelated." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "A NoneType broadcast error in Trainer is not the same bug as a MacOS bus error for a CLIP model." - }, - { - "left": "issue:45310", - "right": "issue:45478", - "accept": true, - "reason": "This is an exact duplicate pair: same title, same model, same from_pretrained failure on transformers>=5.4.0." + "reason": "Trainer broadcast NoneType error vs MacOS bus error on CLIP; unrelated." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "A local-path import warning and a tensor-parallel OOM while loading Qwen2-57B are unrelated." + "reason": "Import warning on local path vs tensor-parallel OOM during model load; unrelated." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype regression during evaluation and a pipeline translation-support error message are different issues." + "reason": "BF16/FP32 dtype regression vs pipeline translation error message; not the same underlying bug." }, { "left": "issue:43550", "right": "issue:45083", "accept": false, - "reason": "Bamba torch.compile/SDPA failure and qwen3_omni_moe feature-length behavior are unrelated bugs." + "reason": "Torch.compile/SDPA failure in Bamba vs qwen3_omni_moe feature-length helper issue." }, { "left": "issue:26413", "right": "issue:31501", "accept": false, - "reason": "Both relate to training startup, but one is a resume_from_checkpoint epoch_iterator error and the other is FineWeb SLM training not starting; too vague to be the same bug." + "reason": "Both affect training startup/resume, but one is checkpoint resume logic and the other is FineWeb training not starting." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same title, same version range, and same concrete `Qwen3.5 Moe from_pretrained` error." }, { "left": "issue:37428", "right": "issue:42175", "accept": false, - "reason": "Missing flash-support import and TensorFlow not included in a torch install are separate packaging/import issues." + "reason": "Missing import in flash attention utils vs installation backend dependency issue; unrelated." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "LoRA/vocabulary collapse with tied embeddings and a category-count limit runtime error are different failures." + "reason": "LoRA/vocab merge collapse vs category cardinality limit error; different bugs." }, { "left": "issue:45459", "right": "issue:46234", "accept": false, - "reason": "Tokenizer protobuf error handling and a missing import_utils availability symbol are unrelated maintenance bugs." + "reason": "Tokenizer protobuf error masking vs missing utility function export; unrelated." }, { "left": "issue:27957", "right": "issue:34689", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta and Llama 3.2 vision model loading are different model-specific issues." + "reason": "Flash Attention 2 support issue vs model loading regression for a different model family." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Failing Flash Attention 2 tests and embedding reinitialization after resize are distinct bugs in different areas." + "reason": "Flash Attention 2 test failures vs embedding reinitialization after resizing; different problems." }, { "left": "issue:45593", "right": "issue:47000", "accept": false, - "reason": "D-FINE auxiliary-loss handling and RTDetrLoss infeasible cost matrix are both detection-related but not the same defect." + "reason": "Different object-detection loss failures: missing auxiliary losses vs infeasible cost matrix." }, { "left": "issue:29870", "right": "issue:32090", "accept": false, - "reason": "BART generate decoder_attention_mask initialization and a Trainer NoneType broadcast error are unrelated." - }, - { - "left": "issue:44360", - "right": "issue:44561", - "accept": false, - "reason": "A DSA indexer ReLU discussion and a v5 import removal breaking trust_remote_code models are not the same underlying issue." + "reason": "BART generate decoder mask bug vs Trainer broadcast NoneType error; unrelated." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "BF16-to-FP32 evaluation dtype drift and undefined video-branch variables in multimodal token counting are different bugs." + "reason": "Embedding dtype regression vs multimodal token counting video-branch undefined names; unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative on the soft edges. The canonical choice of issue:45310 over issue:45478 is supported by the identical title, earlier creation date, and inbound reference. The rejected soft pairs are also phrased cautiously as related but distinct bugs. No overstatement stands out." + "feedback": "The summary is grounded in the packet and stays conservative on the soft-edge pairs. The canonical duplicate call on issue:45310 / issue:45478 is well supported by the identical title, version range, and failure mode. The rest of the verdicts are phrased cautiously and do not overclaim duplication or shared root cause." }, "evaluator_used": true, "retried": false, @@ -10609,7 +10640,6 @@ "issue:34567", "issue:39692", "issue:40444", - "issue:41720", "issue:42490", "issue:43257", "issue:43278", @@ -10621,6 +10651,7 @@ "issue:43873", "issue:43927", "issue:44186", + "issue:44360", "issue:44361", "issue:44448", "issue:44464", @@ -10648,6 +10679,7 @@ "issue:47276" ], "soft_pairs": [ + "issue:44360|issue:44561", "issue:45290|issue:45381", "issue:25251|issue:32090", "issue:29870|issue:33290", @@ -10657,9 +10689,9 @@ "issue:45290|issue:46655", "issue:40444|issue:43526", "issue:43278|issue:43381", - "issue:43257|issue:43299", "issue:44855|issue:46399", "issue:44530|issue:47276", + "issue:43257|issue:43299", "issue:43927|issue:45698", "issue:44977|issue:45229", "issue:42490|issue:43873", @@ -10667,82 +10699,87 @@ "issue:44561|issue:44779", "issue:29127|issue:34567", "issue:33357|issue:39692", - "issue:43873|issue:45229", "issue:44871|issue:44977", + "issue:43873|issue:45229", "issue:29942|issue:30064", "issue:44186|issue:45198", - "issue:44448|issue:44466", "issue:43381|issue:43701", + "issue:44448|issue:44466", "issue:44361|issue:45198", "issue:45588|issue:45823", - "issue:46421|issue:46858", "issue:43278|issue:43575", - "issue:41720|issue:45229" + "issue:46421|issue:46858" ], - "prepared_review_unit_hash": "8df23c443c73637cbac20dcce13d26731c8a7152ba3f01181d20ffbd10a06b37", + "prepared_review_unit_hash": "69a7297450500ab21512dbaccc2ec6658df7866b1bfdac1f2e01a521de082b33", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36853, - "estimated_input_tokens": 9214, - "estimated_eval_tokens": 18684 + "serialized_chars": 36835, + "estimated_input_tokens": 9209, + "estimated_eval_tokens": 18674 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft pairs are mostly broad thematic neighbors, not true duplicates. I would not merge any of them; they describe different code paths, models, or failure modes.", - "confidence": 0.92, - "canonical_issue_reason": "issue:45920 is the broadest and most representative regression in the set: it affects multiple models, has a clear tokenizer-focused failure mode, and reads like a real umbrella bug rather than a narrow edge case.", + "summary": "These are mostly unrelated issues sharing only broad subsystem similarity (tokenization, generation, cache, vision, trainer, etc.). I would reject all soft edges; none look like the same underlying bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best single issue to anchor the cluster because it is concrete, cross-model, and more general than the other narrowly scoped bugs.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "Different problems: DSA indexer/ReLU discussion vs trust_remote_code breakage from removing is_torch_fx_available." + }, { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Different subsystems: chat template/tool-call handling vs Qwen2.5-VL video vision position IDs." + "reason": "Different code paths: chat templating with tool-call messages vs Qwen2.5-VL video vision_position_ids." }, { "left": "issue:25251", "right": "issue:32090", "accept": false, - "reason": "Unrelated bugs: pipeline output structure vs Trainer `_gpu_broadcast_one` NoneType error." + "reason": "Pipeline top_k nesting bug is unrelated to Trainer _gpu_broadcast_one NoneType error." }, { "left": "issue:29870", "right": "issue:33290", "accept": false, - "reason": "Different failure classes: BART generate-time mask initialization vs deepspeed+adafactor OOM." + "reason": "BART generate decoder_attention_mask assignment bug vs DeepSpeed AdaFactor OOM." }, { "left": "issue:44464", "right": "issue:46424", "accept": false, - "reason": "Both involve cache/generation, but one is inconsistent compiled chunked generation output and the other is a cache-length config limitation." + "reason": "Both cache/generation related, but one is compiled chunked generation inconsistency and the other is max_cache_length honoring for quantized cache." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different models and root causes: Pegasus v4/v5 output drift vs Qwen3.5 flash-attention generation failure." + "reason": "Different models and failures: Pegasus v4/v5 output mismatch vs Qwen3.5 flash-attention generation failure." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer broadcast type error and optimizer OOM are unrelated." + "reason": "Trainer broadcast NoneType vs optimizer OOM; no shared bug." }, { "left": "issue:45290", @@ -10754,139 +10791,121 @@ "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "Both are image-related, but one is multi-image VL fine-tuning and the other is BeitImageProcessorFast label reduction." + "reason": "Different processors and symptoms: Qwen2.5-VL iterable dataset multi-image failure vs BeitImageProcessorFast reduce_labels bug." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "Eval-mode dtype change and gradient-checkpointing-in-eval are distinct issues, not the same bug." - }, - { - "left": "issue:43257", - "right": "issue:43299", - "accept": false, - "reason": "Both mention Qwen3 MoE loading, but one is accelerate/deepspeed conversion and the other is a general v5 loading regression." + "reason": "Training/eval dtype mismatch is unrelated to gradient checkpointing being disallowed in eval mode." }, { "left": "issue:44855", "right": "issue:46399", "accept": false, - "reason": "Both are decorator/script-related import problems, but they affect different models and code paths." + "reason": "Python parser/torch.jit import issue vs Qwen3.5 kernel registration/decorator error." }, { "left": "issue:44530", "right": "issue:47276", "accept": false, - "reason": "PagedAttention cache crash and a warning-message issue are clearly unrelated." + "reason": "PagedAttentionCache crash on Qwen3.5 is unrelated to a warning message in continuous batching." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MOE, but one is accelerate+deepspeed weight conversion and the other is v5 model loading regression; not the same concrete bug." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Both are save/load regressions, but one loses custom token IDs and the other loads the wrong remote custom module." + "reason": "Both are save/load related, but one loses custom token IDs in config and the other loads the wrong custom module after save_pretrained." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Flash-attention generation bug and multi-GPU CUDA OOM are different problems." + "reason": "Qwen3.5 flash-attention generation bug is unrelated to Gemma4 multi-GPU CUDA OOM." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Local-path trailing-slash import warning is unrelated to quantization offloading behavior." + "reason": "Local-path import warning and quantized offloading failure are different issues." }, { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "Tokenizer ID regression across models is not the same as silently ignoring `return_tensors` when `text_kwargs` is passed." + "reason": "AutoTokenizer wrong IDs regression is unrelated to return_tensors being ignored with explicit text_kwargs." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Remote-code breakage from removing a utility function is unrelated to DeepSeek tokenizer output changes." + "reason": "Trust_remote_code breakage from API removal vs DeepSeek tokenizer regression; different failure modes." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "LayoutLMv3 error messaging and TrainerState token-count tracking are unrelated." + "reason": "LayoutLMv3 error messaging and TrainerState token counter updating are unrelated." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "MacOS CLIP bus error and SigLIP2 documentation/example errors are not the same bug." + "reason": "Different vision-related issues: MacOS bus error on CLIP model vs SigLIP2 documentation/example errors." }, { - "left": "issue:43873", - "right": "issue:45229", + "left": "issue:44871", + "right": "issue:44977", "accept": false, - "reason": "Quantized offloading issues and multi-GPU OOM are different memory-path failures." + "reason": "Gemma eos_token_id config inconsistency is unrelated to Qwen3.5 flash-attention generation failure." }, { - "left": "issue:44871", - "right": "issue:44977", + "left": "issue:43873", + "right": "issue:45229", "accept": false, - "reason": "Gemma eos-token config mismatch is unrelated to Qwen3.5 flash-attention generation issues." + "reason": "Quantization offloading not working is not the same as Gemma4 multi-GPU OOM." }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Flash Attention 2 tests and void segmentation-map processing are unrelated." + "reason": "Flash Attention 2 test failures are unrelated to image processor void segmentation map processing." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "Different tokenizer families and different failure modes: NER padding/truncation crash vs Wav2Vec2 save/tokenization failure." - }, - { - "left": "issue:44448", - "right": "issue:44466", - "accept": false, - "reason": "Both are v5 regressions, but one is generation output drift and the other is serialization of tied weights; not the same code-path bug." - }, - { - "left": "issue:43381", - "right": "issue:43701", - "accept": false, - "reason": "Gradient checkpointing in eval mode and resume-from-checkpoint key mismatch are unrelated." + "reason": "LayoutLMv2 tokenizer NER/padding crash and Wav2Vec2 save_pretrained/tokenization failure are different tokenizer bugs." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "MLuke tokenizer AttributeError and Wav2Vec2 save/tokenization failure do not share the same underlying defect." + "reason": "MLukeTokenizer AttributeError and Wav2Vec2 save_pretrained/tokenization failure are unrelated." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "Flash-attention sinkless-model crash and Gemma4 device-mapping mismatch are different forward-path bugs." - }, - { - "left": "issue:46421", - "right": "issue:46858", - "accept": false, - "reason": "Both involve cache/compile, but one is spurious recompilation from lazy init and the other is a device-tensor seq-length arithmetic bug." + "reason": "Flash-attention integration s_aux=None crash is unrelated to Gemma4 PLE device mismatch." }, { - "left": "issue:43278", - "right": "issue:43575", + "left": "issue:43381", + "right": "issue:43701", "accept": false, - "reason": "Eval dtype drift and tensor-parallel OOM are unrelated." + "reason": "Eval-mode gradient checkpointing restriction vs resume_from_checkpoint key mismatch are different trainer issues." }, { - "left": "issue:41720", - "right": "issue:45229", + "left": "issue:46421", + "right": "issue:46858", "accept": false, - "reason": "Auto device-mapping assert on A800 and Gemma4 multi-GPU OOM are different GPU failure modes." + "reason": "Chunked-prefill recompilation from lazy cache init differs from non-CUDA static-cache seq_length slice-arithmetic bug." } ] }, @@ -10915,6 +10934,7 @@ "issue:33357", "issue:34689", "issue:35141", + "issue:41720", "issue:42490", "issue:42915", "issue:42994", @@ -10938,18 +10958,17 @@ "issue:44442", "issue:44479", "issue:44568", - "issue:44625", "issue:44743", "issue:44977", "issue:45072", "issue:45198", + "issue:45229", "issue:45372", "issue:45406", "issue:45412", "issue:45440", "issue:45491", "issue:45563", - "issue:45584", "issue:45588", "issue:45593", "issue:45803", @@ -10960,6 +10979,7 @@ "issue:47276" ], "soft_pairs": [ + "issue:41720|issue:45229", "issue:43976|issue:45440", "issue:44206|issue:44479", "issue:45198|issue:45803", @@ -10969,236 +10989,235 @@ "issue:44265|issue:45198", "issue:44077|issue:47276", "issue:44977|issue:45563", - "issue:44568|issue:45491", "issue:43412|issue:45588", + "issue:44568|issue:45491", "issue:43653|issue:44743", - "issue:42915|issue:42994", "issue:43756|issue:45440", + "issue:42915|issue:42994", "issue:43065|issue:43994", - "issue:29127|issue:33357", "issue:43575|issue:43701", + "issue:29127|issue:33357", "issue:42490|issue:43381", "issue:43334|issue:43531", "issue:43575|issue:44387", - "issue:45491|issue:46693", "issue:43065|issue:45412", + "issue:45491|issue:46693", "issue:43994|issue:45412", "issue:29127|issue:35141", "issue:45406|issue:46701", "issue:45372|issue:45588", "issue:43526|issue:47217", "issue:44442|issue:45072", - "issue:45440|issue:45593", - "issue:44625|issue:45584" + "issue:45440|issue:45593" ], - "prepared_review_unit_hash": "56f0ba7702ba433f3aded65edbc6317d059aa3a2833f589c9020bfc06eec0a99", + "prepared_review_unit_hash": "7538b3f87669dc3b474b3c4e3a6d4ecade91aad3671ac1d24a496ebd4e8a3e81", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36867, - "estimated_input_tokens": 9217, - "estimated_eval_tokens": 18690 + "serialized_chars": 36828, + "estimated_input_tokens": 9207, + "estimated_eval_tokens": 18670 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the paired issues share only shallow title similarity and span unrelated regressions/features across generation, tokenization, quantization, export, vision, and warnings. I would not merge any of the soft edges.", + "summary": "These are mostly unrelated issues that only share broad Transformers vocabulary (generation, loading, quantization, VL, OOM, etc.). I do not see any soft pair that looks like the same underlying bug/change, so all soft edges should be rejected.", "confidence": 0.95, - "canonical_issue_reason": "No clear canonical issue: the items do not share a common root cause or subsystem, so none can safely subsume the others.", + "canonical_issue_reason": "No single issue cleanly represents a duplicate cluster here; the items span many unrelated subsystems and failure modes.", "canonical_pr_reason": null, - "best_issue_reason": "If one issue must stand in as the most representative, issue 44387 is the broadest and most active concrete regression report, but it still does not duplicate the rest of the cluster.", + "best_issue_reason": "If one issue must anchor triage, issue 45229 is the strongest standalone report: it is concrete, high-activity, and describes a specific reproducible multi-GPU OOM. It is not a duplicate hub, just the most substantial issue in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Both are OOM-related, but one is a Qwen3 auto device-mapping cudaErrorAssert and the other is Gemma4 multi-GPU CUDA OOM; different models, triggers, and code paths." + }, { "left": "issue:43976", "right": "issue:45440", "accept": false, - "reason": "Different problems: Python-version breakage vs DeepSeekV3 implementation divergence." + "reason": "Python-version compatibility breakage vs a model divergence bug; no shared failure mode beyond Transformers version context." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "Unrelated regressions in different models/code paths; one is audio feature extraction, the other is video input handling." + "reason": "Feature extractor argument regression vs video-input regression across several Qwen VL models; unrelated components and symptoms." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Both touch preprocessing/saving, but the root causes differ: tokenization/save_pretrained failure vs a bare except swallowing errors." + "reason": "Tokenizer save/tokenization failure vs a bare-except swallowing errors in a batch feature converter; different classes of bugs." }, { "left": "issue:29870", "right": "issue:34689", "accept": false, - "reason": "Different models and failure modes: BART generate() bug vs Llama 3.2 vision model loading regression." + "reason": "BART generate() UnboundLocalError vs Llama 3.2 Vision model loading breakage; unrelated models and code paths." }, { "left": "issue:43611", "right": "issue:46121", "accept": false, - "reason": "Base-model-prefix loading issue is unrelated to rope-parameter validation on JSON-loaded lists." + "reason": "base_model_prefix loading regression vs rope validation helper TypeError; no concrete overlap in behavior or fix." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "Tokenizer special-token registration bug is not the same as AutoTokenizer failing to load a specific tokenizer class." + "reason": "BigBirdTokenizer special-token decode bug vs AutoTokenizer failing on FastSpeech2ConformerTokenizer; both tokenizer-related but distinct failures." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export + torch_compilable_check is a different export path from wav2vec2 save/tokenization failures." + "reason": "torch.export/exportability issue with torch_compilable_check vs Wav2Vec2 save/tokenization failure; separate subsystems." }, { "left": "issue:44077", "right": "issue:47276", "accept": false, - "reason": "Optional post_init enforcement and a misleading synced_gpus warning are unrelated." + "reason": "patchtsmixer post_init policy issue vs a misleading synced_gpus warning message; one is a model class constraint, the other is logging text." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Flash-attention generation bug and a stale paged-generate warning are different issues." + "reason": "Qwen3.5 flash-attention generation failure vs a stale warning in paged generate(); different bug surface and no shared fix." }, { - "left": "issue:44568", - "right": "issue:45491", + "left": "issue:43412", + "right": "issue:45588", "accept": false, - "reason": "Tokenizer special-token insertion bug is unrelated to Gemma3 NaN embeddings from sliding-window attention." + "reason": "Executorch export failure on gemma3n vs flash_attention crash on s_aux=None; both export/integration-adjacent but clearly different code paths." }, { - "left": "issue:43412", - "right": "issue:45588", + "left": "issue:44568", + "right": "issue:45491", "accept": false, - "reason": "Executable export guard/unsupported op issue does not match the flash-attention sinkless-model crash." + "reason": "Tokenizer special-token behavior vs Gemma3 NaN embeddings under sliding-window attention; unrelated layers of the stack." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "Tokenizer special-token bug and recurrent-state reset in Qwen3.5 are different code paths." + "reason": "Tokenizer decode/special-token issue vs Qwen3.5 recurrent state reset in cached generation; no substantive overlap." }, { - "left": "issue:42915", - "right": "issue:42994", + "left": "issue:43756", + "right": "issue:45440", "accept": false, - "reason": "Fine-grained FP8 config failure is not the same as quantized model saving failure." + "reason": "Smollm3 RoPE-layer count mismatch vs DeepSeekV3MoE implementation divergence; both model-architecture issues, but different models and fixes." }, { - "left": "issue:43756", - "right": "issue:45440", + "left": "issue:42915", + "right": "issue:42994", "accept": false, - "reason": "RoPE layer count mismatch and DeepSeekV3 implementation divergence are distinct model-spec issues." + "reason": "Qwen3Moe FP8 config failure vs quantized model saving failure; both quantization-adjacent but not the same bug." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Dummy Conv2d implementation detail is unrelated to SigLIP2 AutoModel/pipeline bad outputs." + "reason": "Sam3PixelDecoder dummy Conv2d concern vs SigLIP2 AutoModel/pipeline incorrect outputs; unrelated model families and symptoms." }, { - "left": "issue:29127", - "right": "issue:33357", + "left": "issue:43575", + "right": "issue:43701", "accept": false, - "reason": "Error-message clarity for LayoutLMv3 is not the same as a MacOS bus error with a CLIP model." + "reason": "Tensor-parallel OOM when loading Qwen2-57B vs resume_from_checkpoint key mismatch; different stages and failure modes." }, { - "left": "issue:43575", - "right": "issue:43701", + "left": "issue:29127", + "right": "issue:33357", "accept": false, - "reason": "Tensor-parallel OOM and resume_from_checkpoint key mismatch are unrelated." + "reason": "LayoutLMv3 box-information error message vs CLIP community-model bus error on macOS; only superficial model-loading/build similarity." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Trailing-slash import warning and gradient-checkpointing-in-eval-mode are different behaviors." + "reason": "Trailing-slash local model import warning vs gradient checkpointing disallowed in eval mode; unrelated behaviors." }, { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Missing pad_token_id in Qwen3-VL is unrelated to sliding_window behavior in Qwen3-MoE." + "reason": "Qwen3-VL missing pad_token_id in config vs sliding_window issue for Qwen3-MoE; different model types and root causes." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "TP OOM on one model is not the same as global int4 quantization reserved-memory regression." + "reason": "Loading a large Qwen2 model with tp causing OOM vs increased CUDA reserved memory under int4 quantization; both OOMs, but not the same underlying regression." }, { - "left": "issue:45491", - "right": "issue:46693", + "left": "issue:43065", + "right": "issue:45412", "accept": false, - "reason": "Both are performance/attention-related, but one is a NaN correctness bug and the other is a generation performance regression." + "reason": "Dummy Conv2d concern in Sam3PixelDecoder vs RT-DETR memory not released on deletion; unrelated model architecture vs lifecycle/memory leak bug." }, { - "left": "issue:43065", - "right": "issue:45412", + "left": "issue:45491", + "right": "issue:46693", "accept": false, - "reason": "Dummy layer implementation and RT-DETR memory-release behavior do not share a root cause." + "reason": "Gemma3 NaNs from all-padding sliding windows vs FA2 long-context performance regression; one is correctness/NaN, the other is performance." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "Bad outputs from AutoModel/pipeline and model deletion memory retention are unrelated." + "reason": "Wrong outputs from SigLIP2 AutoModel/pipeline vs RT-DETR memory leak after deletion; entirely different symptoms." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "Error messaging in LayoutLMv3 has nothing to do with embedding reinitialization after resize." + "reason": "Error-message clarity for LayoutLMv3 boxes vs token embedding resize reinitialization with untied embeddings; unrelated functionality." }, { "left": "issue:45406", "right": "issue:46701", "accept": false, - "reason": "Gemma4 processor loading failure and DiffusionGemma pipeline wiring are separate problems." + "reason": "Gemma4 processor attribute error in serve vs DiffusionGemmaGenerationOutput pipeline failure; same broad model family only, but not the same bug." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Missing import from mistral_common is unrelated to flash_attention handling of s_aux=None." + "reason": "Gemma 4 processor loading breakage from mistral_common import vs flash_attention crash on sink-less models; different failure points and remedies." }, { "left": "issue:43526", "right": "issue:47217", "accept": false, - "reason": "BEiT image label reduction bug is unrelated to VL image-placeholder parsing." + "reason": "BeitImageProcessorFast reduce_labels bug vs VL image-placeholder-in-user-text failure; both involve vision inputs, but not the same defect." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Tokenizer loading failure and bfloat16 dtype mismatches are different subsystems." + "reason": "Tokenizer loading failure vs dtype mismatch in bfloat16 inference; unrelated tokenizer vs inference dtype bug." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "DeepSeekV3 divergence and D-FINE auxiliary-loss omission are unrelated model bugs." - }, - { - "left": "issue:44625", - "right": "issue:45584", - "accept": false, - "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription handling are unrelated." + "reason": "DeepseekV3MoE implementation divergence vs D-FINE auxiliary-loss omission when denoising is off; different models and behaviors." } ] }, @@ -11243,11 +11262,11 @@ "issue:44442", "issue:44462", "issue:44514", + "issue:44625", "issue:44743", "issue:44811", "issue:44857", "issue:44898", - "issue:44977", "issue:45245", "issue:45250", "issue:45325", @@ -11258,6 +11277,7 @@ "issue:45459", "issue:45491", "issue:45563", + "issue:45584", "issue:45588", "issue:45712", "issue:45834", @@ -11271,25 +11291,26 @@ "issue:47312" ], "soft_pairs": [ + "issue:44625|issue:45584", "issue:43421|issue:45491", "issue:45459|issue:45712", "issue:45412|issue:46369", - "issue:45372|issue:45406", "issue:43066|issue:44462", "issue:45250|issue:46640", "issue:43526|issue:45834", - "issue:45372|issue:45459", + "issue:45372|issue:45406", "issue:45245|issue:45563", "issue:27957|issue:33290", + "issue:45372|issue:45459", "issue:33290|issue:34689", "issue:43824|issue:44279", "issue:43526|issue:46693", - "issue:27957|issue:29127", "issue:33290|issue:35141", + "issue:27957|issue:29127", "issue:43994|issue:46369", "issue:43381|issue:43873", - "issue:42915|issue:43931", "issue:44857|issue:45588", + "issue:42915|issue:43931", "issue:45588|issue:46821", "issue:44514|issue:45381", "issue:44442|issue:44898", @@ -11297,205 +11318,204 @@ "issue:45325|issue:46697", "issue:46121|issue:46399", "issue:44743|issue:44811", - "issue:44977|issue:45245", "issue:43262|issue:44206" ], - "prepared_review_unit_hash": "264f6ef9c9bb215e67fa67d47ef33de1bb1f6ac4e67c1f3bdc2ee1b9c6a26b8d", + "prepared_review_unit_hash": "c95c4429e965b2a40020438db7ec170e00aedcd9f70729efb08a2a3678f7802d", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35449, - "estimated_input_tokens": 8863, - "estimated_eval_tokens": 17982 + "serialized_chars": 36041, + "estimated_input_tokens": 9011, + "estimated_eval_tokens": 18278 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a heterogeneous mix of unrelated Transformers bugs and regressions across tokenizers, vision, audio, quantization, and Flash Attention; none of the soft pairs look like the same underlying defect closely enough to deduplicate or merge.", - "confidence": 0.94, - "canonical_issue_reason": "If a representative issue must be chosen, issue 46693 is the strongest anchor: it has the most discussion and inbound references and describes a concrete regression with clear impact.", - "canonical_pr_reason": null, - "best_issue_reason": "Issue 46693 is the most actionable and broadly representative among otherwise unrelated reports, with high activity and a specific performance-regression title.", + "summary": "This cluster is highly heterogeneous: the soft-similarity links mostly connect superficially related transformer/model issues, but they do not describe the same underlying bug or change. I would reject all soft edges and treat the set as unrelated reports.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue fits the cluster; the items span unrelated bugs across tokenization, vision, audio, quantization, and Flash Attention paths with only broad keyword overlap.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "Issue 44514 is the most concrete and narrowly scoped bug report in the set, with a clear failing path (`Qwen2_5_VLProcessor.apply_chat_template` on batched input with `padding=False`).", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Different components and failures: Qwen3.5 config propagation vs Whisper empty-transcription generation after token alignment." + }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "Different bugs: runtime special-token post-processor sync vs Gemma3 NaN embeddings from sliding-window padding." + "reason": "Tokenizers backend post-processor sync is unrelated to Gemma3 NaN embeddings from sliding-window attention." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "One is tokenizer error masking on missing protobuf; the other is leftover dummy classes leaking into exports." + "reason": "Tokenizer error masking from protobuf handling is unrelated to leftover dummy class cleanup." }, { "left": "issue:45412", "right": "issue:46369", "accept": false, - "reason": "Both mention memory, but one is RT-DETR GC retention and the other is SAM3-LiteText overflow; different code paths." - }, - { - "left": "issue:45372", - "right": "issue:45406", - "accept": false, - "reason": "Different Gemma4 failures: missing ReasoningEffort import vs processor missing _tokenizer during serve loading." + "reason": "Both mention memory, but one is RT-DETR cleanup/GC and the other is SAM model memory overflow; different bug paths." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Both are tokenizer-loading issues, but one is wrong decoder type in v5 and the other ignores tokenizer.json from the repo." + "reason": "Both are tokenizer-loading themed, but one is decoder-type mismatch in v5 and the other is ignoring repository `tokenizer.json`." }, { "left": "issue:45250", "right": "issue:46640", "accept": false, - "reason": "Too generic; no evidence they are the same Flash Attention 2 bug or fixable in one change." + "reason": "Same broad Flash Attention label, but one is a general issue and the other is a specific regression/performance complaint." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to Kosmos2.5 OCR long-input index errors." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to Kosmos2.5 OCR index errors." }, { "left": "issue:45372", - "right": "issue:45459", + "right": "issue:45406", "accept": false, - "reason": "ImportError for Gemma4 processor loading is unrelated to tokenizer error handling when protobuf is absent." + "reason": "Both involve Gemma 4 loading, but one is a missing `ReasoningEffort` import and the other is a missing `_tokenizer` attribute in serve." }, { "left": "issue:45245", "right": "issue:45563", "accept": false, - "reason": "Different symptoms and subsystems: category-limit RuntimeError vs stale generate warning for num_return_sequences." + "reason": "Category-limit runtime error is unrelated to a stale warning in paged generation." }, { "left": "issue:27957", "right": "issue:33290", "accept": false, - "reason": "XLMRoberta Flash Attention 2 support is unrelated to Adafactor OOM in DeepSpeed." + "reason": "XLMRoberta Flash Attention setup and Adafactor/DeepSpeed OOM are different subsystems and different failures." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma 4 processor import failure is unrelated to protobuf-backed tokenizer error masking." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "Training OOM with Adafactor/DeepSpeed is unrelated to Llama 3.2 Vision model-loading breakage." + "reason": "Optimizer/deepspeed OOM is unrelated to model-loading breakage for Llama 3.2 Vision." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "Specific missing Qwen2_5_VL class import is not the same as a generic dependency issue." + "reason": "Importing a specific Qwen2.5-VL class is not the same as a generic dependency issue with transformers." }, { "left": "issue:43526", "right": "issue:46693", "accept": false, - "reason": "Label reduction in BeitImageProcessorFast is unrelated to FA2 long-context generation regression." + "reason": "Label reduction in BEiT image processing is unrelated to Flash Attention long-context performance regression." }, { - "left": "issue:27957", - "right": "issue:29127", + "left": "issue:33290", + "right": "issue:35141", "accept": false, - "reason": "Flash Attention support for XLMRoberta and LayoutLMv3 error-message clarity are unrelated." + "reason": "DeepSpeed OOM and token-embedding reinitialization are unrelated bugs." }, { - "left": "issue:33290", - "right": "issue:35141", + "left": "issue:27957", + "right": "issue:29127", "accept": false, - "reason": "Optimizer OOM and embedding reinitialization after resize are different code paths and fixes." + "reason": "Flash Attention support for XLMRoberta is unrelated to LayoutLMv3 error-message clarity." }, { "left": "issue:43994", "right": "issue:46369", "accept": false, - "reason": "SigLIP2 nonsense outputs and SAM3-LiteText memory overflow are unrelated model-specific bugs." + "reason": "SigLIP2 auto/pipeline misbehavior is unrelated to SAM memory overflow." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Gradient checkpointing in eval mode and quantization offloading failures are distinct behaviors." + "reason": "Gradient-checkpointing eval-mode restriction is unrelated to quantization offloading behavior." }, { - "left": "issue:42915", - "right": "issue:43931", + "left": "issue:44857", + "right": "issue:45588", "accept": false, - "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are separate model-loading problems." + "reason": "AMP/CUDA loss crash in LwDetr is not the same as Flash Attention sink-less model failure on `s_aux=None`." }, { - "left": "issue:44857", - "right": "issue:45588", + "left": "issue:42915", + "right": "issue:43931", "accept": false, - "reason": "AMP/CUDA crash in LwDetrImageLoss is unrelated to flash_attention.py crashing on sink-less models." + "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are different model/config issues." }, { "left": "issue:45588", "right": "issue:46821", "accept": false, - "reason": "Flash-attention s_aux handling and TimesFM window_size AttributeError are unrelated." + "reason": "Flash Attention sink-less crash and TimesFM `window_size` attribute error do not share the same code path." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Batched apply_chat_template padding crash and wrong vision_position_ids are different Qwen2.5-VL bugs." + "reason": "Both are Qwen2.5-VL, but one is batched chat template padding behavior and the other is video vision position IDs." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "AutoTokenizer failing on a specific tokenizer class is unrelated to Perceiver interpolation failure." + "reason": "Tokenizer loading for FastSpeech2Conformer is unrelated to Perceiver interpolation behavior." }, { "left": "issue:43316", "right": "issue:47312", "accept": false, - "reason": "Gemma3TextConfig API mismatch is unrelated to a kernels version compatibility issue." + "reason": "Gemma3TextConfig API mismatch is unrelated to a kernel-compatibility report." }, { "left": "issue:45325", "right": "issue:46697", "accept": false, - "reason": "Qwen2.5-VL rope-index scaling and Swin2SR over-padding are different preprocessing/model bugs." + "reason": "Qwen2.5-VL rope/position-id scaling bug is unrelated to Swin2SR image padding logic." }, { "left": "issue:46121", "right": "issue:46399", "accept": false, - "reason": "Rope-validation JSON-list TypeError and Qwen3.5 kernel registration decorator failure are unrelated." + "reason": "Rope-validation list typing error is unrelated to a missing decorator causing Qwen3.5 kernel registration failure." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Qwen3.5 recurrent-state reset and Whisper batch_decode skip_special_tokens handling do not share a code path." - }, - { - "left": "issue:44977", - "right": "issue:45245", - "accept": false, - "reason": "Qwen3.5 flash-attention generation bug is unrelated to the category-count RuntimeError." + "reason": "Recurrent state reset in modular Qwen3.5 is unrelated to Whisper `batch_decode` skipping special tokens." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Audio chat-template sample-rate defaulting and LasrFeatureExtractor center-arg crash are different feature bugs." + "reason": "Default audio sampling-rate handling and unsupported `center` argument in a feature extractor are different bugs." } ] }, @@ -11555,28 +11575,29 @@ "issue:44871", "issue:44913", "issue:44964", + "issue:44977", "issue:44987", "issue:45137", "issue:45216", + "issue:45245", "issue:45440", "issue:45759", - "issue:45803", "issue:45859", "issue:45907", "issue:46178", "issue:46234", - "issue:46506", - "issue:47276" + "issue:46506" ], "soft_pairs": [ "issue:43646|issue:44077", + "issue:44977|issue:45245", "issue:43761|issue:45859", "issue:43720|issue:44442", "issue:42175|issue:46234", "issue:25251|issue:28282", + "issue:43873|issue:44871", "issue:27957|issue:29870", "issue:44077|issue:44479", - "issue:43873|issue:44871", "issue:42490|issue:44743", "issue:43643|issue:44913", "issue:43316|issue:45759", @@ -11592,215 +11613,207 @@ "issue:44855|issue:45137", "issue:43606|issue:43701", "issue:43701|issue:46506", - "issue:43931|issue:45216", "issue:43761|issue:44079", "issue:44190|issue:44496", - "issue:44442|issue:45803", - "issue:27957|issue:32090", - "issue:42490|issue:47276" + "issue:43931|issue:45216", + "issue:27957|issue:32090" ], - "prepared_review_unit_hash": "4491986075fa04547603eacc0991d31bb614b84b0f5c67573bf6ca5af85b8da3", + "prepared_review_unit_hash": "3c82ebf1ef61628ac00076f2d7c51479bb82371f7be10aef6d9be27a9958559f", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36469, - "estimated_input_tokens": 9118, - "estimated_eval_tokens": 18492 + "soft_pair_count": 28, + "serialized_chars": 36100, + "estimated_input_tokens": 9025, + "estimated_eval_tokens": 18306 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are not duplicates; they span unrelated failures across configs, loaders, model-specific regressions, docs, and runtime environments. All soft pairs should stay separate.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "This cluster is not a true duplicate set: the soft pairs mostly share generic terms like loading, config, offload, or regression, but they describe different code paths and failure modes. I would reject every soft edge and treat the cluster as heterogeneous.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue is a strong canonical duplicate target here. If forced to pick a loose representative, issue 43316 is the broadest of the config/API-discrepancy reports, but the cluster does not cohere around one underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good global anchor for this cluster; the set is too heterogeneous to treat any one report as the canonical duplicate.", + "best_issue_reason": "Issue 43316 is the closest thing to a representative because it frames a generic API discrepancy, but it is still not a suitable true canonical since the surrounding issues span unrelated subsystems and symptoms.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Different bugs: custom model initialization behavior in v5 vs a patchtsmixer post_init policy issue." + "reason": "Both concern model init/post-init behavior, but one is about Transformers 5 breaking custom initialization and the other about a specific model class allowing optional post_init; not the same bug." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Different problems: flash-attention generation failure for Qwen3.5 versus a categorical cardinality runtime limit." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "Different model/code paths: CLIP hidden_states regression vs Qwen CPU-offload rotary embedding compatibility." + "reason": "Both involve model behavior regressions, but one is CLIPVision hidden_states not returned and the other is Qwen3.5 MoE rotary embedding CPU offload compatibility." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "Unrelated failures: accelerate loading of packed BitNet weights vs tokenizer loading for FastSpeech2ConformerTokenizer." + "reason": "Packed-weight loading in accelerate versus AutoTokenizer failing for a specific tokenizer class are unrelated failures." }, { "left": "issue:42175", "right": "issue:46234", "accept": false, - "reason": "Packaging/backend dependency mismatch is unrelated to a missing import_utils helper function." + "reason": "Both mention backend/import packaging, but one is TensorFlow being absent from an install extra and the other is a missing availability helper symbol." }, { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Pipeline output nesting bug is unrelated to a missing PyTorch installation error." + "reason": "Pipeline output nesting and an AutoModel ImportError are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization offload behavior and Gemma eos token configuration mismatch do not describe the same underlying bug." }, { "left": "issue:27957", "right": "issue:29870", "accept": false, - "reason": "Flash Attention 2 support for XLM-Roberta is unrelated to BART generate() decoder_attention_mask assignment." + "reason": "Both are generation-related model issues, but XLM-RoBERTa Flash Attention 2 support and BART decoder_attention_mask assignment are different code paths." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "One is about post_init support policy; the other is a Qwen video-input regression." - }, - { - "left": "issue:43873", - "right": "issue:44871", - "accept": false, - "reason": "Quantization/offloading behavior and eos_token_id config mismatch are different problems." + "reason": "Optional post_init gating is unrelated to the Qwen video-input regression." }, { "left": "issue:42490", "right": "issue:44743", "accept": false, - "reason": "Local-path import warning and recurrent-state reset with cache are unrelated." + "reason": "A trailing-slash local model path warning and Qwen recurrent-state reset are unrelated." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Missing fields under trust_remote_code and rotary_pct not persisting on reload are distinct config bugs." + "reason": "Both involve config persistence, but one is missing fields from trust_remote_code loading and the other is rotary_pct reverting on reload; different root causes." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Both involve composite configs, but they target different API gaps and code paths." + "reason": "Both are about composite config API shape, but one is a Gemma3TextConfig discrepancy and the other is AutoModelForCausalLM not unwrapping text_config for Qwen multimodal configs." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 parity with JAX is unrelated to stale docs still referencing removed pipeline APIs." + "reason": "SigLIP2 HF-vs-JAX divergence and docs still referencing pipeline() are entirely different issues." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "AddedToken special-argument duplication is unrelated to list_repo_files not forwarding revision." + "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are unrelated API bugs." }, { "left": "issue:44987", "right": "issue:46178", "accept": false, - "reason": "Loading physical-intelligence/fast failing and an import_utils backend mapping error are different issues." + "reason": "Loading a specific model on transformers>=5.1.0 and a backend mapping error for tensorflow_text are not the same problem." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Gemma3TextConfig API discrepancy is unrelated to Llama4Config missing pad_token_id." + "reason": "Both are config attribute inconsistencies, but Gemma3TextConfig API mismatch and Llama4Config missing pad_token_id are separate model-specific issues." }, { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "FSDP CPU RAM-efficient loading failure and Qwen3-VL weight-shape mismatch are not the same bug." + "reason": "FSDP CPU RAM efficient loading and Qwen3-VL weight-shape mismatch are different loading failures." }, { "left": "issue:44964", "right": "issue:45440", "accept": false, - "reason": "Phi-4 multimodal loading failure and DeepSeekV3MoE implementation divergence are distinct." + "reason": "Phi-4 multimodal loading and DeepseekV3MoE divergence are unrelated model families and code paths." }, { "left": "issue:42617", "right": "issue:43575", "accept": false, - "reason": "A training script failure is unrelated to TP loading OOM on Qwen2-57B." + "reason": "A script startup failure and tp-related OOM on Qwen2-57B are not duplicates." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to resume_from_checkpoint key mismatch." + "reason": "torch.compile SDPA failure for Bamba is unrelated to resume_from_checkpoint key mismatch." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "TP OOM when loading Qwen2-57B and a key error loading MobileLLM-125M are unrelated." + "reason": "Tensor-parallel OOM on Qwen2-57B and key errors loading MobileLLM-125M are different issues." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 import/parser error is unrelated to a DeepSpeed ZeRO3 deque IndexError." + "reason": "Python 3.13 indentation/@torch.jit.script parsing issue and DeepSpeed ZeRO3 deque underflow are unrelated." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "CPU offload device mismatch is unrelated to resume_from_checkpoint key mismatch." + "reason": "CPU offload device mismatch and checkpoint-resume key mismatch do not share a code path." }, { "left": "issue:43701", "right": "issue:46506", "accept": false, - "reason": "Checkpoint key mismatch and CPU/CUDA result differences are different failure modes." - }, - { - "left": "issue:43931", - "right": "issue:45216", - "accept": false, - "reason": "Qwen3-VL shape mismatch and Qwen3.5 save_pretrained regression are separate issues." + "reason": "Checkpoint key mismatch and differing CPU/CUDA outputs in PPDocLayoutV3 are unrelated." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "Both touch outputs, but one is a hidden_states regression and the other is generic ModelOutput key assignment." + "reason": "Hidden_states being None in CLIPVisionModel and ModelOutput key assignment with prior None are different bugs." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading failure is unrelated to an unrecognized model/config error." + "reason": "Local dataset loading in an image-classification script and an unrecognized-model config.json error are unrelated." }, { - "left": "issue:44442", - "right": "issue:45803", + "left": "issue:43931", + "right": "issue:45216", "accept": false, - "reason": "Tokenizer loading failure and a bare-except bug in Fuyu tensor conversion are unrelated." + "reason": "Weight-shape mismatch on model load and a broken save_pretrained regression for Qwen3.5 are different failure modes." }, { "left": "issue:27957", "right": "issue:32090", "accept": false, - "reason": "Flash Attention support for XLM-Roberta is unrelated to a Trainer GPU broadcast NoneType error." - }, - { - "left": "issue:42490", - "right": "issue:47276", - "accept": false, - "reason": "Local model path warning and synced_gpus warning wording are unrelated." + "reason": "Flash Attention 2 support for XLM-RoBERTa and a Trainer _gpu_broadcast_one NoneType error are not the same underlying bug." } ] }, @@ -11828,6 +11841,7 @@ "issue:22355", "issue:25251", "issue:27957", + "issue:42490", "issue:42907", "issue:43065", "issue:43232", @@ -11847,17 +11861,16 @@ "issue:44077", "issue:44162", "issue:44188", + "issue:44442", "issue:44464", "issue:44496", "issue:44514", "issue:44521", "issue:44534", - "issue:44610", "issue:44617", "issue:44792", "issue:44871", "issue:44913", - "issue:45381", "issue:45584", "issue:45698", "issue:45803", @@ -11866,20 +11879,22 @@ "issue:46234", "issue:46291", "issue:46365", - "issue:46369", "issue:46421", "issue:46640", "issue:46829", "issue:46932", - "issue:47217" + "issue:47217", + "issue:47276" ], "soft_pairs": [ + "issue:44442|issue:45803", + "issue:42490|issue:47276", "issue:46829|issue:46932", "issue:43295|issue:44871", "issue:43316|issue:43716", "issue:44871|issue:45584", - "issue:43232|issue:43388", "issue:43295|issue:43825", + "issue:43232|issue:43388", "issue:43761|issue:44188", "issue:16998|issue:22355", "issue:44521|issue:47217", @@ -11892,199 +11907,197 @@ "issue:44464|issue:46421", "issue:43404|issue:43441", "issue:44162|issue:44792", - "issue:43352|issue:46640", "issue:43065|issue:44617", + "issue:43352|issue:46640", "issue:45803|issue:46365", "issue:25251|issue:27957", "issue:43540|issue:44514", - "issue:43572|issue:46178", - "issue:44610|issue:46369", - "issue:45381|issue:47217" + "issue:43572|issue:46178" ], - "prepared_review_unit_hash": "76d74e60a7c38cb00be282cf54e4034da774104008b42814ad58ac9195ebbe75", + "prepared_review_unit_hash": "57a8e92d915f9b5093ba09f01870880c15e831f1fb7e7fd4bdc600ffb43d23c6", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35362, - "estimated_input_tokens": 8841, - "estimated_eval_tokens": 17938 + "serialized_chars": 35341, + "estimated_input_tokens": 8836, + "estimated_eval_tokens": 17928 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items span unrelated Transformers bugs and regressions across different models, config paths, and utility functions. None of the soft pairs look like the same underlying bug/change, so I would not deduplicate any of them.", - "confidence": 0.93, - "canonical_issue_reason": "Issue #44521 is the strongest canonical issue candidate: it is open, has the most discussion and inbound references, and describes a concrete, reproducible multimodal chat-template bug.", + "summary": "This cluster is very heterogeneous: most items are unrelated model/config/regression bugs rather than duplicates. I would not merge any of the soft pairs. If a single issue must represent the set, #44521 is the best candidate because it is open, has the most inbound references/discussion, and describes a concrete multimodal bug.", + "confidence": 0.94, + "canonical_issue_reason": "#44521 is the strongest representative: it is open, has the highest external interest in this set, and centers on a clearly reproducible multimodal bug. There is no convincing duplicate group strong enough to make another issue a better canonical target.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #44521 is the best issue target overall because it is well-scoped, currently active, and likely to attract/collect related reports about multimodal template handling.", + "best_issue_reason": "#44521 is the best standalone issue in the set because it has the broadest impact signals and a precise failure mode that is likely actionable for maintainers.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Different failure classes: tokenizer load/import error vs exception-handling bug in tensor conversion." + }, + { + "left": "issue:42490", + "right": "issue:47276", + "accept": false, + "reason": "Both mention warnings, but they concern unrelated warning messages and code paths." + }, { "left": "issue:46829", "right": "issue:46932", "accept": false, - "reason": "Different bugs: one is a VideoPrism output-field mismatch, the other is Gemma4 dtype casting for quantized embedders." + "reason": "Different model families and bugs: output field wiring vs quantized dtype casting." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Unrelated regressions: processor/tokenizer image handling vs Gemma-3 eos_token_id config mismatch." + "reason": "One is a processor/tokenizer regression; the other is an eos_token_id config mismatch." }, { "left": "issue:43316", "right": "issue:43716", "accept": false, - "reason": "Different model-specific API/dtype problems; no shared code-path or symptom." + "reason": "Gemma3 config API inconsistency is unrelated to Mistral-3 image dtype mismatch." }, { "left": "issue:44871", "right": "issue:45584", "accept": false, - "reason": "Gemma-3 eos_token config issue and Whisper empty-transcription failure are distinct features and failure modes." + "reason": "Both are generation/config issues, but the concrete bugs are different models and different root causes." }, { - "left": "issue:43232", - "right": "issue:43388", + "left": "issue:43295", + "right": "issue:43825", "accept": false, - "reason": "Generation kwargs update after sync_gpus is unrelated to gather_for_metrics dropping tuple labels." + "reason": "Regression in custom processor usage is unrelated to a pipeline error-message wording issue." }, { - "left": "issue:43295", - "right": "issue:43825", + "left": "issue:43232", + "right": "issue:43388", "accept": false, - "reason": "Custom processor/tokenizer regression is not the same as the pipeline translation-task error message bug." + "reason": "Generation kwargs update after sync_gpus is unrelated to metrics gathering on the last batch." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "CLIPVision hidden_states output bug is unrelated to torch.compile attention-kernel branching divergence." + "reason": "VisionModel hidden_states regression and attention-kernel divergence are different subsystems and symptoms." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "Model max length behavior and missing transformers.onnx module are entirely separate issues." + "reason": "Model max length question and missing transformers.onnx module are unrelated." }, { "left": "issue:44521", "right": "issue:47217", "accept": false, - "reason": "Both are multimodal, but one is about assistant mask generation and the other about image placeholders in user text; different bugs." + "reason": "Both touch multimodal chat templates, but one is an assistant_mask bug and the other is a failure when the image placeholder appears in user text." }, { "left": "issue:46234", "right": "issue:46291", "accept": false, - "reason": "Missing import_utils availability helper is unrelated to kernels import/version requirements." + "reason": "Missing import_utils helper and pipeline import failure from kernels versioning are distinct issues." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "General custom model initialization breakage is not the same as GPTNeoX rotary_pct reload loss." + "reason": "Custom model initialization breakage is not the same as GPTNeoX rotary_pct not persisting on reload." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "Both involve model loading, but one is gguf_file handling and the other is missing model_type in config.json; different root causes." + "reason": "gguf_file loading breaks for tokenizer/model loading, while the other is an unrecognized-model/config.json issue." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "Optional post_init policy for patchtsmixer is unrelated to non-persistent buffer initialization in v5." + "reason": "Optional post_init deprecation is unrelated to junk-filled non-persistent buffers." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Saving dequantized Ministral/Devstral models is a different failure from loading the wrong custom module after save_pretrained." + "reason": "Dequantized save failure is a quantization bug, not a custom-module resolution bug after save/load." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Gemma-3 eos_token configuration and sinq quantized save/load breakage are separate issues." + "reason": "Eos-token mismatch and sinq quantization round-trip failure are different problems." }, { "left": "issue:44464", "right": "issue:46421", "accept": false, - "reason": "Chunked generation output inconsistency is not the same as cache lazy-init causing recompilation." + "reason": "Compiled-forward generation inconsistency is not the same as chunked prefill recompilation/cache initialization." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "Weight tying in Mistral3ForConditionalGeneration is unrelated to FlashAttention support failure in Ministral-3." + "reason": "Weight tying in Mistral3 is unrelated to FlashAttention support gating for Ministral-3." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 workflow breakage and a janus image-generation test failure are different subsystems and symptoms." + "reason": "ESM2 broken workflow and janus image-generation test failure are unrelated." }, { - "left": "issue:43352", - "right": "issue:46640", + "left": "issue:43065", + "right": "issue:44617", "accept": false, - "reason": "A specific model not supporting Flash Attention 2 is not the same as a generic Flash Attention 2 issue." + "reason": "Dummy Conv2d in a decoder is not the same as CUDA OOM in Sam3Video." }, { - "left": "issue:43065", - "right": "issue:44617", + "left": "issue:43352", + "right": "issue:46640", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to Sam3Video CUDA OOM." + "reason": "Flash Attention support error for NemotronH is not clearly the same as a generic Flash Attention 2 issue." }, { "left": "issue:45803", "right": "issue:46365", "accept": false, - "reason": "Bare except swallowing errors in FuyuBatchFeature is unrelated to importing Trainer failing with datasets installed." + "reason": "Bare except swallowing errors is unrelated to Trainer import failure caused by datasets." }, { "left": "issue:25251", "right": "issue:27957", "accept": false, - "reason": "Pipeline top_k nesting bug and XLMRoberta FlashAttention support are unrelated." + "reason": "Pipeline top_k nesting change and XLMRoberta Flash Attention support are different behaviors." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Different Qwen multimodal bugs: video-input ValueError vs batched apply_chat_template crash with padding=False." + "reason": "Video input ValueError in Qwen3OmniMoe is not the same as batched apply_chat_template padding=False crash." }, { "left": "issue:43572", "right": "issue:46178", "accept": false, - "reason": "StableLmConfig pad_token_idx regression is unrelated to BACKENDS_MAPPING missing tensorflow_text." - }, - { - "left": "issue:44610", - "right": "issue:46369", - "accept": false, - "reason": "OmDet-Turbo processor image-size mismatch is unrelated to Sam3-litetext memory overflow." - }, - { - "left": "issue:45381", - "right": "issue:47217", - "accept": false, - "reason": "Qwen2.5-VL video vision_position_ids bug is not the same as image-placeholder-in-user-text failures." + "reason": "Missing config field after v5 update and backend mapping error for tensorflow_text are unrelated." } ] }, @@ -12128,7 +12141,7 @@ "issue:44242", "issue:44479", "issue:44534", - "issue:44617", + "issue:44610", "issue:44743", "issue:44792", "issue:44855", @@ -12137,10 +12150,9 @@ "issue:44977", "issue:45127", "issue:45200", - "issue:45245", + "issue:45381", "issue:45405", "issue:45542", - "issue:45584", "issue:45588", "issue:45600", "issue:45632", @@ -12150,20 +12162,22 @@ "issue:46170", "issue:46202", "issue:46234", + "issue:46369", "issue:46424", "issue:46693", "issue:46697", - "issue:46858", "issue:46894", "issue:47217" ], "soft_pairs": [ + "issue:44610|issue:46369", + "issue:45381|issue:47217", "issue:44792|issue:44977", "issue:44479|issue:47217", "issue:45542|issue:46234", "issue:44534|issue:44913", - "issue:43504|issue:45698", "issue:43525|issue:43531", + "issue:43504|issue:45698", "issue:44792|issue:45600", "issue:45632|issue:46424", "issue:45803|issue:46170", @@ -12172,8 +12186,8 @@ "issue:45127|issue:45200", "issue:44242|issue:46894", "issue:41950|issue:43475", - "issue:43673|issue:44743", "issue:43650|issue:46202", + "issue:43673|issue:44743", "issue:43931|issue:44060", "issue:44855|issue:45588", "issue:44792|issue:45405", @@ -12181,194 +12195,186 @@ "issue:43388|issue:43688", "issue:29870|issue:35141", "issue:46082|issue:46697", - "issue:43827|issue:43994", - "issue:45245|issue:46858", - "issue:44617|issue:45584" + "issue:43827|issue:43994" ], - "prepared_review_unit_hash": "d737e9ba7950589c944f485348ee1317e2d415264a9331efcca1f4c7fef65f40", + "prepared_review_unit_hash": "2dca31212ce3fc3ca6cf8585c9052108d28bd79da74c8656183a759f613f2659", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35720, - "estimated_input_tokens": 8930, - "estimated_eval_tokens": 18116 + "serialized_chars": 35141, + "estimated_input_tokens": 8786, + "estimated_eval_tokens": 17828 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose similarity bucket of unrelated Transformers issues across generation, video/multimodal, config, cache, and docs paths. None of the soft pairs look like the same underlying bug or a single mergeable PR change, so I would not deduplicate any of them.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue fits this cluster: the items are mostly unrelated reports, so collapsing them into one representative issue would be misleading.", + "summary": "This cluster is a noisy mix of unrelated issue reports across many models, pipelines, and config/load paths. None of the soft-paired items look like true duplicates, so there is no credible single canonical issue or PR here.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No best issue among these as a cluster representative; the strongest candidates are still on different code paths and different products/models.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44610", + "right": "issue:46369", + "accept": false, + "reason": "Processor image-size mismatch vs memory overflow; different model and failure mode." + }, + { + "left": "issue:45381", + "right": "issue:47217", + "accept": false, + "reason": "Video vision_position_ids regression vs image-placeholder-in-text failure; different code paths." + }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Both involve generation failures, but one is a janus test failure and the other is a Qwen3.5 flash-attention bug; different models and failure modes." + "reason": "Janus image-generation test failure vs Qwen3.5 flash-attention generation bug." }, { "left": "issue:44479", "right": "issue:47217", "accept": false, - "reason": "Both touch VL models, but one is a v5.3.0 video-input regression and the other is an image-placeholder placement bug in user text." + "reason": "v5.3.0 video-input regression across Qwen VL models vs text-placeholder handling." }, { "left": "issue:45542", "right": "issue:46234", "accept": false, - "reason": "TensorBoard/TF backend installation is unrelated to the missing import_utils helper for Falcon Mamba availability." + "reason": "TensorBoard/TF backend packaging issue vs missing import_utils symbol." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk and GPTNeoX rotary_pct reload behavior are unrelated configuration/runtime bugs." + "reason": "Non-persistent buffer initialization issue vs GPTNeoX config reload bug." }, { - "left": "issue:43504", - "right": "issue:45698", + "left": "issue:43525", + "right": "issue:43531", "accept": false, - "reason": "Beit legacy-field loading and custom-module selection after save/load are different serialization/loading problems." + "reason": "Missing pad_token_id in Llama4Config vs Qwen3-MoE sliding_window behavior." }, { - "left": "issue:43525", - "right": "issue:43531", + "left": "issue:43504", + "right": "issue:45698", "accept": false, - "reason": "Missing pad_token_id on Llama4Config is unrelated to Qwen3-MoE sliding_window behavior." + "reason": "Legacy field breaks Beit preset load vs wrong custom module loaded after save_pretrained." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "A janus test failure and stale auto_mappings references are separate CI/runtime issues." + "reason": "Janus test failure vs stale auto_mappings entry for removed Sam3LiteText configs." }, { "left": "issue:45632", "right": "issue:46424", "accept": false, - "reason": "Local trust_remote_code cache collisions are unrelated to max_cache_length only applying to Quantized Cache." + "reason": "Leaf-dir cache collision vs max_cache_length cache policy bug." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "Swallowing KeyboardInterrupt in Fuyu tensor conversion is unrelated to a Windows ROCm import/dependency problem." + "reason": "Bare except in tensor conversion vs Windows ROCm import dependency issue." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Embedding reinitialization after resize and trailing-slash local path import warnings are different code paths." + "reason": "Embedding resize/reinit bug vs trailing-slash local model import warning." }, { "left": "issue:46693", "right": "issue:47217", "accept": false, - "reason": "FA2 long-context performance regression is unrelated to image placeholders appearing in user text." + "reason": "FA2 long-context performance regression vs VL image-placeholder bug." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "LoRA merge collapse with extended vocab and Gemma 4 mm_token_type_ids defaults are different model behaviors." + "reason": "LoRA/vocab collapse with tied embeddings vs mm_token_type_ids defaulting." }, { "left": "issue:44242", "right": "issue:46894", "accept": false, - "reason": "Both concern loss handling, but the first is router/load-balancing loss gating while the second is a training-loss double-shift pattern in ASR." + "reason": "Missing load-balancing loss vs ASR training double-shift bug." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Video-classification pipeline processor selection and SAM3 video encoder output attributes are unrelated multimodal bugs." + "reason": "Video-classification processor mismatch vs missing vision encoder attribute." }, { - "left": "issue:43673", - "right": "issue:44743", + "left": "issue:43650", + "right": "issue:46202", "accept": false, - "reason": "Cache missing in chunked_prefill and recurrent states resetting with cache are distinct state-management bugs." + "reason": "Unrelated placeholder issue vs Gemma pipeline roundtrip bug." }, { - "left": "issue:43650", - "right": "issue:46202", + "left": "issue:43673", + "right": "issue:44743", "accept": false, - "reason": "A generic data request and a Gemma pipeline roundtrip issue are not the same defect." + "reason": "Chunked-prefill cache missing vs recurrent state reset; both cache-related but different bugs." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Qwen3-VL weight-shape loading failure and Qwen3-Next tied-weights warning are different model/config problems." + "reason": "Qwen3-VL weight-shape mismatch vs Qwen3-Next tied-weights warning." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Python 3.13 import parsing error and flash_attention crashing on s_aux=None are unrelated." + "reason": "Python 3.13 JIT parse error vs flash_attention s_aux=None crash." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "Janus test failure and a PEFT version bump issue have no shared underlying bug." + "reason": "Janus test failure vs unreleased PEFT version constraint." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Local dataset loading in a training script and Phi-4 multimodal model loading are unrelated." + "reason": "Local dataset script loading issue vs Phi-4 multimodal model load failure." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Dropping tuple label elements in gather_for_metrics and incorrect auxiliary-loss normalization are different training/evaluation bugs." + "reason": "Metric gather truncation vs auxiliary-loss normalization." }, { "left": "issue:29870", "right": "issue:35141", "accept": false, - "reason": "Bart generate() UnboundLocalError and output-embedding reinitialization after resize are separate issues." + "reason": "BART generate mask error vs embedding resize reinit." }, { "left": "issue:46082", "right": "issue:46697", "accept": false, - "reason": "LlamaConfig head_dim validation and Swin2SR image over-padding are unrelated." - }, - { - "left": "issue:43827", - "right": "issue:43994", - "accept": false, - "reason": "Docs still referencing pipeline() and SigLIP2 producing bad outputs through AutoModel/pipeline are different problems." - }, - { - "left": "issue:45245", - "right": "issue:46858", - "accept": false, - "reason": "Category-count limit and cache seq-length device-tensor arithmetic are unrelated runtime failures." - }, - { - "left": "issue:44617", - "right": "issue:45584", - "accept": false, - "reason": "Sam3Video OOM and Whisper empty-transcription failure are distinct model-specific issues." + "reason": "Llama head_dim validation vs Swin2SR over-padding." } ] }, @@ -12434,6 +12440,7 @@ "issue:46655", "issue:46697", "issue:46832", + "issue:46858", "issue:46897", "issue:46932", "issue:47013", @@ -12441,28 +12448,30 @@ "issue:47321" ], "soft_pairs": [ + "issue:44617|issue:45584", + "issue:45245|issue:46858", "issue:45072|issue:46932", "issue:43644|issue:44913", "issue:45588|issue:47013", "issue:43335|issue:44855", "issue:44734|issue:45406", - "issue:46399|issue:47312", "issue:25251|issue:29870", + "issue:46399|issue:47312", "issue:44869|issue:46832", "issue:45381|issue:46369", "issue:41720|issue:45245", "issue:45632|issue:46617", - "issue:43550|issue:44190", "issue:43827|issue:47321", + "issue:43550|issue:44190", "issue:43854|issue:44792", "issue:45563|issue:45584", "issue:39692|issue:47321", "issue:45381|issue:46655", - "issue:43065|issue:44610", "issue:44617|issue:46369", + "issue:43065|issue:44610", "issue:44479|issue:44734", - "issue:44933|issue:45362", "issue:43335|issue:44190", + "issue:44933|issue:45362", "issue:39692|issue:44190", "issue:44779|issue:46489", "issue:43295|issue:45276", @@ -12470,212 +12479,221 @@ "issue:43526|issue:46697", "issue:43540|issue:44734" ], - "prepared_review_unit_hash": "236a96931c17e05d9429519545054a310d2c54c15005f75b9c0062acc3c2e7a6", + "prepared_review_unit_hash": "dbe82f58e1a7f1f37d96b9b1906780283a4d7e66ae8ce95932148ecca5110917", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35608, - "estimated_input_tokens": 8902, - "estimated_eval_tokens": 18060 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36838, + "estimated_input_tokens": 9210, + "estimated_eval_tokens": 18676 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated regressions across different models/subsystems. The only clear duplicate-like pair is the DeepSeek tokenizer regression pair; the rest are superficial similarity matches and should remain separate.", - "confidence": 0.9, - "canonical_issue_reason": "Issue 44779 is the broadest and most generic report in the only true duplicate-like pair; issue 46489 reads like a follow-up gap in the same DeepSeek tokenizer regression.", + "summary": "This cluster is heterogeneous: the soft pairs are mostly same-subsystem or same-model-family near-misses, but not the same concrete bug/change. I would not merge any of the candidate pairs.", + "confidence": 0.88, + "canonical_issue_reason": "No canonical issue: the items span unrelated models, regressions, docs, and runtime failures, so there isn\u2019t a single duplicate target for the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "44779 is the best representative issue because it states the regression at the highest level, while 46489 is a narrower variant/follow-up.", + "best_issue_reason": "issue:44734 is the strongest standalone bug report here: it describes a specific crash in `transformers serve` with a clear failing code path and a likely actionable fix.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45072", - "right": "issue:46932", + "left": "issue:44617", + "right": "issue:45584", "accept": false, - "reason": "Both mention dtype/bfloat16 problems, but they affect different models and code paths (SwitchTransformers/TimmWrapperModel vs Gemma4 quantized embedders)." + "reason": "Different bugs: Sam3Video OOM vs Whisper empty-transcription crash." }, { - "left": "issue:43644", - "right": "issue:44913", + "left": "issue:45245", + "right": "issue:46858", "accept": false, - "reason": "One is about junk-filled non-persistent buffers; the other is a config reload bug for rotary_pct. Different mechanisms." + "reason": "Both involve runtime errors, but one is a category-limit error and the other is a cache-length/device-tensor arithmetic bug." + }, + { + "left": "issue:45072", + "right": "issue:46932", + "accept": false, + "reason": "Both mention dtype issues, but they affect different models and different code paths." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Unrelated failures: non-persistent buffer initialization vs GPTNeoX config reload behavior." }, { "left": "issue:45588", "right": "issue:47013", "accept": false, - "reason": "Different failures: flash_attention s_aux=None crash vs xLSTM state unpacking mismatch." + "reason": "Both are model/runtime crashes, but the underlying faults are different (`s_aux` handling vs xLSTM return shape unpacking)." }, { "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "Sparse-layer construction bug vs Python 3.13 IndentationError from @torch.jit.script parsing." + "reason": "Completely different problems: SwitchTransformers config bug vs Python 3.13 import/`torch.jit` parsing error." }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "KV-cache continuation tensor indexing in serve is unrelated to Gemma4Processor missing _tokenizer." + "reason": "Serve crash with tensor indexing is not the same as a missing `_tokenizer` attribute on Gemma4Processor." }, { - "left": "issue:46399", - "right": "issue:47312", + "left": "issue:25251", + "right": "issue:29870", "accept": false, - "reason": "A missing decorator on apply_rotary_pos_emb is not the same as broad kernel-version incompatibility." + "reason": "Pipeline output nesting and Bart `generate()` mask initialization are unrelated bugs." }, { - "left": "issue:25251", - "right": "issue:29870", + "left": "issue:46399", + "right": "issue:47312", "accept": false, - "reason": "Pipeline top_k nesting bug and Bart generate decoder_attention_mask UnboundLocalError are different issues." + "reason": "One is a missing decorator registration crash; the other is a broad kernels version incompatibility." }, { "left": "issue:44869", "right": "issue:46832", "accept": false, - "reason": "Whisper trailing replacement-character crash is unrelated to RT-DETR feature-level TypeError." + "reason": "Whisper trailing-character decoding crash is unrelated to RT-DETR feature-level handling." }, { "left": "issue:45381", "right": "issue:46369", "accept": false, - "reason": "Different models and failures: Qwen2.5-VL vision position ids vs SAM3-litetext memory overflow." + "reason": "Same broad family of model issues, but one is video position IDs and the other is memory overflow." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "cudaErrorAssert during auto device mapping is unrelated to the 2^24 category-count limit." + "reason": "Different failures in different subsystems: Qwen3 device-mapping assert vs category-count limit." }, { "left": "issue:45632", "right": "issue:46617", "accept": false, - "reason": "Both involve trust_remote_code cache paths, but one is a leaf-directory collision and the other is a FileNotFoundError; not enough evidence of the same bug." + "reason": "Both touch trust_remote_code cache loading, but one is a leaf-directory collision and the other is a FileNotFoundError; not clearly the same bug." }, { - "left": "issue:43550", - "right": "issue:44190", + "left": "issue:43827", + "right": "issue:47321", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba-9B-v2 is unrelated to loading a local dataset in the training script." + "reason": "Docs still referencing pipeline() has nothing to do with AMD Quark class updates." }, { - "left": "issue:43827", - "right": "issue:47321", + "left": "issue:43550", + "right": "issue:44190", "accept": false, - "reason": "Docs still referencing pipeline() after v5 removals is unrelated to an AMD quark class update bug." + "reason": "torch.compile/SDPA failure is unrelated to loading a local dataset in the image-classification example." }, { "left": "issue:43854", "right": "issue:44792", "accept": false, - "reason": "Unit-test loading failure for GLM-4.7-Flash and janus image-generation test failure are distinct model/test problems." + "reason": "Two separate model/test failures with no shared code-path evidence." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "A stale warning for num_return_sequences is unrelated to Whisper empty-transcription handling." + "reason": "A stale warning in paged generate is not the same as Whisper empty-transcription generation failure." }, { "left": "issue:39692", "right": "issue:47321", "accept": false, - "reason": "SigLIP2 docs/example errors do not match an AMD quark class issue." + "reason": "SigLIP2 doc example errors and AMD Quark class updates are unrelated." }, { "left": "issue:45381", "right": "issue:46655", "accept": false, - "reason": "Both are video-related, but Qwen2.5-VL vision_position_ids and SmolVLM padding are different code paths and symptoms." + "reason": "Both are video-processing issues, but they concern different models and different defects (position IDs vs padding calculation)." }, { - "left": "issue:43065", - "right": "issue:44610", + "left": "issue:44617", + "right": "issue:46369", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to OmDet-Turbo processor size mismatch." + "reason": "Same SAM3 area, but one is OOM in video and the other is memory overflow in litetext; not the same bug." }, { - "left": "issue:44617", - "right": "issue:46369", + "left": "issue:43065", + "right": "issue:44610", "accept": false, - "reason": "Both mention memory pressure in SAM3-family models, but one is explicit CUDA OOM in video and the other is a litetext overflow report; not the same bug." + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to OmDet-Turbo input-size mismatch." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Qwen video-input regression and serve KV-cache indexing crash are different failures." + "reason": "Both are regressions, but one is Qwen video-input behavior and the other is a serve/KV-cache indexing crash." }, { - "left": "issue:44933", - "right": "issue:45362", + "left": "issue:43335", + "right": "issue:44190", "accept": false, - "reason": "Missing image_utils import and Qwen3.5 chat crash are unrelated." + "reason": "Config-layer creation bug is unrelated to local dataset loading in the training script." }, { - "left": "issue:43335", - "right": "issue:44190", + "left": "issue:44933", + "right": "issue:45362", "accept": false, - "reason": "Sparse-layer config bug is unrelated to loading a local dataset." + "reason": "Missing import from image_utils is unrelated to Qwen3.5 chat crashes." }, { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "SigLIP2 docs errors and dataset loading are different problem areas." + "reason": "SigLIP2 docs errors and dataset loading failure do not share the same underlying bug." }, { "left": "issue:44779", "right": "issue:46489", - "accept": true, - "reason": "Both report DeepSeek tokenizer regressions on transformers v5+; 46489 reads like a specific follow-up/variant of the broader 44779 bug." + "accept": false, + "reason": "Both concern DeepSeek tokenization, but one is a general v5 regression and the other is a specific gap in PR #44801; not the same concrete defect." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Processor.tokenizer regression and resize_token_embeddings not updating Gemma4 embed/output layers are different API issues." + "reason": "Processor/tokenizer regression and Gemma4 resize-token-embeddings propagation bug are different issues." }, { "left": "issue:44242", "right": "issue:46897", "accept": false, - "reason": "MoE load-balancing loss gating and Florence2 training-loss double-shift are not the same underlying bug." + "reason": "Load-balancing-loss omission and Florence2 double-shift are different training-loss bugs." }, { "left": "issue:43526", "right": "issue:46697", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels and Swin2SR over-padding images are unrelated image-processing bugs." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to Swin2SR over-padding." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-input ValueError and serve KV-cache continuation crash are different subsystems and failures." + "reason": "Qwen3OmniMoe video-input ValueError is not the same as the serve KV-cache indexing crash." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative on the soft-edge pairs. The one accepted pair (44779/46489) is plausibly the same DeepSeek tokenizer regression, and the rest are reasonably rejected as different bug classes or code paths. No obvious evidence overstating. " - }, - "evaluator_used": true, + "evaluator_result": null, + "evaluator_used": false, "retried": false, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -12742,51 +12760,51 @@ "issue:47321" ], "soft_pairs": [ - "issue:45127|issue:45357", "issue:43673|issue:46421", + "issue:45127|issue:45357", "issue:43854|issue:44964", "issue:45276|issue:45584", "issue:47145|issue:47321", "issue:44462|issue:45907", - "issue:44521|issue:45381", "issue:43643|issue:44496", + "issue:44521|issue:45381", + "issue:43994|issue:47321", "issue:44112|issue:45600", "issue:43317|issue:47312", - "issue:43994|issue:47321", "issue:43611|issue:47321", "issue:27957|issue:35141", - "issue:45290|issue:46752", - "issue:44246|issue:45003", "issue:42175|issue:44279", - "issue:43673|issue:46858", + "issue:44246|issue:45003", + "issue:45290|issue:46752", "issue:44112|issue:44964", - "issue:43404|issue:45216", + "issue:43673|issue:46858", "issue:42796|issue:44521", - "issue:44964|issue:45600", "issue:32090|issue:35141", + "issue:44964|issue:45600", "issue:44737|issue:45803", + "issue:43404|issue:45216", "issue:44496|issue:47217", - "issue:44938|issue:45542", "issue:43531|issue:45600", - "issue:45632|issue:46234", + "issue:44938|issue:45542", "issue:43335|issue:45593", + "issue:45632|issue:46234", "issue:36246|issue:41628", "issue:36246|issue:38617" ], - "prepared_review_unit_hash": "85bd0756e39fa4c64c5ed079ddcb932c697952e8a413d7c99fb2c75096e566a8", + "prepared_review_unit_hash": "678e35c832599605a19a66725bdd6e35730aaaa52105bee0225b65999801fc9d", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36907, + "serialized_chars": 36905, "estimated_input_tokens": 9227, "estimated_eval_tokens": 18710 }, @@ -12795,192 +12813,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a true duplicate set: the soft-linked pairs mostly point to unrelated bugs across model loading, config exports, chat templating, caching, CI, and dependency packaging. None of the candidate pairs should be merged as the same underlying issue.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue fits the cluster; the items do not share one root cause or code path, so collapsing them would be misleading.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers issues. None of the soft pairs look like the same concrete bug/change, so I reject all soft edges. If a single issue must represent the cluster, issue:44521 is the least-bad anchor because it has the most engagement, but it is not a true duplicate hub.", + "confidence": 0.77, + "canonical_issue_reason": "issue:44521 is the most central/referenced item here, but the cluster is not a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "No issue is a good global representative because the cluster is a loose text-similarity grab bag rather than one bug family.", + "best_issue_reason": "issue:44521 has the strongest engagement and broadest relevance among these mostly unrelated bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45127", - "right": "issue:45357", + "left": "issue:43673", + "right": "issue:46421", "accept": false, - "reason": "Different problems: LoRA/embedding tying vs Qwen3.5 visual encoder save_pretrained keys." + "reason": "Both mention chunked prefill/cache, but they describe different failures: missing cache in v5.0.0 vs lazy-init recompilation." }, { - "left": "issue:43673", - "right": "issue:46421", + "left": "issue:45127", + "right": "issue:45357", "accept": false, - "reason": "Both involve cache, but one is a missing GenerationMixin cache in chunked prefill and the other is lazy cache init causing recompilation." + "reason": "Different models and different problems: LoRA/extended-vocab collapse vs incorrect visual encoder keys in save_pretrained." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "Both are model-loading failures, but they affect different models and appear to have different root causes." + "reason": "Both are model-loading failures, but for different models and likely different root causes." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Unrelated code paths: Gemma4 resize_token_embeddings vs Whisper empty-transcription handling." + "reason": "One is about resize_token_embeddings not updating outputs; the other is Whisper empty-transcription handling." }, { "left": "issue:47145", "right": "issue:47321", "accept": false, - "reason": "Completely different areas: FP8 checkpoint scaling vs AMD quark class updates." + "reason": "FP8 frozen-checkpoint scaling and AMD quark class updates are unrelated." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "Tokenizer file selection bug vs list_repo_files revision forwarding; not the same issue." + "reason": "Tokenizer.json selection and list_repo_files revision forwarding are different repository-access bugs." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Both involve config loading, but one is trust_remote_code field loss and the other is missing model_type in config.json." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "Both are multimodal/template issues, but one is assistant mask generation and the other is vision position ids." + "reason": "Both are multimodal/VL-related, but the actual failures differ: assistant masks vs vision_position_ids." }, { - "left": "issue:43643", - "right": "issue:44496", + "left": "issue:43994", + "right": "issue:47321", "accept": false, - "reason": "Both involve config loading, but one is trust_remote_code field loss and the other is missing model_type in config.json." + "reason": "Nonsensical SigLIP outputs and AMD quark class updates are unrelated." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "CI stale device override failure is unrelated to removed Sam3LiteText references in auto_mappings." + "reason": "Both touch CI, but they are separate failures in different subsystems." }, { "left": "issue:43317", "right": "issue:47312", "accept": false, - "reason": "Dequantized model offload failure is unrelated to kernel compatibility." - }, - { - "left": "issue:43994", - "right": "issue:47321", - "accept": false, - "reason": "Unrelated: SigLIP2 inference corruption vs AMD quark class update." + "reason": "Dequantized model offload and kernels 0.16 compatibility are unrelated." }, { "left": "issue:43611", "right": "issue:47321", "accept": false, - "reason": "Base_model_prefix loading regression and AMD quark class update are different bugs." + "reason": "base_model_prefix loading regression and AMD quark class update are unrelated." }, { "left": "issue:27957", "right": "issue:35141", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta is unrelated to token embedding reinitialization on resize." + "reason": "Flash Attention support for XLMRoberta and post_init reinitializing output embeddings are different bugs." }, { - "left": "issue:45290", - "right": "issue:46752", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "Same API, but different edge cases: tool-call assistant messages vs empty input list." + "reason": "One is a missing TensorFlow backend in extras; the other is a generic dependency issue." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import performance issue vs unsafe sys.modules access; different failure modes and fixes." - }, - { - "left": "issue:42175", - "right": "issue:44279", - "accept": false, - "reason": "Packaging/backends issue is too broad and not the same as a general dependency complaint." + "reason": "Import performance and unsafe sys.modules access are not the same bug." }, { - "left": "issue:43673", - "right": "issue:46858", + "left": "issue:45290", + "right": "issue:46752", "accept": false, - "reason": "Both are cache-related, but one is missing cache initialization and the other is device-tensor slice arithmetic in static cache." + "reason": "Both are apply_chat_template crashes, but the triggering inputs and failure modes differ." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "GraniteSpeech CI failure is unrelated to Phi-4 multimodal loading." + "reason": "Different model-specific CI/load failures with no shared concrete code path." }, { - "left": "issue:43404", - "right": "issue:45216", + "left": "issue:43673", + "right": "issue:46858", "accept": false, - "reason": "Different regressions: Mistral3 lm_head tying vs Qwen3.5 save_pretrained checkpoint contents." + "reason": "Both involve cache/prefill, but one is lazy initialization and the other is device-tensor arithmetic in static cache." }, { "left": "issue:42796", "right": "issue:44521", "accept": false, - "reason": "Tokenizer BPE-marker handling and multimodal assistant masks are unrelated." + "reason": "Tokenizer BPE-marker handling and assistant-mask computation are separate tokenizer/chat-template issues." }, { - "left": "issue:44964", - "right": "issue:45600", + "left": "issue:32090", + "right": "issue:35141", "accept": false, - "reason": "Phi-4 load failure vs removed Sam3LiteText mappings; different subsystems." + "reason": "Trainer GPU broadcast NoneType error is unrelated to embedding resize/reinitialization." }, { - "left": "issue:32090", - "right": "issue:35141", + "left": "issue:44964", + "right": "issue:45600", "accept": false, - "reason": "Trainer broadcast TypeError is unrelated to embedding resize reinitialization." + "reason": "Model loading failure and auto_mappings CI breakage are different problems." }, { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "XLNet CPU placement bug and Fuyu bare-except bug are unrelated." + "reason": "XLNet device placement bug and bare-except error masking are unrelated." }, { - "left": "issue:44496", - "right": "issue:47217", + "left": "issue:43404", + "right": "issue:45216", "accept": false, - "reason": "Unrecognized model/config issue vs VL image-placeholder handling; different causes." + "reason": "Mistral3 weight tying and Qwen3.5 save_pretrained regression are different model-specific bugs." }, { - "left": "issue:44938", - "right": "issue:45542", + "left": "issue:44496", + "right": "issue:47217", "accept": false, - "reason": "Python 3.14 import failure is unrelated to tensorboard-only/no-TensorFlow backend setup." + "reason": "Unrecognized model config and image-placeholder handling in VL models are not the same issue." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to auto_mappings CI breakage." + "reason": "Qwen3-MoE sliding_window behavior and removed-config CI breakage are unrelated." }, { - "left": "issue:45632", - "right": "issue:46234", + "left": "issue:44938", + "right": "issue:45542", "accept": false, - "reason": "trust_remote_code cache collision and missing import_utils symbol are unrelated." + "reason": "Python 3.14 import failure and TensorBoard-without-TF backend error are different dependency/runtime issues." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "Sparse-layer creation and D-FINE auxiliary-loss disabling are different model bugs." + "reason": "Sparse-layer creation and auxiliary-loss disabling are unrelated model-architecture bugs." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "trust_remote_code cache-path collisions and missing import_utils helper are unrelated." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they concern different missing symbols and likely different export gaps." + "reason": "Both are import errors, but for different symbols and likely different export paths." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Different missing imports in different modules; no shared underlying bug is evident." + "reason": "Qwen2_5_VLImageProcessor import failure and layer_type_validation import failure are unrelated." } ] }, @@ -13006,10 +13024,10 @@ "nodes": [ "issue:39692", "issue:41950", + "issue:42796", "issue:42907", "issue:42994", "issue:43295", - "issue:43299", "issue:43381", "issue:43441", "issue:43526", @@ -13034,7 +13052,6 @@ "issue:45020", "issue:45042", "issue:45083", - "issue:45216", "issue:45335", "issue:45405", "issue:45446", @@ -13056,8 +13073,8 @@ "soft_pairs": [ "issue:44496|issue:45042", "issue:45083|issue:46506", - "issue:43976|issue:44987", "issue:43295|issue:45335", + "issue:43976|issue:44987", "issue:44295|issue:46506", "issue:41950|issue:44016", "issue:39692|issue:46932", @@ -13071,11 +13088,11 @@ "issue:44016|issue:44220", "issue:44279|issue:46515", "issue:46082|issue:46121", - "issue:44016|issue:44062", "issue:43650|issue:44220", + "issue:44016|issue:44062", "issue:44792|issue:46424", - "issue:42907|issue:43441", "issue:42994|issue:47332", + "issue:42907|issue:43441", "issue:43441|issue:45588", "issue:44327|issue:46752", "issue:43976|issue:44279", @@ -13083,231 +13100,244 @@ "issue:45335|issue:45584", "issue:41950|issue:45020", "issue:43295|issue:44779", - "issue:43299|issue:45216", "issue:43901|issue:44496", "issue:44279|issue:45405", - "issue:43901|issue:47217" + "issue:43901|issue:47217", + "issue:42796|issue:43650", + "issue:39692|issue:43901" ], - "prepared_review_unit_hash": "86efcebee845475882c8aa8e260190a9a38d104bef4f771af6d6ff4140d742c8", + "prepared_review_unit_hash": "48a62fd27b2ce58754e68ac3d2ddcb3bac55dcae2d098e01ac67a2c9186b5552", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 37361, - "estimated_input_tokens": 9341, - "estimated_eval_tokens": 18938 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 34, + "serialized_chars": 37038, + "estimated_input_tokens": 9260, + "estimated_eval_tokens": 18776 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a true duplicate set: it mixes unrelated regressions, docs issues, model-specific failures, and environment/version incompatibilities. No soft pair looks like the same underlying bug or change.", - "confidence": 0.93, + "summary": "This cluster is mostly a loose thematic grouping of Transformers bugs across different models, pipelines, and version regressions. I do not see any soft pair that is clearly the same underlying issue, so I would not merge any of these as duplicates.", + "confidence": 0.9, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#45020 is the best umbrella representative because it broadly captures recent Transformers regressions and is the closest high-level fit for this heterogeneous set, even though it is not a true duplicate canonical.", + "best_issue_reason": "issue:42994 is the cleanest, most general bug report in the set: a concrete quantized-model saving failure with likely broader applicability than the more model-specific offloading variants.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "Unrecognized-model config error vs PIL image processor torchvision dependency regression; unrelated failure modes." + "reason": "Both are transformer/model loading or vision-related failures, but one is an unrecognized model/config issue and the other is PIL image-processor torchvision gating; different code paths and symptoms." }, { "left": "issue:45083", "right": "issue:46506", "accept": false, - "reason": "Different subsystems and symptoms: helper length computation in qwen3_omni_moe vs CPU/CUDA output mismatch in PPDocLayoutV3." + "reason": "These are both numerical/output anomalies, but they affect different models and different helpers; no sign they are the same bug or could be fixed by one change." }, { - "left": "issue:43976", - "right": "issue:44987", + "left": "issue:43295", + "right": "issue:45335", "accept": false, - "reason": "One is a Python-version incompatibility, the other is a model-loading failure for a specific repo; not the same bug." + "reason": "One is a regression in processor/tokenizer access for custom multimodal code; the other is a T5Gemma embedding resize inconsistency. Different components and failure modes." }, { - "left": "issue:43295", - "right": "issue:45335", + "left": "issue:43976", + "right": "issue:44987", "accept": false, - "reason": "Processor/tokenizer regression in multimodal custom code vs token embedding resize not propagating to decoder embeddings." + "reason": "Both mention version compatibility, but one is a Python-version support break and the other is a model-specific loading failure in transformers>=5.1.0; not the same underlying bug." }, { "left": "issue:44295", "right": "issue:46506", "accept": false, - "reason": "Buffer access error for position_ids is unrelated to CPU/CUDA result differences in a document layout model." + "reason": "One is about reading a registered buffer for position_ids; the other is CPU/CUDA divergence in a document layout model. Different mechanisms and fixes." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Pipeline processor lookup bug vs notebook syntax error; no shared underlying code path." + "reason": "A pipeline processor lookup bug and a notebook syntax error are unrelated; they do not share a concrete code path." }, { "left": "issue:39692", "right": "issue:46932", "accept": false, - "reason": "Documentation example/model-processor mismatch and quantization failure vs Gemma4 dtype casting for quantized embedders." + "reason": "SigLIP2 docs/quantization example problems and Gemma4 quantized embedder dtype casting are different models and different failure modes." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Both touch flex_attention, but one is compiled flex_attention failing on torch>=2.9 while the other is a bad version check for AuxRequest import." + "reason": "Both involve version/platform checks around attention code, but one is a compiled flex_attention failure on torch>=2.9 and the other is an incorrect PyTorch version guard for AuxRequest import." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "Different model/config loading issues: missing model_type vs not unwrapping text_config in composite multimodal configs." + "reason": "Both concern model/config loading, but one is an unrecognized model error and the other is composite Qwen text_config unwrapping; not the same bug." }, { "left": "issue:43673", "right": "issue:46424", "accept": false, - "reason": "Chunked prefill cache missing in GenerationMixin vs cache max length only honored for quantized cache; related area, different bug." + "reason": "One is a missing GenerationMixin cache in chunked_prefill; the other is cache max length handling for quantized cache only. Related area, but distinct defects." }, { "left": "issue:42994", "right": "issue:47333", "accept": false, - "reason": "Both are save failures, but one is quantized model saving and the other is disk-offloaded Qwen3 model saving; not the same issue." + "reason": "Both are saving/offloading-related, but one is a generic quantized model save failure and the other is a specific disk-offloaded Qwen3-30B-A3B save failure; too model-specific to treat as one bug." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug vs outdated TextClassificationPipeline docs; unrelated." + "reason": "A BeitImageProcessorFast label reduction bug is unrelated to text-classification docs mentioning return_all_scores." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Cache behavior regression vs ANSI codes in loading_report output; no common bug." + "reason": "Generation cache behavior and ANSI escape output in loading_report are unrelated subsystems." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient checkpointing eval-mode restriction is unrelated to D-FINE auxiliary loss behavior." + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are different training-time concerns with no shared code path." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error vs audio feature extraction helper issue; different artifacts and code paths." + "reason": "A notebook syntax error and a _torch_extract_fbank_features issue are unrelated." }, { "left": "issue:44279", "right": "issue:46515", "accept": false, - "reason": "Generic dependency issue vs torch 2.4.1 incompatibility; too broad and not the same concrete failure." + "reason": "A dependency issue with transformers and an incompatibility with torch 2.4.1 are both environment-related, but not the same concrete failure." }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "LlamaConfig head_dim validation vs rope validation helper TypeError; both config-related but distinct bugs." + "reason": "One is a LlamaConfig validation rule for head_dim; the other is rope-parameter dict conversion with JSON-loaded lists. Different validation logic." }, { - "left": "issue:44016", - "right": "issue:44062", + "left": "issue:43650", + "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error is unrelated to AddedToken special-argument duplication." + "reason": "A vague 'ADD THE DATA' issue and an audio feature extraction bug are not the same problem." }, { - "left": "issue:43650", - "right": "issue:44220", + "left": "issue:44016", + "right": "issue:44062", "accept": false, - "reason": "Placeholder issue title with no concrete bug vs fbank feature extraction problem; no match." + "reason": "Notebook syntax errors and AddedToken keyword-argument duplication are unrelated." }, { "left": "issue:44792", "right": "issue:46424", "accept": false, - "reason": "Janus image-generation test failure vs cache max length handling; unrelated." + "reason": "A janus image-generation test failure and cache length handling for quantized cache affect different model paths." }, { - "left": "issue:42907", - "right": "issue:43441", + "left": "issue:42994", + "right": "issue:47332", "accept": false, - "reason": "Saving dequantized Ministral/Devstral models vs FlashAttention failure in Ministral-3; same family name but different concrete problems." + "reason": "Both involve saving offloaded/quantized models, but the reproductions target different architectures and storage/offload setups; not safely mergeable as one PR." }, { - "left": "issue:42994", - "right": "issue:47332", + "left": "issue:42907", + "right": "issue:43441", "accept": false, - "reason": "Quantized model save failure vs disk-offloaded Qwen3-VL save failure; different model setups and failure modes." + "reason": "Saving dequantized Ministrals/Devstrals and FlashAttention failures in Ministral-3 are separate bugs in different parts of the model stack." }, { "left": "issue:43441", "right": "issue:45588", "accept": false, - "reason": "Both mention flash attention, but one is a model-specific v5 RC failure and the other is an s_aux=None crash in the flash_attention integration." + "reason": "Both mention flash attention, but one is a Ministral FlashAttention failure and the other is an AttributeError on s_aux=None for sink-less models; different concrete failures." }, { "left": "issue:44327", "right": "issue:46752", "accept": false, - "reason": "QA decode_spans top-k bounds error vs empty chat-template list IndexError; unrelated runtime errors." + "reason": "A QA pipeline top-k bounds crash and apply_chat_template on empty list are different index/error conditions in different APIs." }, { "left": "issue:43976", "right": "issue:44279", "accept": false, - "reason": "Python version support problem vs generic dependency issue; not the same underlying defect." + "reason": "Python-version compatibility and a generic dependency issue are too broad and not the same underlying defect." }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Missing Qwen2.5-VL export/import vs nonexistent import from image_utils; different missing-symbol bugs." + "reason": "ImportError for Qwen2_5_VLForConditionalGeneration and a nonexistent import from image_utils are both import problems, but for different symbols and modules." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Token embedding resize propagation bug vs Whisper empty-transcription generation regression; unrelated." + "reason": "T5Gemma embedding resize behavior and Whisper empty-transcription generation after align_special_tokens are unrelated model behaviors." }, { "left": "issue:41950", "right": "issue:45020", "accept": false, - "reason": "Pipeline image-processor lookup bug vs broad remote_code regression; not the same concrete code-path issue." + "reason": "A video-classification processor lookup bug and remote_code breakages are not the same code path." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "processor.tokenizer/custom multimodal regression vs Deepseek tokenizer correctness change; different tokenization problems." + "reason": "One is a processor/tokenizer regression with images; the other is DeepSeek tokenizer output changes in v5. Different tokenizer bugs." }, { - "left": "issue:43299", - "right": "issue:45216", + "left": "issue:43901", + "right": "issue:44496", "accept": false, - "reason": "Qwen3VL MoE loading breakage vs Qwen3.5 checkpoint save corruption; different model lifecycle stages." + "reason": "Docs behavior mismatch for return_all_scores and an unrecognized model/config error do not describe the same issue." }, { - "left": "issue:43901", - "right": "issue:44496", + "left": "issue:44279", + "right": "issue:45405", "accept": false, - "reason": "Docs mismatch for return_all_scores vs unrecognized model config; unrelated." + "reason": "A dependency issue and a bumped MIN_PEFT_VERSION not yet on PyPI are both packaging-related, but not the same bug." }, { "left": "issue:43901", "right": "issue:47217", "accept": false, - "reason": "Docs wording issue vs VL models failing when image placeholder is in user text; different problem domains." + "reason": "Text-classification docs and VL image-placeholder handling are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "A tokenizer BPE-marker bug and a placeholder request issue are unrelated; no shared code path." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "SigLIP2 documentation/quantization errors and text-classification docs mentioning return_all_scores are separate documentation or example issues, not the same bug." } ] }, @@ -13331,9 +13361,7 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-533", "nodes": [ - "issue:39692", "issue:41950", - "issue:42796", "issue:43299", "issue:43381", "issue:43479", @@ -13342,7 +13370,7 @@ "issue:43650", "issue:43653", "issue:43854", - "issue:43901", + "issue:43873", "issue:43931", "issue:43976", "issue:44016", @@ -13364,6 +13392,7 @@ "issue:45375", "issue:45405", "issue:45542", + "issue:45588", "issue:45593", "issue:45812", "issue:45823", @@ -13381,224 +13410,231 @@ "issue:47405" ], "soft_pairs": [ - "issue:42796|issue:43650", - "issue:39692|issue:43901", + "issue:43299|issue:45216", "issue:44779|issue:45812", - "issue:43577|issue:45216", "issue:43479|issue:45375", "issue:46614|issue:47085", - "issue:46123|issue:46369", "issue:41950|issue:45823", + "issue:46123|issue:46369", + "issue:43577|issue:45216", "issue:44466|issue:45276", "issue:43976|issue:45405", "issue:44075|issue:44368", "issue:43653|issue:45593", "issue:44462|issue:45081", - "issue:47145|issue:47405", "issue:43381|issue:47000", + "issue:47145|issue:47405", "issue:43299|issue:45357", "issue:43650|issue:46209", - "issue:43299|issue:43854", "issue:43644|issue:47145", "issue:44351|issue:45278", - "issue:46082|issue:46399", + "issue:43299|issue:43854", "issue:44683|issue:45542", + "issue:46082|issue:46399", "issue:44016|issue:44291", - "issue:46082|issue:46512", "issue:43931|issue:47145", + "issue:46082|issue:46512", "issue:45081|issue:45907", - "issue:44242|issue:45200" + "issue:44242|issue:45200", + "issue:45588|issue:47000", + "issue:43873|issue:47145" ], - "prepared_review_unit_hash": "c43430b69c2fcd4bf59453d0b119cb81905f1b81bb68a6799832ab578dce5dd5", + "prepared_review_unit_hash": "71978428e799810ce435bb3fd8bd104b1694703f057f06e639733ba8c28cf08e", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 36084, - "estimated_input_tokens": 9021, - "estimated_eval_tokens": 18298 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35807, + "estimated_input_tokens": 8952, + "estimated_eval_tokens": 18160 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous; the soft pairs are mostly broad topical similarities rather than true duplicates. I reject all soft edges. If a single canonical issue is needed, issue 45812 is the best representative because it is a broad, user-facing tokenizer regression with clear impact and stronger signal than the others.", - "confidence": 0.41, - "canonical_issue_reason": "Issue 45812 is the strongest representative of the set: it describes a broad, reproducible tokenizer regression affecting many Granite models and has the clearest systemic impact among the listed issues.", + "summary": "This cluster is very heterogeneous: the candidate pairs mostly share only broad themes like tokenizers, loading, or regressions, but not the same concrete bug. I found no pair with enough evidence to merge as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45216 is the clearest concrete regression report with a specific version boundary and a well-scoped model-saving failure, making it the best representative issue in this set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45812 is the best overall issue candidate because it is specific, high-signal, and broadly impactful; the other issues are mostly narrower model/config/doc regressions or unrelated failures.", + "best_issue_reason": "Issue 45216 is the most actionable and specific of the reports; it has a precise regression claim and is easier to anchor than the broader or more ambiguous items.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:42796", - "right": "issue:43650", - "accept": false, - "reason": "Different problems: tokenizer BPE marker handling vs a vague non-technical issue title; no shared code path or bug." - }, - { - "left": "issue:39692", - "right": "issue:43901", + "left": "issue:43299", + "right": "issue:45216", "accept": false, - "reason": "Both are documentation/example-related, but they concern different pipelines and different errors; not the same underlying issue." + "reason": "Both involve Qwen model regressions, but one is a loading failure for Qwen3VL MoE and the other is an incorrect save_pretrained output for Qwen3.5; different code paths." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and describe different failure modes/code paths." - }, - { - "left": "issue:43577", - "right": "issue:45216", - "accept": false, - "reason": "Different bugs: Blip2 dtype propagation vs Qwen3.5 save_pretrained checkpoint key regression." + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes (incorrect tokenization vs wrong IDs for Granite)." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both involve config fields/defaults, but they target different models and different missing fields; not a duplicate." + "reason": "Both concern config handling, but they are different models and different missing/defaulted fields; no same underlying bug is evident." }, { "left": "issue:46614", "right": "issue:47085", "accept": false, - "reason": "Unrelated: stale phonemizer backend state vs tokenizer thread-safety regression." + "reason": "One is a stale language backend in a specific tokenizer class, the other is a broad thread-safety regression in fast tokenizers; not the same issue." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "These are unrelated subsystems: a pipeline processor lookup bug versus a device-mapping mismatch in Gemma4 forward passes." }, { "left": "issue:46123", "right": "issue:46369", "accept": false, - "reason": "Different subsystems and symptoms: batch result dropping vs model memory overflow." + "reason": "Different symptoms and code paths: partial-batch handling in a pipeline versus a memory overflow in a model invocation." }, { - "left": "issue:41950", - "right": "issue:45823", + "left": "issue:43577", + "right": "issue:45216", "accept": false, - "reason": "Different runtime bugs in different pipelines/models; only superficial broad inference-related similarity." + "reason": "Both touch loading/saving dtype or regression behavior, but one is a BLIP2 dtype propagation issue and the other is a Qwen3.5 save_pretrained regression." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Both touch Gemma4 serialization/embedding behavior, but one is lm_head serialization and the other is resize_token_embeddings propagation; separate code paths." + "reason": "Both mention tied weights or embeddings, but one is serialization inconsistency and the other is resize_token_embeddings not updating Gemma4 internals." }, { "left": "issue:43976", "right": "issue:45405", "accept": false, - "reason": "Completely different issues: Python version compatibility vs PEFT version pin mismatch." + "reason": "A Python version compatibility break and a dependency version bump are unrelated." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Unrelated: optimizer argument handling vs a model warning about tied embeddings." + "reason": "Different bugs: optimizer argument handling versus a model config warning about tied embeddings." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "Different bugs in different components: tokenizer special-token registration vs D-FINE auxiliary loss handling." + "reason": "Tokenizer special-token registration is unrelated to D-FINE auxiliary-loss handling." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer-related, but one is AutoTokenizer repository file selection and the other is a Mistral regex patch crash; not the same bug." - }, - { - "left": "issue:47145", - "right": "issue:47405", - "accept": false, - "reason": "Both are silent corruption concerns, but one is FP8 weight_scale handling and the other is wrong Auto class/model_type loading." + "reason": "Both are tokenizer-loading related, but one ignores tokenizer.json while the other crashes in a Mistral regex patch; different failures." }, { "left": "issue:43381", "right": "issue:47000", "accept": false, - "reason": "Different failures: gradient checkpointing in eval mode vs RTDetrLoss assignment infeasibility." + "reason": "Gradient checkpointing in eval mode and RTDetr loss infeasibility are separate model/training bugs." }, { - "left": "issue:43299", - "right": "issue:45357", + "left": "issue:47145", + "right": "issue:47405", "accept": false, - "reason": "Both concern Qwen multimodal models, but one is a loading failure and the other is incorrect save_pretrained key serialization." + "reason": "Both are silent corruption/loading issues, but one is FP8 weight_scale application and the other is wrong Auto class/model_type conversion skipping." }, { - "left": "issue:43650", - "right": "issue:46209", + "left": "issue:43299", + "right": "issue:45357", "accept": false, - "reason": "No meaningful overlap: a placeholder/noise issue title vs an FP8 inference failure on B200." + "reason": "Same broad Qwen3.5 family, but one is a load-time model initialization failure and the other is incorrect save_pretrained key serialization." }, { - "left": "issue:43299", - "right": "issue:43854", + "left": "issue:43650", + "right": "issue:46209", "accept": false, - "reason": "Both are model loading issues, but for different models and different failure mechanisms." + "reason": "No overlap beyond being user-reported failures; one is placeholder text, the other is an FP8 inference issue on B200." }, { "left": "issue:43644", "right": "issue:47145", "accept": false, - "reason": "Both mention loading/corruption risks, but the concrete bugs are different: non-persistent buffers vs FP8 compressed-tensors weight scaling." + "reason": "Different loading/saving corruption mechanisms; non-persistent buffers filled with junk is not the same as missing FP8 weight scaling." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "One is a specific missing import; the other is a broad upgrade-related import failure report. Too vague to treat as the same bug." + "reason": "A specific missing export cannot be merged with a broad report of many import errors without evidence they share the same root cause." }, { - "left": "issue:46082", - "right": "issue:46399", + "left": "issue:43299", + "right": "issue:43854", "accept": false, - "reason": "Different model/config issues: LlamaConfig head_dim validation vs Qwen3.5 kernel registration crash." + "reason": "Different models and different load failures; there is no indication they stem from the same bug." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "Unrelated: flex_attention compiler failure vs TensorBoard/TensorFlow backend installation issue." + "reason": "Compiled flex_attention on torch 2.9 is unrelated to tensorboard/TensorFlow backend installation issues." }, { - "left": "issue:44016", - "right": "issue:44291", + "left": "issue:46082", + "right": "issue:46399", "accept": false, - "reason": "Notebook syntax error and unexpected init_empty_weights argument are unrelated." + "reason": "Different components and failure modes: LlamaConfig validation versus a Qwen3.5 kernel registration error." }, { - "left": "issue:46082", - "right": "issue:46512", + "left": "issue:44016", + "right": "issue:44291", "accept": false, - "reason": "Both are config-related, but one is head_dim validation and the other is stale config.dtype after from_config; different problems." + "reason": "A notebook syntax error and an init_empty_weights TypeError are unrelated." }, { "left": "issue:43931", "right": "issue:47145", "accept": false, - "reason": "Different domains entirely: pipeline documentation mismatch vs FP8 checkpoint corruption." + "reason": "A Qwen3-VL weight-shape mismatch is not the same as FP8 frozen checkpoints missing weight_scale application." + }, + { + "left": "issue:46082", + "right": "issue:46512", + "accept": false, + "reason": "Both concern config/dtype semantics, but one is head_dim validation and the other is stale model.config.dtype after from_config." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Different API bugs: tokenizer regex crash vs list_repo_files not forwarding revision." + "reason": "Tokenizer regex patching and list_repo_files revision forwarding are unrelated API bugs." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Different model behaviors: missing load-balancing loss vs Gemma4 mm_token_type_ids defaulting." + "reason": "Both involve missing defaults in model behavior, but they affect different models and different inputs/loss logic." + }, + { + "left": "issue:45588", + "right": "issue:47000", + "accept": false, + "reason": "An attention integration AttributeError and a cost-matrix infeasibility in RTDetrLoss are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:47145", + "accept": false, + "reason": "Quantized offloading behavior is not the same bug as FP8 compressed-tensors loading without applying weight_scale." } ] }, @@ -13635,12 +13671,13 @@ "issue:43673", "issue:43761", "issue:43784", - "issue:43873", + "issue:43824", "issue:43881", "issue:43901", "issue:43976", "issue:44016", "issue:44186", + "issue:44188", "issue:44246", "issue:44261", "issue:44263", @@ -13656,7 +13693,6 @@ "issue:45372", "issue:45397", "issue:45542", - "issue:45588", "issue:45600", "issue:45656", "issue:45712", @@ -13672,238 +13708,238 @@ "issue:47217" ], "soft_pairs": [ - "issue:45588|issue:47000", - "issue:43873|issue:47145", "issue:45103|issue:45372", "issue:45859|issue:46851", - "issue:42994|issue:43317", "issue:44336|issue:46851", + "issue:42994|issue:43317", "issue:43441|issue:45834", "issue:33290|issue:45656", "issue:45542|issue:45800", - "issue:44671|issue:44779", "issue:44623|issue:44749", - "issue:42175|issue:43976", + "issue:44671|issue:44779", "issue:44246|issue:47217", + "issue:42175|issue:43976", "issue:44016|issue:45823", "issue:44623|issue:45397", - "issue:44186|issue:47000", "issue:43525|issue:43881", + "issue:44186|issue:47000", "issue:43479|issue:43525", - "issue:43784|issue:45542", "issue:43673|issue:43761", - "issue:43525|issue:43643", + "issue:43784|issue:45542", "issue:44261|issue:44263", + "issue:43525|issue:43643", "issue:43901|issue:44246", "issue:45325|issue:46851", - "issue:44589|issue:45362", "issue:43575|issue:47145", "issue:45600|issue:45712", + "issue:42796|issue:43441", + "issue:44188|issue:44336", + "issue:44589|issue:45362", "issue:46399|issue:46846", - "issue:42796|issue:43441" + "issue:43824|issue:43976" ], - "prepared_review_unit_hash": "9432ca99dd8ee98e793292ac424af20cade60de2eddd98e24bff13c60bfb1fc1", + "prepared_review_unit_hash": "835948a0d236e3c72f98b780e1bc98bc7c973129c9bf498884ac887ec66c88f4", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36265, - "estimated_input_tokens": 9067, - "estimated_eval_tokens": 18390 + "serialized_chars": 36300, + "estimated_input_tokens": 9075, + "estimated_eval_tokens": 18406 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are distinct bugs/regressions across unrelated models, tokenizer paths, config objects, and backend/offload code. I did not find any soft pair that clearly describes the same underlying fixable bug, so I rejected all candidate duplicates.", - "confidence": 0.89, - "canonical_issue_reason": "No single canonical duplicate emerges from this set. If one issue must serve as the anchor, #43873 is the broadest and most referenced report around quantization/offloading, but it still does not subsume the other issues.", + "summary": "This cluster is highly heterogeneous: the issues span unrelated models, config bugs, tokenizer errors, loading/offload problems, docs, and CI cleanup. None of the soft pairs look like the same underlying bug or change, so all are rejected.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#43873 is the strongest standalone issue in the set: it has the most discussion and inbound references, a clear user-facing failure mode, and a broad enough title to be a useful triage anchor.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:45588", - "right": "issue:47000", - "accept": false, - "reason": "Different subsystems and failures: FlashAttention sink-less crash vs RTDetr loss infeasible-cost-matrix error." - }, - { - "left": "issue:43873", - "right": "issue:47145", - "accept": false, - "reason": "Both mention quantization/offload, but one is load/offloading behavior and the other is silent FP8 checkpoint corruption with weight scales." - }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "Unrelated bugs: auto-docstring annotation parsing vs mistral_common import failure during Gemma 4 processor loading." + "reason": "Different failure modes and code paths: auto-docstring annotation handling vs a mistral_common import error in Gemma 4 processor loading." }, { "left": "issue:45859", "right": "issue:46851", "accept": false, - "reason": "Same model family, but different code paths: CPU offload rotary embedding compatibility vs variable-length position_ids handling." + "reason": "Same model family but different bugs: CPU-offload rotary embedding compatibility vs variable-length position_ids handling." }, { - "left": "issue:42994", - "right": "issue:43317", + "left": "issue:44336", + "right": "issue:46851", "accept": false, - "reason": "Both involve quantization/loading, but saving failure and dequantized offload loading failure are distinct bugs." + "reason": "Unrelated: ANSI output handling in loading_report vs Qwen3.5 variable-length input behavior." }, { - "left": "issue:44336", - "right": "issue:46851", + "left": "issue:42994", + "right": "issue:43317", "accept": false, - "reason": "ANSI escape handling in loading_report is unrelated to Qwen3.5 variable-length position_ids." + "reason": "Both quantization-related, but one is checkpoint saving and the other is dequantized model loading with device_map auto/offload." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "Different model families and failures: FlashAttention crash vs long-ocr index error." + "reason": "Completely different models and failures: FlashAttention import/runtime issue vs long OCR input index error." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Both touch deepspeed/optimizer behavior, but OOM with Adafactor is not the same as optimizer.step being called twice." + "reason": "Both mention deepspeed, but one is OOM with Adafactor and the other is optimizer.step being called twice; not the same bug." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "Packaging/backend dependency issue vs torch/transformers version incompatibility; no shared bug path." - }, - { - "left": "issue:44671", - "right": "issue:44779", - "accept": false, - "reason": "Different symptoms and components: CamemBERT MLM output regression vs Deepseek tokenizer correctness regression." + "reason": "Different problems: missing TensorFlow backend dependency vs general torch/transformers incompatibility." }, { "left": "issue:44623", "right": "issue:44749", "accept": false, - "reason": "processor.save_pretrained missing files is unrelated to the reported data-filtering slowdown." + "reason": "No shared underlying issue: missing files on processor.save_pretrained vs a slowdown report after upgrading Transformers." }, { - "left": "issue:42175", - "right": "issue:43976", + "left": "issue:44671", + "right": "issue:44779", "accept": false, - "reason": "Different failure domains: missing TensorFlow backend in an extra install vs Python version support breakage." + "reason": "Both are incorrect model outputs, but for different models and causes; not the same code-path bug." }, { "left": "issue:44246", "right": "issue:47217", "accept": false, - "reason": "General import latency issue vs multimodal image-placeholder parsing failure." + "reason": "Import performance issue vs VL image-placeholder handling bug; unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Different scope: packaging backend selection vs Python version compatibility regression." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error and Gemma4 device-mismatch are unrelated issues." + "reason": "Notebook syntax error and Gemma4 device mismatch are unrelated." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Missing processor save files is not the same as Gemma-4 ZeRO3 from_pretrained failure." - }, - { - "left": "issue:44186", - "right": "issue:47000", - "accept": false, - "reason": "LayoutLMv2 tokenizer batching/NER crash is a different code path from RTDetr loss infeasibility." + "reason": "Processor save_pretrained file emission bug vs Gemma-4 zero3 from_pretrained issue; different code paths." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Llama4Config pad_token_id AttributeError and glm-4v-9b loading failure are distinct config/load bugs." + "reason": "Both are model-loading/config-related, but one is a missing pad_token_id attribute and the other is a GLM-4V loading failure; not enough to treat as one bug." }, { - "left": "issue:43479", - "right": "issue:43525", + "left": "issue:44186", + "right": "issue:47000", "accept": false, - "reason": "Both are config-related, but one is default multimodal sub-config initialization and the other is a missing pad_token_id attribute." + "reason": "Tokenizer padding/NER crash vs RTDetrLoss infeasible cost matrix; unrelated." }, { - "left": "issue:43784", - "right": "issue:45542", + "left": "issue:43479", + "right": "issue:43525", "accept": false, - "reason": "Importing sentence-transformers NameError is unrelated to TensorFlow backend packaging." + "reason": "Both involve config defaults, but for different models and different missing-field behavior." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Chunked prefill cache regression and CLIP hidden_states regression affect different model/execution paths." + "reason": "Different regressions in different APIs: GenerationMixin chunked_prefill cache vs CLIPVisionModel hidden_states output." }, { - "left": "issue:43525", - "right": "issue:43643", + "left": "issue:43784", + "right": "issue:45542", "accept": false, - "reason": "Llama4Config missing pad_token_id is not the same as trust_remote_code returning incomplete AutoConfig fields." + "reason": "Import-time NameError in sentence-transformers integration vs TensorFlow backend installation problem; unrelated." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "Different models and failures: missing rms_norm_eps precision issue vs torch.split return-value bug." + "reason": "Both are model-specific issues, but one is an rms_norm_eps precision/config problem and the other is torch.split return-value handling; not the same bug." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Both mention missing config fields, but one is a model config attribute omission and the other is trust_remote_code AutoConfig returning incomplete objects; too different to merge." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Documentation mismatch and import slowness are unrelated." + "reason": "Docs mismatch for return_all_scores vs import performance issue; unrelated." }, { "left": "issue:45325", "right": "issue:46851", "accept": false, - "reason": "Both involve Qwen3.5/position handling, but one is still-image temporal scaling and the other is variable-length input support." - }, - { - "left": "issue:44589", - "right": "issue:45362", - "accept": false, - "reason": "Float8 storage lookup error is unrelated to Qwen3.5 chat crash." + "reason": "Both touch position_ids, but they affect different Qwen models and distinct logic paths." }, { "left": "issue:43575", "right": "issue:47145", "accept": false, - "reason": "Qwen2 TP OOM on load and FP8 frozen-checkpoint weight_scale omission are different quantization/offload failures." + "reason": "OOM during tensor-parallel loading vs silent FP8 checkpoint scaling corruption; unrelated." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "One is stale auto-mapping references; the other is leftover dummy classes leaking into namespace and CI." + "reason": "Both are cleanup/maintenance-type reports, but they concern different files and different repository checks." + }, + { + "left": "issue:42796", + "right": "issue:43441", + "accept": false, + "reason": "Tokenizer BPE-marker handling and FlashAttention failure are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile vs ANSI code emission in loading_report; unrelated." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Storage-object lookup failure vs Qwen3.5 chat crash; unrelated." }, { "left": "issue:46399", "right": "issue:46846", "accept": false, - "reason": "Both are Qwen3.5 issues, but one is a rotary-embedding registration error and the other is a TP plan omission causing OOM/shape errors." + "reason": "Same model family, but one is a kernel registration/decorator crash and the other is a tensor-parallel plan omission causing OOM/shape errors; different fixes." }, { - "left": "issue:42796", - "right": "issue:43441", + "left": "issue:43824", + "right": "issue:43976", "accept": false, - "reason": "Tokenizer BPE-marker handling and FlashAttention runtime failure are different bugs." + "reason": "Import error for a model class vs Python version compatibility; unrelated." } ] }, @@ -13937,10 +13973,9 @@ "issue:43550", "issue:43596", "issue:43723", - "issue:43824", "issue:43976", "issue:44162", - "issue:44188", + "issue:44242", "issue:44246", "issue:44295", "issue:44297", @@ -13957,6 +13992,7 @@ "issue:45561", "issue:45563", "issue:45588", + "issue:45593", "issue:46291", "issue:46399", "issue:46424", @@ -13976,245 +14012,232 @@ "issue:47333" ], "soft_pairs": [ - "issue:44188|issue:44336", - "issue:43824|issue:43976", "issue:45362|issue:47333", "issue:44336|issue:44373", "issue:44246|issue:44496", "issue:43723|issue:44162", "issue:42617|issue:43976", - "issue:47176|issue:47177", "issue:44297|issue:45397", + "issue:47176|issue:47177", "issue:46291|issue:46399", "issue:45083|issue:46851", "issue:46655|issue:46726", - "issue:42491|issue:43299", "issue:46399|issue:46851", + "issue:42491|issue:43299", "issue:45003|issue:47217", "issue:45072|issue:46829", "issue:43526|issue:45003", "issue:43540|issue:46752", "issue:44295|issue:45397", "issue:45103|issue:46399", - "issue:45003|issue:46821", "issue:44246|issue:44871", "issue:45588|issue:46851", + "issue:45003|issue:46821", "issue:45561|issue:45563", "issue:44336|issue:46424", - "issue:46614|issue:46665", "issue:43550|issue:46506", + "issue:46614|issue:46665", "issue:36246|issue:43596", "issue:43540|issue:46655", - "issue:42915|issue:47312" + "issue:42915|issue:47312", + "issue:44242|issue:45593" ], - "prepared_review_unit_hash": "74aad7c4de1b86efa19db97b4afee640c360b28b9560b7bd14ab908389e984f0", + "prepared_review_unit_hash": "7ee6d2124596d96fe4f0e2b8016b64a68c040d20bfbdb34ae94ea17e0934851a", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36164, - "estimated_input_tokens": 9041, - "estimated_eval_tokens": 18338 + "soft_pair_count": 29, + "serialized_chars": 35808, + "estimated_input_tokens": 8952, + "estimated_eval_tokens": 18160 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are unrelated regressions across imports, Qwen/VL models, kernels, tokenizer behavior, testing, and even security issues. I did not find any soft pair that clearly describes the same underlying bug/change, so all proposed merges are rejected.", - "confidence": 0.31, - "canonical_issue_reason": "issue:43824 is the strongest representative because it is a clear, high-signal breaking import regression with broad discussion and references; the rest of the cluster is too mixed to form a clean duplicate set.", + "summary": "This cluster is a mixed bag of mostly unrelated Transformers issues spanning Qwen, tokenizers, model loading, kernels, security, CI, and docs. None of the soft pairs look like the same underlying bug or change, so there is no credible duplicate canonical here.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43824 is the best single issue to anchor this cluster since it is a prominent core-library import failure and has the most obvious cross-user impact among the listed issues.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44188", - "right": "issue:44336", - "accept": false, - "reason": "Different subsystems and symptoms: attention kernel divergence vs ANSI codes in loading_report." - }, - { - "left": "issue:43824", - "right": "issue:43976", - "accept": false, - "reason": "Both are version-related breakages, but one is a specific missing Qwen import and the other is a Python compatibility problem; not the same bug." - }, { "left": "issue:45362", "right": "issue:47333", "accept": false, - "reason": "Different Qwen3.5 chat crash vs disk-offloaded model save failure; unrelated code paths." + "reason": "Different Qwen runtime problems: chat/template crash vs offloading save failure; no shared code path." }, { "left": "issue:44336", "right": "issue:44373", "accept": false, - "reason": "Unrelated: terminal ANSI output vs a docstring issue for position_ids." + "reason": "Unrelated doc/report formatting vs a docstring typo; not the same bug." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import slowdowns and unrecognized model config errors are different failure modes." + "reason": "Import slowness and unrecognized-model config errors are different failure modes." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer loading regression and ESM2 breakage are separate model-specific problems." + "reason": "Tokenizer loading regression vs ESM2 model breakage; different subsystems." }, { "left": "issue:42617", "right": "issue:43976", "accept": false, - "reason": "A script execution issue and a Python version incompatibility are not the same underlying bug." + "reason": "A script/runtime failure and a Python-version compatibility issue are not the same underlying problem." }, { - "left": "issue:47176", - "right": "issue:47177", + "left": "issue:44297", + "right": "issue:45397", "accept": false, - "reason": "Both are security issues, but command-substitution bypass and tar/zip slip are distinct vulnerabilities." + "reason": "Tokenizer save metadata mismatch vs Gemma-4 ZeRO/from_pretrained failure; different paths." }, { - "left": "issue:44297", - "right": "issue:45397", + "left": "issue:47176", + "right": "issue:47177", "accept": false, - "reason": "Tokenizer save metadata mismatch is unrelated to Gemma-4 zero3 loading." + "reason": "Distinct security issues: shell-substitution allowlist bypass vs archive path traversal/RCE." }, { "left": "issue:46291", "right": "issue:46399", "accept": false, - "reason": "Kernel dependency import failure vs missing decorator during rotary pos emb registration; different concrete problems." + "reason": "Both involve kernels, but one is an import-time dependency/version constraint and the other is a missing registration decorator in Qwen3.5." }, { "left": "issue:45083", "right": "issue:46851", "accept": false, - "reason": "Both involve Qwen models, but one is an audio feature-length helper and the other is variable-length position_ids handling." + "reason": "Both mention Qwen and positions, but one is a feature-extract length helper and the other is variable-length position_ids handling." }, { "left": "issue:46655", "right": "issue:46726", "accept": false, - "reason": "Video padding miscalculation and missing vision backend reporting are separate issues." + "reason": "Video padding bug vs missing-vision-backend processor error; not the same bug." }, { - "left": "issue:42491", - "right": "issue:43299", + "left": "issue:46399", + "right": "issue:46851", "accept": false, - "reason": "Both mention Qwen3 MoE/version changes, but one is LoRA reuse across branches and the other is Qwen3VL MoE loading failure." + "reason": "Different Qwen3.5 failures: kernel registration vs variable-length input handling." }, { - "left": "issue:46399", - "right": "issue:46851", + "left": "issue:42491", + "right": "issue:43299", "accept": false, - "reason": "Different failures: kernel registration error vs position_ids handling for variable-length inputs." + "reason": "Both are Qwen MoE regressions, but one is LoRA compatibility across hf4.x/hf5.x and the other is model loading on transformers v5; too different to merge." }, { "left": "issue:45003", "right": "issue:47217", "accept": false, - "reason": "Unsafe sys.modules access and VL image-placeholder handling are unrelated." + "reason": "Unsafe sys.modules access is unrelated to VL image-placeholder parsing." }, { "left": "issue:45072", "right": "issue:46829", "accept": false, - "reason": "Different model behaviors: bf16 dtype mismatch in inference vs incorrect hidden_states output packaging." + "reason": "Different inference bugs: dtype mismatch vs incorrect output field assignment." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to sys.modules access in modeling_utils." + "reason": "BEIT label reduction logic and sys.modules access are unrelated." }, { "left": "issue:43540", "right": "issue:46752", "accept": false, - "reason": "Qwen3OmniMoe video input error and empty chat-template list crash are different code paths." + "reason": "Qwen3OmniMoe video-input error vs empty-list chat template crash; different code paths." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Registering position_ids as a buffer causing read errors is unrelated to zero3 from_pretrained on Gemma-4." + "reason": "Buffer access for position_ids vs Gemma-4 ZeRO loading; not the same underlying issue." }, { "left": "issue:45103", "right": "issue:46399", "accept": false, - "reason": "Auto-docstring crash with future annotations is unrelated to Qwen3.5 kernel registration failures." - }, - { - "left": "issue:45003", - "right": "issue:46821", - "accept": false, - "reason": "sys.modules access bug and TimesFM window_size AttributeError do not share the same underlying defect." + "reason": "Auto-docstring future-annotations crash is unrelated to Qwen kernel registration." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import performance issue and Gemma eos_token_id inconsistency are different problems." + "reason": "Import-time performance regression and eos_token_id config inconsistency are unrelated." }, { "left": "issue:45588", "right": "issue:46851", "accept": false, - "reason": "Flash-attention s_aux handling and Qwen3.5 variable-length position_ids are not the same bug." + "reason": "Flash-attention sink handling vs variable-length position_ids; different failures." + }, + { + "left": "issue:45003", + "right": "issue:46821", + "accept": false, + "reason": "sys.modules access and TimesFM window_size errors are unrelated." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "Pytest-xdist race in testing utils and stale paged-generate warning are unrelated." + "reason": "A test-file race and a stale warning in generate() are different issues." }, { "left": "issue:44336", "right": "issue:46424", "accept": false, - "reason": "ANSI code emission and cache max length semantics are separate concerns." + "reason": "ANSI output formatting bug vs cache-length logic bug; no shared code path." }, { - "left": "issue:46614", - "right": "issue:46665", + "left": "issue:43550", + "right": "issue:46506", "accept": false, - "reason": "Phonemizer backend state bug and from_pretrained behavior inside Python classes are different issues." + "reason": "torch.compile/SDPA failure and CPU-vs-CUDA divergence are distinct model/runtime problems." }, { - "left": "issue:43550", - "right": "issue:46506", + "left": "issue:46614", + "right": "issue:46665", "accept": false, - "reason": "Torch.compile SDPA failure in Bamba and CPU/CUDA result divergence in PPDocLayoutV3 are unrelated." + "reason": "Phonemizer backend caching and from_pretrained-in-class behavior are different issues." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "Missing Qwen image processor import and DeepSpeed/BertModel index error are unrelated." + "reason": "Qwen2_5_VLImageProcessor import error is unrelated to BertModel ZeRO init indexing." }, { - "left": "issue:43540", - "right": "issue:46655", + "left": "issue:42915", + "right": "issue:47312", "accept": false, - "reason": "Different multimodal video-processing bugs in Qwen3OmniMoe vs SmolVLM." + "reason": "Qwen3Moe FP8 config failure and kernels 0.16 incompatibility are different regressions." }, { - "left": "issue:42915", - "right": "issue:47312", + "left": "issue:44242", + "right": "issue:45593", "accept": false, - "reason": "Qwen3Moe FP8 failure and kernels 0.16 incompatibility are different issues." + "reason": "Both concern missing losses, but they affect different models and mechanisms, so they are not the same bug." } ] }, @@ -14247,11 +14270,11 @@ "issue:43352", "issue:43475", "issue:43901", + "issue:44038", "issue:44060", "issue:44077", "issue:44162", "issue:44220", - "issue:44242", "issue:44295", "issue:44351", "issue:44451", @@ -14273,13 +14296,13 @@ "issue:45459", "issue:45561", "issue:45588", - "issue:45593", "issue:45600", "issue:45701", "issue:45759", "issue:46132", "issue:46170", "issue:46369", + "issue:46506", "issue:46655", "issue:46726", "issue:46752", @@ -14287,7 +14310,6 @@ "issue:47321" ], "soft_pairs": [ - "issue:44242|issue:45593", "issue:44779|issue:45292", "issue:41950|issue:44220", "issue:44479|issue:46655", @@ -14301,224 +14323,225 @@ "issue:45292|issue:45701", "issue:43295|issue:45292", "issue:44295|issue:44749", - "issue:43352|issue:45250", - "issue:44451|issue:46752", "issue:45137|issue:45600", - "issue:44162|issue:45200", + "issue:43352|issue:45250", "issue:44466|issue:45335", + "issue:44162|issue:45200", + "issue:44451|issue:46752", "issue:46170|issue:47321", "issue:44995|issue:45200", "issue:44060|issue:44077", "issue:44877|issue:45375", - "issue:45003|issue:45561", "issue:43475|issue:46369", + "issue:45003|issue:45561", "issue:44987|issue:45459", "issue:44479|issue:46170", "issue:44995|issue:45588", + "issue:44038|issue:46506", "issue:44623|issue:46726" ], - "prepared_review_unit_hash": "292d41143ab4c69d3e3e66c8333a4fb9659794ee73f986e41acdf0b27fff3990", + "prepared_review_unit_hash": "876797a17597f25408310bcbf1dd637103861bcedc21e2d8fc352ff035f39cfd", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35734, - "estimated_input_tokens": 8934, - "estimated_eval_tokens": 18124 + "serialized_chars": 35692, + "estimated_input_tokens": 8923, + "estimated_eval_tokens": 18102 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft edges are all false positives: they share broad keywords or subsystems, but each pair points to a different concrete bug, model, or code path. No true duplicate cluster emerges from these candidates.", - "confidence": 0.9, - "canonical_issue_reason": null, + "summary": "A heterogeneous cluster of Transformers bug reports: mostly model-specific regressions around v5/v5.3, multimodal processors/configs, DeepSpeed, imports, and tokenizer behavior. The items share broad theme overlap but do not look like true duplicates of one underlying bug.", + "confidence": 0.22, + "canonical_issue_reason": "Issue 44479 is the broadest, most representative regression report in the set: it describes a v5.3.0 video-input regression affecting multiple related Qwen VL model variants, which is closer to the center of the cluster than the narrower single-model or unrelated infrastructure bugs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the strongest standalone issue here: it is a concrete, high-signal regression report with clearly named affected Qwen VL variants and a specific failure mode.", + "best_issue_reason": "Issue 44479 is the best overall representative because it is a concrete, user-facing regression with multiple affected models and a clear version boundary, making it a stronger triage anchor than the many narrower or unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44242", - "right": "issue:45593", - "accept": false, - "reason": "Both mention missing losses, but one is router load-balancing loss and the other is auxiliary losses for D-FINE; different models and mechanisms." - }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "One is tokenizer output corruption in v5, the other is resize_token_embeddings not updating output embeddings; different bugs." + "reason": "Tokenizer output changes and embedding-resize wiring are different code paths and different failure modes." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification pipeline image-processor lookup and _torch_extract_fbank_features are unrelated code paths." + "reason": "Both mention pipelines/features, but one is about processor selection and the other is a feature-extraction bug." }, { "left": "issue:44479", "right": "issue:46655", "accept": false, - "reason": "Both are video-related regressions, but one is a Qwen VL video-input regression and the other is a SmolVLM video padding bug; not the same fix." + "reason": "Both are video-related regressions, but they affect different model families and the reported defects are different." }, { "left": "issue:46726", "right": "issue:47217", "accept": false, - "reason": "Unrecognized image processor due to missing backend is different from failures when the image placeholder appears in user text." + "reason": "AutoProcessor backend detection and VL image-placeholder handling are distinct problems." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Docs mentioning return_all_scores and unsafe sys.modules access are unrelated." + "reason": "Docs mismatch for pipeline arguments is unrelated to unsafe sys.modules access." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Both involve DeepSpeed, but one is a Qwen3 MOE weight-conversion issue and the other is a ZeRO3 deque underflow crash." + "reason": "Both involve DeepSpeed, but one is Qwen3 MoE weight conversion and the other is a ZeRO3 deque crash." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both touch Qwen multimodal configs, but one is a missing config field dropped by strict validation and the other is wrong text_config unwrapping." + "reason": "Both touch composite Qwen configs, but one is strict-schema field loss and the other is from_config unwrapping behavior." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Different missing exports: Qwen2_5_VLImageProcessor versus HybridCache." + "reason": "Different missing exports: Qwen2.5 VL image processor vs HybridCache." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Different import failures from different packages and causes." + "reason": "Different packages and causes: Transformers export vs huggingface_hub import." }, { "left": "issue:42886", "right": "issue:46132", "accept": false, - "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior are unrelated." + "reason": "Offline cache loading and AttentionInterface registration are unrelated." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "resize_token_embeddings/output_embeddings mismatch is not the same as version-dependent tokenization changes." + "reason": "Embedding resize behavior is not the same bug as tokenization changes across versions." }, { "left": "issue:43295", "right": "issue:45292", "accept": false, - "reason": "Processor.tokenizer/image-to-tokenizer regression is different from embedding resize not propagating to output embeddings." + "reason": "Processor/tokenizer regression is unrelated to token embedding resizing." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "position_ids buffer access error and a 5.3.0 slowdown are different bugs." + "reason": "Position_ids buffer access and a filtering slowdown are unrelated." }, { - "left": "issue:43352", - "right": "issue:45250", + "left": "issue:45137", + "right": "issue:45600", "accept": false, - "reason": "A model-specific unsupported Flash Attention 2 warning is not the same as the generic Flash Attention 2.0 issue." + "reason": "DeepSpeed ZeRO3 runtime failure and removed auto-mapping config references are different issues." }, { - "left": "issue:44451", - "right": "issue:46752", + "left": "issue:43352", + "right": "issue:45250", "accept": false, - "reason": "Model loading failure and apply_chat_template on an empty list are unrelated." + "reason": "A model explicitly not supporting Flash Attention 2.0 is not the same as the generic Flash Attention issue." }, { - "left": "issue:45137", - "right": "issue:45600", + "left": "issue:44466", + "right": "issue:45335", "accept": false, - "reason": "ZeRO3 deque crash and stale auto_mappings references are different failure modes." + "reason": "Serialization of tied weights and t5gemma embedding resize behavior are different bugs." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage and Gemma 4 token-id defaults are unrelated model-specific issues." + "reason": "ESM2 breakage and Gemma 4 multimodal token-type defaults are unrelated." }, { - "left": "issue:44466", - "right": "issue:45335", + "left": "issue:44451", + "right": "issue:46752", "accept": false, - "reason": "Both concern tied-embedding behavior, but one is serialization-dependent and the other is decoder.embed_tokens not updating after resize." + "reason": "Model loading failure and empty-list chat-template IndexError are unrelated." }, { "left": "issue:46170", "right": "issue:47321", "accept": false, - "reason": "Windows ROCm import/type errors and an AMD quark class update issue are not the same bug." + "reason": "Windows ROCm import/dependency failure and AMD quark-class update are different concerns." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "Stale indexer cache on second forward and missing mm_token_type_ids defaults are unrelated." + "reason": "Moe cache crash on second forward pass is unrelated to Gemma 4 token-type IDs." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Qwen3-Next tied-weights warning and patchtsmixer post_init policy are different model/config issues." + "reason": "Incorrect tied-weights warnings and patchtsmixer post_init policy are different code paths." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Strict config rejection for granite_speech and missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig are different config-validation problems." + "reason": "Both involve config strictness, but they are distinct models and opposite schema problems." }, { - "left": "issue:45003", - "right": "issue:45561", + "left": "issue:45459", + "right": "issue:46132", "accept": false, - "reason": "Unsafe sys.modules access and a pytest-xdist captured_info.txt race are unrelated." + "reason": "Tokenizer error masking from protobuf handling is unrelated to AttentionInterface registration." }, { - "left": "issue:43475", - "right": "issue:46369", - "accept": false, - "reason": "SAM 3 Video missing attribute and SAM3 LiteText memory overflow are distinct SAM-family bugs." + "left": "issue:45561", + "right": "issue:45003", + "accept": false, + "reason": "A pytest-xdist temp-file race is unrelated to sys.modules access in modeling_utils." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Loading physical-intelligence/fast and masking tokenizer errors when protobuf is absent are different failure paths." + "reason": "Loading a specific model failing and protobuf-related tokenizer error masking are different root causes." }, { "left": "issue:44479", "right": "issue:46170", "accept": false, - "reason": "Qwen VL video regression and Windows ROCm torch.distributed import problems are unrelated." + "reason": "Video-input regression and Windows ROCm import failure are unrelated." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "GlmMoeDsa stale cache on second forward and flash_attention.py crashing on s_aux=None are different concrete bugs." + "reason": "Second-forward cache crash and flash_attention s_aux handling are different runtime bugs." + }, + { + "left": "issue:44038", + "right": "issue:46506", + "accept": false, + "reason": "Qwen3-VL-Moe regression and CPU/CUDA output divergence are not the same issue." }, { "left": "issue:44623", "right": "issue:46726", "accept": false, - "reason": "Missing files from processor.save_pretrained and AutoProcessor complaining about an unrecognized image processor are not the same issue." + "reason": "Missing save_pretrained files and backend-missing image processor warnings affect different phases of processor handling." } ] }, @@ -14556,7 +14579,6 @@ "issue:43906", "issue:43931", "issue:43994", - "issue:44038", "issue:44162", "issue:44188", "issue:44242", @@ -14572,6 +14594,7 @@ "issue:45397", "issue:45412", "issue:45563", + "issue:45593", "issue:45656", "issue:45685", "issue:45987", @@ -14581,18 +14604,17 @@ "issue:46170", "issue:46202", "issue:46431", - "issue:46506", "issue:46710", "issue:46726", "issue:46752", "issue:46851", "issue:46858", "issue:47020", + "issue:47068", "issue:47145", "issue:47312" ], "soft_pairs": [ - "issue:44038|issue:46506", "issue:45685|issue:46431", "issue:29870|issue:30333", "issue:44877|issue:46170", @@ -14613,124 +14635,119 @@ "issue:44361|issue:44749", "issue:45987|issue:46082", "issue:44368|issue:45563", - "issue:46752|issue:47020", "issue:46132|issue:47145", + "issue:46752|issue:47020", "issue:45103|issue:46858", "issue:44483|issue:46202", - "issue:43994|issue:46726" + "issue:43994|issue:46726", + "issue:45593|issue:47068" ], - "prepared_review_unit_hash": "fe3cc3becf9cea9dcf9e0b2642eebad631be238089e1d32dd31b99c652b002a5", + "prepared_review_unit_hash": "17af3efcc080e5389ead5bd18ae9fdc3db5d6c4cecfad2ac3e34a0f2774a24d5", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35535, - "estimated_input_tokens": 8884, - "estimated_eval_tokens": 18024 + "serialized_chars": 35614, + "estimated_input_tokens": 8904, + "estimated_eval_tokens": 18064 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous and does not contain any clear duplicate issue/PR nucleus. All soft pairs appear to describe different bugs or regressions, so I reject every candidate pairing.", - "confidence": 0.84, - "canonical_issue_reason": "issue:46202 is the most concrete, self-contained user-facing bug report in the set, with a specific roundtrip failure and an easily described reproduction.", + "summary": "These items are a loose collection of unrelated bug reports across tokenizers, generation, config loading, vision models, and distributed training. None of the soft pairs look like the same underlying bug or a mergeable PR-level duplicate.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:46202 is the best representative issue because it is narrowly scoped, reproducible, and actionable compared with the other broader or more ambiguous reports.", + "best_issue_reason": "issue:43906 is the most actionable standalone report because it is explicitly framed as an isolated reproduction, but it is not a canonical duplicate target for the rest of this set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44038", - "right": "issue:46506", - "accept": false, - "reason": "Different models and failure modes: Qwen3-VL-Moe vs PPDocLayoutV3 CPU/CUDA divergence." - }, { "left": "issue:45685", "right": "issue:46431", "accept": false, - "reason": "Unrelated subsystems: MPS histogram error in MoE code vs missing chat template on a processor." + "reason": "MPS histogram/int error in MoE is unrelated to a chat-template processor missing a chat template." }, { "left": "issue:29870", "right": "issue:30333", "accept": false, - "reason": "Completely different problems: Bart generation UnboundLocalError vs MLflow reporting after failed training." + "reason": "Bart generation UnboundLocalError and MLflow job-state reporting are different subsystems and failure modes." }, { "left": "issue:44877", "right": "issue:46170", "accept": false, - "reason": "Different areas and symptoms: strict config loading for granite_speech vs Windows/ROCm torch.distributed import issues." + "reason": "Strict config loading for granite_speech is unrelated to Windows/ROCm torch.distributed import issues." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "One is MoE router-loss bookkeeping; the other is an AMP/CUDA crash in LwDetrImageLoss." + "reason": "Load-balancing loss omission and LwDetrImageLoss AMP crash are different code paths and bugs." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Different code paths: gemma-4 ZeRO-3 loading vs double optimizer stepping with DeepSpeed." + "reason": "Gemma-4 Zero3 from_pretrained failure is unrelated to Deepspeed optimizer stepping twice." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "Different models and bugs: missing fpn_position_embeddings in SAM 3 video vs Qwen3OmniMoe video-input validation." + "reason": "Missing fpn_position_embeddings in SAM 3 video output is not the same as Qwen3OmniMoe video-input validation." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "A warning about tie_word_embeddings is not the same as a flash-attention generation bug." + "reason": "A tied-embeddings warning during LoRA fine-tuning is not the same bug as flash-attention generation failure." }, { "left": "issue:43653", "right": "issue:44162", "accept": false, - "reason": "Tokenizer special-token registration is unrelated to ESM2 being broken." + "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are different tokenizer/model issues." }, { "left": "issue:42994", "right": "issue:47312", "accept": false, - "reason": "Quantized model saving failure is unrelated to kernel version incompatibility." + "reason": "Quantized model save failure and kernel version incompatibility are not the same underlying problem." }, { "left": "issue:43723", "right": "issue:46710", "accept": false, - "reason": "Tokenizer loading in v5 is unrelated to an accuracy regression after a version upgrade." + "reason": "Tokenizer loading in v5 and DeepSeek accuracy regression after upgrade are distinct regressions." }, { "left": "issue:42199", "right": "issue:46851", "accept": false, - "reason": "DETR cardinality/background-class error is unrelated to Qwen3.5 variable-length position_ids handling." + "reason": "DETR-derived cardinality logic and Qwen3.5 variable-length position_ids are unrelated model-specific bugs." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "Both are SAM3-related, but one is a dummy Conv2d implementation detail and the other is a model-type warning; not the same bug." + "reason": "Dummy Conv2d in SAM3 pixel decoder is unrelated to sam3_video/sam3_tracker config-type mismatch." }, { "left": "issue:45412", "right": "issue:46123", "accept": false, - "reason": "Memory not released for RT-DETR has no direct relation to MaskGenerationPipeline final-batch behavior." + "reason": "RT-DETR memory release and MaskGenerationPipeline partial-batch dropping are different issues." }, { "left": "issue:42886", @@ -14742,67 +14759,73 @@ "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Weight-shape mismatch when loading Qwen3-VL is unrelated to a configuration warning about tie_word_embeddings." + "reason": "Qwen3-VL weight-shape mismatch is not the same as a Qwen3.5 tied-embeddings warning." }, { "left": "issue:43906", "right": "issue:46202", "accept": false, - "reason": "The left issue is only an isolated reproduction of an older report; there is no evidence it is the same tokenizer/pipeline roundtrip bug as the right issue." + "reason": "An isolated reproduction of another issue is too generic to match a Gemma pipeline roundtrip failure on \u2581; no concrete shared code path is evident." }, { "left": "issue:44188", "right": "issue:46202", "accept": false, - "reason": "Torch.compile attention-kernel branching divergence is unrelated to Gemma pipeline roundtrip behavior." + "reason": "Torch.compile attention-kernel divergence is unrelated to Gemma tokenizer/pipeline roundtrip behavior." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer AttributeError is unrelated to a slowdown regression after upgrading Transformers." + "reason": "MLLukeTokenizer AttributeError and Transformers slowdown after upgrade are different bugs." }, { "left": "issue:45987", "right": "issue:46082", "accept": false, - "reason": "StaticCache sequence-length return type bug is unrelated to LlamaConfig head_dim validation." + "reason": "StaticCache seq-length return type and LlamaConfig head_dim validation are separate config/cache issues." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "A tied-embeddings warning is unrelated to a stale num_return_sequences warning in paged generate()." + "reason": "Two warning reports with different triggers and semantics; not the same bug." }, { - "left": "issue:46752", - "right": "issue:47020", + "left": "issue:46132", + "right": "issue:47145", "accept": false, - "reason": "Empty-list chat-template crash and XLMRobertaTokenizer Unigram construction error are different tokenizer issues." + "reason": "AttentionInterface.register behavior change and FP8 weight_scale loading corruption are unrelated." }, { - "left": "issue:46132", - "right": "issue:47145", + "left": "issue:46752", + "right": "issue:47020", "accept": false, - "reason": "AttentionInterface.register behavior change is unrelated to FP8 weight_scale being skipped on load." + "reason": "Empty-list chat-template IndexError and XLMRobertaTokenizer Unigram init type mismatch are different tokenizer bugs." }, { "left": "issue:45103", "right": "issue:46858", "accept": false, - "reason": "Auto-docstring annotation handling and static-cache seq-length arithmetic on non-CUDA backends are different bugs." + "reason": "Docstring annotation parsing crash and static-cache slice arithmetic on non-CUDA backends are unrelated." }, { "left": "issue:44483", "right": "issue:46202", "accept": false, - "reason": "Chat/completions request rejection is unrelated to a Gemma pipeline symbol roundtrip failure." + "reason": "Chat/completions request rejection is unrelated to Gemma pipeline roundtrip failures." }, { "left": "issue:43994", "right": "issue:46726", "accept": false, - "reason": "Siglip2 incorrect outputs with AutoModel/pipeline is unrelated to missing-vision-backend processor reporting." + "reason": "SigLIP2 nonsense outputs and missing vision-backend processor errors are different auto-model/processor issues." + }, + { + "left": "issue:45593", + "right": "issue:47068", + "accept": false, + "reason": "D-FINE auxiliary-loss omission and Deepspeed loss host sync are different training bugs." } ] }, @@ -14846,6 +14869,7 @@ "issue:44297", "issue:44373", "issue:44623", + "issue:44661", "issue:44749", "issue:44869", "issue:44871", @@ -14859,7 +14883,6 @@ "issue:45397", "issue:45478", "issue:45561", - "issue:45593", "issue:45656", "issue:46123", "issue:46132", @@ -14872,11 +14895,10 @@ "issue:46851", "issue:46877", "issue:47000", - "issue:47068", + "issue:47059", "issue:47217" ], "soft_pairs": [ - "issue:45593|issue:47068", "issue:44188|issue:46421", "issue:43701|issue:44749", "issue:44749|issue:45656", @@ -14889,225 +14911,226 @@ "issue:44964|issue:45137", "issue:39692|issue:46726", "issue:43784|issue:46399", - "issue:46463|issue:46877", "issue:44373|issue:46851", + "issue:46463|issue:46877", "issue:28282|issue:30990", "issue:44871|issue:46132", - "issue:45561|issue:47217", "issue:43723|issue:46326", - "issue:44188|issue:46851", + "issue:45561|issue:47217", "issue:45230|issue:45397", - "issue:45375|issue:45478", + "issue:44188|issue:46851", "issue:44869|issue:45356", "issue:46877|issue:47000", "issue:44038|issue:44623", "issue:43232|issue:43673", - "issue:45310|issue:45375", + "issue:45375|issue:45478", "issue:41628|issue:44933", - "issue:36010|issue:41628" + "issue:45310|issue:45375", + "issue:36010|issue:41628", + "issue:44661|issue:47059" ], - "prepared_review_unit_hash": "d6a34fd83ea9a84df0336c83474cb36f18cbfdb75953e9a936c5f704b9032e64", + "prepared_review_unit_hash": "f5f18bc565a52db4079178c1005c133bbac08f0ef0cf12d10bc2b691b2cd48bf", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36163, - "estimated_input_tokens": 9041, - "estimated_eval_tokens": 18338 + "serialized_chars": 36174, + "estimated_input_tokens": 9044, + "estimated_eval_tokens": 18344 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a mix of distinct Transformers regressions and docs/issues across different subsystems (imports, tokenizers, generation, vision, deepspeed, save/load, and performance). None of the soft pairs look like the same concrete bug or change, so I would not merge any of them as duplicates.", - "confidence": 0.8, - "canonical_issue_reason": "issue:43646 is the broadest and most central runtime-regression report in the set ('Transformers 5.0.0 breaks custom model initialization'). There is no clear duplicate set here, but this is the best umbrella representative among the unrelated issues.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports and regressions. The clearest duplicate-looking item is the Qwen3.5 MoE `from_pretrained` error pair, with 45310 as the better canonical issue and 45478 as a later near-duplicate.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45310 is the earliest and cleanest representative of the near-identical Qwen3.5 MoE `from_pretrained` failure; issue:45478 repeats the same title/symptom later.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best issue representative because it is broad, clearly framed, and captures the main v5 compatibility regression theme better than the narrower model- or component-specific reports.", + "best_issue_reason": "issue:45310 is the strongest global representative here because it has a clear concrete regression, a close duplicate, and is more coherent than the many unrelated issues in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:45593", - "right": "issue:47068", - "accept": false, - "reason": "Both touch training/loss code and deepspeed, but one is about auxiliary losses in D-FINE and the other is a CUDA-tensor conditional causing host sync; different bugs." - }, { "left": "issue:44188", "right": "issue:46421", "accept": false, - "reason": "Both mention compile/cache/performance behavior, but one is attention-kernel divergence under torch.compile and the other is lazy cache init causing recompilation; not the same fix." + "reason": "Different bugs: one is attention-kernel divergence under `torch.compile`, the other is cache lazy-init causing recompilation in chunked prefill." }, { "left": "issue:43701", "right": "issue:44749", "accept": false, - "reason": "Checkpoint key mismatch and slower filtering after upgrade are unrelated failure modes." + "reason": "Unrelated symptoms: checkpoint key mismatch vs a preprocessing slowdown after upgrading Transformers." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "One is a slowdown after upgrading transformers; the other is deepspeed causing optimizer.step() to run twice." + "reason": "One is a performance regression; the other is a deepspeed optimizer step-counting bug." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Both are save_pretrained-related, but one is tokenizer metadata mismatch and the other is missing processor files; different code paths." + "reason": "Both are save-related, but they affect different APIs and failures: tokenizer config mismatch vs processor files missing." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d and MaskGenerationPipeline partial-batch dropping are unrelated." + "reason": "Different models and failure modes: SAM3 decoder placeholder Conv2d vs MaskGenerationPipeline dropping a final partial batch." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Custom model initialization breakage and ESM2-specific breakage are separate regressions." + "reason": "Broad v5 custom init breakage is not the same as the ESM2-specific breakage described in 44162." }, { "left": "issue:43475", "right": "issue:46821", "accept": false, - "reason": "Missing fpn_position_embeddings on a SAM3 vision output object is unrelated to TimesFM window_size attribute errors." + "reason": "Different code paths and attributes: SAM3 vision encoder output missing a field vs TimesFM `window_size` attribute access." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "One is a missing field on a model output object; the other is trust_remote_code returning incomplete AutoConfig fields." + "reason": "SAM3 missing `fpn_position_embeddings` is unrelated to `trust_remote_code=True` dropping fields in `AutoConfig`." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Missing processor save files and duplicate optimizer steps under deepspeed are different problems." + "reason": "Different APIs and root causes: processor serialization vs deepspeed optimizer behavior." }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Phi-4 multimodal load failure and a DeepSpeed ZeRO3 deque error are unrelated." + "reason": "Different subsystems and bugs: Phi-4 multimodal loading vs DeepSpeed ZeRO3 deque underflow." }, { "left": "issue:39692", "right": "issue:46726", "accept": false, - "reason": "SigLIP2 doc-example errors and AutoProcessor backend-missing messaging are different issues; one is documentation/example correctness, the other is runtime error handling." + "reason": "Docs-example errors are not the same as missing-vision-backend handling in AutoProcessor." }, { "left": "issue:43784", "right": "issue:46399", "accept": false, - "reason": "Sentence-transformers import NameError and Qwen3.5 kernel registration ValueError do not share the same underlying bug." + "reason": "Import-time `nn` NameError in sentence-transformers is unrelated to Qwen3.5 kernel registration failure." }, { - "left": "issue:46463", - "right": "issue:46877", + "left": "issue:44373", + "right": "issue:46851", "accept": false, - "reason": "Incorrect T5 decoder validation error text and auto_docstring stdout logging are unrelated." + "reason": "A wrong docstring for `position_ids` is not the same as a runtime variable-length input bug." }, { - "left": "issue:44373", - "right": "issue:46851", + "left": "issue:46463", + "right": "issue:46877", "accept": false, - "reason": "A wrong docstring for position_ids is not the same as a runtime failure on variable-length inputs using position_ids." + "reason": "Incorrect validation error text in T5 is unrelated to docstring lint output being printed at import time." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "PyTorch-missing ImportError and Sentence Transformers hanging while loading are different import/runtime issues." + "reason": "Missing PyTorch import error and a loading hang are different failure modes." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Gemma eos_token_id config mismatch and AttentionInterface.register behavior change are unrelated." - }, - { - "left": "issue:45561", - "right": "issue:47217", - "accept": false, - "reason": "A pytest-xdist race on captured_info.txt is unrelated to vision-language prompt placeholder handling." + "reason": "Gemma EOS config mismatch does not match the behavior change in `AttentionInterface.register`." }, { "left": "issue:43723", "right": "issue:46326", "accept": false, - "reason": "Tokenizer loading in v5 and continue_final_message for mistral_common tokenizers are different tokenizer regressions." + "reason": "Tokenizer loading in v5 and mistral_common final-message handling are separate code paths." }, { - "left": "issue:44188", - "right": "issue:46851", + "left": "issue:45561", + "right": "issue:47217", "accept": false, - "reason": "Attention-kernel divergence under torch.compile and Qwen3.5 variable-length position_ids failures are distinct code paths." + "reason": "A pytest-xdist file race is unrelated to VL models failing on image placeholders in user text." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "Generic bug report versus gemma-4 zero3 from_pretrained is far too broad to be the same bug." + "reason": "`Bug report` is too generic to be a duplicate of a specific Zero3 `from_pretrained` issue." }, { - "left": "issue:45375", - "right": "issue:45478", + "left": "issue:44188", + "right": "issue:46851", "accept": false, - "reason": "Both are Qwen3.5 MoE related, but one is a missing config field silently dropped by @strict and the other is a from_pretrained error; not enough to treat as the same concrete fix." + "reason": "Both mention generation-related behavior, but one is about compiled attention divergence and the other about Qwen3.5 variable-length `position_ids`." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper word timestamp crash on trailing replacement character and Kimi tokenizer codec-handling regression are different tokenizer/decoding bugs." + "reason": "Whisper timestamp decoding crash and Kimi tokenizer regression are different tokenization/decoding bugs." }, { "left": "issue:46877", "right": "issue:47000", "accept": false, - "reason": "Doc-lint diagnostics printing at import time and RTDetrLoss cost-matrix infeasibility are unrelated." + "reason": "Doc-lint stdout noise is unrelated to RTDetrLoss cost-matrix infeasibility." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe compatibility and processor save_pretrained missing files are different issues, even though both touch multimodal tooling." + "reason": "Qwen3-VL-Moe loading problems are not the same as processor save-pretrained file omissions." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "GenerationMixin cache update after sync_gpus and chunked_prefill cache-missing behavior are related to generation internals but not the same concrete bug." + "reason": "Both concern generation internals, but one is a sync-gpus kwargs update bug and the other is a missing cache in chunked prefill." }, { - "left": "issue:45310", - "right": "issue:45375", + "left": "issue:45375", + "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5 MoE, but one is a generic from_pretrained error and the other is a specific missing config field; likely related area, not same bug." + "reason": "These are related in model family, but one is a missing config field silently dropped and the other is a broader `from_pretrained` error report." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "AutoImageProcessor import failure and a non-existent image_utils import are different import errors." + "reason": "Different import failures: missing `AutoImageProcessor` export vs a non-existent import from `image_utils`." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "A Qwen3.5 MoE loading error is not the same as a missing `deepstack_visual_indexes` config field." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "GenerationMixin import failure and AutoImageProcessor import failure are separate missing-symbol issues." + "reason": "Both are import errors, but for different symbols and modules (`GenerationMixin` vs `AutoImageProcessor`)." + }, + { + "left": "issue:44661", + "right": "issue:47059", + "accept": false, + "reason": "Model-mapping code failure and CLI completion breakage in huggingface_hub are unrelated." } ] }, @@ -15154,7 +15177,6 @@ "issue:44351", "issue:44496", "issue:44568", - "issue:44661", "issue:44743", "issue:44792", "issue:44843", @@ -15181,7 +15203,6 @@ "issue:47085" ], "soft_pairs": [ - "issue:44661|issue:47059", "issue:43873|issue:44945", "issue:42371|issue:45092", "issue:45081|issue:45103", @@ -15190,15 +15211,15 @@ "issue:44857|issue:46877", "issue:46202|issue:46701", "issue:45593|issue:46877", - "issue:44038|issue:45310", "issue:43232|issue:44792", + "issue:44038|issue:45310", "issue:41628|issue:44351", "issue:42175|issue:44933", "issue:44222|issue:45310", "issue:44945|issue:45753", "issue:45561|issue:47059", - "issue:45405|issue:46877", "issue:30333|issue:33290", + "issue:45405|issue:46877", "issue:44743|issue:47068", "issue:42898|issue:44568", "issue:42831|issue:44222", @@ -15213,229 +15234,223 @@ "issue:42673|issue:43704", "issue:44186|issue:47085" ], - "prepared_review_unit_hash": "be1cd401590cb822da9fec212919dbcd7f284589c2fcea71192a611e7623d683", + "prepared_review_unit_hash": "40f2a97d79900cf27b0aec6a94a951f894a5c359474de963ade6b2d2db0fbd65", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36970, - "estimated_input_tokens": 9243, - "estimated_eval_tokens": 18742 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36097, + "estimated_input_tokens": 9025, + "estimated_eval_tokens": 18306 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "There is one strong duplicate pair: issue:42673 and issue:43704 share the exact title and describe the same Qwen3ForCausalLM VRAM leak. The rest of the soft pairs are only broadly similar and should stay separate.", - "confidence": 0.97, - "canonical_issue_reason": "issue:42673 is the better canonical report because it has the same title as issue:43704, but a much fuller body and far more discussion.", + "summary": "This cluster is mostly a loose set of unrelated issues. Only one soft pair looks like the same underlying bug: the duplicated Qwen3ForCausalLM VRAM leak report. The rest are too semantically different to merge as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "issue:42673 is the earlier, more detailed report of the same Qwen3ForCausalLM VRAM leak seen again in issue:43704, so it is the better canonical issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42673 is the best issue to keep as the main report: same bug as issue:43704, with the fuller description and more discussion.", + "best_issue_reason": "issue:42673 is the best global issue candidate because it is the clearest true duplicate pair in the cluster and represents a concrete, repeated bug report.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44661", - "right": "issue:47059", - "accept": false, - "reason": "Tokenizer mapping extension vs CLI completion breakage from a hub API change are different bugs." - }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to incorrect outputs under pipeline parallelism." + "reason": "Both involve model/runtime behavior, but one is quantization offloading and the other is pipeline-parallel output correctness; different bugs and code paths." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 settings guidance and InternVL2 meta-init incompatibility are separate issues." + "reason": "TF32 settings and old InternVL2 meta-init incompatibility are unrelated failures." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Mistral regex/tokenizer loading and auto_docstring future-annotations handling are different code paths." + "reason": "Mistral regex loading crash vs auto_docstring future-annotations crash; different subsystems and root causes." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Embedding dtype drift and DeepSpeed ZeRO3 deque error are unrelated failures." + "reason": "Embedding dtype drift in evaluate and a DeepSpeed ZeRO3 deque IndexError are not the same issue." }, { "left": "issue:45406", "right": "issue:46202", "accept": false, - "reason": "Gemma processor attribute crash and Gemma token roundtrip regression are different bugs." + "reason": "Both mention Gemma, but one is a missing processor attribute and the other is a tokenizer roundtrip failure; not the same bug." }, { "left": "issue:44857", "right": "issue:46877", "accept": false, - "reason": "AMP/CUDA loss crash and import-time docstring stdout noise are unrelated." + "reason": "A CUDA float16 loss crash is unrelated to auto_docstring printing diagnostics at import time." }, { "left": "issue:46202", "right": "issue:46701", "accept": false, - "reason": "Both are Gemma-related, but they describe different pipeline failures." + "reason": "Gemma tokenizer roundtrip failure and DiffusionGemma image-text pipeline failure are different concrete problems." }, { "left": "issue:45593", "right": "issue:46877", "accept": false, - "reason": "D-FINE auxiliary-loss logic and auto_docstring logging behavior are separate subsystems." + "reason": "D-FINE auxiliary-loss behavior and auto_docstring stdout logging are unrelated." }, { - "left": "issue:44038", - "right": "issue:45310", + "left": "issue:43232", + "right": "issue:44792", "accept": false, - "reason": "Both involve Qwen MoE models, but they are different model variants and load failures." + "reason": "Generation kwargs handling after sync_gpus is unrelated to a janus test failure." }, { - "left": "issue:43232", - "right": "issue:44792", + "left": "issue:44038", + "right": "issue:45310", "accept": false, - "reason": "Generation kwargs update bug and janus image-generation test failure are not the same bug." + "reason": "Both are Qwen/MoE-related, but they describe different model variants and different from_pretrained failures." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they concern different exported symbols and fixes." + "reason": "Two separate import errors for different symbols; no shared underlying bug." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Package extra/backend selection and a missing image_utils import are unrelated." + "reason": "Missing TensorFlow in an extra dependency install and a nonexistent image_utils import are different packaging/API issues." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "FP8 save_pretrained behavior and Qwen3.5 MoE from_pretrained loading are different bugs." + "reason": "FP8 save_pretrained on MoE and Qwen3.5 from_pretrained failure are not the same code-path problem." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Pipeline-parallel output correctness and a Qwen3_5 image infinite loop are different failures." + "reason": "Pipeline-parallel output correctness and Qwen3_5 image-specific infinite loop are unrelated." }, { "left": "issue:45561", "right": "issue:47059", "accept": false, - "reason": "pytest-xdist file race and huggingface_hub CLI incompatibility are unrelated." + "reason": "A pytest-xdist file race and a huggingface_hub CLI compatibility error are unrelated." }, { - "left": "issue:45405", - "right": "issue:46877", + "left": "issue:30333", + "right": "issue:33290", "accept": false, - "reason": "Dependency version bump and import-time docstring stdout output are unrelated." + "reason": "MLFlow reporting on failed training and a DeepSpeed Adafactor OOM are different failures." }, { - "left": "issue:30333", - "right": "issue:33290", + "left": "issue:45405", + "right": "issue:46877", "accept": false, - "reason": "MLflow reporting after failure and Adafactor OOM in deepspeed are distinct problems." + "reason": "A version bump complaint for PEFT and auto_docstring stdout diagnostics are unrelated." }, { "left": "issue:44743", "right": "issue:47068", "accept": false, - "reason": "Recurrent-state reset in modular_qwen3_5 is unrelated to host synchronization in deepspeed loss computation." + "reason": "Qwen3.5 recurrent-state reset logic and a DeepSpeed host-sync warning are different bugs." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Tokenization cleanup behavior and missing BOS/EOS insertion are different regressions." + "reason": "Both are tokenizer regressions, but one is clean_up_tokenization_spaces semantics and the other is add_special_tokens BOS/EOS behavior." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "Generic FineGrainedFP8 accuracy issue and FP8 save_pretrained for MoE are not the same bug." + "reason": "Both mention FP8, but one is accuracy and the other is save_pretrained serialization; not mergeable as one fix." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Missing nn import during sentence-transformers import and missing Qwen2_5_VL export are different issues." + "reason": "Separate import errors for sentence-transformers and Qwen2.5-VL symbols." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "TensorFlow backend extra packaging and leftover dummy PT classes are separate hygiene issues." + "reason": "Dependency backend packaging and leftover dummy class cleanup are different maintenance tasks." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version compatibility and missing image_utils import are unrelated." + "reason": "Python version compatibility and an image_utils import regression are unrelated." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Unrecognized model config and xdist test-file race are unrelated." + "reason": "Unrecognized model/config issue and xdist race condition are not the same bug." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Sentence-transformers loading hang and embedding reinitialization are different bugs." + "reason": "SentenceTransformers loading hang and embedding reinitialization during resize are different model-init problems." }, { "left": "issue:44568", "right": "issue:47085", "accept": false, - "reason": "Tokenizer BOS/EOS regression and tokenizer thread-safety regression are different code paths." + "reason": "Tokenizer add_special_tokens behavior and tokenizer thread-safety are distinct issues." }, { "left": "issue:44843", "right": "issue:47085", "accept": false, - "reason": "HF_HUB_OFFLINE breakage in _patch_mistral_regex is unrelated to tokenizer thread safety." + "reason": "Offline-mode mistral regex patching and tokenizer thread-safety are different root causes." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Different tokenizer regressions: clean_up_tokenization_spaces behavior vs wrong Granite token IDs." + "reason": "Both are tokenizer regressions, but one concerns cleanup space behavior and the other wrong token IDs for Granite models." }, { "left": "issue:42673", "right": "issue:43704", "accept": true, - "reason": "Same title and same underlying Qwen3ForCausalLM VRAM leak across multiple dataloader threads; 42673 is the richer report." + "reason": "Same title, same Qwen3ForCausalLM VRAM leak symptom, and clearly the same underlying bug reported twice." }, { "left": "issue:44186", "right": "issue:47085", "accept": false, - "reason": "LayoutLMv2 NER padding crash and fast tokenizer thread safety are unrelated." + "reason": "LayoutLMv2 tokenizer crashes on NER padding/truncation is unrelated to tokenizers thread-safety." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The duplicate decision for issue:42673 / issue:43704 is well supported by the packet (identical title, and 42673 has more discussion and a much fuller body), and the soft-edge rejects are generally conservative. However, the rationale overstates evidence in a few places, especially by mentioning \u2018more discussion/references\u2019 when the packet does not show any reference signal for either issue. I\u2019d accept only if the reasons are tightened to stay strictly within the provided fields (e.g., \u2018fuller body and more discussion\u2019)." + "accept": true, + "feedback": "The summary is grounded in the packet: issue:42673 and issue:43704 are the only clearly duplicated reports, and the rest of the soft pairs are rejected conservatively with reasons that do not overstate the evidence. No pull-request pairing concerns apply here." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -15510,223 +15525,223 @@ "issue:44060|issue:45125", "issue:45600|issue:47059", "issue:30333|issue:34689", - "issue:43479|issue:44079", + "issue:27957|issue:30333", "issue:44016|issue:46231", + "issue:43479|issue:44079", "issue:42898|issue:44779", - "issue:27957|issue:30333", "issue:42673|issue:47085", - "issue:43479|issue:44869", - "issue:30333|issue:36331", "issue:42898|issue:43295", + "issue:30333|issue:36331", + "issue:43479|issue:44869", "issue:42898|issue:44448", "issue:46077|issue:46123", "issue:22355|issue:28282", "issue:44295|issue:45656", "issue:42222|issue:45200", - "issue:44734|issue:45103", "issue:43874|issue:45092", + "issue:44734|issue:45103", "issue:42796|issue:43723", "issue:44075|issue:46231", "issue:43116|issue:44190", - "issue:45405|issue:45712", - "issue:44222|issue:44297" + "issue:44222|issue:44297", + "issue:45405|issue:45712" ], - "prepared_review_unit_hash": "aa4a5a607203a66aaa6d725eb075514351103f0b952c3afbcbe1e5834a866bca", + "prepared_review_unit_hash": "f64e5b0d62efa82dccf73c2d5d98e3fcc0520f30a65b0e113e4d9e55ae1cdd25", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35680, - "estimated_input_tokens": 8920, - "estimated_eval_tokens": 18096 + "serialized_chars": 35685, + "estimated_input_tokens": 8922, + "estimated_eval_tokens": 18100 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are unrelated issue reports that only share broad subsystem themes (tokenizers, multimodal models, version regressions, docs). None of the soft-similarity pairs looks like the same underlying bug/change, so all soft edges should be rejected.", - "confidence": 0.31, - "canonical_issue_reason": "issue:43723 is the clearest broad, user-facing v5 regression here (tokenizer loading via AutoTokenizer.from_pretrained) and is the best single anchor if one must be chosen, though the cluster does not form a true duplicate set.", + "summary": "The cluster is broad and mostly heterogeneous, with only loose surface similarity around Transformers v5 regressions, tokenizers, and model loading. None of the soft-edge pairs look like the same concrete bug/change, so I would not merge any of them.", + "confidence": 0.76, + "canonical_issue_reason": "issue:43723 is the broadest and most actionable representative of the tokenizer/loading regression subset in this cluster, especially the v5-related failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43723 is the most representative standalone issue: it targets a core API, has a clear regression framing, and is easier to generalize from than the more model-specific reports.", + "best_issue_reason": "issue:43723 is the best single issue anchor for triage because it targets a general AutoTokenizer/v5 loading problem that is closer to several nearby regressions than the more model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43479", "right": "issue:46829", "accept": false, - "reason": "Different bugs: Phi4MultimodalConfig default initialization vs VideoPrism output field misassignment." + "reason": "Different model families and different failure modes: config initialization vs output field mix-up." }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "Different subsystems: processor.save_pretrained missing files vs Qwen3.5 MoE from_pretrained failure." + "reason": "One is processor save_pretrained file omission; the other is a Qwen3.5 MoE from_pretrained error." }, { "left": "issue:42673", "right": "issue:46693", "accept": false, - "reason": "Different failures: Qwen3ForCausalLM VRAM leak in multithreaded loading vs FA2 long-context performance regression." + "reason": "VRAM leak in multi-threaded dataloaders is unrelated to FA2 long-context performance regression." }, { "left": "issue:39692", "right": "issue:46909", "accept": false, - "reason": "Both are broken docs examples, but for different models/scripts and different concrete errors." + "reason": "Both are documentation examples, but they are different examples with different concrete errors." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer-related v5 regressions, but one is loading failure and the other is add_special_tokens behavior for a specific tokenizer." + "reason": "Both involve tokenizers in v5, but one is loading failure and the other is add_special_tokens behavior for a specific model." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Different model-init issues: tied-weights warning vs missing _tp_plan for tensor parallelism." + "reason": "Different Qwen3 variants and different bugs: tied-weights warning vs missing tensor-parallel plan." }, { "left": "issue:45600", "right": "issue:47059", "accept": false, - "reason": "Unrelated: auto_mappings CI break from removed configs vs CLI failure caused by huggingface_hub API changes." + "reason": "auto_mappings config cleanup is unrelated to a CLI completion API break in huggingface_hub." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "Different domains: MLFlow reporting on failed training jobs vs Llama 3.2 Vision model-loading regression." + "reason": "MLflow reporting on failed training jobs is unrelated to a Llama model-loading regression." }, { - "left": "issue:43479", - "right": "issue:44079", + "left": "issue:27957", + "right": "issue:30333", "accept": false, - "reason": "Different code paths: multimodal config defaults vs ModelOutput key assignment semantics." + "reason": "Flash Attention model support and MLflow failure handling are different subsystems and bugs." }, { "left": "issue:44016", "right": "issue:46231", "accept": false, - "reason": "Different documentation problems: notebook syntax error vs wrong parameter name in docs." + "reason": "Notebook syntax error and docs parameter-name mismatch are not the same underlying issue." }, { - "left": "issue:42898", - "right": "issue:44779", + "left": "issue:43479", + "right": "issue:44079", "accept": false, - "reason": "Both are tokenizer/v5 regressions, but they affect different models and different behaviors." + "reason": "Config defaulting bug and ModelOutput key assignment bug are distinct code paths." }, { - "left": "issue:27957", - "right": "issue:30333", + "left": "issue:42898", + "right": "issue:44779", "accept": false, - "reason": "Different problems entirely: Flash Attention 2 with XLMRoberta vs MLFlow reporting bug." + "reason": "Both are tokenizer regressions in v5, but they describe different model behavior changes and likely different fixes." }, { "left": "issue:42673", "right": "issue:47085", "accept": false, - "reason": "Both mention threading, but one is a VRAM leak in a model path and the other is fast tokenizer thread safety." + "reason": "VRAM leak in dataloader threads and tokenizer thread-safety regression are different concurrency problems." }, { - "left": "issue:43479", - "right": "issue:44869", + "left": "issue:42898", + "right": "issue:43295", "accept": false, - "reason": "Different multimodal areas: config default initialization vs Whisper timestamp decoding crash." + "reason": "General clean_up_tokenization_spaces change and processor.tokenizer/images regression are not the same bug." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "Different training issues: MLFlow status reporting vs Trainer compute_loss signature mismatch." + "reason": "Failed-job MLflow reporting and CustomTrainer compute_loss signature mismatch are unrelated training issues." }, { - "left": "issue:42898", - "right": "issue:43295", + "left": "issue:43479", + "right": "issue:44869", "accept": false, - "reason": "Both are v5 regressions, but one is tokenization cleanup behavior and the other is processor/tokenizer image-handling regression." + "reason": "Multimodal config defaults are unrelated to Whisper timestamp decoding on trailing replacement characters." }, { "left": "issue:42898", "right": "issue:44448", "accept": false, - "reason": "Both are version-related output changes, but for different models and different tokenizer/serialization paths." + "reason": "Different output changes for different models; not a single concrete code-path problem." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Both involve masks, but one is Gemma4 bidirectional attention and the other is batch-finalization in a generation pipeline." + "reason": "Bidirectional attention mask construction and partial-batch pipeline bookkeeping are unrelated." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Both are import errors, but one is a missing transformers.onnx module and the other is missing PyTorch dependency detection." + "reason": "Importing transformers.onnx and missing PyTorch for AutoModel are separate dependency/import failures." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Different bugs: position_ids buffer access error vs optimizer.step being invoked twice with deepspeed." + "reason": "position_ids buffer read error and deepspeed double optimizer step are unrelated." }, { "left": "issue:42222", "right": "issue:45200", "accept": false, - "reason": "Both are multimodal model bugs, but they concern different models and different config/input assumptions." + "reason": "Broken vitpose model files and Gemma 4 mm_token_type_ids defaults are different model-specific issues." }, { - "left": "issue:44734", - "right": "issue:45103", + "left": "issue:43874", + "right": "issue:45092", "accept": false, - "reason": "Completely unrelated: serving KV-cache tensor indexing crash vs auto-docstring crash with future annotations." + "reason": "Missing image-patch API and remote-code meta-init incompatibility are both multimodal, but not the same bug." }, { - "left": "issue:43874", - "right": "issue:45092", + "left": "issue:44734", + "right": "issue:45103", "accept": false, - "reason": "Different model-specific multimodal initialization issues with no shared concrete code path." + "reason": "Serve KV-cache indexing crash and auto-docstring future-annotations crash are unrelated." }, { "left": "issue:42796", "right": "issue:43723", "accept": false, - "reason": "Both are tokenizer-related, but one is BPE marker handling and the other is AutoTokenizer.from_pretrained loading failure." + "reason": "Tokenizer BPE marker handling and tokenizer loading in v5 are different failures." }, { "left": "issue:44075", "right": "issue:46231", "accept": false, - "reason": "Different maintenance issues: optimizer argument handling vs a documentation parameter-name typo." + "reason": "Optimizer SGD args bug is not a documentation parameter-name typo." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Different example-script failures: multi-label classification output vs local dataset loading for image classification." + "reason": "Multi-label classification output bug and local dataset loading failure are separate example-script issues." }, { - "left": "issue:45405", - "right": "issue:45712", + "left": "issue:44222", + "right": "issue:44297", "accept": false, - "reason": "Unrelated: dependency version bump not yet released vs leftover dummy classes leaking into checks." + "reason": "FP8 MoE save_pretrained and tokenizer_class metadata mismatch are different save_pretrained paths." }, { - "left": "issue:44222", - "right": "issue:44297", + "left": "issue:45405", + "right": "issue:45712", "accept": false, - "reason": "Different save_pretrained bugs in separate subsystems: FP8 MoE model saving vs tokenizer_class metadata mismatch." + "reason": "PEFT version pinning and leftover dummy class cleanup are unrelated repository maintenance issues." } ] }, @@ -15808,17 +15823,17 @@ "issue:44222|issue:44623", "issue:45245|issue:46132", "issue:44933|issue:45712", - "issue:44297|issue:45310", "issue:44373|issue:44485", - "issue:43723|issue:44977", + "issue:44297|issue:45310", "issue:43122|issue:46209", + "issue:43723|issue:44977", "issue:41628|issue:42757", "issue:43232|issue:47276", "issue:43504|issue:47059", "issue:46614|issue:46932", "issue:45071|issue:46082", - "issue:43475|issue:46305", "issue:43232|issue:45834", + "issue:43475|issue:46305", "issue:44222|issue:44484", "issue:30990|issue:38175", "issue:44295|issue:44623", @@ -15826,219 +15841,219 @@ "issue:44263|issue:45092", "issue:44327|issue:47059", "issue:46369|issue:46909", - "issue:44190|issue:46082", "issue:43232|issue:46424", - "issue:42907|issue:46050", - "issue:44263|issue:44373" + "issue:44263|issue:44373", + "issue:44190|issue:46082", + "issue:42907|issue:46050" ], - "prepared_review_unit_hash": "ad1a1d04b6059242ff4b2de6ce8c3e96b13ce474263c89b3adfa9a83f41cb42e", + "prepared_review_unit_hash": "253db94a4441074f64da4fa1138731e4e33a7d4d6781ddd03610623bc50f554f", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36297, - "estimated_input_tokens": 9075, - "estimated_eval_tokens": 18406 + "serialized_chars": 36293, + "estimated_input_tokens": 9074, + "estimated_eval_tokens": 18404 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are distinct Transformers bugs spanning loading, tokenization, generation, quantization, config validation, and model-specific regressions. None of the soft pairs look like the same underlying issue, so there is no clear duplicate canonicalization target.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue fits the cluster well; the items are too diverse and do not converge on one concrete bug.", + "summary": "This cluster is noisy and heterogeneous: the issues share superficial keywords like loading/import/save_pretrained/tokenization, but they point to distinct bugs, models, and code paths. I do not see a safe canonical issue or PR in this set.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No global best issue stands out. If forced, 44623 is the broadest save/load-related report, but it is still too narrow and not representative of the rest of the cluster.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46209", "right": "issue:46431", "accept": false, - "reason": "Different bugs: FP8 inference on B200 vs processor chat-template handling." + "reason": "FP8 inference on B200 vs missing chat template; unrelated failure modes and code paths." }, { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "Different code paths: model output mismatch across versions vs AttentionInterface registration semantics." + "reason": "Different problems: Pegasus output regression vs AttentionInterface.register behavior change." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer config mismatch is unrelated to DeepSpeed causing double optimizer steps." + "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to deepspeed optimizer stepping twice." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FP8 accuracy regression is not the same as missing files in processor.save_pretrained." + "reason": "FineGrainedFP8 accuracy issue is not the same as processor.save_pretrained missing files." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Config default initialization bug is unrelated to non-persistent buffer junk in v5." + "reason": "Phi4MultimodalConfig init bug and non-persistent buffer junk are different bugs in different components." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "FP8 MoE save_pretrained failure is a different serialization issue from processor file saving." + "reason": "Both mention saving, but one is FP8 moe serialization and the other is processor file output." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category-count runtime error is unrelated to AttentionInterface registration behavior." + "reason": "Category limit runtime error is unrelated to AttentionInterface.register semantics." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Missing image_utils import and leftover dummy classes are separate import/repo-cleanup problems." - }, - { - "left": "issue:44297", - "right": "issue:45310", - "accept": false, - "reason": "Tokenizer save metadata mismatch is not the same as Qwen3.5 MoE from_pretrained error." + "reason": "Both are import-related, but they concern different modules and different root causes." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Wrong docstring for position_ids is unrelated to GLM-5 RoPE implementation." + "reason": "Wrong docstring for position_ids is not the GLM-5 RoPE implementation bug." }, { - "left": "issue:43723", - "right": "issue:44977", + "left": "issue:44297", + "right": "issue:45310", "accept": false, - "reason": "AutoTokenizer loading failure and flash-attention generation regression affect different components." + "reason": "Tokenizer save_pretrained mismatch and Qwen3.5 MoE from_pretrained error are distinct issues." }, { "left": "issue:43122", "right": "issue:46209", "accept": false, - "reason": "Tokenization version regression is unrelated to FP8 inference failure on B200." + "reason": "Tokenization differences across versions are unrelated to FP8 inference on B200." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading in v5 is not the same as Qwen3.5 flash-attention generation failure." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols from different dependencies." + "reason": "Both are import errors, but for different symbols and dependency paths." }, { "left": "issue:43232", "right": "issue:47276", "accept": false, - "reason": "Generation state update bug is not the same as a misleading warning message." + "reason": "Sync-gpus generation bug and misleading warning text are related only broadly, not the same concrete bug." }, { "left": "issue:43504", "right": "issue:47059", "accept": false, - "reason": "BEiT legacy-field loading failure is unrelated to huggingface_hub CLI completion breakage." + "reason": "Legacy field loading in Beit is unrelated to the hub CLI completion attribute regression." }, { "left": "issue:46614", "right": "issue:46932", "accept": false, - "reason": "Stale phonemizer backend state is unrelated to Gemma4 dtype casting for quantized embedders." + "reason": "Stale phonemizer backend state is unrelated to Gemma4 dtype casting in quantized embedders." }, { "left": "issue:45071", "right": "issue:46082", "accept": false, - "reason": "PretrainedConfig type-checking regression is different from LlamaConfig head_dim validation." + "reason": "Type-checking regression and LlamaConfig head_dim validation are different problems." }, { - "left": "issue:43475", - "right": "issue:46305", + "left": "issue:43232", + "right": "issue:45834", "accept": false, - "reason": "SAM3 missing attribute and Moonshine off-by-one are different model-specific defects." + "reason": "Generation kwargs handling and Kosmos2.5 OCR index error are unrelated." }, { - "left": "issue:43232", - "right": "issue:46424", + "left": "issue:43475", + "right": "issue:46305", "accept": false, - "reason": "Generation sync_gpus handling is unrelated to cache max_length semantics." + "reason": "SAM3 missing attribute and MoonshineStreaming off-by-one are different model/runtime bugs." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "FP8 save_pretrained bug is not the same as the default max_shard_size question." + "reason": "FP8 save_pretrained moe bug is unrelated to the 50GB max_shard_size question." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "SentenceTransformers loading hang and zero probabilities in SigLIP2 are different model issues." + "reason": "Sentence Transformers loading stall and SigLIP zero probabilities are unrelated model-specific issues." }, { "left": "issue:44295", "right": "issue:44623", "accept": false, - "reason": "Reading a registered buffer is unrelated to missing processor save files." + "reason": "Reading position_ids after buffer registration is not the same as missing processor save files." }, { "left": "issue:44360", "right": "issue:46762", "accept": false, - "reason": "Missing ReLU in the DSA indexer is not the same as numerical differences in Minimax MSA attention." + "reason": "DSA indexer lacking ReLU and Minimax MSA numerical differences are different implementation issues." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "torch.split return handling in GlmMoeDsaIndexer is unrelated to remote-code checkpoint meta-init incompatibility." + "reason": "torch.split return values in GlmMoeDsaIndexer is unrelated to InternVL2 meta-init incompatibility." }, { "left": "issue:44327", "right": "issue:47059", "accept": false, - "reason": "QA decode_spans top-k failure is unrelated to the CLI completion attribute error." + "reason": "QA decode_spans top_k bounds error and hub CLI completion error are unrelated." }, { "left": "issue:46369", "right": "issue:46909", "accept": false, - "reason": "SAM3-LiteText memory overflow is unrelated to a broken documentation example." + "reason": "SAM3-LiteText memory overflow is not the broken DIA documentation example." }, { - "left": "issue:44190", - "right": "issue:46082", + "left": "issue:43232", + "right": "issue:46424", "accept": false, - "reason": "Local dataset loading in the training script is unrelated to LlamaConfig head_dim validation." + "reason": "Generation sync_gpus handling and cache max_cache_length behavior are different bugs." }, { - "left": "issue:43232", - "right": "issue:46424", + "left": "issue:44263", + "right": "issue:44373", "accept": false, - "reason": "These concern different generation/cache behaviors, not the same underlying bug." + "reason": "GlmMoeDsaIndexer torch.split behavior and a docstring fix for position_ids do not match." }, { - "left": "issue:42907", - "right": "issue:46050", + "left": "issue:44190", + "right": "issue:46082", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and sinq round-trip failure are both save/load related but are different models and failure modes." + "reason": "Local dataset loading script issue is unrelated to LlamaConfig head_dim validation." }, { - "left": "issue:44263", - "right": "issue:44373", + "left": "issue:42907", + "right": "issue:46050", "accept": false, - "reason": "A torch.split runtime issue is unrelated to a wrong docstring for position_ids." + "reason": "Both involve save/load around quantization, but they are different models and failure modes, so not a duplicate pair." } ] }, @@ -16115,14 +16130,14 @@ "issue:45092|issue:45357", "issue:44279|issue:46328", "issue:43986|issue:46726", + "issue:22355|issue:27957", "issue:43408|issue:46402", "issue:46305|issue:46655", - "issue:22355|issue:27957", "issue:45381|issue:46305", "issue:42831|issue:45310", "issue:41628|issue:43824", - "issue:44483|issue:45406", "issue:44263|issue:44360", + "issue:44483|issue:45406", "issue:22355|issue:28218", "issue:44964|issue:47059", "issue:42831|issue:44038", @@ -16130,11 +16145,11 @@ "issue:44829|issue:46002", "issue:44188|issue:44373", "issue:43784|issue:45600", - "issue:39401|issue:44568", "issue:42796|issue:43122", + "issue:39401|issue:44568", "issue:43576|issue:43784", - "issue:43644|issue:46305", "issue:43986|issue:47217", + "issue:43644|issue:46305", "issue:42831|issue:45753", "issue:43756|issue:44060", "issue:43408|issue:46369", @@ -16143,214 +16158,214 @@ "issue:45381|issue:46909", "issue:42886|issue:42994" ], - "prepared_review_unit_hash": "188e36ca2ac30aa9f2c3a346529943f17829260d777e53bc19c68eeecaaa272b", + "prepared_review_unit_hash": "d5be645108d01bdab29aa085ce3cdc089e40b1b7eb6be1616bd7ddc663e86032", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36445, - "estimated_input_tokens": 9112, - "estimated_eval_tokens": 18480 + "serialized_chars": 36442, + "estimated_input_tokens": 9111, + "estimated_eval_tokens": 18478 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a noisy similarity cluster: many items are superficially related to Transformers v5 regressions, tokenizers, or multimodal models, but they describe different code paths and failure modes. I would not merge any of the soft-edge pairs as duplicates.", - "confidence": 0.31, - "canonical_issue_reason": "issue:43122 is the broadest and most representative issue here: a version-wide tokenization regression with concrete repro, while many nearby items are narrower model-specific or API-specific bugs.", + "summary": "No clear duplicate cluster emerges here: the soft pairs mostly share only broad subsystem labels (tokenizers, multimodal loading, sam3, DSA, etc.) but describe different concrete failures. I would reject all soft edges and avoid picking a single canonical issue/PR for the set.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43122 is the best standalone issue to anchor the cluster because it is the most general, clearly phrased, and closest to the common tokenizer/regression theme. It is not a perfect duplicate target for the whole cluster, but it is the strongest representative issue.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen/vision-related regressions, but one is remote-code/meta-init incompatibility and the other is incorrect visual encoder keys on save_pretrained. Different bugs and code paths." + "reason": "Both involve model-loading behavior, but one is remote-code/meta-init incompatibility and the other is incorrect visual encoder keys on save_pretrained; different bugs and code paths." }, { "left": "issue:44279", "right": "issue:46328", "accept": false, - "reason": "Completely different topics: a vague dependency issue vs a SequenceFeatureExtractor.pad performance problem." + "reason": "Generic dependency complaint vs a SequenceFeatureExtractor.pad performance issue; no shared failure mode." }, { "left": "issue:43986", "right": "issue:46726", "accept": false, - "reason": "Both involve missing vision backends in AutoProcessor flows, but one is a video-model crash without torchvision and the other is an unrecognized image processor error. Too different to be the same bug." + "reason": "Both mention missing vision backends in AutoProcessor, but one is a crash for video models without torchvision and the other is an unrecognized-image-processor error; not the same concrete bug." + }, + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "Importing transformers.onnx is unrelated to XLMRoberta Flash Attention 2 behavior." }, { "left": "issue:43408", "right": "issue:46402", "accept": false, - "reason": "Both mention SAM3, but one is a model-type warning/misconfiguration and the other is missing text encoder weight loading. Different underlying failures." + "reason": "Same sam3 family, but one is a model-type warning and the other is missing text encoder weights; different problems." }, { "left": "issue:46305", "right": "issue:46655", "accept": false, - "reason": "Both are off-by-one-ish processing bugs, but they affect different models and different processors/masks. Not the same underlying issue." - }, - { - "left": "issue:22355", - "right": "issue:27957", - "accept": false, - "reason": "No real overlap: an import error for transformers.onnx versus an XLMRoberta Flash Attention 2 issue." + "reason": "Both are off-by-one style issues in different video/streaming components and different models; insufficient evidence of one underlying bug." }, { "left": "issue:45381", "right": "issue:46305", "accept": false, - "reason": "Different model families and different bugs: Qwen2.5-VL vision_position_ids versus MoonshineStreaming sliding-window masking." + "reason": "Different models and different position/padding logic; only a broad video-processing similarity." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "A FineGrainedFP8 accuracy issue is not the same as a Qwen3.5 MoE from_pretrained error, even if both are regressions." + "reason": "FineGrainedFP8 accuracy regression and Qwen3.5 MoE from_pretrained error are unrelated." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but for unrelated symbols and likely unrelated export paths." + "reason": "Both are import errors, but for different symbols/models with no shared root cause shown." }, { - "left": "issue:44483", - "right": "issue:45406", + "left": "issue:44263", + "right": "issue:44360", "accept": false, - "reason": "Different subsystems: chat/completions request handling versus Gemma4Processor attribute initialization." + "reason": "Both reference the DSA indexer, but one is about torch.split outputs and the other about a missing ReLU; not the same bug." }, { - "left": "issue:44263", - "right": "issue:44360", + "left": "issue:44483", + "right": "issue:45406", "accept": false, - "reason": "Both concern a DSA indexer, but one is about torch.split return handling and the other is a missing ReLU. Related area, different bugs." + "reason": "Chat/completions request handling and Gemma4Processor _tokenizer attribute crash are different code paths." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "Importing transformers.onnx and tokenizer extra spacing after added tokens are unrelated problems." + "reason": "ONNX import failure and tokenizer extra-space behavior are unrelated." }, { "left": "issue:44964", "right": "issue:47059", "accept": false, - "reason": "Different layers entirely: a multimodal model load failure versus a huggingface_hub CLI test failure." + "reason": "Model-loading failure and huggingface_hub CLI completion regression are unrelated." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "Both are Qwen/MoE-related regressions, but the concrete failures differ: FP8 accuracy versus a from_pretrained error." + "reason": "FineGrainedFP8 accuracy issue does not match Qwen3-VL-Moe loading bug." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "Different bugs: missing rms_norm_eps in MLA layernorm versus remote-code checkpoints breaking under meta initialization." + "reason": "A layernorm epsilon precision mismatch is not the same as remote-code checkpoint meta-initialization incompatibility." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "Both are training stability/NaN issues, but they affect different mechanisms and code paths; not plausibly one merged fix." + "reason": "Training degeneration with flash_attention_3 and NaNs after checkout/init are different failure modes." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Attention-kernel divergence under torch.compile is unrelated to a docstring mistake for position_ids." + "reason": "torch.compile kernel branching bug and docstring typo are unrelated." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "Both are integration/import regressions, but one is a missing nn symbol when importing sentence-transformers and the other is stale auto_mappings references. Different failures." + "reason": "Import-time NameError in sentence-transformers differs from stale auto_mappings references and CI breakage." }, { - "left": "issue:39401", - "right": "issue:44568", + "left": "issue:42796", + "right": "issue:43122", "accept": false, - "reason": "Both are tokenizer behavior regressions, but they concern different models and different symptoms: offset_mapping versus special tokens not added." + "reason": "Both are tokenizer regressions, but one is a specific BPE-marker handling bug and the other is a broad version-to-version tokenization change; not enough to treat as the same issue." }, { - "left": "issue:42796", - "right": "issue:43122", + "left": "issue:39401", + "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer regressions, but one is a BPE-marker handling issue for Ministral and the other is a broader tokenization change across versions. Not the same concrete bug." + "reason": "Wrong offset_mapping and missing BOS/EOS insertion are different tokenizer behaviors." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "A broken CLI command and a NameError during import are unrelated." + "reason": "Broken transformers env CLI and import-time NameError are unrelated." }, { - "left": "issue:43644", - "right": "issue:46305", + "left": "issue:43986", + "right": "issue:47217", "accept": false, - "reason": "Non-persistent buffer initialization junk and a sliding-window mask off-by-one are unrelated issues." + "reason": "Both involve multimodal model input handling, but the former is missing torchvision during AutoProcessor loading and the latter is image-placeholder handling in user text; different bugs." }, { - "left": "issue:43986", - "right": "issue:47217", + "left": "issue:43644", + "right": "issue:46305", "accept": false, - "reason": "Both are VL/processor-related, but one is a missing torchvision crash and the other is image placeholder handling in user text. Different code paths." + "reason": "Non-persistent buffer junk and MoonshineStreaming mask off-by-one are unrelated." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "A numerical accuracy issue in FineGrainedFP8 is not the same as a model-specific infinite loop on a specific image." + "reason": "FP8 accuracy regression and Qwen3_5 infinite-loop-on-image bug do not share a concrete code path." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Different model internals: RoPE layer dropping versus incorrect tied-weights warning." + "reason": "RoPE-layer count bug and tied-weights warning bug are separate model-definition issues." }, { "left": "issue:43408", "right": "issue:46369", "accept": false, - "reason": "A warning about sam3_video/sam3_tracker mismatch is not the same as a memory overflow in sam3-litetext." + "reason": "Both are sam3-related, but one is a warning mismatch and the other is memory overflow; different failures." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "One is offline cache loading failure, the other is a wrong tokenizer decoder type in v5. Different tokenizer problems." + "reason": "Offline cache loading failure and wrong tokenizer decoder type are distinct tokenizer problems." }, { "left": "issue:43540", "right": "issue:46909", "accept": false, - "reason": "A Qwen3OmniMoe video-input ValueError and a broken DIA docs example are unrelated." + "reason": "Qwen3OmniMoe video processing ValueError and a broken DIA documentation example are unrelated." }, { "left": "issue:45381", "right": "issue:46909", "accept": false, - "reason": "A Qwen2.5-VL video position bug is unrelated to a documentation example issue." + "reason": "Qwen2.5-VL video position IDs bug and a documentation example issue are unrelated." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Tokenizer cache loading and quantized model saving are different subsystems and failures." + "reason": "Tokenizer cache/offline loading and quantized model saving are unrelated." } ] }, @@ -16424,9 +16439,9 @@ "issue:47059" ], "soft_pairs": [ - "issue:27957|issue:30990", - "issue:42757|issue:45278", "issue:39379|issue:46172", + "issue:42757|issue:45278", + "issue:27957|issue:30990", "issue:43646|issue:46002", "issue:30990|issue:33357", "issue:43122|issue:46326", @@ -16450,221 +16465,233 @@ "issue:44038|issue:44297", "issue:43824|issue:44351", "issue:44038|issue:44295", - "issue:45216|issue:45478", "issue:44360|issue:44485", + "issue:45216|issue:45478", "issue:43116|issue:45200", "issue:43122|issue:44779", "issue:44038|issue:44222", "issue:46172|issue:46762" ], - "prepared_review_unit_hash": "022d9814bbe7556d846546161dbcba231abffd68cb2466b0df22789b74c45a89", + "prepared_review_unit_hash": "0df8eae8f07b6b699b9b30d79f29b2767e42b4af05c56bd94a94b0619f2535b7", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36861, - "estimated_input_tokens": 9216, - "estimated_eval_tokens": 18688 + "serialized_chars": 36859, + "estimated_input_tokens": 9215, + "estimated_eval_tokens": 18686 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are unrelated Transformer bug reports spanning tokenizers, imports, multimodal models, and training/runtime regressions. I do not see any soft pair that is clearly the same underlying bug/change, so all soft edges are rejected.", - "confidence": 0.74, - "canonical_issue_reason": "No single issue cleanly represents a duplicate cluster here, but issue 44521 is the strongest standalone candidate: it is open, has the most inbound references, and describes a concrete multimodal apply_chat_template failure with clear impact.", + "summary": "The cluster is heterogeneous, with a few loose theme overlaps around import and tokenizer regressions, but no clear duplicate pairs. The best representative issue is the broad import-regression report.", + "confidence": 0.64, + "canonical_issue_reason": "issue:45278 is the broadest and most central issue in the set, capturing the shared import-regression theme more clearly than the narrower model- or tokenizer-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44521 is the best overall issue candidate because it is active, well-discussed, and specific enough to be actionable, whereas most other items are narrower version- or model-specific reports.", + "best_issue_reason": "issue:45278 is the best overall representative because it is recent, broad, and plausibly umbrellas several neighboring import-failure reports without overcommitting to a specific model or API path.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:27957", - "right": "issue:30990", + "left": "issue:39379", + "right": "issue:46172", "accept": false, - "reason": "Different problems: Flash Attention 2 behavior for XLMRoberta vs Sentence Transformers loading hangs." + "reason": "Different multimodal code paths: mllama mask handling vs DeepSeek-V4 tensor transpose logic." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "Both mention imports, but one is a specific missing symbol and the other is a broad upgrade regression with many import failures; not the same bug." + "reason": "Both are import-related, but one is a specific missing symbol and the other is a broad upgrade-related import breakage; not the same bug." }, { - "left": "issue:39379", - "right": "issue:46172", + "left": "issue:27957", + "right": "issue:30990", "accept": false, - "reason": "Different model/code paths: mllama masking logic vs deepseek-v4 dtype/transpose handling." + "reason": "XLM-RoBERTa Flash Attention 2 support and Sentence Transformers loading hang are unrelated failures." }, { "left": "issue:43646", "right": "issue:46002", "accept": false, - "reason": "Custom model initialization regression and NaNs in classification heads are distinct failure modes." + "reason": "Custom model initialization breakage and NaNs in classification heads are different failure modes." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Loading hang for Sentence Transformers is unrelated to a Mac bus error on a community CLIP model." + "reason": "A loading hang and a MacOS bus error are different symptoms with no clear shared code-path." }, { "left": "issue:43122", "right": "issue:46326", "accept": false, - "reason": "Both are tokenizer-related, but one is a general tokenization regression and the other is a specific continue_final_message bug for mistral_common-based tokenizers." + "reason": "Both concern tokenization, but one is a version-to-version tokenization change and the other is a continue_final_message bug for mistral_common tokenizers." }, { "left": "issue:44295", "right": "issue:44297", "accept": false, - "reason": "One is a buffer/position_ids access issue; the other is tokenizer_config save_pretrained metadata mismatch." + "reason": "Reading a registered buffer and tokenizer_class mismatch in saved config are separate save/load issues." }, { "left": "issue:43329", "right": "issue:47059", "accept": false, - "reason": "Multimodal token counting bug is unrelated to a CLI incompatibility with huggingface_hub 1.22.0." + "reason": "Undefined variables in multimodal token counting and a CLI completion attribute error are unrelated." }, { "left": "issue:39401", "right": "issue:44488", "accept": false, - "reason": "Different root causes: Qwen3 offset mapping vs failure to load a specific sleng-bert model." + "reason": "Qwen3 offset mapping is a tokenizer alignment bug; loading cjvt/sleng-bert is a model-loading issue." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Qwen3.5 num_labels propagation and flash_attention_3 training degeneration are separate issues." + "reason": "num_labels propagation and flash_attention_3 training degeneration are unrelated model/config bugs." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and loading Phi-4 multimodal instruct are different model-specific bugs." + "reason": "Saving dequantized Ministral/Devstral models and loading Phi-4 multimodal are different model-specific failures." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Tokenizer offset mapping bug is not the same as multimodal assistant mask generation in apply_chat_template." + "reason": "Wrong offset mappings and zero assistant masks are different tokenizer/template defects." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Python-version incompatibility is unrelated to incorrect LLM output under pipeline parallelism." + "reason": "Python version compatibility and pipeline-parallel output corruption are unrelated." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "A single missing AutoImageProcessor import is not the same underlying problem as a broad 5.5.0 import regression." + "reason": "Both are import problems, but one is a missing AutoImageProcessor symbol and the other is a broad post-upgrade import regression; not the same concrete bug." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "Tokenizer mapping None-handling and AutoProcessor cached_file kwargs forwarding are different API bugs." + "reason": "TOKENIZER_MAPPING_NAMES handling and AutoProcessor kwargs forwarding are distinct code paths." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Dequantized model save failure and pipeline-parallel output corruption are unrelated." + "reason": "Quantized model save failure and incorrect pipeline-parallel output are unrelated." }, { "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "Both touch tokenization, but one is a tokenizer regression across versions and the other is multimodal chat-template masking." + "reason": "General tokenizer output differences and multimodal assistant-mask generation are not the same bug." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "Missing transformers.onnx module import and slow/fast tokenizer inconsistency are unrelated." + "reason": "onnx import failure and slow/fast tokenizer inconsistency are unrelated." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Cached_file kwarg forwarding is unrelated to chat template crashes on tool-call assistant messages." + "reason": "cached_file argument forwarding and chat-template crashes with tool calls are different API paths." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 API guidance and a torch.split return-value issue do not share a bug path." + "reason": "TF32 API guidance and a torch.split return-value bug in a DSA indexer are unrelated." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "Both are SAM3 video-related, but one is a model-type warning and the other is a missing output attribute; not the same concrete defect." + "reason": "Both are SAM3 video-related, but one is a model-type warning and the other is a missing attribute on an encoder output; not clearly the same bug." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Different failures: save_pretrained for dequantized models vs invalid GenerationConfig for GLM-5." + "reason": "Model save/load failure and GLM-5 GenerationConfig validation are unrelated." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Model saving for dequantized Ministrals/Devstrals and gguf_file loading breakage are separate code paths." + "reason": "Dequantized model saving and gguf_file loading failures affect different code paths." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe loading issue is unrelated to tokenizer save_pretrained class metadata mismatch." + "reason": "Qwen3-VL-Moe loading and tokenizer_class mismatch are different problems." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Different missing imports: Qwen2.5-VL model class vs HybridCache symbol." + "reason": "Missing Qwen2_5_VL export and missing HybridCache export are separate import regressions." }, { "left": "issue:44038", "right": "issue:44295", "accept": false, - "reason": "Qwen3-VL-Moe loading and position_ids buffer access are unrelated." + "reason": "Qwen3-VL-Moe model loading and position_ids buffer access are unrelated." }, { - "left": "issue:45216", - "right": "issue:45478", + "left": "issue:44360", + "right": "issue:44485", "accept": false, - "reason": "Both involve Qwen3.5 and version 5.x, but one is a save_pretrained regression and the other is a from_pretrained load failure; not enough evidence they are the same bug." + "reason": "A missing ReLU in the DSA indexer and GLM-5 RoPE implementation are different algorithmic issues." }, { - "left": "issue:44360", - "right": "issue:44485", + "left": "issue:45216", + "right": "issue:45478", "accept": false, - "reason": "DSA indexer activation-function discussion and GLM-5 RoPE implementation are unrelated." + "reason": "Both involve Qwen3.5 and 5.4.0+, but one is save_pretrained checkpoint corruption and the other is a from_pretrained error; too little evidence they are the same bug." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Empty results in the classification example script is unrelated to Gemma 4 mm_token_type_ids defaulting behavior." + "reason": "Example-script multi-label classification output and Gemma 4 mm_token_type_ids defaults are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they concern different models and different failure descriptions; not enough to treat as one duplicate." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Qwen3-VL-Moe loading and FP8 MoE save_pretrained are different features and failure modes." }, { "left": "issue:46172", "right": "issue:46762", "accept": false, - "reason": "Deepseek-v4 transpose/dtype bug and Minimax attention numerical differences are different model implementations and failure modes." + "reason": "DeepSeek-V4 tensor-transpose logic and Minimax attention numerical mismatch are different numerical bugs." } ] }, @@ -16741,8 +16768,8 @@ "issue:42175|issue:44938", "issue:43278|issue:44945", "issue:39401|issue:46614", - "issue:42222|issue:47312", "issue:39379|issue:43232", + "issue:42222|issue:47312", "issue:42757|issue:44351", "issue:36296|issue:43116", "issue:43493|issue:45753", @@ -16752,9 +16779,9 @@ "issue:46231|issue:46877", "issue:43976|issue:45446", "issue:44483|issue:47217", - "issue:45357|issue:45478", - "issue:42898|issue:46614", "issue:42673|issue:45072", + "issue:42898|issue:46614", + "issue:45357|issue:45478", "issue:43611|issue:46002", "issue:44625|issue:46002", "issue:38617|issue:45278", @@ -16763,184 +16790,184 @@ "issue:45083|issue:45446", "issue:44610|issue:46909" ], - "prepared_review_unit_hash": "9afb5d1856040b8fa3bfb8ee9839077c39df736928016f553f77e5e5e63c4d2a", + "prepared_review_unit_hash": "f47be8e0d08d5e6c12d17001569c2077fea3066366996660801c474fcaaf48a8", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 25, - "serialized_chars": 34085, - "estimated_input_tokens": 8522, - "estimated_eval_tokens": 17300 + "serialized_chars": 34081, + "estimated_input_tokens": 8521, + "estimated_eval_tokens": 17298 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is a grab-bag of unrelated issues; the soft-similarity pairs share keywords or subsystem names but do not look like the same underlying bug/change.", - "confidence": 0.89, - "canonical_issue_reason": "No single canonical issue fits this cluster: the items cover many unrelated bugs, feature requests, docs, and environment problems.", + "summary": "This cluster is largely heterogeneous: most soft-similarity pairs are not true duplicates and only share broad subsystem or regression themes. The closest-to-related items are the import/version-regression reports, but even those point to different missing symbols or different failure modes.", + "confidence": 0.2, + "canonical_issue_reason": "issue:45278 is the broadest representative of the import/regression subset in this set, but there is no strong single canonical duplicate because the cluster spans many unrelated bugs.", "canonical_pr_reason": null, - "best_issue_reason": "No issue here is a safe umbrella for the others; the cluster does not form one coherent duplicate target.", + "best_issue_reason": "issue:45278 is the most general and representative issue here; it best matches the recurring pattern of post-upgrade import breakages, while the rest are narrower or unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "Import failure in transformers.onnx vs LayoutLMv3 box/error-message clarity; unrelated code paths." + "reason": "Different problems: missing transformers.onnx module vs LayoutLMv3 box-content validation/error messaging." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "Packaging/backend dependency omission vs Python 3.14 load failure; different root causes." + "reason": "Both mention environment/loading issues, but one is about missing TensorFlow in a torch install and the other is Python 3.14 loading failure; not the same bug." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "BF16-to-FP32 eval dtype regression vs incorrect output under pipeline parallelism; not the same bug." + "reason": "BF16/FP32 dtype drift in evaluation is unrelated to incorrect output under pipeline parallelism." }, { "left": "issue:39401", "right": "issue:46614", "accept": false, - "reason": "Qwen3 tokenizer offset mapping vs Wav2Vec2 phonemizer backend state; different components and symptoms." + "reason": "Qwen3 tokenizer offset mapping and stale phonemizer language backend are different code paths and failure modes." }, { - "left": "issue:42222", - "right": "issue:47312", + "left": "issue:39379", + "right": "issue:43232", "accept": false, - "reason": "VitPose model breakage vs kernels 0.16 incompatibility; no evidence of one concrete shared failure." + "reason": "mllama mask handling vs generation kwargs update after sync_gpus are unrelated defects." }, { - "left": "issue:39379", - "right": "issue:43232", + "left": "issue:42222", + "right": "issue:47312", "accept": false, - "reason": "Both touch generation, but one is mllama mask handling and the other sync_gpus/_update_model_kwargs; different fixes." + "reason": "VitPose model breakage and kernels 0.16 incompatibility are not the same underlying issue." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Missing imports from different packages/symbols; likely version skew, not the same defect." + "reason": "Both are import errors, but they involve different missing symbols and different packages; too broad to treat as one bug." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor-parallel training bug vs example script multi-label empty results; unrelated scenarios." + "reason": "Tensor parallel training bug is unrelated to an example script returning empty multi-label results." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 HF/JAX discrepancy vs Qwen3.5 image-specific infinite loop; unrelated model bugs." + "reason": "SigLIP2 JAX parity discrepancy and Qwen3.5 infinite-loop-on-image are different model-specific bugs." }, { "left": "issue:44995", "right": "issue:46697", "accept": false, - "reason": "Stale indexer-cache crash vs over-padding in an image processor; different code paths." + "reason": "Stale indexer cache crash and image over-padding are unrelated." }, { "left": "issue:44188", "right": "issue:46328", "accept": false, - "reason": "torch.compile attention-kernel divergence vs feature extractor padding performance; not the same issue." + "reason": "Attention-kernel divergence under torch.compile is unrelated to a padding performance issue." }, { "left": "issue:28348", "right": "issue:46640", "accept": false, - "reason": "GPT2 FlashAttention 2 support request vs a vague FA2 issue; too little to treat as one concrete bug." + "reason": "FlashAttention 2 support request for GPT2LMHeadModel is not the same as a separate Flash Attention 2 issue." }, { "left": "issue:46231", "right": "issue:46877", "accept": false, - "reason": "Doc parameter-name mismatch vs import-time stdout diagnostics; unrelated documentation/tooling issues." + "reason": "Documentation parameter-name mismatch and import-time doc-lint stdout spam are different issues." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python-version compatibility problem vs incorrect PyTorch version gate for an import; different failures." + "reason": "Both involve version checks, but one is a Python-version compatibility break and the other is a PyTorch version guard for AuxRequest import." }, { "left": "issue:44483", "right": "issue:47217", "accept": false, - "reason": "API request rejection vs VL placeholder-in-user-text handling; not the same underlying bug." + "reason": "Chat/completions request rejection and image-placeholder handling in VL models are different API/model behaviors." }, { - "left": "issue:45357", - "right": "issue:45478", + "left": "issue:42673", + "right": "issue:45072", "accept": false, - "reason": "save_pretrained visual-key serialization bug vs from_pretrained load error; separate Qwen3.5 regressions." + "reason": "VRAM leak in dataloader threads and bfloat16 dtype mismatches are unrelated." }, { "left": "issue:42898", "right": "issue:46614", "accept": false, - "reason": "Tokenizer cleanup behavior change in v5 vs stale phonemizer language backend; unrelated." + "reason": "Tokenizer cleanup behavior change and stale phonemizer backend state are different tokenizer-related bugs." }, { - "left": "issue:42673", - "right": "issue:45072", + "left": "issue:45357", + "right": "issue:45478", "accept": false, - "reason": "VRAM leak in multi-threaded Qwen3.5 inference vs dtype mismatches in other models; different problems." + "reason": "Both concern Qwen3.5, but one is save_pretrained key serialization and the other is from_pretrained loading failure; not the same concrete defect." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "base_model_prefix loading regression vs NaNs in classification heads; different code paths and symptoms." + "reason": "base_model_prefix loading breakage and NaNs in classification heads are unrelated." }, { "left": "issue:44625", "right": "issue:46002", "accept": false, - "reason": "Qwen3.5 num_labels propagation bug vs classification-head NaNs; not the same change or bug." + "reason": "Qwen3.5 num_labels propagation and NaNs after checkout/init are different bugs." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "One missing import symbol vs broad upgrade-related import errors; too general to call duplicates." + "reason": "The latter is a broad umbrella for many import errors, but this specific missing-symbol ImportError is not the same concrete issue." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "Optimizer argument handling vs trainer train/evaluate lifecycle failure; unrelated behavior." + "reason": "Unused SGD args and trainer.evaluate() failing after trainer.train() are unrelated." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Generic tokenizer decoder-type regression vs Deepseek-specific incorrect outputs; insufficient overlap for duplication." + "reason": "Tokenizer decoder-type regression and Deepseek tokenizer incorrect results are too broad and not clearly the same bug." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "Feature-extractor length helper bug vs PyTorch version-check import bug; unrelated." + "reason": "Qwen3_omni_moe feature-length helper behavior and flex_attention version guard are unrelated." }, { "left": "issue:44610", "right": "issue:46909", "accept": false, - "reason": "Processor output size mismatch vs broken documentation example; different classes of issue." + "reason": "OmDet-Turbo processor sizing bug and a broken documentation example are not the same underlying issue." } ] }, @@ -16984,6 +17011,7 @@ "issue:43874", "issue:43937", "issue:43992", + "issue:44038", "issue:44117", "issue:44291", "issue:44355", @@ -16994,7 +17022,6 @@ "issue:45071", "issue:45081", "issue:45092", - "issue:45362", "issue:45405", "issue:45446", "issue:45478", @@ -17016,7 +17043,7 @@ "soft_pairs": [ "issue:43937|issue:44485", "issue:43116|issue:43493", - "issue:45362|issue:45478", + "issue:44038|issue:45478", "issue:43066|issue:46710", "issue:44355|issue:46231", "issue:42673|issue:42796", @@ -17024,221 +17051,221 @@ "issue:42371|issue:43874", "issue:43116|issue:43825", "issue:43618|issue:46829", - "issue:43992|issue:44704", "issue:43827|issue:45446", - "issue:46328|issue:47116", + "issue:43992|issue:44704", "issue:45405|issue:45446", + "issue:46328|issue:47116", "issue:33453|issue:39401", "issue:43824|issue:46981", - "issue:44291|issue:45071", "issue:39379|issue:43688", + "issue:44291|issue:45071", "issue:44117|issue:45588", "issue:43673|issue:46328", "issue:43064|issue:43673", - "issue:44117|issue:45081", "issue:45092|issue:46762", + "issue:44117|issue:45081", "issue:43937|issue:45563", "issue:44466|issue:46002", "issue:44380|issue:46077", "issue:42222|issue:47000", "issue:39379|issue:46897" ], - "prepared_review_unit_hash": "9e289385c487839c5152e3fe3336a304da77f5494ab4ee4bea14b2f3c1fb0509", + "prepared_review_unit_hash": "47b05f058c35f42463c721ff612f65161b37e2c6d99dd69dd643afd422a5149f", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36148, - "estimated_input_tokens": 9037, - "estimated_eval_tokens": 18330 + "serialized_chars": 36133, + "estimated_input_tokens": 9034, + "estimated_eval_tokens": 18324 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the issues span tokenizer bugs, generation/cache regressions, model-specific vision/attention problems, docs, and training/runtime errors. None of the soft pairs look like the same underlying bug/change, so all are rejected as duplicates.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical issue cleanly covers this set because the reports are unrelated. If forced to anchor the cluster, issue 42371 is the most central/high-traffic report with a concrete user-facing regression, but it does not represent the rest of the items.", + "summary": "This cluster is a loose grab bag of mostly Transformers v5 regressions and model-specific bugs (tokenizers, loading, generation, attention, docs, and a few performance issues). None of the soft pairs look like the same underlying bug or a mergeable PR-level duplicate.", + "confidence": 0.89, + "canonical_issue_reason": "issue:43066 is the broadest core regression here: a v5 tokenizer/decoder loading problem in the main library path, which best matches the cluster\u2019s recurring tokenizer/loading breakages.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 42371 is the strongest anchor candidate: it has the highest discussion activity and a specific, actionable regression. Even so, it is not a good umbrella for the other issues in this cluster.", + "best_issue_reason": "issue:43066 is the clearest representative issue for the cluster\u2019s main theme (v5-era tokenizer/loading regressions) without being overly model-specific.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Same model family, but one is a GenerationConfig validity error and the other is a RoPE implementation discussion; different code paths." + "reason": "Both mention GLM-5, but one is a GenerationConfig validity error and the other is a RoPE implementation discussion; different code paths." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Different subsystems: example script multi-label output vs SigLIP2 implementation discrepancy." + "reason": "Multi-label classification script output and SigLIP2 numerical discrepancy are unrelated bugs in different areas." }, { - "left": "issue:45362", + "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5, but one is chat-time crashing and the other is a from_pretrained MoE loading error." + "reason": "Both are Qwen3-family v5 issues, but one is Qwen3-VL-Moe loading and the other is Qwen3.5-Moe from_pretrained; not the same defect." }, { "left": "issue:43066", "right": "issue:46710", "accept": false, - "reason": "Tokenizer decoder type regression is unrelated to an accuracy drop in DeepSeek-R1 after upgrade." + "reason": "Tokenizer decoder type regression and an accuracy regression after upgrade are distinct symptoms with no clear shared code-path." }, { "left": "issue:44355", "right": "issue:46231", "accept": false, - "reason": "Compiled Python file errors and a docs parameter-name mismatch are not the same bug." + "reason": "Compiled Python file errors and a docs parameter-name mismatch are unrelated." }, { "left": "issue:42673", "right": "issue:42796", "accept": false, - "reason": "VRAM leak in multi-threaded loading and tokenizer BPE-marker handling are distinct problems." + "reason": "VRAM leakage in multithreaded Qwen3ForCausalLM and tokenizer BPE marker handling are different bugs." }, { "left": "issue:46726", "right": "issue:46981", "accept": false, - "reason": "One is missing-vision-backend behavior, the other is repeated HTTP 429 retries in AutoProcessor; different failure modes." + "reason": "AutoProcessor missing-vision-backend messaging and repeated HTTP 429 retries are separate processor/network behaviors." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API settings and a missing image-patch method in GLM46V are unrelated." + "reason": "TF32 settings guidance and a GLM46V image-patch API crash are unrelated." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Example script multi-label classification bug vs an incorrect pipeline() translation error message; different surfaces and fixes." + "reason": "Classification example emptiness and a pipeline translation error message are different example/UI issues." }, { "left": "issue:43618", "right": "issue:46829", "accept": false, - "reason": "Both are output-structure issues, but for different models and different tensors; not the same underlying bug." + "reason": "Both are output-field assignment bugs, but they affect different model classes and different outputs; not the same concrete fix." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() and a flex_attention import version check are unrelated." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing encoder weights on load vs AutoProcessor kwarg propagation; separate loading paths." + "reason": "Missing embed_tokens.weight in UMT5Encoder and cached_file kwargs passthrough are separate loading bugs." }, { - "left": "issue:43827", + "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "Docs still referencing removed pipeline() vs a PyTorch version check bug in flex_attention." + "reason": "PEFT minimum-version bump and PyTorch version-check logic are unrelated maintenance issues." }, { "left": "issue:46328", "right": "issue:47116", "accept": false, - "reason": "SequenceFeatureExtractor.pad inefficiency is not the same as a streaming pipeline regression." - }, - { - "left": "issue:45405", - "right": "issue:45446", - "accept": false, - "reason": "A dependency version bump issue and a PyTorch import/version check bug are different changes." + "reason": "SequenceFeatureExtractor.pad performance waste and streaming pipeline regression do not share a concrete failure mode." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Both tokenizer-related, but one is loading regression and the other is wrong offset_mapping for Qwen3." + "reason": "Both are tokenizer-related, but one is loading regression and the other is wrong offset_mapping for Qwen3; different defects." }, { "left": "issue:43824", "right": "issue:46981", "accept": false, - "reason": "ImportError for a missing model class is unrelated to HTTP 429 retry behavior." + "reason": "ImportError for Qwen2.5-VL and repeated 429 retries during config fetch are unrelated." }, { - "left": "issue:44291", - "right": "issue:45071", + "left": "issue:39379", + "right": "issue:43688", "accept": false, - "reason": "Meta-init argument handling and PretrainedConfig type checking are distinct v5 regressions." + "reason": "Incorrect mllama mask handling and auxiliary-loss normalization in OLMoE/GPT Oss are different model-training bugs." }, { - "left": "issue:39379", - "right": "issue:43688", + "left": "issue:44291", + "right": "issue:45071", "accept": false, - "reason": "Different model families and different math errors: masked-row handling vs auxiliary-loss normalization." + "reason": "Unexpected _is_hf_initialized arg during init_empty_weights and PretrainedConfig type-checking breakage are separate v5 compatibility issues." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping None handling is unrelated to a flash_attention crash on s_aux=None." + "reason": "Tokenizer mapping None handling and flash_attention sink-less-model crash involve different code paths and failure causes." }, { "left": "issue:43673", "right": "issue:46328", "accept": false, - "reason": "Cache-missing generation regression and SequenceFeatureExtractor padding inefficiency do not share a code path." + "reason": "Generation cache missing during chunked_prefill and a pad() performance issue are unrelated." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "Trainer/FSDP2 optimizer-state corruption is separate from GenerationMixin cache behavior." + "reason": "Wrong optimizer states on nonzero ranks in Trainer.train and missing generation cache are distinct distributed-training vs generation bugs." }, { - "left": "issue:44117", - "right": "issue:45081", + "left": "issue:45092", + "right": "issue:46762", "accept": false, - "reason": "Tokenizer mapping assumptions and Mistral regex patch crashes are different tokenizer bugs." + "reason": "Remote-code meta-initialization incompatibility and Minimax attention numerical differences are not the same bug." }, { - "left": "issue:45092", - "right": "issue:46762", + "left": "issue:44117", + "right": "issue:45081", "accept": false, - "reason": "Remote-code meta-init incompatibility and numerical attention differences are unrelated." + "reason": "Tokenizer mapping assumptions and Mistral regex patch crashes are separate tokenizer-loading problems." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "Invalid GenerationConfig and a stale warning in paged generate are different issues." + "reason": "Invalid GenerationConfig and a stale num_return_sequences warning are different generation-config issues." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "Serialization inconsistency for lm_head.weight is unrelated to NaNs in classification heads after checkout/init." + "reason": "Serialization inconsistency for lm_head.weight and NaNs in classification heads are unrelated." }, { "left": "issue:44380", "right": "issue:46077", "accept": false, - "reason": "Attention scaling being ignored and causal-mask construction under bidirectional attention are separate attention bugs." + "reason": "GPT2 attention scaling ignored in SDPA/FlashAttention and Gemma4 causal-mask construction are different attention bugs." }, { "left": "issue:42222", "right": "issue:47000", "accept": false, - "reason": "Vitpose model breakage and RTDetrLoss infeasible cost matrix are unrelated model-specific failures." + "reason": "VitPose loading breakage and RTDetrLoss infeasible cost matrix are unrelated model failures." }, { "left": "issue:39379", "right": "issue:46897", "accept": false, - "reason": "Similar pattern name only; the bugs affect different models and different loss/normalization paths." + "reason": "mllama mask handling and Florence2 double-shift training loss are different bugs, only superficially similar as training regressions." } ] }, @@ -17312,8 +17339,8 @@ "issue:47110" ], "soft_pairs": [ - "issue:43937|issue:44977", "issue:42617|issue:45341", + "issue:43937|issue:44977", "issue:43408|issue:43742", "issue:43450|issue:43986", "issue:42222|issue:43441", @@ -17323,9 +17350,9 @@ "issue:43208|issue:46894", "issue:39401|issue:44779", "issue:44829|issue:45216", - "issue:44410|issue:45310", "issue:44877|issue:46231", "issue:45071|issue:47059", + "issue:44410|issue:45310", "issue:45632|issue:46981", "issue:45341|issue:45561", "issue:45341|issue:46328", @@ -17339,211 +17366,211 @@ "issue:44964|issue:45278", "issue:42371|issue:44261", "issue:44075|issue:45656", - "issue:42371|issue:42831", - "issue:44261|issue:46762" + "issue:44261|issue:46762", + "issue:42371|issue:42831" ], - "prepared_review_unit_hash": "1cd8ad3bb19cf0564236c5c04707b7468a836a2d891fd79930c2428ea85a89da", + "prepared_review_unit_hash": "d47b1e91de0fb01f5d78cd1abcad573ec92ddf1697e01217012ed5e9ec936c44", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36056, - "estimated_input_tokens": 9014, - "estimated_eval_tokens": 18284 + "serialized_chars": 36052, + "estimated_input_tokens": 9013, + "estimated_eval_tokens": 18282 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items span several unrelated regression themes (imports, tokenization, model loading, chat APIs, training, and testing utilities). None of the soft pairs looks like the same concrete bug/change, so I would keep them separate. The broad import-regression report is the best anchor for the cluster.", - "confidence": 0.77, - "canonical_issue_reason": "#45278 is the broadest umbrella report in the set: it describes multiple import failures after upgrade, which makes it the most useful anchor for nearby import-regression reports.", + "summary": "The cluster is heterogeneous: most items are distinct model-, tokenizer-, processor-, training-, or testing-specific bugs, and none of the soft pairs look like true duplicates.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45278 is the broadest umbrella regression report in the set (many import errors after upgrading), and it best serves as a representative anchor for this otherwise mixed cluster.", "canonical_pr_reason": null, - "best_issue_reason": "#45278 is the best overall representative because it is broad, user-facing, and closest to the recurring import-error theme seen in several nearby issues.", + "best_issue_reason": "issue:45278 is the most general, user-facing issue here and the closest thing to a cluster-level representative; the rest are mostly narrower, unrelated bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43937", - "right": "issue:44977", + "left": "issue:42617", + "right": "issue:45341", "accept": false, - "reason": "Both are model-specific generation bugs, but the models and failure modes differ: GenerationConfig validation vs flash-attention generation behavior." + "reason": "Script startup failure vs a small testing_utils bug; different code paths and symptoms." }, { - "left": "issue:42617", - "right": "issue:45341", + "left": "issue:43937", + "right": "issue:44977", "accept": false, - "reason": "Unrelated areas: a parallel script failure versus a small testing_utils bug." + "reason": "Both mention generation, but they are different models and different failures (invalid config vs flash-attention generation bug)." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "Different loading problems and different symptoms; one is a model-type warning mismatch, the other a KeyError on model load." + "reason": "Model-type warning/mismatch vs key error while loading another model; not the same underlying bug." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Both touch video processing, but one is an output-shape bug and the other is a crash when torchvision is missing." + "reason": "Video processor batch-shape bug vs AutoProcessor failure without torchvision; separate processor/loading issues." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "Distinct model families and bugs: VitPose breakage versus Ministral flash-attention failure." + "reason": "Vitpose loading break and Ministral FlashAttention crash are unrelated model-specific problems." }, { "left": "issue:44263", "right": "issue:46762", "accept": false, - "reason": "Different concrete problems: torch.split return handling versus numeric mismatch in attention implementation." + "reason": "Both involve tensor math, but one is a torch.split indexing issue and the other is attention numerical mismatch." }, { "left": "issue:42222", "right": "issue:46752", "accept": false, - "reason": "VitPose model breakage is unrelated to apply_chat_template crashing on an empty list." + "reason": "Model loading failure vs empty chat-template IndexError; no shared concrete code path." }, { "left": "issue:45071", "right": "issue:46512", "accept": false, - "reason": "Both involve config/dtype, but they are different code paths and symptoms; one is type checking, the other stale config.dtype after from_config()." + "reason": "Type-checking regression vs stale dtype metadata after from_config; distinct config/model state bugs." }, { "left": "issue:43208", "right": "issue:46894", "accept": false, - "reason": "Different training bugs in different model families; no shared code-path is evident." + "reason": "xLSTM training blockers vs CohereASR loss double-shift bug; same broad domain, different defect." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they report different failures on different models and do not clearly point to one bug." + "reason": "Tokenizer offset_mapping bug vs Deepseek tokenization regression; related area, but not the same issue." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "Training degeneracy with flash-attention is not the same as a save_pretrained checkpoint regression." - }, - { - "left": "issue:44410", - "right": "issue:45310", - "accept": false, - "reason": "Both are Qwen3-related, but one is missing projections in a model layer while the other is a from_pretrained error for a different model variant." + "reason": "FlashAttention training degeneration and save_pretrained checkpoint regression are different failures." }, { "left": "issue:44877", "right": "issue:46231", "accept": false, - "reason": "A config-loading strictness bug and a documentation parameter-name typo are not the same issue." + "reason": "Strict config loading vs documentation typo in set_training; unrelated." }, { "left": "issue:45071", "right": "issue:47059", "accept": false, - "reason": "PretrainedConfig type checking and a CLI compatibility error with huggingface_hub are unrelated." + "reason": "PretrainedConfig typing regression vs CLI completion attribute error; no common bug." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing qwen3next projections vs Qwen3.5 Moe from_pretrained error; different model internals and failure modes." }, { "left": "issue:45632", "right": "issue:46981", "accept": false, - "reason": "Both mention loading paths, but one is a cache-path collision and the other is HTTP 429 retry behavior across config files." + "reason": "Local cache-path collision vs repeated 429 retry behavior; separate AutoProcessor problems." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Same file area, but the reported problems differ: an unspecified testing_utils bug versus an xdist race on captured_info.txt." + "reason": "Both touch testing utilities, but one is a small bug and the other is a pytest-xdist race condition." }, { "left": "issue:45341", "right": "issue:46328", "accept": false, - "reason": "A testing_utils bug and a SequenceFeatureExtractor.pad performance issue are unrelated." + "reason": "Testing_utils bug vs SequenceFeatureExtractor.pad performance issue; not the same defect." }, { "left": "issue:44704", "right": "issue:47110", "accept": false, - "reason": "Different AutoProcessor/special-tokens bugs with different failure surfaces; not the same concrete defect." + "reason": "Kwargs not forwarded in AutoProcessor.from_pretrained vs special_tokens list/dict error; different API paths." }, { "left": "issue:39401", "right": "issue:43421", "accept": false, - "reason": "Tokenizer offset_mapping regression and runtime post-processor update are related broadly, but not the same code-path bug." + "reason": "Tokenizer offset mapping bug vs runtime post-processor update issue; different functionality." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols and different import paths." + "reason": "Both are import errors, but they concern different missing symbols and likely separate regressions." }, { "left": "issue:43122", "right": "issue:46298", "accept": false, - "reason": "Tokenizer behavior regression and noisy logging are unrelated." + "reason": "Tokenizer behavior regression vs noisy logging message; unrelated." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "A generation-config validation issue is not the same as an unrecognized-model/config.json failure." + "reason": "GenerationConfig validation for GLM vs unrecognized model/config loading failure; different root causes." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "ImportError in configuration_utils and an audio feature extraction issue do not share the same underlying bug." + "reason": "ImportError vs audio feature extraction bug; no common code path." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Both hit chat/completions, but one is general request rejection while the other is a streaming-inference failure on a specific model." + "reason": "Both hit chat/completions, but one is request acceptance and the other is streaming inference failure; distinct bugs." }, { "left": "issue:44964", "right": "issue:45278", "accept": false, - "reason": "Both are load/import regressions after upgrade, but one is a multimodal model failure and the other is a broad import-error report." + "reason": "Both are load/import regressions, but one is a specific multimodal model load failure and the other is a broad upgrade-related import issue." }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 API guidance and MLA layernorm epsilon precision are separate numerical/configuration topics." + "reason": "TF32 API warning and MLA q_a_layernorm epsilon precision bug are unrelated." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "Both concern optimizers, but one ignores SGD args while the other double-calls optimizer.step under deepspeed." + "reason": "SGD args ignored vs optimizer.step being called twice under deepspeed; different optimizer bugs." }, { - "left": "issue:42371", - "right": "issue:42831", + "left": "issue:44261", + "right": "issue:46762", "accept": false, - "reason": "TF32 behavior and FineGrainedFP8 accuracy are different numerical settings and failures." + "reason": "Different numerical issues in different components: missing rms_norm_eps vs MSA attention mismatch." }, { - "left": "issue:44261", - "right": "issue:46762", + "left": "issue:42371", + "right": "issue:42831", "accept": false, - "reason": "Layernorm epsilon precision concerns are unrelated to MSA attention numerical differences." + "reason": "TF32 settings warning vs FineGrainedFP8 accuracy regression; not the same problem." } ] }, @@ -17621,29 +17648,29 @@ "issue:44484|issue:45310", "issue:43066|issue:43122", "issue:44077|issue:46981", - "issue:44410|issue:45478", "issue:43931|issue:44410", + "issue:44410|issue:45478", "issue:44263|issue:44485", "issue:43673|issue:45341", "issue:41628|issue:45071", + "issue:43208|issue:46897", "issue:43097|issue:43452", "issue:43931|issue:44038", - "issue:43208|issue:46897", "issue:43441|issue:44829", "issue:43012|issue:43577", - "issue:46752|issue:47116", "issue:43874|issue:44261", + "issue:46752|issue:47116", "issue:27957|issue:43352", "issue:44484|issue:45478", "issue:44117|issue:44987", "issue:44263|issue:46172", - "issue:43957|issue:44351", "issue:44360|issue:46172", + "issue:43957|issue:44351", "issue:43531|issue:45341", "issue:43097|issue:45588", "issue:43582|issue:45341", - "issue:44038|issue:44230", "issue:44261|issue:44485", + "issue:44038|issue:44230", "issue:43986|issue:45341", "issue:43097|issue:46050", "issue:43986|issue:45372", @@ -17651,20 +17678,20 @@ "issue:44246|issue:46981", "issue:45440|issue:46172" ], - "prepared_review_unit_hash": "24c22cfe7a8edfcfca9098f1a439d1e161478cedd25c0e80aaa130e869b7ca81", + "prepared_review_unit_hash": "208c57134b883f97679d9efc04172dc99a0e320b19e13605bdf0ed7b929f07e1", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37241, + "serialized_chars": 37244, "estimated_input_tokens": 9311, "estimated_eval_tokens": 18878 }, @@ -17673,192 +17700,204 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a loose thematic cluster around Transformers v5 regressions, model-specific loading bugs, and attention/tokenization edge cases. I do not see any pair that is clearly the same underlying bug, so I would not merge any soft-edge candidates.", - "confidence": 0.82, - "canonical_issue_reason": null, + "summary": "This cluster is mostly false-positive similarity around unrelated Transformers issues. The only clear duplicate-looking item is the Qwen3.5 MoE from_pretrained regression (45310/45478); I\u2019d use 45310 as canonical because it is the earlier, clearer representative. No PRs are present.", + "confidence": 0.63, + "canonical_issue_reason": "Issue 45310 is the best canonical issue because it is the earlier of an exact duplicate-title pair with 45478 and clearly describes a concrete, reproducible Qwen3.5 MoE from_pretrained regression.", "canonical_pr_reason": null, - "best_issue_reason": "No true canonical duplicate emerges; if a single anchor is needed, issue #43097 is the broadest and most central regression report with downstream references.", + "best_issue_reason": "45310 is the strongest overall issue representative in this set: it is specific, bug-oriented, and has an exact duplicate elsewhere in the items.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Different failures: removed tie_embeddings_and_encoder_decoder vs unexpected position-id key warnings. Same release family, not the same bug." + "reason": "Both are v5 regressions, but one is about a removed config flag and the other about unexpected position-id keys; different bugs." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "Unrelated topics: save_pretrained shard size default vs Qwen3.5 MoE from_pretrained error." + "reason": "Max shard size question vs. Qwen3.5 MoE from_pretrained error; unrelated symptoms and code paths." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Both touch tokenizer behavior, but one is a decoder-type mismatch and the other is a version-to-version tokenization change. Not the same concrete defect." + "reason": "Both concern tokenizer behavior changes, but one is a decoder-type issue and the other is a broader tokenization regression." }, { "left": "issue:44077", "right": "issue:46981", "accept": false, - "reason": "Completely different areas: PatchTSMixer post_init validation vs AutoProcessor HTTP 429 retry behavior." + "reason": "patchtsmixer post_init validation and HTTP 429 retry behavior are unrelated." }, { - "left": "issue:44410", - "right": "issue:45478", + "left": "issue:43931", + "right": "issue:44410", "accept": false, - "reason": "Different Qwen3-family model bugs with different symptoms and likely different code paths." + "reason": "Different model families and failure modes: VL weight-shape mismatch vs. missing Qwen3next projections." }, { - "left": "issue:43931", - "right": "issue:44410", + "left": "issue:44410", + "right": "issue:45478", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch is not the same as missing projections in qwen3next layer 0." + "reason": "Different model families; missing projections in qwen3next is not the same as Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "Separate GLM issues: torch.split return values vs RoPE implementation details." + "reason": "DSA indexer split semantics and GLM-5 RoPE implementation are separate model internals issues." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Generation cache/chunked_prefill regression is unrelated to a testing_utils bug." + "reason": "Generation cache/chunked_prefill regression vs. a testing_utils bug; no shared concrete code path." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Import error for AutoImageProcessor is unrelated to PretrainedConfig type-checking breakage." + "reason": "ImportError for AutoImageProcessor is unrelated to PretrainedConfig type checking." + }, + { + "left": "issue:43208", + "right": "issue:46897", + "accept": false, + "reason": "xLSTM training bugs and Florence2 double-shift loss bug are distinct model-specific failures." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "Different regression classes: removed config attribute vs gguf_file / from_pretrained loading breakage." + "reason": "Removed tie_embeddings behavior and gguf_file/from_pretrained loading breakage are different regressions." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "Both involve Qwen3 variants, but the reported loading failures are on different models and do not clearly share one root cause." - }, - { - "left": "issue:43208", - "right": "issue:46897", - "accept": false, - "reason": "xLSTM training blockers and Florence2 loss-shift bug are unrelated." + "reason": "Same broad Qwen3-VL area, but the concrete failures differ: one is a load shape mismatch, the other a Qwen3-VL-Moe bug." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "FlashAttention-related, but one is a model compatibility failure and the other is degenerate training behavior; not the same bug." + "reason": "FlashAttention 2 support crash and FlashAttention 3 degenerate training are different problems." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "Both are dtype/precision related, but one is a PyTorch warning on compile and the other is a BLIP2 loading dtype propagation bug." + "reason": "One is a PyTorch warning during compile, the other is incorrect dtype propagation when loading BLIP2." }, { - "left": "issue:46752", - "right": "issue:47116", + "left": "issue:43874", + "right": "issue:44261", "accept": false, - "reason": "Empty-list chat template crash and streaming pipeline regression are unrelated." + "reason": "Missing image-processor method vs. missing rms_norm_eps in MLA q_a_layernorm; unrelated." }, { - "left": "issue:43874", - "right": "issue:44261", + "left": "issue:46752", + "right": "issue:47116", "accept": false, - "reason": "Image-processor multimodal token counting and MLA layernorm epsilon are unrelated." + "reason": "Empty-chat-template IndexError is not the same as a streaming pipeline regression." }, { "left": "issue:27957", "right": "issue:43352", "accept": false, - "reason": "Both mention Flash Attention 2, but one is support for XLMRoberta and the other is a model that explicitly does not support FA2 yet." + "reason": "Both mention Flash Attention 2, but on different models and with different support/status issues." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "save_pretrained shard-size discussion is unrelated to Qwen3.5 MoE loading errors." + "reason": "A save_pretrained shard-size question is unrelated to the Qwen3.5 MoE from_pretrained error." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "The tokenizer-mapping None bug could affect loading, but the second issue is too model-specific to conclude the same root cause from the title alone." + "reason": "Tokenizer_MAPPING_NAMES returning None is a different failure from a specific model-loading regression." }, { "left": "issue:44263", "right": "issue:46172", "accept": false, - "reason": "Different DeepSeek/GLM implementation bugs with no clear shared code path." + "reason": "DSA torch.split behavior and deepseek-v4 comb transpose are unrelated code paths." }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "DSA indexer ReLU complaint and DeepSeek v4 comb transpose issue are unrelated." + "reason": "Missing ReLU in the DSA indexer is unrelated to the deepseek-v4 dtype/transpose bug." + }, + { + "left": "issue:43957", + "right": "issue:44351", + "accept": false, + "reason": "Meta-device model loading issues and missing HybridCache import are not the same bug." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to a testing utility bug." + "reason": "Qwen3-MoE sliding_window handling and a testing_utils bug are unrelated." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "Config attribute removal and flash_attention s_aux=None crash are different issues." + "reason": "Removed tie_embeddings behavior is unrelated to flash_attention.py crashing on s_aux=None." }, { "left": "issue:43582", "right": "issue:45341", "accept": false, - "reason": "Apple Silicon allocator TypeError is unrelated to testing_utils." - }, - { - "left": "issue:44038", - "right": "issue:44230", - "accept": false, - "reason": "Qwen3-VL-Moe loading bug and fp8 support request are not the same change or bug." + "reason": "Apple Silicon caching_allocator_warmup TypeError is unrelated to testing_utils." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Layernorm epsilon precision bug is unrelated to GLM-5 RoPE implementation." + "reason": "LayerNorm epsilon precision issue vs. RoPE implementation bug; different model components." }, { - "left": "issue:43986", - "right": "issue:45341", + "left": "issue:44038", + "right": "issue:44230", "accept": false, - "reason": "AutoProcessor/torchvision crash is unrelated to testing_utils." + "reason": "A Qwen3-VL-Moe runtime bug and an fp8 support request are not the same underlying issue." }, { - "left": "issue:45753", - "right": "issue:46369", + "left": "issue:43986", + "right": "issue:45372", "accept": false, - "reason": "Qwen3_5 infinite-loop-on-image and sam3-litetext memory overflow are unrelated." + "reason": "AutoProcessor without torchvision crash and ReasoningEffort import error are separate processor-loading problems." }, { "left": "issue:44246", "right": "issue:46981", "accept": false, - "reason": "Import performance sometimes is not the same as repeated HTTP 429 retries during from_pretrained." + "reason": "Slow import and repeated 429 retries are both performance-related, but they are different mechanisms and code paths." }, { "left": "issue:45440", "right": "issue:46172", "accept": false, - "reason": "Native DeepSeekV3MoE divergence and DeepSeek-v4 comb transpose issue are distinct model-implementation problems." + "reason": "DeepseekV3MoE divergence from remote implementation is not the same as the deepseek-v4 comb transpose bug." + }, + { + "left": "issue:45753", + "right": "issue:46369", + "accept": false, + "reason": "Qwen3_5 infinite loop on an image and SAM3 LiteText memory overflow are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "Removed tie_embeddings behavior and quantized save/load corruption are different regression classes." } ] }, @@ -17942,8 +17981,8 @@ "issue:44038|issue:44410", "issue:38617|issue:42757", "issue:46655|issue:46664", - "issue:44936|issue:46365", "issue:43208|issue:43550", + "issue:44936|issue:46365", "issue:46298|issue:46752", "issue:43064|issue:44315", "issue:46172|issue:46710", @@ -17953,211 +17992,217 @@ "issue:38617|issue:44351", "issue:43502|issue:43992", "issue:44451|issue:47116", - "issue:44938|issue:46178", "issue:44380|issue:46421", + "issue:44938|issue:46178", "issue:43064|issue:47116", "issue:43012|issue:43550", "issue:42907|issue:45753", - "issue:43066|issue:43906", "issue:45712|issue:46298", + "issue:43066|issue:43906", "issue:42907|issue:44655" ], - "prepared_review_unit_hash": "9d2c53a37ff4a5c8bf0c54d4cf0d56e9ea7dfd970561b1d70b5faa68643b0b86", + "prepared_review_unit_hash": "e42ef0affa779d49b0bfa5171ddd12a3949b4ab12ac121a0e2c096335f1defea", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36086, - "estimated_input_tokens": 9022, - "estimated_eval_tokens": 18300 + "serialized_chars": 36089, + "estimated_input_tokens": 9023, + "estimated_eval_tokens": 18302 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is highly heterogeneous; the suggested soft pairs are mostly related only by broad subsystem or keyword overlap, not the same underlying bug. No PRs are present. If one issue must stand in as the most representative umbrella report, the broad regression ticket is the best fit.", - "confidence": 0.22, - "canonical_issue_reason": "issue:47116 is the broadest, most umbrella-like report here (a Transformers v5 regression) and is more representative than the highly specific bug reports, though the cluster overall is not a true duplicate set.", + "summary": "These items are mostly unrelated transformer bugs; the soft links are mostly superficial similarity around imports, model families, or training/inference features. I would not merge any of the provided soft pairs.", + "confidence": 0.86, + "canonical_issue_reason": "issue:43873 is the strongest standalone issue in the set: it has high discussion/inbound activity, is still open, and describes a concrete, impactful bug with clear reproduction around quantization/offloading.", "canonical_pr_reason": null, - "best_issue_reason": "issue:47116 is the most suitable single representative issue because it describes a general v5 regression and has the broadest scope among otherwise unrelated reports.", + "best_issue_reason": "issue:43873 is the best representative issue because it appears well-developed, broadly relevant, and actionable compared with the rest of the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "Different failures: a removed config API/flag vs quantization offloading behavior. Same release theme, not same bug." + "reason": "Different problems: one is a v5 API removal/compatibility complaint, the other is a quantization/offloading failure. No shared code path or fix." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Both involve Qwen-family models, but one is missing projections in layer 0 and the other is a missing config field silently dropped by strict handling. Different code paths." + "reason": "Both are Qwen-related, but one is missing projections in a layer and the other is a config field being dropped. Different bugs and likely different files." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "An AddedToken keyword-argument error is unrelated to TOKENIZER_MAPPING_NAMES returning None during from_pretrained." + "reason": "Unrelated tokenizer constructor error vs tokenizer-mapping lookup behavior. Same area of code, but not the same concrete defect." }, { "left": "issue:44117", "right": "issue:46701", "accept": false, - "reason": "Tokenizer mapping fallback logic and diffusion image-text pipeline failure are distinct problems." + "reason": "Tokenizer mapping fallback bug and a DiffusionGemma pipeline failure are separate issues with no clear common root cause." }, { "left": "issue:43673", "right": "issue:46123", "accept": false, - "reason": "Generation cache initialization during chunked prefill is unrelated to MaskGenerationPipeline dropping the last partial batch." + "reason": "Both mention generation/pipeline behavior, but one is cache availability in chunked prefill and the other is final partial-batch handling in mask generation. Different code paths." }, { "left": "issue:43012", "right": "issue:46821", "accept": false, - "reason": "A PyTorch bfloat16 precision warning is not the same as a TimesFM window_size AttributeError." + "reason": "A PyTorch precision warning and a TimesFM window_size AttributeError are unrelated." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Both mention GLM components, but one is an image processor method missing and the other is a torch.split value/shape issue in a different indexer path." + "reason": "Both are GLM-family issues, but one is a missing image-processor method and the other is a torch.split return-value problem in a different component." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Different model-specific bugs: Qwen3-VL-Moe behavior vs qwen3next missing attention projections." + "reason": "Different Qwen variants and different failure modes: one is a Transformers 5.0/Qwen3-VL-Moe bug, the other is missing layer projections in qwen3next." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols from different modules with different fixes." + "reason": "Both are import errors, but they involve different symbols, modules, and dependencies. Not the same underlying bug." }, { "left": "issue:46655", "right": "issue:46664", "accept": false, - "reason": "Same model family, but padding calculation and num_frames semantics are different video-processor bugs." + "reason": "Same model family, but one is padding math and the other is num_frames semantics in the video processor. Related area, different defects." }, { - "left": "issue:44936", - "right": "issue:46365", + "left": "issue:43208", + "right": "issue:43550", "accept": false, - "reason": "Trainer.evaluate() failing after train() is unrelated to failing to import Trainer because datasets is installed." + "reason": "xLSTM training bugs and a Bamba torch.compile/SDPA failure are separate model-specific issues." }, { - "left": "issue:43208", - "right": "issue:43550", + "left": "issue:44936", + "right": "issue:46365", "accept": false, - "reason": "xLSTM training bugs and Bamba torch.compile/SDPA failure are different model/code-path issues." + "reason": "Trainer.evaluate after train and inability to import Trainer due to datasets are different failures with different causes." }, { "left": "issue:46298", "right": "issue:46752", "accept": false, - "reason": "Noisy logging and apply_chat_template IndexError on empty input are unrelated." + "reason": "A noisy logging message and an empty-list apply_chat_template crash are unrelated." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption is not the same as Liger Kernel not being applied with model_init." + "reason": "Different training-stack problems: FSDP2/PEFT/optimizer-state corruption versus Liger Kernel not being applied with model_init." }, { "left": "issue:46172", "right": "issue:46710", "accept": false, - "reason": "A dtype/transpose bug in deepseek-v4 is distinct from an accuracy regression after upgrading Transformers." + "reason": "A deepseek-v4 tensor operation issue and a broad accuracy regression after upgrade are not the same concrete bug." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "Missing torchvision during AutoProcessor loading and save_pretrained shard-size behavior are unrelated." + "reason": "AutoProcessor/torchvision crash and save_pretrained shard size behavior are unrelated." }, { "left": "issue:42890", "right": "issue:46298", "accept": false, - "reason": "Seed-related test flakiness is unrelated to a noisy logging message." + "reason": "Missing set_seed in tests and a noisy logging message have nothing in common beyond being maintenance issues." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "Passing kwargs to cached_file and auto_docstring crashes under future annotations are different bugs." + "reason": "AutoProcessor kwargs forwarding and auto_docstring annotation handling are different subsystems and bugs." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they concern different missing names and different compatibility surfaces." + "reason": "Both are import-related, but they expose different missing exports and likely different release-compatibility problems. Not mergeable as one fix." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "local_files_only still allowing API calls is a network/offline bug, not the same as a missing embed_tokens.weight during loading." + "reason": "local_files_only still making API requests and a missing embed_tokens weight in from_pretrained are unrelated loading issues." }, { "left": "issue:44451", "right": "issue:47116", "accept": false, - "reason": "A model-loading failure for ScandiBERT is not the same as a streaming pipeline regression in v5." + "reason": "A model load regression for ScandiBERT and a streaming pipeline regression are distinct failures." }, { - "left": "issue:44938", - "right": "issue:46178", + "left": "issue:44380", + "right": "issue:46421", "accept": false, - "reason": "Python 3.14 loading failure and an invalid tensorflow_text backend mapping are different compatibility issues." + "reason": "Attention scaling ignored in GPT2 backends and chunked-prefill cache lazy init causing recompilation are separate runtime issues." }, { - "left": "issue:44380", - "right": "issue:46421", + "left": "issue:44938", + "right": "issue:46178", "accept": false, - "reason": "GPT2 attention-scaling being ignored under SDPA/FlashAttention is unrelated to cache lazy initialization causing recompilation." + "reason": "Python 3.14 import failure and a backend mapping error for tensorflow_text are unrelated compatibility problems." }, { "left": "issue:43064", "right": "issue:47116", "accept": false, - "reason": "Optimizer-state corruption in Trainer/FSDP2/PEFT is not the same as a streaming pipeline regression." + "reason": "Wrong optimizer state on distributed ranks and a streaming pipeline regression do not share the same code path or fix." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "A precision warning for bfloat16 compilation and a Bamba SDPA compile failure are different problems." + "reason": "Precision warning versus compile-time SDPA failure; different symptoms and likely different fixes." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and a Qwen3_5 infinite loop on one image are unrelated." + "reason": "Saving dequantized Ministrals/Devstrals and Qwen3_5 infinite loop on one image are distinct model-specific bugs." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Left is a repo hygiene/dummy-object cleanup issue; right is logging noise. No overlap in underlying bug." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 is not shown to be the same as the isolated reproduction of issue 38071." + "reason": "Wrong tokenizer decoder type and a reproduction of a different issue are not the same defect." }, { - "left": "issue:45712", - "right": "issue:46298", + "left": "issue:42907", + "right": "issue:44655", "accept": false, - "reason": "Leftover dummy classes in dummy_pt_objects.py and noisy logging are unrelated maintenance vs behavior bugs." + "reason": "Dequantized model saving and pipeline save_pretrained behavior are both save-related, but they concern different object types and failure modes." } ] }, @@ -18237,8 +18282,8 @@ "issue:43012|issue:44368", "issue:43992|issue:44484", "issue:43352|issue:43976", - "issue:43827|issue:47116", "issue:44734|issue:47116", + "issue:43827|issue:47116", "issue:44077|issue:46365", "issue:44162|issue:46172", "issue:44380|issue:46132", @@ -18250,8 +18295,8 @@ "issue:42175|issue:46365", "issue:43012|issue:47276", "issue:43756|issue:44485", - "issue:44222|issue:45478", "issue:36010|issue:42757", + "issue:44222|issue:45478", "issue:43064|issue:44380", "issue:42890|issue:47059", "issue:42831|issue:44261", @@ -18261,20 +18306,20 @@ "issue:44625|issue:46365", "issue:45276|issue:46077" ], - "prepared_review_unit_hash": "fbb335d96a5968864eec31aa178c2bf0b277d12856925f6629df0a8bc21678f6", + "prepared_review_unit_hash": "836fae356273cdc1733c4f2735d264ab7a89e30ce9f1a64a3de42ed73f427531", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36355, + "serialized_chars": 36354, "estimated_input_tokens": 9089, "estimated_eval_tokens": 18434 }, @@ -18283,186 +18328,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a grab bag of unrelated Transformers issues. None of the soft pairs look like the same underlying bug/change, so all suggested merges should be rejected. There is no clear canonical duplicate hub across the full set; the strongest standalone issue is a broad v5 runtime regression.", - "confidence": 0.93, - "canonical_issue_reason": null, + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues (imports, model/config bugs, backend regressions, warnings, docs, and packaging). None of the soft pairs look like the same underlying bug or change, so all are rejected.", + "confidence": 0.94, + "canonical_issue_reason": "No single issue is a true duplicate anchor for the whole set; issue:43097 is the broadest v5-breaking-change report and the best representative issue if one must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": "issue:47116 is a high-signal, broadly impactful runtime regression in a core v5 code path (streaming pipeline), with a concrete failure mode and likely wider user impact than the more niche reports.", + "best_issue_reason": "Issue:43097 is the most generally framed, high-impact regression in the set; the others are narrower model-, backend-, or packaging-specific bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "Different models and failures: xLSTM training blockers vs BeitImageProcessorFast reduce_labels behavior." + "reason": "xLSTM training bugs vs BeitImageProcessorFast label reduction; different models and failure modes." }, { "left": "issue:45584", "right": "issue:46365", "accept": false, - "reason": "Whisper empty-transcription generation bug vs importing Trainer failing due to datasets; unrelated code paths." + "reason": "Whisper empty-transcription generation bug vs Trainer import conflict with datasets; unrelated subsystems." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "Both mention pipeline, but one is a v5 translation-support error message and the other is save_pretrained for Pipeline objects." + "reason": "pipeline() error-message regression vs Pipeline save_pretrained serialization; different problems." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Different warnings in different areas: bfloat16 compile precision warning vs tie_word_embeddings warning in a finetuning setup." + "reason": "bfloat16 compile warning vs tie_word_embeddings warning during Qwen fine-tuning; unrelated warnings." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "UMT5Encoder weight-loading bug vs a save_pretrained shard-size question; not the same underlying issue." + "reason": "Missing embed_tokens.weight on load vs max_shard_size default on save; load-path vs save-path issue." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Nemotron FlashAttention support bug vs Python version compatibility problem; unrelated." + "reason": "Flash Attention support for Nemotron vs Python version compatibility; not the same bug." }, { - "left": "issue:43827", + "left": "issue:44734", "right": "issue:47116", "accept": false, - "reason": "Docs still referencing pipeline() vs a streaming pipeline runtime regression; docs cleanup and code bug are distinct." + "reason": "KV-cache continuation indexing in /v1/responses vs streaming pipeline regression; different serving paths." }, { - "left": "issue:44734", + "left": "issue:43827", "right": "issue:47116", "accept": false, - "reason": "Both are serve/pipeline regressions, but they fail in different code paths and with different symptoms." + "reason": "Docs still referencing pipeline() vs runtime streaming regression; docs issue is separate from code bug." }, { "left": "issue:44077", "right": "issue:46365", "accept": false, - "reason": "PatchTSMixer post_init restriction vs Trainer import broken by datasets; unrelated." + "reason": "patchtsmixer post_init validation vs Trainer import blocked by datasets; unrelated." }, { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "ESM2 breakage vs deepseek-v4 tensor transpose issue; different models and code paths." + "reason": "ESM2 broken vs deepseek-v4 comb transpose bug; different models and code paths." }, { "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "GPT2 attention scaling ignored in backends vs AttentionInterface.register behavior change; related subsystem, different bugs." + "reason": "GPT2 attention scaling ignored in SDPA/FlashAttention vs AttentionInterface.register behavior change; distinct attention issues." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Gemma3Config missing vocab_size vs Apple Silicon caching_allocator_warmup TypeError; unrelated." + "reason": "Gemma3Config missing vocab_size vs Apple Silicon caching_allocator_warmup TypeError; config bug vs platform/runtime bug." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "Nemotron FA2 support vs GPT2 attention scaling behavior; not the same bug." + "reason": "Nemotron FlashAttention 2 unsupported vs GPT2 attention scaling ignored; both backend-related but not the same failure." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols from different packages." + "reason": "Different missing imports from different packages/symbols; same symptom class, but not the same underlying issue." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "Attention backend scaling bug vs degenerate zero-loss classification bug; unrelated." + "reason": "Attention scaling backend bug vs zero-loss classification bug; unrelated." }, { "left": "issue:43097", "right": "issue:46298", "accept": false, - "reason": "Removed tie_embeddings option vs noisy logging message; different concerns." + "reason": "Removed API without indication vs noisy logging message; unrelated severity and root cause." }, { "left": "issue:42175", "right": "issue:46365", "accept": false, - "reason": "Packaging/backends install issue vs importing Trainer failing due to datasets; not the same failure." + "reason": "Missing TensorFlow backend in torch install vs Trainer import conflict with datasets; different packaging problems." }, { "left": "issue:43012", "right": "issue:47276", "accept": false, - "reason": "Both are warnings, but one is about bfloat16 compilation precision and the other about synced_gpus messaging." + "reason": "bfloat16 compile warning vs misleading synced_gpus warning in batching; unrelated warnings." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Both mention RoPE, but they concern different models and likely different implementation defects." + "reason": "Smollm3 RoPE layer drop vs GLM-5 RoPE implementation; both RoPE-related but clearly different bugs." }, { - "left": "issue:44222", - "right": "issue:45478", + "left": "issue:36010", + "right": "issue:42757", "accept": false, - "reason": "FP8 save_pretrained on moe vs Qwen3.5 Moe from_pretrained error; related area but different operations and failures." + "reason": "Different import errors for different symbols/modules; not the same missing-export regression." }, { - "left": "issue:36010", - "right": "issue:42757", + "left": "issue:44222", + "right": "issue:45478", "accept": false, - "reason": "Different import errors from different libraries and symbols." + "reason": "FP8 save_pretrained MoE bug vs Qwen3.5 MoE from_pretrained error; different stages and symptoms." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Trainer/FSDP2/PEFT optimizer-state bug is unrelated to GPT2 attention scaling." + "reason": "Wrong optimizer states with FSDP2+PEFT vs GPT2 attention scaling bug; unrelated code paths." }, { "left": "issue:42890", "right": "issue:47059", "accept": false, - "reason": "Seed-related test flakiness vs huggingface_hub CLI attribute mismatch; unrelated." + "reason": "Flaky tests due to missing set_seed vs CLI failure from huggingface_hub API change; different causes." }, { "left": "issue:42831", "right": "issue:44261", "accept": false, - "reason": "Both are precision/accuracy complaints, but one is FineGrainedFP8 accuracy and the other is MLA q_a_layernorm eps propagation." + "reason": "FineGrainedFP8 accuracy issue vs missing rms_norm_eps precision drift; different precision bugs." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "Chat-completions request rejection vs batched apply_chat_template crash; both chat-related but not the same bug." + "reason": "Chat completions API rejection vs batched processor crash; different endpoints and models." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "Both are CLIP regressions, but one is missing attentions and the other is missing hidden_states; separate outputs and code paths." + "reason": "CLIP attentions missing vs CLIPVision hidden_states missing; same family, but different outputs and regressions." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "ImportError for a moved utility vs local_files_only still making API calls; unrelated." + "reason": "ImportError for layer_type_validation vs local_files_only still making API requests; unrelated import vs networking bug." }, { "left": "issue:44625", "right": "issue:46365", "accept": false, - "reason": "Config propagation bug in Qwen3.5 vs Trainer import conflict; no common underlying change." + "reason": "Qwen3.5 num_labels propagation vs Trainer import conflict with datasets; unrelated." }, { "left": "issue:45276", "right": "issue:46077", "accept": false, - "reason": "Gemma4 embedding resize bug vs Gemma4 bidirectional-attention masking bug; same model family but different concrete defects." + "reason": "Gemma4 embedding resize bug vs Gemma4 causal mask construction bug; same model family but different concrete failures." } ] }, @@ -18535,8 +18580,8 @@ "issue:47337" ], "soft_pairs": [ - "issue:43299|issue:43867", "issue:44655|issue:47321", + "issue:43299|issue:43867", "issue:44655|issue:47332", "issue:43618|issue:44380", "issue:42994|issue:44655", @@ -18569,232 +18614,232 @@ "issue:36296|issue:43575", "issue:43618|issue:44077" ], - "prepared_review_unit_hash": "e3e18d094b92ec03bf3c86ae33baccdacbe8a872882801396f2ee8acd1124d54", + "prepared_review_unit_hash": "6867ec84c3726f0bb4b5f9dcc053640abc1fc88104a82589eabfb9ddbf4c6932", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 33, - "serialized_chars": 36446, - "estimated_input_tokens": 9112, - "estimated_eval_tokens": 18480 + "serialized_chars": 36444, + "estimated_input_tokens": 9111, + "estimated_eval_tokens": 18478 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No strong duplicates here: the items span unrelated bugs across loading, saving, training, model configs, Python-version compatibility, and specific model families. I rejected all soft-edge pairs because they do not appear to be the same underlying bug/change.", - "confidence": 0.94, - "canonical_issue_reason": "issue:45216 is the clearest, most self-contained regression report in the set: it names the affected model family, the failing API (`save_pretrained`), and the version boundary (`since version 5.4.0`).", + "summary": "These artifacts are mostly unrelated issue reports; the soft-similarity pairs do not look like true duplicates or the same fix. No PRs are present. The most representative issue is the active, high-activity Gemma3Config bug.", + "confidence": 0.91, + "canonical_issue_reason": "issue:36683 is the most central standalone report here: open, very high discussion activity, and a concrete, narrowly described runtime error (`Gemma3Config` missing `vocab_size`).", "canonical_pr_reason": null, - "best_issue_reason": "issue:45216 is the strongest standalone issue for triage because it is concrete, reproducible, and describes a clearly scoped regression.", + "best_issue_reason": "issue:36683 is the best issue candidate because it is specific, actionable, and active; the rest are either narrower, closed, or clearly unrelated to one another.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43299", - "right": "issue:43867", + "left": "issue:44655", + "right": "issue:47321", "accept": false, - "reason": "Both are loading-related, but they point to different failure modes and likely different code paths/models." + "reason": "Both mention framework/model issues, but one is about `save_pretrained` on Pipeline objects and the other is an AMD Quark class update bug; different code paths." }, { - "left": "issue:44655", - "right": "issue:47321", + "left": "issue:43299", + "right": "issue:43867", "accept": false, - "reason": "One is about `save_pretrained` on Pipeline objects; the other is an AMD quark class update issue. Unrelated." + "reason": "Both are loading failures, but they concern different models and different failure modes; not the same underlying bug." }, { "left": "issue:44655", "right": "issue:47332", "accept": false, - "reason": "Both mention saving, but one is Pipeline serialization and the other is disk-offloaded Qwen3-VL model saving; different bugs." + "reason": "Both involve saving, but one is generic Pipeline saving and the other is disk-offloaded Qwen3-VL saving; too different to merge." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "CLIP attentions assignment and GPT2 attention scaling under SDPA/FlashAttention are distinct model/backend bugs." + "reason": "Different models and different regressions: CLIP attentions assignment vs GPT2 attention scaling under SDPA/FlashAttention." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Quantized model saving failure is not the same as saving Pipeline objects." + "reason": "Quantized model saving failure is a different bug from Pipeline `save_pretrained` failure." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "`num_labels` propagation and Gemma4 bidirectional attention masking are unrelated configuration bugs." + "reason": "Different models and different config propagation issues (`num_labels` vs bidirectional attention masking)." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Different missing-attribute failures in different model classes; no shared underlying bug." + "reason": "Both are missing-attribute errors, but on unrelated model classes and different missing fields." }, { "left": "issue:43352", "right": "issue:45593", "accept": false, - "reason": "Flash Attention support gating and D-FINE auxiliary-loss behavior are unrelated." + "reason": "Nemotron Flash Attention 2 support and D-FINE auxiliary-loss behavior are unrelated." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "A missing `vocab_size` attribute and a `post_init` policy change are different issues." + "reason": "`vocab_size` missing on Gemma3Config is unrelated to patchtsmixer `post_init` handling." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both are version/platform compatibility reports, but they affect different Python versions and likely different breakages." + "reason": "Both are Python-version compatibility complaints, but they concern different versions and different breakages; not the same bug." }, { "left": "issue:44355", "right": "issue:46877", "accept": false, - "reason": "Compiled-file import errors and auto-docstring stdout diagnostics are different import-time problems." + "reason": "Compiled-file import errors and auto-docstring stdout diagnostics are unrelated." }, { "left": "issue:46077", "right": "issue:46512", "accept": false, - "reason": "Bidirectional attention masking and stale `config.dtype` after `from_config` are unrelated behaviors." + "reason": "Different config/state bugs: attention-mask construction vs stale `config.dtype` after `from_config`." }, { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "Both involve loading/saving workflows, but one is sorted-state-dict loading and the other is incorrect checkpoint saving." + "reason": "Both are model loading/saving regressions, but the titles point to different mechanisms and scopes; not enough to treat as the same issue." }, { "left": "issue:44263", "right": "issue:45440", "accept": false, - "reason": "Different MoE-related model bugs with different symptoms and code paths." + "reason": "`torch.split()` behavior in one MoE indexer and DeepSeekV3 implementation divergence are distinct problems." }, { "left": "issue:44938", "right": "issue:47059", "accept": false, - "reason": "Python 3.14 load failure and a CLI completion incompatibility with `huggingface_hub` are unrelated dependency issues." + "reason": "Python 3.14 import failure and a CLI completion regression from `huggingface_hub` are unrelated." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "Attention metadata assignment and `save_pretrained` regression are not the same bug." + "reason": "CLIP attentions regression and Qwen3.5 checkpoint-saving regression are unrelated." }, { "left": "issue:36683", "right": "issue:47110", "accept": false, - "reason": "A missing model config attribute and a `special_tokens` dict/list type error are unrelated." + "reason": "A missing config attribute and a `special_tokens` type error are different bugs." }, { "left": "issue:36683", "right": "issue:44625", "accept": false, - "reason": "Different config propagation/attribute problems in different model families." + "reason": "Both are config-related, but one is a missing `vocab_size` attribute and the other is `num_labels` propagation for Qwen3.5; different fixes." }, { "left": "issue:29870", "right": "issue:36296", "accept": false, - "reason": "BART generate-time decoder mask initialization and tensor-parallel training are unrelated." + "reason": "Bart generation `decoder_attention_mask` bug and tensor-parallel training bug are unrelated." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "General compiled-file import failures versus a TorchScript parsing/indentation problem in DebertaV2." + "reason": "Compiled Python file errors and a Python 3.13 `@torch.jit.script` parsing issue are different import-time failures." }, { "left": "issue:44908", "right": "issue:46851", "accept": false, - "reason": "Learning-rate scheduler kwargs handling and variable-length Qwen3.5 position_ids handling are unrelated." + "reason": "Scheduler kwargs handling and Qwen3.5 variable-length position IDs are unrelated." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "`@torch.no_grad` on cache updates and broad import errors after upgrade are different issues." + "reason": "`@torch.no_grad` decoration for cache updates and broad import errors after upgrading Transformers are not the same bug." }, { "left": "issue:43867", "right": "issue:46202", "accept": false, - "reason": "Loading sorted state_dicts and Gemma pipeline roundtrip with `\u2581` are unrelated serialization/tokenization issues." + "reason": "A generic state-dict-sorting load error and Gemma pipeline roundtrip failure are different code paths." }, { "left": "issue:43992", "right": "issue:47317", "accept": false, - "reason": "A missing embedding weight during load and a PPFormulaNet training-loss bug are different failures." + "reason": "UMT5 encoder weight loading and PPFormulaNet training-loss bug are unrelated." }, { "left": "issue:44075", "right": "issue:46365", "accept": false, - "reason": "SGD argument handling and Trainer import being blocked by `datasets` are unrelated." + "reason": "SGD args being ignored is not the same as an import failure caused by `datasets`." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "Both are loading errors, but they concern different models and different missing/miswired state." + "reason": "Both mention loading, but they are different Qwen/UMT5 issues with different missing pieces." }, { "left": "issue:29870", "right": "issue:33453", "accept": false, - "reason": "Generate-time decoder mask bug and tokenizer loading regression do not share the same root cause." + "reason": "Bart generation error and tokenizer loading regression are unrelated." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption and a cache-strategy typo are unrelated." + "reason": "FSDP/PEFT optimizer-state corruption and a cache-strategy typo are not the same bug." }, { "left": "issue:44393", "right": "issue:47000", "accept": false, - "reason": "Qwen3-VL bbox hallucination/output issues and RTDetrLoss infeasible-cost-matrix errors are different detection/vision bugs." + "reason": "Qwen3-VL bbox hallucination/error and RTDetrLoss infeasible cost matrix are both vision-related, but they are different failures in different models." }, { "left": "issue:44393", "right": "issue:47337", "accept": false, - "reason": "A model output bug and a flaky bbox-range test failure are not the same underlying issue." + "reason": "A Qwen3-VL bounding-box issue and a flaky bbox-range test are not the same underlying defect." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 moe save_pretrained and missing tensor-parallel plan are separate MoE issues." + "reason": "FP8 MoE save_pretrained failure and missing `_tp_plan` for tensor parallelism are different issues." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Tensor-parallel training bug and TP load OOM are different problems; one is correctness, the other capacity." + "reason": "Tensor-parallel training bug and tensor-parallel OOM on Qwen2-57B are different problems; OOM alone does not imply the same defect." }, { "left": "issue:43618", "right": "issue:44077", "accept": false, - "reason": "CLIP attentions assignment and `patchtsmixer` post_init policy are unrelated." + "reason": "CLIP attentions regression and patchtsmixer `post_init` policy are unrelated." } ] }, @@ -18830,6 +18875,7 @@ "issue:42898", "issue:43010", "issue:43065", + "issue:43232", "issue:43502", "issue:43525", "issue:43576", @@ -18868,15 +18914,15 @@ ], "soft_pairs": [ "issue:41669|issue:44246", - "issue:43867|issue:45375", "issue:43010|issue:43976", + "issue:43867|issue:45375", "issue:42890|issue:45561", "issue:43867|issue:45357", "issue:28282|issue:33453", - "issue:42548|issue:44351", "issue:43010|issue:45850", - "issue:45125|issue:45478", + "issue:42548|issue:44351", "issue:43618|issue:44977", + "issue:45125|issue:45478", "issue:45125|issue:45375", "issue:44360|issue:45440", "issue:43502|issue:44062", @@ -18884,242 +18930,252 @@ "issue:42890|issue:44112", "issue:42890|issue:44964", "issue:44734|issue:46617", - "issue:44387|issue:44556", - "issue:43723|issue:44556", "issue:42898|issue:46664", - "issue:43824|issue:46617", + "issue:43723|issue:44556", + "issue:44387|issue:44556", "issue:43010|issue:43065", + "issue:43824|issue:46617", "issue:45071|issue:47317", "issue:43576|issue:46132", "issue:43502|issue:44485", "issue:44355|issue:44749", - "issue:30333|issue:33453", "issue:45850|issue:47110", "issue:43874|issue:46617", + "issue:30333|issue:33453", + "issue:30990|issue:31795", "issue:41628|issue:42548", - "issue:30990|issue:31795" + "issue:43010|issue:43232" ], - "prepared_review_unit_hash": "706cb7bf7548507b481b2b5dd445faeeea2edd30988c27f171d3be02a38447c4", + "prepared_review_unit_hash": "ed495b25fc46412d2dab85413feef3413d34029380871716db6c2498706d5e64", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 35979, - "estimated_input_tokens": 8995, - "estimated_eval_tokens": 18246 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36824, + "estimated_input_tokens": 9206, + "estimated_eval_tokens": 18668 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are unrelated transformer regressions, import/load failures, CI flakes, docs, or model-specific bugs. I do not see any soft pair that clearly shares the same concrete underlying bug/change, so I reject all soft edges.", - "confidence": 0.87, - "canonical_issue_reason": "No single canonical issue: the items span unrelated import, loading, CI, docs, and model-specific regressions.", + "summary": "This cluster is mostly a grab bag of unrelated import/load regressions, tokenizer/model compatibility bugs, and test flakes. Only one pair looks plausibly like the same underlying issue: import-time slowness around `transformers` import.", + "confidence": 0.56, + "canonical_issue_reason": "issue:44246 is the broadest user-facing report in the only obvious duplicate pair, and it best represents the import-performance problem that issue:41669 appears to explain more concretely.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44246 is the closest representative only because it is the broadest import-performance complaint, but the cluster is too mixed to justify a true canonical issue.", + "best_issue_reason": "issue:44246 is the best overall representative because it states the externally visible symptom in a general way and can absorb the more specific import-cost complaint in issue:41669.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41669", "right": "issue:44246", - "accept": false, - "reason": "Both mention import slowness, but 41669 is a specific import-* cleanup while 44246 is a generic intermittent slowdown report." + "accept": true, + "reason": "Both concern `transformers` import taking too long; 41669 gives a likely cause (`import *` in models) for the same performance problem described in 44246." }, { - "left": "issue:43867", - "right": "issue:45375", + "left": "issue:43010", + "right": "issue:43976", "accept": false, - "reason": "Different Qwen issues: one is a load/state_dict sorting failure, the other is a missing vision config field." + "reason": "Different problems: one is a missing `@torch.no_grad` decoration on cache/layer updates, the other is Python version compatibility for Transformers 5.1.0." }, { - "left": "issue:43010", - "right": "issue:43976", + "left": "issue:43867", + "right": "issue:45375", "accept": false, - "reason": "Unrelated problems: cache/layer update no_grad decoration vs Python 3.9 compatibility." + "reason": "Different failure modes: sorted `state_dict` loading vs a missing Qwen3.5 vision config field silently dropped by strict validation." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Test flakiness from missing seeds is not the same as an xdist file-write race." + "reason": "One is a flaky model integration test needing `set_seed()`, the other is an xdist race on `captured_info.txt`; both are test issues but not the same bug." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Load failure from sorted state_dict is distinct from incorrect visual encoder keys on save_pretrained." + "reason": "Both involve loading/saving models, but the concrete problems differ: `state_dict` sorting vs incorrect Qwen3.5 visual encoder keys on save." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "PyTorch-missing import error and tokenizer-loading regression are different failure modes." - }, - { - "left": "issue:42548", - "right": "issue:44351", - "accept": false, - "reason": "Both are import errors, but for different top-level symbols and likely different export paths." + "reason": "Different import/load failures: missing PyTorch for AutoModel vs a tokenizer-loading regression." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "Specific no_grad cache update fix does not match a broad 'many models break' regression." + "reason": "`@torch.no_grad` on cache updates is a specific fix; the other is a broad complaint that 5.0 breaks too many models." }, { - "left": "issue:45125", - "right": "issue:45478", + "left": "issue:42548", + "right": "issue:44351", "accept": false, - "reason": "Same model family, but tensor-parallel metadata missing is not the same as a generic from_pretrained error." + "reason": "Both are import errors, but they concern different missing symbols (`PreTrainedModel` vs `HybridCache`) and likely different code paths." }, { "left": "issue:43618", "right": "issue:44977", "accept": false, - "reason": "CLIP attentions assignment and Qwen3.5 flash-attention generation are unrelated bugs." + "reason": "CLIP attentions not being assigned is unrelated to Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3.5 MoE, but one is about tensor-parallel plan metadata and the other is a `from_pretrained` error; not the same concrete bug." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Tensor-parallel support gap and missing vision config field are separate Qwen3.5-MoE problems." + "reason": "Different Qwen3.5 MoE issues: missing tensor-parallel plan vs missing vision config field." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "The DSA/ReLU discussion is unrelated to DeepSeekV3 implementation divergence." + "reason": "The DSA indexer/ReLU discussion is unrelated to divergence between native and remote DeepSeekV3 implementations." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "Offline-loading network leakage and AddedToken keyword duplication are different code paths." + "reason": "`local_files_only=True` still making requests is unrelated to `AddedToken()` receiving duplicate `special` arguments." }, { "left": "issue:43525", "right": "issue:47317", "accept": false, - "reason": "Llama4Config pad_token_id error and PPFormulaNet training-loss bug are unrelated." + "reason": "A missing `pad_token_id` on `Llama4Config` is unrelated to a PPFormulaNet training-loss bug." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Seed-related test instability is not the same as the stale device-override CI failure." + "reason": "One is a seed-related flaky test, the other is a CI device-override test failure; different artifacts and failure mechanisms." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "A flaky integration test and a model-load failure for Phi-4 multimodal are not the same issue." + "reason": "Seeded test flakiness is unrelated to failure loading `microsoft/Phi-4-multimodal-instruct`." }, { "left": "issue:44734", "right": "issue:46617", "accept": false, - "reason": "KV-cache continuation indexing and cache FileNotFoundError are separate serving/load bugs." + "reason": "KV-cache continuation tensor indexing in serve is unrelated to loading a trust-remote-code model from cache." }, { - "left": "issue:44387", - "right": "issue:44556", + "left": "issue:42898", + "right": "issue:46664", "accept": false, - "reason": "Int4 memory regression and checkpoint reload incompatibility are different problems." + "reason": "Tokenization-space cleanup behavior in v5 is unrelated to the SmolVLM `num_frames` meaning inconsistency." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Tokenizer loading regression and post-upgrade checkpoint reload failure do not share a concrete root cause." + "reason": "Both are loading-related, but one is tokenizer loading via `AutoTokenizer`, while the other is checkpoint reload compatibility after upgrading from v4.57 to v5.2/5.3." }, { - "left": "issue:42898", - "right": "issue:46664", + "left": "issue:44387", + "right": "issue:44556", "accept": false, - "reason": "Tokenization-space behavior change and video num_frames semantics are unrelated API changes." + "reason": "Int4 quantization OOM from extra CUDA reserved memory is a different issue from checkpoint reload incompatibility after upgrade." }, { - "left": "issue:43824", - "right": "issue:46617", + "left": "issue:43010", + "right": "issue:43065", "accept": false, - "reason": "Top-level import error for Qwen2.5-VL and trust_remote_code cache FileNotFoundError are different failures." + "reason": "`no_grad` decoration on cache updates is unrelated to a dummy `nn.Conv2d` in `Sam3PixelDecoder`." }, { - "left": "issue:43010", - "right": "issue:43065", + "left": "issue:43824", + "right": "issue:46617", "accept": false, - "reason": "Cache update no_grad decoration and a dummy Conv2d in Sam3PixelDecoder are unrelated changes." + "reason": "Missing `Qwen2_5_VLForConditionalGeneration` import and trust-remote-code cache loading are different import/load failures." }, { "left": "issue:45071", "right": "issue:47317", "accept": false, - "reason": "PretrainedConfig type-checking regression is unrelated to the PPFormulaNet training-loss bug." + "reason": "PretrainedConfig type-checking regression is unrelated to PPFormulaNet training loss." }, { "left": "issue:43576", "right": "issue:46132", "accept": false, - "reason": "Broken transformers env CLI and AttentionInterface.register behavior change are distinct API issues." + "reason": "A broken `transformers env` command is not the same as `AttentionInterface.register` changing registered-function behavior." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Local-files-only network leak is unrelated to the GLM-5 RoPE discussion." + "reason": "Unexpected API requests with `local_files_only=True` is unrelated to GLM-5 RoPE implementation concerns." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled-file import errors and a 5.3 upgrade slowdown are not the same bug." - }, - { - "left": "issue:30333", - "right": "issue:33453", - "accept": false, - "reason": "MLflow reporting on failed jobs and tokenizer-loading regression are unrelated." + "reason": "Compiled Python file errors are unrelated to the slowdown when filtering data after upgrading Transformers." }, { "left": "issue:45850", "right": "issue:47110", "accept": false, - "reason": "Broad model-breakage complaint and special_tokens keys error are too different to merge." + "reason": "A broad complaint that 5.0 breaks many models is too general to equate with the specific `special_tokens` list-vs-dict attribute error." }, { "left": "issue:43874", "right": "issue:46617", "accept": false, - "reason": "Missing image-processor helper and cache FileNotFoundError are separate issues." + "reason": "Missing `get_number_of_image_patches` in GLM46V is unrelated to trust-remote-code cache loading." }, { - "left": "issue:41628", - "right": "issue:42548", + "left": "issue:30333", + "right": "issue:33453", "accept": false, - "reason": "Cannot import different exported classes; no shared concrete root cause is evident." + "reason": "MLflow reporting on failed training jobs is unrelated to tokenizer loading regression." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "A stuck loading report and a documentation confusion issue are not duplicates." + "reason": "Sentence Transformers getting stuck loading is unrelated to confusing `input_ids`/`past_key_values` documentation." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different parts of the package." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Cache/layer `no_grad` decoration and generation after `sync_gpus` are different generation-path issues." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted soft-edge pair (issue:41669 / issue:44246) is plausibly the same import-time slowness problem, while the other pairs are rejected with appropriately cautious reasoning. No overstatement stands out." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -19144,8 +19200,6 @@ "issue:36296", "issue:38617", "issue:42757", - "issue:43010", - "issue:43232", "issue:43299", "issue:43352", "issue:43475", @@ -19157,6 +19211,7 @@ "issue:43827", "issue:43901", "issue:44038", + "issue:44077", "issue:44222", "issue:44230", "issue:44273", @@ -19167,6 +19222,7 @@ "issue:44485", "issue:44492", "issue:44556", + "issue:44861", "issue:44961", "issue:45071", "issue:45125", @@ -19187,7 +19243,6 @@ ], "soft_pairs": [ "issue:44222|issue:44230", - "issue:43010|issue:43232", "issue:46143|issue:46144", "issue:46664|issue:46726", "issue:43502|issue:44222", @@ -19212,233 +19267,234 @@ "issue:45362|issue:45850", "issue:45375|issue:47110", "issue:43525|issue:47110", - "issue:43673|issue:44492", "issue:44485|issue:44961", - "issue:44315|issue:46209", + "issue:43673|issue:44492", "issue:43352|issue:44961", + "issue:44315|issue:46209", "issue:44961|issue:45468", - "issue:44273|issue:45542" + "issue:44273|issue:45542", + "issue:44077|issue:44861" ], - "prepared_review_unit_hash": "2a0069a0a8b9301d045a22c9457e4344f7b613dedf5c29dfcfc2f2f0e2d8a08b", + "prepared_review_unit_hash": "adb099f2c93389d8eb40b4ed249a7c07fb6ef94717109f4a696c75c12ad18d52", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36713, - "estimated_input_tokens": 9179, - "estimated_eval_tokens": 18614 + "serialized_chars": 36727, + "estimated_input_tokens": 9182, + "estimated_eval_tokens": 18620 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of false-positive similarity matches. Many items mention v5 regressions or model-specific breakages, but the paired issues usually concern different code paths, models, or symptom types. No soft pair looks like the same underlying bug/change strongly enough to merge.", - "confidence": 0.82, - "canonical_issue_reason": "issue:45850 is the broadest umbrella issue in the set: it captures the general \u2018v5 breaks many models\u2019 regression theme that overlaps with several of the more specific version-breakage reports, even though it is not a duplicate of them.", + "summary": "This cluster is highly heterogeneous: the issues span unrelated regressions, model-specific bugs, docs, import errors, and CLI failures. No soft pair looks like the same underlying defect with enough specificity to merge.", + "confidence": 0.93, + "canonical_issue_reason": "There is no true duplicate anchor here. If one issue must represent the cluster, 43576 is the broadest and most discussed v5 regression report, with inbound references and overlap in the general upgrade-breakage theme, but it should not be treated as a duplicate hub.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 is the best representative issue for this cluster because it is the most general regression report and the closest to the common high-level theme across the false-positive matches.", + "best_issue_reason": "43576 is the best representative issue only because it is broad, active, and centrally situated among several v5-related regressions. It is still not a good canonical duplicate target because the cluster does not describe one single bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "Both mention FP8/MoE, but one is about save_pretrained and the other about inference support; different concrete failures." - }, - { - "left": "issue:43010", - "right": "issue:43232", - "accept": false, - "reason": "Cache/update-method no_grad concern is unrelated to generation kwargs after sync_gpus." + "reason": "Both mention FP8/moe, but one is a save_pretrained bug and the other is inference support; different code paths and fixes." }, { "left": "issue:46143", "right": "issue:46144", "accept": false, - "reason": "Both are about RoFormer, but one is kwargs propagation and the other is attention-interface usage; different bugs." + "reason": "One is about forwarding kwargs in RoFormer model methods; the other is about the attention implementation not using the attention interface. Related area, but not the same defect." }, { "left": "issue:46664", "right": "issue:46726", "accept": false, - "reason": "Video frame semantics vs missing vision-backend image processor error are unrelated." + "reason": "Both concern vision/video processing, but one is an inconsistent num_frames semantic and the other is a missing vision backend causing an unrecognized image processor error." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "local_files_only network calls are unrelated to FP8 save_pretrained behavior." + "reason": "local_files_only API requests and FP8 moe save_pretrained are unrelated bugs." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Both are version-regression reports, but they affect different models and loading paths." + "reason": "Both are version-v5 compatibility complaints, but one is Qwen3VL MoE loading and the other is checkpoint reload compatibility after upgrading; too broad to be the same bug." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLflow job reporting and tensor-parallel training are unrelated." + "reason": "MLflow reporting on failed jobs and tensor parallel training are unrelated." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs still referencing pipeline() has nothing to do with the unrelated racoon issue." + "reason": "Docs still referencing pipeline() and a random 'racoon' issue are clearly unrelated." }, { "left": "issue:29870", "right": "issue:31795", "accept": false, - "reason": "BART generation error and forward-doc confusion are different issues." + "reason": "Bart generate() UnboundLocalError and documentation confusion for input_ids/past_key_values are different problems." }, { "left": "issue:33357", "right": "issue:33453", "accept": false, - "reason": "MacOS CLIP bus error and tokenizer-loading regression are unrelated." + "reason": "MacOS CLIP bus error and tokenizer loading regression are separate failures." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "Both involve Qwen3 MoE, but one is a broad v5 breakage and the other is a missing tensor-parallel plan; not the same bug." + "reason": "Qwen3-VL-Moe loading breakage and missing _tp_plan for tensor parallelism are not the same concrete bug." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled Python file errors and PretrainedConfig type-checking are unrelated." + "reason": "Compiled Python file errors and PretrainedConfig type checking regressions differ in symptom, cause, and fix." }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "Network requests with local_files_only and racoon are unrelated." + "reason": "local_files_only network leakage and a meaningless 'racoon' issue are unrelated." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "Broken env CLI and missing Qwen2.5-VL import are different problems." + "reason": "Broken transformers env command and missing Qwen2_5_VL import are different CLI/import issues." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "CLI env failure and PretrainedConfig type-checking are unrelated." + "reason": "env command breakage and PretrainedConfig type-checking regression do not share a concrete code path." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "FP8 inference support and DeepSeekV4 weight conversion are unrelated." + "reason": "FP8 inference support for Qwen models and a deepseekv4 weight-conversion question are unrelated." }, { "left": "issue:43576", "right": "issue:47059", "accept": false, - "reason": "Both are CLI-related, but one is a transformers env command regression and the other is an upstream huggingface_hub Typer compatibility failure." + "reason": "Broken env command and CLI test failure with huggingface_hub are distinct CLI incompatibilities." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "An import error in huggingface_hub may be adjacent, but there is not enough evidence that it is the same bug as the broken env command." + "reason": "Missing is_offline_mode import and the transformers env command failure are unrelated." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Forward-doc confusion and a missing configuration_utils symbol are unrelated." + "reason": "Forward-doc confusion and missing layer_type_validation import are different issue types and code paths." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor-parallel training bug and TextClassificationPipeline doc mismatch are unrelated." + "reason": "Tensor parallel training bug and TextClassificationPipeline docs mismatch are not the same underlying defect." }, { "left": "issue:43475", "right": "issue:46617", "accept": false, - "reason": "SAM3 video encoder attribute error and trust_remote_code cache FileNotFoundError are unrelated." + "reason": "Missing fpn_position_embeddings in SAM 3 Video and trust_remote_code cache FileNotFoundError are unrelated." }, { "left": "issue:44351", "right": "issue:46617", "accept": false, - "reason": "HybridCache import failure and cached trust_remote_code FileNotFoundError are different code paths." + "reason": "Missing HybridCache import and cache-loading FileNotFoundError are different import/loading failures." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger Kernel application and DSA indexer ReLU discussion are unrelated." + "reason": "Liger Kernel not applied in model_init and a DSA indexer ReLU discussion are unrelated." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "Specific Qwen3.5-35B chat crash is not the same concrete bug as the broad v5 breaks-many-models report." + "reason": "Qwen3.5-35B chat crash and the vague '5.0 breaks too many models' complaint are not the same concrete bug." }, { "left": "issue:45375", "right": "issue:47110", "accept": false, - "reason": "Qwen3.5 MoE config field loss and special_tokens list/dict error are unrelated." + "reason": "Missing deepstack_visual_indexes in a specific config and special_tokens list/keys error are different model/config bugs." }, { "left": "issue:43525", "right": "issue:47110", "accept": false, - "reason": "Llama4Config pad_token_id missing and special_tokens list/dict error are unrelated." + "reason": "Llama4Config missing pad_token_id and special_tokens list/keys error are separate attribute/type problems." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation and the unrelated 'racoon' issue are not duplicates." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "Chunked-prefill cache missing and a cache-strategy typo are unrelated." + "reason": "Missing GenerationMixin cache during chunked_prefill and a cache-strategy typo are not the same issue." }, { - "left": "issue:44485", + "left": "issue:43352", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation and racoon are unrelated." + "reason": "Nemotron Flash Attention 2.0 support and the unrelated 'racoon' issue are different problems." }, { "left": "issue:44315", "right": "issue:46209", "accept": false, - "reason": "Liger Kernel/model_init and FP8 inference on B200 are unrelated." - }, - { - "left": "issue:43352", - "right": "issue:44961", - "accept": false, - "reason": "Nemotron FlashAttention support and racoon are unrelated." + "reason": "Liger Kernel not applied in model_init and FP8 inference on B200 are distinct performance/runtime issues." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "Racoon and Gemma-4 audio positional encoding are unrelated." + "reason": "The 'racoon' issue has no meaningful technical overlap with Gemma4AudioRelPositionalEncoding." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Lazy loading and tensorboard-without-TensorFlow backend error are unrelated." + "reason": "Lazy loading behavior and undefined tf backend due to tensorboard-only install are unrelated." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "patchtsmixer post_init policy and _get_tied_weight_keys list-handling crash are separate code changes." } ] }, @@ -19469,13 +19525,13 @@ "issue:42890", "issue:43298", "issue:43408", + "issue:43519", "issue:43525", "issue:43572", "issue:43576", "issue:43673", "issue:43867", "issue:43976", - "issue:44077", "issue:44246", "issue:44273", "issue:44295", @@ -19505,16 +19561,16 @@ "issue:46172", "issue:46218", "issue:46328", + "issue:46655", "issue:46664", "issue:47059", "issue:47110", "issue:47176" ], "soft_pairs": [ - "issue:44077|issue:44861", "issue:43867|issue:47110", - "issue:45381|issue:46664", "issue:44623|issue:45230", + "issue:45381|issue:46664", "issue:44273|issue:45003", "issue:44485|issue:46664", "issue:41669|issue:43673", @@ -19531,8 +19587,8 @@ "issue:43408|issue:44315", "issue:44492|issue:45850", "issue:41669|issue:46218", - "issue:44492|issue:46328", "issue:44295|issue:45230", + "issue:44492|issue:46328", "issue:43976|issue:44393", "issue:44297|issue:45230", "issue:42890|issue:44908", @@ -19546,252 +19602,253 @@ "issue:43976|issue:46218", "issue:43298|issue:46095", "issue:43976|issue:47176", - "issue:31795|issue:33453" + "issue:31795|issue:33453", + "issue:43519|issue:46655" ], - "prepared_review_unit_hash": "ecf465bd30488aafdc0f5a9f4146e6bdb3e8ea4227341413de6fb559d5717d39", + "prepared_review_unit_hash": "bfcc5a85855ff685deb7531150187f48d25da199869870c9fd23b9188fb6436e", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 36, - "serialized_chars": 37322, - "estimated_input_tokens": 9331, - "estimated_eval_tokens": 18918 + "serialized_chars": 37880, + "estimated_input_tokens": 9470, + "estimated_eval_tokens": 19196 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a grab bag of unrelated Transformers issues. A handful touch v5/regression/loading themes, but none of the soft pairs look like the same concrete bug, so I would keep them separate. The best anchor issue is the checkpoint reload regression after upgrading to v5.", - "confidence": 0.52, - "canonical_issue_reason": "Issue 44556 is the strongest anchor: it reports a concrete, reproducible v4.57\u2192v5.2/v5.3 checkpoint reload regression with clear version context, unlike the others which are either broader, model-specific, or vague.", + "summary": "The cluster is mostly a set of unrelated issues with a few loose v5 migration/loading/import-time similarities; none of the soft pairs looks like the same underlying bug or change, so all soft edges should be rejected. The best overall issue anchor is the checkpoint reload regression after upgrading to v5, since it is concrete and broadly representative of the migration breakages in this set.", + "confidence": 0.7, + "canonical_issue_reason": "issue:44556 is the clearest, most concrete v5 regression in the set: it describes a real checkpoint reload failure after upgrading from v4.57 to v5.2/v5.3, which makes it a useful duplicate-triage anchor for neighboring migration/loading reports.", "canonical_pr_reason": null, - "best_issue_reason": "Same as canonical: it is the most actionable and triage-friendly regression report in the cluster, with a specific failure mode and upgrade path.", + "best_issue_reason": "issue:44556 is the strongest global representative among these issues because it is specific, actionable, and centrally about v5 compatibility/loading failures rather than a vague or model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44077", - "right": "issue:44861", - "accept": false, - "reason": "Config validation for patchtsmixer and a helper crash on tied-weight keys are different defects." - }, { "left": "issue:43867", "right": "issue:47110", "accept": false, - "reason": "State-dict loading failure and a special_tokens dict/list AttributeError are unrelated code paths." + "reason": "Different bugs: state_dict loading error vs special_tokens/list-vs-dict handling." }, { - "left": "issue:45381", - "right": "issue:46664", + "left": "issue:44623", + "right": "issue:45230", "accept": false, - "reason": "Wrong vision_position_ids and inconsistent num_frames semantics are both video-related but not the same bug." + "reason": "processor.save_pretrained missing files is specific; 45230 is just a generic bug report." }, { - "left": "issue:44623", - "right": "issue:45230", + "left": "issue:45381", + "right": "issue:46664", "accept": false, - "reason": "Missing files on processor save_pretrained is specific; \"Bug report\" is too generic to match it." + "reason": "Same model family, but distinct video-processing bugs: wrong vision_position_ids vs num_frames semantics." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Both involve loading/import behavior, but one is a general lazy-loading complaint and the other is a specific sys.modules access bug." + "reason": "Both touch loading/import behavior, but one is lazy loading not working and the other is unsafe sys.modules access." }, { "left": "issue:44485", "right": "issue:46664", "accept": false, - "reason": "GLM-5 RoPE implementation and video num_frames semantics are unrelated." + "reason": "GLM-5 RoPE implementation is unrelated to SmolVLM video padding/frames issues." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Import-time slowdown from wildcard imports is different from a chunked_prefill cache regression." + "reason": "Import-* slowdown is a performance issue; 43673 is a generation cache regression during chunked prefill." }, { "left": "issue:44466", "right": "issue:44556", "accept": false, - "reason": "Serialization inconsistency and checkpoint reload failure after upgrade are related only at a broad level, not the same defect." + "reason": "Tied-weight serialization inconsistency is a different code path from checkpoint reload compatibility after v5 upgrade." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "Missing pad_token_id in Llama4Config is a concrete config bug, while the other issue is a broad complaint about many models breaking in v5." + "reason": "Llama4 pad_token_id AttributeError is model-specific; 45850 is only a broad 'many models break' complaint." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "A missing pad_token_idx field is a specific config regression, not the same as a generic v5 breakage report." + "reason": "StableLm pad_token_idx missing is a specific config bug, not the same as the broad v5 breakage report." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "2D bounding-box hallucination/error and an infinite loop on a specific image are different failure modes." + "reason": "Different Qwen3_5 issues: 2D bbox hallucination/error vs infinite loop on a specific image." }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "A multimodal output bug and a PretrainedConfig type-checking regression are not the same underlying problem." + "reason": "Qwen3-VL bounding-box output bug is unrelated to PretrainedConfig type-checking failures." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both mention upgrade incompatibilities, but one is checkpoint reload after v5 upgrade and the other is remote-code/meta-init incompatibility for InternVL2." + "reason": "Both are v5 compatibility issues, but one is generic checkpoint reload after upgrade and the other is InternVL2 remote-code/meta-init incompatibility." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 API guidance and a typo in cache strategies are unrelated." + "reason": "TF32 behavior API change is unrelated to a cache-strategy typo." }, { "left": "issue:45850", "right": "issue:47059", "accept": false, - "reason": "A broad \"v5 breaks many models\" complaint and a CLI/huggingface_hub compatibility failure are different issues." + "reason": "Broad v5 breakage claim is not the same as the CLI test failure with huggingface_hub 1.22.0." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Slow import and lazy-loading malfunction overlap conceptually, but the reports do not point to the same bug." + "reason": "Both involve import/lazy-loading, but one is intermittent slow import and the other is lazy loading not functioning correctly." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Pipeline save_pretrained support and tied-weight-key handling are separate save/load paths." + "reason": "Saving Pipeline objects is a different problem from _get_tied_weight_keys crashing on a list." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "A model-type warning for sam3_* and Liger Kernel not applying under model_init are unrelated." + "reason": "SAM3 tracker/model-type warning is unrelated to Liger Kernel not being applied via model_init." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "A typo in cache strategy docs/code is not the same as a broad v5 breakage complaint." + "reason": "A typo in cache strategies is not the same bug as the broad 'many models break in v5' report." }, { "left": "issue:41669", "right": "issue:46218", "accept": false, - "reason": "Import-star performance regression and a module import failure for Qwen3_5 are different problems." + "reason": "Import-* slowdown is a performance regression; 46218 is a missing Qwen3_5 import/export path." }, { - "left": "issue:44492", - "right": "issue:46328", + "left": "issue:44295", + "right": "issue:45230", "accept": false, - "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad performance issue are unrelated." + "reason": "position_ids buffer access error is specific; 45230 is too generic to indicate the same issue." }, { - "left": "issue:44295", - "right": "issue:45230", + "left": "issue:44492", + "right": "issue:46328", "accept": false, - "reason": "Position_ids buffer access bug and a generic bug report are not the same concrete defect." + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad inefficiency are unrelated." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python version support regression and Qwen3-VL bbox output problems are unrelated." + "reason": "Python version compatibility problem is not the same as a Qwen3-VL output error." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer_class mismatch on save and a generic bug report do not match." + "reason": "tokenizer_class mismatch in save_pretrained is specific; 45230 is just a generic bug report." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "Missing set_seed in tests and inverse_sqrt scheduler kwargs handling are unrelated." + "reason": "Missing set_seed in tests and inverse_sqrt scheduler kwargs being ignored are different test vs scheduler bugs." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "AutoProcessor kwargs passthrough and scheduler kwargs handling are different subsystems." + "reason": "AutoProcessor kwargs passthrough to cached_file is unrelated to scheduler kwarg handling." }, { "left": "issue:44908", "right": "issue:45071", "accept": false, - "reason": "Scheduler kwargs being ignored and PretrainedConfig type checking are distinct defects." + "reason": "Scheduler kwargs bug and PretrainedConfig type-checking regression are unrelated subsystems." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Import performance regression and a small testing_utils bug are not the same issue." + "reason": "Import-star performance regression is different from a testing_utils bug." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "The transformers env command and scheduler keyword handling are unrelated." + "reason": "Broken transformers env command is unrelated to inverse_sqrt scheduler kwargs." }, { "left": "issue:43576", "right": "issue:47176", "accept": false, - "reason": "Broken env command and shell-command allowlist bypass are different security/CLI concerns." + "reason": "CLI env command breakage is unrelated to a shell-operator allowlist bypass security issue." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-star slowdown and inability to import HybridCache are different import-path failures." + "reason": "Import-star slowdown is unrelated to missing HybridCache import." }, { "left": "issue:46018", "right": "issue:46172", "accept": false, - "reason": "DeepSeek-V4 gating behavior and a comb.to(dtype).transpose(-1, -2) bug are unrelated." + "reason": "DeepSeek-V4 gating issue is not the same as the comb.to(...).transpose(...) bug." }, { "left": "issue:43976", "right": "issue:46218", "accept": false, - "reason": "Python 3.9+ compatibility and Qwen3_5 import failure are different issues." + "reason": "Python version incompatibility and Qwen3_5 import failure are different failure modes." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching not supporting audio models and a weight-conversion-script question are not duplicates." + "reason": "Continuous batching audio support is unrelated to DeepSeek-V4 weight conversion format/support." }, { "left": "issue:43976", "right": "issue:47176", "accept": false, - "reason": "Python version compatibility and a shell operator bypass are unrelated." + "reason": "Python compatibility issue is unrelated to a shell command-substitution allowlist bypass." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion around forward arguments and tokenizer loading regression are clearly different problems." + "reason": "Documentation confusion in model.forward is unrelated to tokenizer loading regression." + }, + { + "left": "issue:43519", + "right": "issue:46655", + "accept": false, + "reason": "Incorrect timestamp calculation in Qwen3VL is unrelated to SmolVLM video padding math." } ] }, @@ -19827,6 +19884,7 @@ "issue:42371", "issue:42913", "issue:43066", + "issue:43298", "issue:43352", "issue:43502", "issue:43519", @@ -19857,7 +19915,6 @@ "issue:45656", "issue:45850", "issue:46218", - "issue:46655", "issue:46894", "issue:47059", "issue:47110", @@ -19865,7 +19922,6 @@ "issue:47177" ], "soft_pairs": [ - "issue:43519|issue:46655", "issue:44315|issue:44829", "issue:44485|issue:47176", "issue:45230|issue:45656", @@ -19876,8 +19932,8 @@ "issue:43519|issue:44485", "issue:44485|issue:47177", "issue:44273|issue:44908", - "issue:44861|issue:47110", "issue:31795|issue:34689", + "issue:44861|issue:47110", "issue:43519|issue:44908", "issue:41669|issue:43704", "issue:43576|issue:46218", @@ -19910,312 +19966,314 @@ "issue:31515|issue:38617", "issue:29870|issue:31515", "issue:30333|issue:31515", - "issue:31515|issue:36296" + "issue:31515|issue:36296", + "issue:43298|issue:44230", + "issue:31515|issue:31795" ], - "prepared_review_unit_hash": "8960120db9b6a423566989896346a96aedb15b676d6de54c5a458cf7433aa857", + "prepared_review_unit_hash": "b5f65af377cb6913a8f7a957e8b763ebe0100aeaa7f9f5b81f457320bcde661f", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 46, - "serialized_chars": 40653, - "estimated_input_tokens": 10164, - "estimated_eval_tokens": 20584 + "soft_pair_count": 47, + "serialized_chars": 40919, + "estimated_input_tokens": 10230, + "estimated_eval_tokens": 20716 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very heterogeneous. The soft pairs are mostly topic-level similarity (e.g. Flash Attention, processors, tokenizers, model loading), but they do not describe the same concrete bug or change, so none should be merged as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "No clear canonical issue: the items cover unrelated bugs/features across model loading, processors, tokenizers, training, CLI, and security, and the soft matches are only superficial.", + "summary": "All listed soft pairs appear to be unrelated issues across different models, features, or failure modes, so I reject every soft edge. There are no PRs in the set.", + "confidence": 0.93, + "canonical_issue_reason": "If one issue must be treated as the representative item, issue 44829 is the clearest, most concrete bug report: it names a specific model, a specific setting (`flash_attention_3`), and a concrete training failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "No issue stands out as a safe global representative because the cluster is not a single defect family.", + "best_issue_reason": "Issue 44829 is the strongest standalone issue for triage quality because it is specific, reproducible, and describes an exact symptom rather than a broad or vague request.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43519", - "right": "issue:46655", - "accept": false, - "reason": "Both are processor-related media bugs, but they affect different models and different calculations (timestamp vs padding)." - }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "One is about Liger kernel application during model_init; the other is about Flash Attention 3 causing degenerate training. Different code paths." + "reason": "Different code paths: model_init/Liger Kernel application vs Flash Attention 3 training degeneration." }, { "left": "issue:44485", "right": "issue:47176", "accept": false, - "reason": "GLM-5 RoPE implementation and a shell-command allowlist bypass are unrelated." + "reason": "Unrelated topics: GLM-5 RoPE implementation vs shell-command allowlist bypass." }, { "left": "issue:45230", "right": "issue:45656", "accept": false, - "reason": "Generic bug report versus a deepspeed optimizer-step double-call bug; no shared underlying fix." + "reason": "Generic bug report vs a specific DeepSpeed double optimizer-step bug; no shared underlying change." }, { "left": "issue:45468", "right": "issue:46894", "accept": false, - "reason": "Both mention training/model behavior, but they target different models and different bugs (audio positional encoding vs loss double-shift)." + "reason": "Different models and symptoms: Gemma-4 audio positional encoding vs CohereASR double-shift training loss bug." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "DSA indexer/ReLU discussion is unrelated to Gemma audio positional encoding." + "reason": "DSA/ReLU discussion is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:43976", "right": "issue:47177", "accept": false, - "reason": "Python version compatibility and tar/zip slip RCE are entirely different problems." + "reason": "Python-version compatibility issue vs tar/zip-slip RCE security bug." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken env command and lazy loading failure are separate v5 issues with different failure modes." + "reason": "Both mention v5, but one is `transformers env` and the other is lazy loading; different failures." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Both are model/component implementation bugs, but one is Qwen3VL timestamp math and the other is GLM-5 RoPE logic." + "reason": "Timestamp calculation bug in Qwen3VL processor is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:44485", "right": "issue:47177", "accept": false, - "reason": "RoPE implementation bug and archive extraction RCE are unrelated." + "reason": "RoPE implementation bug vs command-substitution allowlist bypass." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading and inverse_sqrt scheduler kwargs are unrelated subsystems." + "reason": "Lazy loading regression and inverse_sqrt scheduler kwargs are different subsystems and symptoms." }, { - "left": "issue:44861", - "right": "issue:47110", + "left": "issue:31795", + "right": "issue:34689", "accept": false, - "reason": "Tied-weight key handling and special_tokens dict/list handling are different API surfaces." + "reason": "Documentation confusion about forward args vs a model-loading regression in Llama 3.2 vision." }, { - "left": "issue:31795", - "right": "issue:34689", + "left": "issue:44861", + "right": "issue:47110", "accept": false, - "reason": "Documentation confusion for forward arguments is not the same as a model-loading regression for Llama 3.2 Vision." + "reason": "Different crashes in different helpers: tied weights keys vs special_tokens dict/list handling." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Processor timestamp calculation and scheduler parameter passing have no shared code path." + "reason": "Processor timestamp math is unrelated to scheduler keyword propagation." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import * cleanup performance work is unrelated to a VRAM leak in multi-threaded inference." + "reason": "Import-star cleanup/performance issue vs VRAM leak in Qwen3ForCausalLM." }, { "left": "issue:43576", "right": "issue:46218", "accept": false, - "reason": "Broken env CLI and missing Qwen3_5ForCausalLM import are separate failures." + "reason": "`transformers env` command failure is unrelated to Qwen3_5 import errors." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both are tokenizer-related, but one is a v4-to-v5 behavior shift and the other is a specific decoder-type bug." + "reason": "Both concern tokenizer v5 behavior, but the reported bugs are different enough to keep separate." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Different model internals: GLM-5 RoPE vs Gemma4AudioRelPositionalEncoding." + "reason": "GLM-5 RoPE implementation and Gemma-4 audio positional encoding are distinct model-specific bugs." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "A VRAM leak across dataloader threads is unrelated to RoPE implementation." + "reason": "VRAM leak in Qwen3ForCausalLM is unrelated to RoPE implementation." }, { "left": "issue:45850", "right": "issue:47177", "accept": false, - "reason": "Broad compatibility complaints do not match a shell-command injection/RCE vulnerability." + "reason": "Broad regressions after v5 vs a tar/zip-slip arbitrary file-write RCE." }, { "left": "issue:45850", "right": "issue:47176", "accept": false, - "reason": "Generic breakage across models is not the same as a command allowlist bypass." + "reason": "General breakage report vs shell-operator allowlist bypass; different vulnerabilities." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API guidance and VRAM leakage are unrelated." + "reason": "TF32 configuration guidance is unrelated to a VRAM leak bug." }, { "left": "issue:43352", "right": "issue:47177", "accept": false, - "reason": "Flash Attention support for NemotronH and archive-slip RCE are unrelated." + "reason": "Flash Attention 2.0 support gap for Nemotron is unrelated to file-write RCE." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch and audio positional encoding are different bugs." + "reason": "CPU offload device mismatch is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "local_files_only network-escape bug and timestamp calculation bug are unrelated." + "reason": "Local-files-only network request bug vs timestamp calculation bug." }, { "left": "issue:44861", "right": "issue:47059", "accept": false, - "reason": "Tied-weight key handling is unrelated to the CLI completion attribute regression." + "reason": "Different failures: tied-weight key handling vs CLI completion attribute error." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp math and missing processor save files are distinct processor bugs." + "reason": "Timestamp bug in Qwen3VL processor vs missing files in processor.save_pretrained." }, { "left": "issue:45850", "right": "issue:46218", "accept": false, - "reason": "Broad model breakage and missing Qwen3_5 import are not the same defect." + "reason": "General v5 regressions vs Qwen3_5 import failure; not the same bug." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "A torch.split indexer issue and Gemma audio positional encoding are unrelated." + "reason": "torch.split return-value bug is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:43704", "right": "issue:45341", "accept": false, - "reason": "VRAM leak in dataloader threads is unrelated to a testing utility bug." + "reason": "VRAM leak in model usage vs a testing_utils typo." }, { "left": "issue:28348", "right": "issue:45250", "accept": false, - "reason": "Both mention Flash Attention 2.0, but one is GPT2LMHeadModel support and the other is a generic request; not the same concrete change." + "reason": "GPT2-specific Flash Attention 2.0 support request vs a generic Flash Attention 2.0 issue title." }, { "left": "issue:43352", "right": "issue:46218", "accept": false, - "reason": "Flash Attention support for NemotronH and a missing Qwen3_5 import are different issues." + "reason": "Nemotron Flash Attention 2.0 support vs Qwen3_5 import failure." }, { "left": "issue:44351", "right": "issue:46218", "accept": false, - "reason": "Missing HybridCache import and missing Qwen3_5ForCausalLM import are similar in form but unrelated symbols/code paths." + "reason": "HybridCache import failure and Qwen3_5 import failure are different modules/problems." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Tied-weight key crash and chat-template tool-call crash are unrelated." + "reason": "Tied-weight-key crash vs chat-template crash with tool calls; different code paths." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "FP8 support request and missing processor save files are different feature/bug scopes." + "reason": "FP8 support request vs missing files on processor.save_pretrained." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs bug and a placeholder 'Unemployment' issue are clearly unrelated." + "reason": "Scheduler kwargs bug is unrelated to the placeholder/irrelevant issue." }, { "left": "issue:44998", "right": "issue:47177", "accept": false, - "reason": "Placeholder issue text has no relation to an archive-slip security bug." + "reason": "Irrelevant placeholder issue vs security RCE." }, { "left": "issue:44998", "right": "issue:47176", "accept": false, - "reason": "Placeholder issue text has no relation to a command-substitution allowlist bypass." + "reason": "Irrelevant placeholder issue vs shell-operator bypass." }, { "left": "issue:28348", "right": "issue:43352", "accept": false, - "reason": "Both are Flash Attention 2.0-related, but they target different models and different support gaps." + "reason": "Same broad Flash Attention theme, but different models and concrete requests." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "Cache-strategy typo and placeholder issue are unrelated." + "reason": "Cache-strategy typo vs irrelevant placeholder issue." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression and placeholder issue are unrelated." + "reason": "CLIPOutput attentions regression vs irrelevant placeholder issue." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "FP8 support request and Qwen3.5 from_pretrained regression are different concerns." + "reason": "FP8 support request for qwen3-vl/qwen3.5 moe vs from_pretrained error in a different version/model." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow checkpoint loading and missing layer_type_validation import are unrelated." + "reason": "Slow checkpoint loading vs ImportError for `layer_type_validation`; unrelated symptoms." }, { "left": "issue:29870", "right": "issue:31515", "accept": false, - "reason": "BART generate() decoder mask bug and slow from_pretrained loading are different code paths." + "reason": "BART generate decoder_attention_mask bug vs slow from_pretrained checkpoint loading." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLFlow reporting on failed jobs and checkpoint loading speed are unrelated." + "reason": "MLflow job-finalization bug vs checkpoint loading performance issue." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Checkpoint loading performance and tensor-parallel training bug are not the same defect." + "reason": "from_pretrained slowdown and tensor-parallel training bug are different problems." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching support for audio models is unrelated to FP8 qwen3-vl/qwen3.5 support." } ] }, @@ -20239,23 +20297,17 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-533", "nodes": [ - "issue:31515", - "issue:31795", "issue:37934", "issue:42936", "issue:43240", - "issue:43298", "issue:43668", "issue:43717", "issue:43813", "issue:44031", - "issue:44230", "issue:46018", "issue:46144" ], "soft_pairs": [ - "issue:43298|issue:44230", - "issue:31515|issue:31795", "issue:43717|issue:46144", "issue:42936|issue:46018", "issue:43668|issue:44031", @@ -20269,118 +20321,106 @@ "issue:37934|issue:43668", "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "b702a52c72c3edd9c29ff10850ba0bcf9311249d37d7b8655397134217926900", + "prepared_review_unit_hash": "d060d6efa04dc84d3ebf55f6253a97de7c60dade99b33fe9b3ee0279a19b4b25", "component_budget": { "node_count": 533, "item_count": 533, "soft_pair_count": 1964, - "serialized_chars": 1018554, - "estimated_input_tokens": 254639, - "estimated_eval_tokens": 509534 + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 }, "budget": { - "node_count": 13, - "item_count": 13, - "soft_pair_count": 14, - "serialized_chars": 11438, - "estimated_input_tokens": 2860, - "estimated_eval_tokens": 5976 + "node_count": 9, + "item_count": 9, + "soft_pair_count": 12, + "serialized_chars": 8611, + "estimated_input_tokens": 2153, + "estimated_eval_tokens": 4562 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the items span unrelated bugs, feature requests, docs, typing, and model-architecture questions. None of the soft pairs look like the same underlying issue/change, so this is not a duplicate cluster.", + "summary": "These issues are not duplicates: they span unrelated models, configs, warnings, and typo/behavior bugs across different code paths. No pair looks like the same underlying defect.", "confidence": 0.97, - "canonical_issue_reason": "No canonical issue: the items do not share one underlying bug or fix path; they are only loosely related by being ML/model-adjacent GitHub issues.", - "canonical_pr_reason": null, - "best_issue_reason": "Issue 42936 is the most concrete standalone bug report in the set, but it is still unrelated to the other items and should not be treated as a cluster representative.", - "best_pr_reason": null, + "canonical_issue_reason": "No single canonical issue stands out; the set is heterogeneous and does not cluster around one bug or feature change.", + "canonical_pr_reason": "No pull requests are present.", + "best_issue_reason": "Issue 42936 is the strongest standalone report because it names a specific model, a concrete missing-use bug, and has the clearest actionable symptom.", + "best_pr_reason": "No pull requests are present.", "soft_edge_verdicts": [ - { - "left": "issue:43298", - "right": "issue:44230", - "accept": false, - "reason": "Both are inference-related, but one is about continuous batching for audio models and the other is fp8 support for specific Qwen models; different features, different failure modes." - }, - { - "left": "issue:31515", - "right": "issue:31795", - "accept": false, - "reason": "One reports slow checkpoint loading in from_pretrained, the other asks for documentation clarification about model.forward arguments." - }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Mamba init_weights behavior and RoFormer attention-interface usage are different model implementations and unrelated code paths." + "reason": "Different model families and different bugs: Mamba init-weights behavior vs RoFormer attention interface usage." }, { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Mask2former ignore_value handling and DeepSeek-V4 shared-expert gating are distinct model bugs in different architectures." + "reason": "Mask2Former ignore_value handling is unrelated to DeepSeek-V4 shared expert gating." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "A ModernBERT config type-hint issue is unrelated to tokenizer regex warning spam across versions." + "reason": "A type-hint issue in ModernBERTConfig is not the same as a tokenizer regex warning regression." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "One concerns a Mask2former parameter not used after definition; the other concerns kwargs not being forwarded to a loss function." + "reason": "One is about a model-specific ignore_value parameter; the other is about kwargs not being forwarded to a loss function." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Loss-function kwargs forwarding and RoFormer attention implementation are separate bugs in different subsystems." + "reason": "Loss-function kwargs forwarding and RoFormer attention-interface usage are separate code paths and bugs." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Mask2former ignore_value handling is unrelated to tokenizer regex warning behavior." + "reason": "Mask2Former ignore_value handling does not match a tokenizer warning regression." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "These target different model components: Mask2former loss handling vs RoFormer attention plumbing." + "reason": "These affect different architectures and distinct implementation layers." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "A model parameter usage bug and a typo in quantization conversion code are not the same underlying issue." + "reason": "A Mask2Former behavior bug is not the same as a quantization-attribute typo." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "Incorrect type hints in ModernBERTConfig do not overlap with a typo in quantization operation naming." + "reason": "Incorrect type annotation in config is unrelated to a typo in a quantization field name." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Loss kwargs propagation and a config type-hint correction are unrelated." + "reason": "Passing kwargs to loss calculation and fixing a config type hint are different defects." }, { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "A question about Llama4TextL2Norm semantics is different from a ModernBERT config annotation bug." + "reason": "RMS-norm implementation question for Llama4TextL2Norm is unrelated to a ModernBERT norm_eps type hint." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Mask2former ignore_value usage is a runtime/model bug; ModernBERT norm_eps is just a type-hint issue." + "reason": "Mask2Former ignore_value usage and ModernBERTConfig norm_eps typing do not describe the same bug." } ] }, @@ -20413,7 +20453,7 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "55c8d09ffaaf340a790528e8d002ae946d59de999a00d281e47a43a32a047549", + "prepared_review_unit_hash": "cac86e5bce768c272656a354c1b0666c7281c2bad37cb1cb977ef06d66c2dadf", "component_budget": { "node_count": 3, "item_count": 3, @@ -20435,7 +20475,7 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three issues are unrelated: truncation behavior in tokenizers, ReduceLROnPlateau-style scheduling, and Wav2Vec2 word offsets. This cluster does not contain a duplicate set, so there is no canonical issue to keep as the representative.", + "summary": "These three issues are unrelated feature requests: token truncation behavior, ReduceLROnPlateau-like scheduler functionality, and Wav2Vec2 word time offsets. The cluster looks spurious; no duplicates to merge.", "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, @@ -20446,19 +20486,19 @@ "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Different topics and code paths: tokenizer truncation control vs. speech model word timestamp extraction." + "reason": "Different areas and requests: tokenizer truncation direction vs Wav2Vec2 word boundary timestamps." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "No shared underlying bug or feature: truncation direction option vs. learning-rate scheduler behavior." + "reason": "Unrelated functionality: truncation strategy option vs learning-rate scheduler behavior." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "Unrelated requests in different subsystems: LR scheduling utility vs. Wav2Vec2 timing offsets." + "reason": "Different code paths and goals: scheduler adaptation vs extracting token/word offsets from audio model outputs." } ] }, @@ -20490,7 +20530,7 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "73dc4b960f1530e903f4e3d51df433789cacf4c633a0423cb5255ea03f5b3331", + "prepared_review_unit_hash": "123b400b83b96fcddc280b4b2633058af1b4c23ae9f3712be93748ff166c9663", "component_budget": { "node_count": 3, "item_count": 3, @@ -20512,24 +20552,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The three issues are related only at a very broad Hugging Face/transformers level, but they describe different problems: Tapas tokenization mismatch, private model download/auth error, and a generic inability to load bert-base-uncased. None look like duplicates of the same underlying bug.", + "summary": "The cluster is not a true duplicate set: one issue is about Tapas tokenization, and the other two are about model loading/download failures, but they point to different failure modes. Issue 16618 is the broadest representative of the loading-error side, while 15980 is specifically about private-model auth messaging.", "confidence": 0.89, - "canonical_issue_reason": "Issue #16618 is the broadest and most central user-facing failure mode among the three; it has the most discussion and a direct model-loading symptom, making it the best representative if one must be chosen.", + "canonical_issue_reason": "Issue 16618 is the most general representative of the cluster\u2019s model-loading problems; the other two are either unrelated (13244) or narrower/auth-specific (15980).", "canonical_pr_reason": null, - "best_issue_reason": "#16618 is the strongest standalone issue because it describes a concrete, reproducible loading failure and is closer to the generic model-loading problem space than the other two, which are narrower and clearly different.", + "best_issue_reason": "16618 is the best issue to anchor the cluster because it captures the broader user-facing load failure theme, whereas 15980 is a specific login/error-message case and 13244 is a different tokenization problem.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tapas tokenization mismatch and failing to load bert-base-uncased are different code paths and symptoms; this is not the same bug." + "reason": "Tapas tokenization mismatch is a different underlying bug from failing to load bert-base-uncased; they affect different code paths and symptoms." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "A private-model authentication/error-message issue is distinct from a generic inability to load a public model; they do not appear to be duplicates." + "reason": "15980 is about an unauthenticated private-model download and bad error messaging, while 16618 is a public-model load failure; not the same concrete bug." } ] }, @@ -20561,7 +20601,7 @@ "issue:19631|issue:22372", "issue:19631|issue:20362" ], - "prepared_review_unit_hash": "6cf63584d699c51f167127a5bc7a44b26637e8aae2247fb7208effe5f2d4a743", + "prepared_review_unit_hash": "d37841b74d599ded462731174b3434f6e1b58fa16d0709548690e4b1ea7a321d", "component_budget": { "node_count": 3, "item_count": 3, @@ -20583,24 +20623,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate feature requests to add different model architectures (EDSR/MDSR, FlexiBERT, Restormer). They are not duplicates of one another; the similarity is only that they are all model-addition requests.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 19631 is the best canonical issue for the cluster because it is the broadest of the three and covers two related model additions, making it the most representative request among otherwise unrelated feature asks.", + "summary": "These are three separate feature requests for different model additions (EDSR/MDSR, FlexiBERT, Restormer). They share only a generic \"Add \" shape, so there is no duplicate relationship here.", + "confidence": 0.97, + "canonical_issue_reason": "issue:19631 is the oldest and most discussed item, so it is the best cluster representative, but it is still a distinct request rather than a duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 19631 is the best overall issue in this set because it is the most substantial and general of the feature requests, with the most discussion and the widest scope.", + "best_issue_reason": "issue:19631 is the strongest representative by age and discussion activity; however, all three issues should remain separate because they request different models.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:19631", - "right": "issue:22372", + "right": "issue:20362", "accept": false, - "reason": "EDSR/MDSR and Restormer are different architectures with different implementation targets; this is only broad topical similarity, not the same underlying request." + "reason": "Different model targets: EDSR/MDSR vs FlexiBERT. The similarity is only the generic request pattern, not the same underlying issue." }, { "left": "issue:19631", - "right": "issue:20362", + "right": "issue:22372", "accept": false, - "reason": "EDSR/MDSR and FlexiBERT are unrelated model additions; they do not describe the same feature or code change." + "reason": "Different model targets: EDSR/MDSR vs Restormer. No shared code-path or concrete duplicate problem is evident." } ] }, @@ -20690,322 +20730,322 @@ "issue:47299|issue:47402", "issue:46226|issue:47299" ], - "prepared_review_unit_hash": "a06fa69153b103e172bfab26e22da0fc1a5380e9a7ee3800d7ae198a867eccb2", + "prepared_review_unit_hash": "bcc7d539f32b36453c8b5b015d51b06f1183f2fbe4908db89edce127392aa048", "component_budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 48, - "serialized_chars": 26226, - "estimated_input_tokens": 6557, - "estimated_eval_tokens": 13370 + "serialized_chars": 26216, + "estimated_input_tokens": 6554, + "estimated_eval_tokens": 13364 }, "budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 48, - "serialized_chars": 26226, - "estimated_input_tokens": 6557, - "estimated_eval_tokens": 13370 + "serialized_chars": 26216, + "estimated_input_tokens": 6554, + "estimated_eval_tokens": 13364 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is not a true duplicate set: it mixes two unrelated feature requests (Flash Attention 2 and GGUF support) with many superficially similar i18n/translation issues. The translation tickets share template wording, but they are for different languages or noisy placeholder variants, so I would not merge any soft edge.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43834 is the most central of the i18n template-style issues and has the broadest soft overlap with the other placeholder translation tickets, though the cluster overall is not genuinely duplicative.", + "summary": "This cluster mixes one unrelated feature request pair (Flash Attention vs GGUF) with many separate docs-translation issues for different languages. The i18n items are superficially similar, but they are not the same underlying change, so none of the soft edges should be merged.", + "confidence": 0.59, + "canonical_issue_reason": "issue:43834 is the cleanest generic i18n template and the most central representative of the translation-request subgroup, but the overall set is not a true duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43834 is the best representative of the dense translation-ticket subgroup, but only as a generic i18n template issue rather than a true canonical duplicate.", + "best_issue_reason": "issue:43834 is the closest thing to a hub among the translation issues, but it still represents a distinct language-specific request rather than a mergeable duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different feature requests in different subsystems: Flash Attention 2 support vs GGUF support." + "reason": "Flash Attention support and GGUF support are different feature requests in different code paths." }, { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Different language translation tickets; similarity is just the shared i18n template." + "reason": "Both are docs translation requests, but for separate language targets; not the same issue." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Different language translation tickets; same boilerplate, not the same change." + "reason": "Both are docs translation requests, but for separate language targets; not the same issue." }, { "left": "issue:27088", "right": "issue:47402", "accept": false, - "reason": "Different language translation tickets; only the template text overlaps." + "reason": "Both are translation-template issues, but they appear to be distinct language-specific requests." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Different language translation tickets; superficial template similarity only." + "reason": "Same translation template, but the issues are language-specific and not the same underlying task." }, { "left": "issue:27088", "right": "issue:30191", "accept": false, - "reason": "Turkish translation vs Persian translation; related topic, but not the same issue." + "reason": "Turkish docs translation and Persian docs translation are separate language requests." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish translation vs generic placeholder translation issue; not the same underlying ticket." + "reason": "Spanish translation and the generic translation template are not the same concrete request." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Turkish and Spanish translation requests are distinct language tasks." + "reason": "Turkish and Spanish docs translations are separate language issues." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Different language translation tickets; shared template wording is not enough." + "reason": "Different target languages; these are separate translation requests, not duplicates." }, { "left": "issue:28936", "right": "issue:30240", "accept": false, - "reason": "Spanish vs Polish translation requests are separate issues." + "reason": "Spanish and Polish docs translations are distinct language-specific issues." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish vs Romanian translation requests are separate language tasks." + "reason": "Spanish and Romanian docs translations are distinct language-specific issues." }, { "left": "issue:30191", "right": "issue:30240", "accept": false, - "reason": "Persian and Polish translation requests are different issues." + "reason": "Persian and Polish translation issues are separate language requests." }, { "left": "issue:28936", "right": "issue:30191", "accept": false, - "reason": "Spanish and Persian translation requests are distinct language contributions." + "reason": "Spanish and Persian translation issues are separate language requests." }, { "left": "issue:30240", "right": "issue:44370", "accept": false, - "reason": "Polish translation request vs generic placeholder translation issue; not the same thing." + "reason": "Polish translation and the generic template are not the same concrete translation task." }, { "left": "issue:30240", "right": "issue:44365", "accept": false, - "reason": "Polish translation request vs generic placeholder translation issue; only boilerplate matches." + "reason": "Polish translation and the generic template are not the same concrete translation task." }, { "left": "issue:30240", "right": "issue:47402", "accept": false, - "reason": "Polish translation request vs generic placeholder translation issue; not a true duplicate." + "reason": "Polish translation and the generic template are not the same concrete translation task." }, { "left": "issue:30240", "right": "issue:43834", "accept": false, - "reason": "Polish translation request vs generic placeholder translation issue; superficial similarity only." + "reason": "Polish translation and the generic template are not the same concrete translation task." }, { "left": "issue:30240", "right": "issue:47299", "accept": false, - "reason": "Polish translation request vs noisy/nonmatching translation ticket; not the same underlying issue." + "reason": "Polish docs translation and the Thai/branded translation-like issue are not the same request." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Romanian translation vs placeholder translation issue; not the same ticket." + "reason": "Romanian translation and the generic template are not the same concrete request." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Romanian translation vs placeholder translation issue; template overlap only." + "reason": "Romanian translation and the generic template are not the same concrete request." }, { "left": "issue:38435", "right": "issue:47402", "accept": false, - "reason": "Romanian translation vs placeholder translation issue; not a true duplicate." + "reason": "Romanian translation and the generic template are not the same concrete request." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "Romanian translation vs placeholder translation issue; only the i18n template is shared." + "reason": "Romanian translation and the generic template are not the same concrete request." }, { "left": "issue:38435", "right": "issue:47299", "accept": false, - "reason": "Romanian translation vs noisy/nonmatching translation ticket; different underlying requests." + "reason": "Romanian translation and the Thai/branded translation-like issue are not the same request." }, { "left": "issue:43834", "right": "issue:44370", "accept": false, - "reason": "Both are template-style translation issues, but not the same language/content request." + "reason": "Both are template-like translation issues, but they do not identify the same language or change." }, { "left": "issue:43834", "right": "issue:44365", "accept": false, - "reason": "Both are template-style translation issues, but they are not the same issue." + "reason": "Both are template-like translation issues, but they do not identify the same language or change." }, { "left": "issue:43834", "right": "issue:47402", "accept": false, - "reason": "Placeholder translation tickets only; no evidence of the same underlying change." + "reason": "Both are template-like translation issues, but they do not identify the same language or change." }, { "left": "issue:43834", "right": "issue:47299", "accept": false, - "reason": "Placeholder translation ticket vs noisy language-specific ticket; not a duplicate." + "reason": "Template-like translation issue versus a different translation-like request; not the same underlying task." }, { "left": "issue:43834", "right": "issue:46226", "accept": false, - "reason": "Generic translation template issue vs unrelated Thai text ticket; no shared underlying task." + "reason": "These look like separate translation/request artifacts, not one mergeable issue." }, { "left": "issue:44365", "right": "issue:44370", "accept": false, - "reason": "Two different placeholder translation issues; similarity is only boilerplate." + "reason": "Both are translation-template issues, but they are not the same language-specific request." }, { "left": "issue:44365", "right": "issue:47402", "accept": false, - "reason": "Template-style translation issues, but not the same language request." + "reason": "Both are translation-template issues, but they are not the same language-specific request." }, { "left": "issue:44365", "right": "issue:47299", "accept": false, - "reason": "Placeholder translation issue vs noisy Thai-language title; not the same issue." + "reason": "Template-like translation issue versus a different translation-like request; not the same underlying task." }, { "left": "issue:44365", "right": "issue:46226", "accept": false, - "reason": "Generic translation template issue vs unrelated Thai text ticket; no duplicate evidence." + "reason": "These appear to be separate translation/request artifacts, not duplicates." }, { "left": "issue:44370", "right": "issue:47402", "accept": false, - "reason": "Two generic placeholder translation issues; boilerplate match only." + "reason": "Both are translation-template issues, but they are not the same language-specific request." }, { "left": "issue:44370", "right": "issue:47299", "accept": false, - "reason": "Template translation issue vs unrelated/noisy Thai title; not the same underlying request." + "reason": "Template-like translation issue versus a different translation-like request; not the same underlying task." }, { "left": "issue:44370", "right": "issue:46226", "accept": false, - "reason": "Placeholder translation issue vs unrelated Thai text ticket; not a duplicate." + "reason": "These appear to be separate translation/request artifacts, not duplicates." }, { "left": "issue:46222", "right": "issue:47299", "accept": false, - "reason": "Noisy Thai-language title vs unrelated placeholder translation issue; similarity is not enough." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:44370", "right": "issue:46222", "accept": false, - "reason": "Generic translation template issue vs unrelated Thai text ticket; different underlying content." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:44365", "right": "issue:46222", "accept": false, - "reason": "Generic translation template issue vs unrelated Thai text ticket; not the same task." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:46222", "right": "issue:47402", "accept": false, - "reason": "Noisy Thai-language title vs placeholder translation issue; not a true duplicate." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:43834", "right": "issue:46222", "accept": false, - "reason": "Generic translation template issue vs unrelated Thai text ticket; no shared underlying change." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:46226", "right": "issue:47402", "accept": false, - "reason": "Unrelated Thai-language issue vs placeholder translation issue; not a duplicate." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:46226", "right": "issue:46931", "accept": false, - "reason": "Two unrelated Thai-language/noisy titles; no evidence they describe the same change." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:44370", "right": "issue:46931", "accept": false, - "reason": "Placeholder translation issue vs unrelated Thai-language title; only superficial similarity." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:44365", "right": "issue:46931", "accept": false, - "reason": "Placeholder translation issue vs unrelated Thai-language title; not the same underlying issue." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:43834", "right": "issue:46931", "accept": false, - "reason": "Generic template translation issue vs unrelated Thai-language title; not a duplicate." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:46931", "right": "issue:47402", "accept": false, - "reason": "Unrelated Thai-language/noisy title vs placeholder translation issue; not the same request." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." }, { "left": "issue:47299", "right": "issue:47402", "accept": false, - "reason": "Noisy custom-language title vs generic placeholder translation issue; no duplicate evidence." + "reason": "Both are translation-like issues, but they do not point to the same concrete language request." }, { "left": "issue:46226", "right": "issue:47299", "accept": false, - "reason": "Two unrelated noisy Thai-language/title variants; not clearly the same underlying issue." + "reason": "These are different translation-like requests with different text/targets, so not the same issue." } ] }, @@ -21109,12 +21149,12 @@ "issue:42503|issue:46653", "issue:45027|issue:46653" ], - "prepared_review_unit_hash": "08ab592370fddb846351ae2f900ce47adabffb1cdc5960af883d26d2ba03c233", + "prepared_review_unit_hash": "3cc60960208b1d71e2fbeaa898c87d00a5e84e3ad84c76ef09c7a875ffae7506", "component_budget": { "node_count": 22, "item_count": 22, "soft_pair_count": 55, - "serialized_chars": 31689, + "serialized_chars": 31690, "estimated_input_tokens": 7923, "estimated_eval_tokens": 16102 }, @@ -21122,7 +21162,7 @@ "node_count": 22, "item_count": 22, "soft_pair_count": 55, - "serialized_chars": 31689, + "serialized_chars": 31690, "estimated_input_tokens": 7923, "estimated_eval_tokens": 16102 }, @@ -21131,342 +21171,342 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are mostly separate model-request issues that only share superficial wording like \"Add X\" or broad ML model categories. I don\u2019t see a true duplicate cluster; the soft links are too loose to merge any pair confidently.", - "confidence": 0.95, - "canonical_issue_reason": "No single canonical issue stands out because the items span unrelated requests across different model families and modalities (keypoint matching, text encoders, vision detectors, audio/TTS/ASR, OCR, etc.).", + "summary": "This cluster is a loose set of model-request/support issues across unrelated models and modalities. The titles are superficially similar, but they do not describe the same underlying change, so there is no good canonical duplicate target here.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "If a representative is needed, issue:36718 is the broadest/high-activity item among the keypoint-matching requests, but it is still not a duplicate of the other issues.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36718", "right": "issue:47004", "accept": false, - "reason": "Related keypoint task, but RoMa matcher vs LoMa detector+matcher are different requests and not the same underlying change." + "reason": "Both are keypoint matcher requests, but RoMa and LoMa are different models/changes." }, { "left": "issue:36718", "right": "issue:41115", "accept": false, - "reason": "Keypoint matcher request vs MiniCPM3 architecture request; different models and code paths." + "reason": "Different domains: keypoint matching vs a MiniCPM3 model architecture request." }, { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "EuroBert config entry vs DEIMv2 model addition; both are model onboarding, not the same model." + "reason": "EuroBert config support and DEIMv2 are unrelated model additions." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert text model config vs Qwen3-TTS audio support; unrelated modalities." + "reason": "EuroBert config and Qwen3-TTS are different model families and tasks." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert config vs RF-DETR model addition; unrelated model families." + "reason": "A config update for EuroBert is not the same as adding RF-DETR." }, { "left": "issue:36817", "right": "issue:42954", "accept": false, - "reason": "EuroBert config vs XiaoMi MiMo; different architectures and targets." + "reason": "EuroBert config work is unrelated to Xiaomi MiMo support." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are distinct vision models; no evidence they share a concrete implementation change." + "reason": "RF-DETR and DEIMv2 are distinct detection model requests." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR object detection vs LFM2.5 Audio 1.5B; different modalities." + "reason": "Different model requests: RF-DETR vs LFM2.5 Audio 1.5B." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR detector vs weight-sparse transformer request; not the same model or fix." + "reason": "RF-DETR and a weight-sparse transformer request are not the same change." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR vision model vs Qwen3-ASR audio support; unrelated tasks." + "reason": "Object detection model support and Qwen3-ASR support are unrelated." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR detection model vs Qwen3-TTS audio support; different modality and code paths." + "reason": "RF-DETR and Qwen3-TTS are different models and tasks." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT text model request vs Qwen3-ASR audio support; unrelated." + "reason": "NeoBERT and Qwen3-ASR are separate model support requests." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT model addition vs OpenAI weight-sparse transformer request; different underlying models." + "reason": "NeoBERT support is unrelated to the OpenAI weight-sparse transformer request." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT text model vs LFM2.5 Audio 1.5B; no shared concrete change." + "reason": "NeoBERT and LFM2.5 Audio are different architectures and tasks." }, { "left": "issue:37015", "right": "issue:37334", "accept": false, - "reason": "NeoBERT addition vs LLMDet support; different model families and implementation areas." + "reason": "NeoBERT and LLMDet are distinct model additions, not one bug/change." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are unrelated model additions; only the generic \"add model\" framing is shared." + "reason": "RF-DETR and NeoBERT are unrelated model requests." }, { "left": "issue:37334", "right": "issue:41115", "accept": false, - "reason": "LLMDet support vs MiniCPM3 architecture; different models and likely different files." + "reason": "LLMDet support and MiniCPM3 architecture work are different model integrations." }, { "left": "issue:37334", "right": "issue:43837", "accept": false, - "reason": "LLMDet vision-language support vs Qwen3-ASR audio support; not the same code-path problem." + "reason": "LLMDet and Qwen3-ASR are unrelated model support requests." }, { "left": "issue:37334", "right": "issue:43671", "accept": false, - "reason": "LLMDet support vs Qwen3-TTS audio support; unrelated features." + "reason": "LLMDet and Qwen3-TTS are different tasks and code paths." }, { "left": "issue:37334", "right": "issue:42503", "accept": false, - "reason": "LLMDet support vs ModernVBERT models; separate model requests." + "reason": "LLMDet support and ModernVBERT models are not the same underlying change." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture vs Qwen3-ASR support; unrelated model families." + "reason": "MiniCPM3 architecture and Qwen3-ASR support are different model requests." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 request vs weight-sparse transformer request; no indication of the same implementation." + "reason": "MiniCPM3 and the OpenAI sparse transformer request do not describe the same change." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 model architecture vs ModernVBERT models; distinct requests." + "reason": "MiniCPM3 and ModernVBERT are unrelated model additions." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 vision model vs Qwen3-TTS audio support; different modalities and changes." + "reason": "DEIMv2 and Qwen3-TTS are distinct model support requests." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT vs ModernVBERT; both are BERT-like, but they are different model families." + "reason": "NeoBERT and ModernVBERT are different model families." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT models vs Qwen3-ASR support; unrelated model classes." + "reason": "ModernVBERT model support is unrelated to Qwen3-ASR support." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT text models vs Qwen3-TTS audio support; not a duplicate." + "reason": "ModernVBERT and Qwen3-TTS are different model additions." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like models with RoPE is a generic architecture request, while Voxtral-4B-TTS-2603 is a specific audio model support request." + "reason": "BERT-like RoPE support and Voxtral TTS support are not the same feature." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT model addition vs generic RoPE architecture support; related only at a very broad transformer level." + "reason": "ModernVBERT models and generic BERT-like RoPE support are related only broadly, not the same change." }, { "left": "issue:42738", "right": "issue:42954", "accept": false, - "reason": "Generic RoPE BERT-like model support vs XiaoMi MiMo; no evidence of the same model or patch." + "reason": "BERT-like RoPE support is unrelated to Xiaomi MiMo." }, { "left": "issue:42738", "right": "issue:46653", "accept": false, - "reason": "Generic architecture request vs a user/person issue; clearly unrelated." + "reason": "BERT-like RoPE support and a personal-name issue are unrelated." }, { "left": "issue:42954", "right": "issue:44205", "accept": false, - "reason": "XiaoMi MiMo vs SAM3-LiteText; different model types and tasks." + "reason": "MiMo and SAM3-LiteText are different model support requests." }, { "left": "issue:37334", "right": "issue:42954", "accept": false, - "reason": "LLMDet support vs XiaoMi MiMo; separate model requests." + "reason": "LLMDet support and Xiaomi MiMo are unrelated model requests." }, { "left": "issue:42954", "right": "issue:43671", "accept": false, - "reason": "XiaoMi MiMo vs Qwen3-TTS support; unrelated model families." + "reason": "Xiaomi MiMo and Qwen3-TTS are different model families." }, { "left": "issue:42503", "right": "issue:42954", "accept": false, - "reason": "ModernVBERT models vs XiaoMi MiMo; different text-model requests." + "reason": "ModernVBERT and Xiaomi MiMo are not the same underlying request." }, { "left": "issue:42954", "right": "issue:45177", "accept": false, - "reason": "XiaoMi MiMo vs DeepSeek-OCR2; different application domains." + "reason": "MiMo and DeepSeek-OCR2 are unrelated model additions." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are related by product line but they are distinct audio directions and not the same underlying change." + "reason": "Qwen3-TTS and Qwen3-ASR are different modalities and separate support requests." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS support vs Voxtral-4B-TTS-2603 support; different TTS models and likely different integrations." + "reason": "Qwen3-TTS and Voxtral TTS are different model support tasks." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Audio TTS support vs OCR model request; unrelated." + "reason": "Qwen3-TTS support and DeepSeek-OCR2 support are unrelated." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS support vs weight-sparse transformer request; different model families." + "reason": "Qwen3-TTS and the sparse transformer request are not the same code-path problem." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR vs LFM2.5 Audio 1.5B; both audio, but different concrete models and tasks." + "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are different audio models." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR audio support vs EUPE vision-language encoder; unrelated." + "reason": "Qwen3-ASR and EUPE are unrelated model support requests." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "ASR model support vs weight-sparse transformer request; not the same code change." + "reason": "ASR model support and weight-sparse transformer support are different changes." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio 1.5B vs SAM3-LiteText; audio model request vs segmentation/text model request." + "reason": "LFM2.5 Audio and SAM3-LiteText are unrelated model requests." }, { "left": "issue:43909", "right": "issue:46645", "accept": false, - "reason": "Audio model request vs Granite SWA+sink option; different models and implementation concerns." + "reason": "LFM2.5 Audio support and Granite SWA+sink options are different changes." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Weight-sparse transformer request vs EUPE encoder addition; distinct model proposals." + "reason": "Weight-sparse transformer support and EUPE are distinct model additions." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 architecture vs Voxtral TTS support; unrelated model families." + "reason": "MiniCPM3 architecture work and Voxtral TTS support are unrelated." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR vs Voxtral TTS; both audio, but different concrete support requests." + "reason": "Qwen3-ASR and Voxtral-4B-TTS are different speech-model requests." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT text model vs Voxtral TTS; unrelated." + "reason": "NeoBERT and Voxtral TTS have no shared underlying bug/change." }, { "left": "issue:37334", "right": "issue:45027", "accept": false, - "reason": "LLMDet support vs Voxtral TTS; different modality and code path." + "reason": "LLMDet support and Voxtral TTS support are unrelated." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "ASR support vs OCR model addition; not a duplicate." + "reason": "Qwen3-ASR and DeepSeek-OCR2 are separate model support requests." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE encoder request vs TIPSv2 vision-language pretraining; separate model additions." + "reason": "EUPE and TIPSv2 are different Meta/DeepMind model requests." }, { "left": "issue:42954", "right": "issue:46653", "accept": false, - "reason": "XiaoMi MiMo vs a personal-name issue; clearly unrelated." + "reason": "MiMo and a name-only issue are clearly not the same change." }, { "left": "issue:42503", "right": "issue:46653", "accept": false, - "reason": "ModernVBERT models vs a personal-name issue; unrelated." + "reason": "ModernVBERT and a name-only issue are unrelated." }, { "left": "issue:45027", "right": "issue:46653", "accept": false, - "reason": "Voxtral TTS support vs a personal-name issue; unrelated." + "reason": "Voxtral TTS support and a name-only issue are unrelated." } ] }, @@ -21496,7 +21536,7 @@ "soft_pairs": [ "issue:38740|issue:46443" ], - "prepared_review_unit_hash": "4b2119e52395d58876ebde2d0e8dfab96eaace7d59ebc67e34364be83d5f6593", + "prepared_review_unit_hash": "5e041c3e70b2184d9600b1e23c70090aaa65e0f482c96f82f2b0bedb26217907", "component_budget": { "node_count": 2, "item_count": 2, @@ -21518,18 +21558,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are unrelated: one is a docs request about adding Pruna as an optimization framework, and the other is a Vision RoPE refactor. The soft linkage is not enough to treat them as the same underlying problem.", - "confidence": 0.96, - "canonical_issue_reason": "Issue #46443 is the better representative only because it is the more substantive engineering task; #38740 is a standalone docs request and not related to the refactor.", + "summary": "The two issues are unrelated: one is a docs request to add Pruna as an optimization framework, and the other is a Vision RoPE refactor. No duplicate or same-underlying-change relationship is evident.", + "confidence": 0.98, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#46443 is the most project-relevant and actionable issue in the pair, while #38740 is documentation-focused and unrelated.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:38740", "right": "issue:46443", "accept": false, - "reason": "Different topics and code areas: a documentation enhancement versus a Vision RoPE refactor. No shared issue target, filenames, or concrete bug/change linkage." + "reason": "Different topics and scopes: documentation/framework listing vs a Vision RoPE code refactor. They do not describe the same bug or change." } ] }, @@ -21559,7 +21599,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "2bdbf9476ec457c5f6fa393a1e214676fb00ecc9bd72705d56e08712247b7717", + "prepared_review_unit_hash": "b3ab2f23bfbf83089200bdd35408b71073ef6a3eb73fbe7fdb47e02fb045dac1", "component_budget": { "node_count": 2, "item_count": 2, @@ -21581,18 +21621,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated: one is a vague request about block decoding, the other is a model/version support request for DeepSeek v3.2. No duplicate cluster should be formed.", - "confidence": 0.98, + "summary": "These two issues are not duplicates: one requests Set Block Decoding, the other asks for DeepSeek v3.2 support. They describe different feature areas and different underlying work.", + "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue 43037 is the more actionable issue because it names a concrete target (DeepSeek v3.2 support), while issue 41084 is too generic to serve as a good representative.", + "best_issue_reason": "No single canonical issue fits this cluster because the issues are about different requests and do not share the same underlying bug or change.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different underlying problems: block decoding behavior vs adding support for a specific model version. They do not describe the same bug or change." + "reason": "Reject: the titles and implied scopes are unrelated; block decoding behavior is a codec/runtime feature request, while DeepSeek v3.2 support is model compatibility work." } ] }, @@ -21622,7 +21662,7 @@ "soft_pairs": [ "issue:41306|issue:41826" ], - "prepared_review_unit_hash": "3b3ed90c7d7e9e17ba6f3d23b3640b2c1ee56afb6bb217e8e64345948aaf40d6", + "prepared_review_unit_hash": "5f3b891ba5f8b465396304c3e6ee07c0dfc862e72ef572ef41a97e1912661d71", "component_budget": { "node_count": 2, "item_count": 2, @@ -21644,18 +21684,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only by a shared \"Tiled\" naming pattern, but they describe different model components and different changes: one is about exposing a `from_pretrained()` flag for TiledFusedLogitsLoss / FLCE, the other is about adding TiledMLP to reduce memory footprint.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 41306 is the more concrete API-oriented request, tied to a specific `from_pretrained()` behavior for TiledFusedLogitsLoss/FLCE.", + "summary": "These are two distinct feature requests in the same general 'Tiled' area, but they target different components and changes: one is about a from_pretrained() flag for TiledFusedLogitsLoss/FLCE, the other is about integrating TiledMLP for lower memory use.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 41306 is the more concrete, narrowly scoped API request and is the better anchor for this cluster; 41826 is a separate TiledMLP feature proposal.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 41306 is the better anchor because it names a specific feature gap and likely has clearer implementation scope than the broader TiledMLP memory-footprint request.", + "best_issue_reason": "Issue 41306 is the best representative issue because it describes a specific user-facing change with a clear code path, whereas 41826 is a different feature area.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41306", "right": "issue:41826", "accept": false, - "reason": "Reject: they concern different modules and different problems; no evidence they describe the same bug or the same code change, beyond both using the Tiled prefix." + "reason": "Different underlying requests: one asks for a from_pretrained() flag for TiledFusedLogitsLoss/FLCE, the other proposes TiledMLP integration for memory reduction. They are related only at a broad subsystem level, not the same bug/change." } ] }, @@ -21685,12 +21725,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "1de907a14b45ac2c76d29582db1a5d95a1fdef7caae4e4a2445e09a5f10122a1", + "prepared_review_unit_hash": "7b7bd12133201876507696bf8cc5698b84410e251b1d434fe7e3f7f85f14c61e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, + "serialized_chars": 1582, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -21698,7 +21738,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, + "serialized_chars": 1582, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -21707,18 +21747,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a very high level (both mention custom trainers in Trainer), but they describe different problems: one is about unused-column removal/customizability, the other about FLOPS estimation. They should not be merged as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "Issue 43598 is the more general Trainer-behavior customization request; 43600 is a separate metric-estimation concern rather than the same underlying bug.", + "summary": "The two issues are related to Trainer customization, but they describe different problems: one is about `remove_unused_column` behavior and the other about FLOPS estimation for custom trainers. They should not be deduplicated.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 43598 is the more concrete Trainer-behavior bug report, centered on a specific setting (`remove_unused_column`) and its customizability impact.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43598 is the better cluster anchor because it targets a broader Trainer control-flow concern and is more representative of the shared customization theme.", + "best_issue_reason": "Issue 43598 is the better anchor for this cluster because it is specific, actionable, and directly tied to a single Trainer code path.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Different underlying problems: column pruning behavior vs FLOPS accounting. Shared focus on custom trainers is too broad to imply the same bug or change." + "reason": "They share a broad theme of custom Trainer support, but one concerns column removal/customizability while the other concerns FLOPS counting. Different underlying bugs and likely different fixes." } ] }, @@ -21748,12 +21788,12 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "ffe4426c3612dbbf42b246d1eb4ba687923ba9d0412807c7f99059072d564d48", + "prepared_review_unit_hash": "a2d7233a7f05c12dcec207254326c83cf7110dd953df637d747e4ebe9ce74263", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -21761,7 +21801,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -21770,18 +21810,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one requests Qwen3-Omni model registration in AutoModel/AutoModelForConditionalGeneration, while the other requests AutoModelForSequenceClassification support for Qwen3.5 (Qwen3_5Config). They target different model families and different auto-class heads.", - "confidence": 0.97, - "canonical_issue_reason": "Issue #43684 is the broader registration request and is the better representative of the cluster, but it is still separate from #44405, which is a distinct Qwen3.5 sequence-classification support request.", + "summary": "These are related only at a broad \u201cadd model support to AutoModel\u201d level, but they target different Qwen model families and different auto classes, so they should not be deduplicated.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 43684 is the broader and more central representative: it requests registration for Qwen3-Omni in AutoModel/AutoModelForConditionalGeneration, whereas the other issue is a narrower sequence-classification request for Qwen3.5.", "canonical_pr_reason": null, - "best_issue_reason": "#43684 is the clearest central issue for this cluster because it is a general model-registration request; #44405 is a different feature for a different configuration/head.", + "best_issue_reason": "43684 is the better cluster representative because it is the more general model-registration request and closer to the likely architectural change, while 44405 is a separate, narrower capability gap.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying changes: Qwen3-Omni registration for AutoModel/AutoModelForConditionalGeneration versus Qwen3.5 AutoModelForSequenceClassification support. They do not describe the same bug or patch." + "reason": "Different underlying changes: Qwen3-Omni registration for AutoModel/AutoModelForConditionalGeneration versus Qwen3.5 sequence-classification support for Qwen3_5Config. Same broad subsystem, but not the same bug or fix." } ] }, @@ -21818,77 +21858,80 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "418a0121804e3b2fc16c4e6166703e7e44c59f693b9a4b58984c5e37c9b30a76", + "prepared_review_unit_hash": "625c8bbbf53745591b6c9087b551d4ac0c02b0dc10e102c75cb869f682556466", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4264, - "estimated_input_tokens": 1066, - "estimated_eval_tokens": 2388 + "serialized_chars": 4260, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4264, - "estimated_input_tokens": 1066, - "estimated_eval_tokens": 2388 + "serialized_chars": 4260, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All four items are extremely low-signal issues with vague or blank titles and no substantive content. There is not enough evidence to confidently merge any pair as the same underlying report; the three \"huggingface \" issues look superficially similar but remain unverified duplicates.", - "confidence": 0.91, - "canonical_issue_reason": "issue:43845 is the earliest of the near-identical \"huggingface \" issues, so it is the best representative if a single canonical issue must be chosen.", + "summary": "Three issues are near-identical placeholders titled \"huggingface\" and likely duplicate each other; \"Fett\" appears unrelated. No PRs are present.", + "confidence": 0.62, + "canonical_issue_reason": "Issue 43845 is the earliest of the repeated \"huggingface\" reports and is the clearest representative of that duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43845 is the strongest global representative because it matches the repeated title pattern and is the earliest among the similar reports.", + "best_issue_reason": "Issue 43845 is the best single issue to keep because it matches the repeated topic and predates the other near-duplicate \"huggingface\" reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Titles are unrelated ('Fett' vs 'huggingface '), and there is no body or target evidence tying them to the same bug." + "reason": "\"Fett\" does not match the repeated \"huggingface\" topic, and there is no concrete evidence they describe the same problem." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Titles are unrelated ('Fett' vs 'huggingface '), with no concrete evidence of a shared underlying issue." + "reason": "\"Fett\" appears unrelated to the \"huggingface\" reports; title similarity is too weak to treat them as the same issue." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Titles are unrelated and the issues contain no detail showing they describe the same problem." + "reason": "No meaningful overlap beyond being closed issues; the topics do not line up." }, { "left": "issue:43845", "right": "issue:43846", - "accept": false, - "reason": "The titles are identical, but the reports are too empty to verify they refer to the same underlying bug rather than separate low-quality submissions." + "accept": true, + "reason": "They have the same title, were opened minutes apart, and nothing in the available metadata distinguishes them as different bugs." }, { "left": "issue:43845", "right": "issue:43859", - "accept": false, - "reason": "Same generic title, but no description or explicit linkage confirms a shared defect." + "accept": true, + "reason": "This appears to be the same repeated \"huggingface\" issue report, with no evidence of a distinct underlying problem." }, { "left": "issue:43846", "right": "issue:43859", - "accept": false, - "reason": "Same generic title, but there is insufficient substantive evidence to safely treat them as duplicates." + "accept": true, + "reason": "Same repeated \"huggingface\" title and no differentiating content; likely duplicate reports of the same issue." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet: the three \"huggingface\" issues share the same title and are close enough in timing to plausibly form a duplicate cluster, while \"Fett\" is not well supported as related. The choice of 43845 as the canonical/earliest representative is reasonable, and the soft-edge verdicts stay mostly conservative." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -21926,106 +21969,106 @@ "issue:44573|issue:47036", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "bc057e4d088e00717ad8505f749a8e09a09afc23187b746d8adce01e5a84bbd4", + "prepared_review_unit_hash": "0037db997ec137ca3beba49762766d561fbfee8119d46ad07a8f83aa320db6c8", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 12, - "serialized_chars": 7355, - "estimated_input_tokens": 1839, - "estimated_eval_tokens": 3934 + "serialized_chars": 7346, + "estimated_input_tokens": 1837, + "estimated_eval_tokens": 3930 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 12, - "serialized_chars": 7355, - "estimated_input_tokens": 1839, - "estimated_eval_tokens": 3934 + "serialized_chars": 7346, + "estimated_input_tokens": 1837, + "estimated_eval_tokens": 3930 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is dominated by spam/low-signal issues and does not appear to contain a single underlying bug or request. The few non-spam titles point to unrelated topics, so none of the soft duplicate links should be accepted.", - "confidence": 0.97, + "summary": "These issues look like unrelated low-signal/spam reports rather than duplicates of one concrete bug or change. No soft edge has enough evidence to merge.", + "confidence": 0.94, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 44418 is the only non-spam title and is the cleanest representative if one must be chosen, but it still lacks enough detail to serve as a true canonical duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are spam-like placeholders; no concrete shared bug or request is described." + "reason": "Both are titled \"\", but there is no issue text or concrete shared defect/change to show they are the same underlying report." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both are spam-like placeholders with no evidence of the same underlying issue." + "reason": "Same spam-like title, but no substantive content links them to the same bug or change." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam, the other is a documentation suggestion; unrelated." + "reason": "One is spam and the other is a vague documentation suggestion; they do not describe the same issue." }, { "left": "issue:44371", "right": "issue:47036", "accept": false, - "reason": "Spam placeholder versus a community benchmark tip; not the same issue." + "reason": "Spam placeholder versus a community benchmark tip; no shared concrete problem." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus a specific title about Google LLC; unrelated." + "reason": "Spam placeholder versus a distinct titled issue about \"Google LLC\"; no evidence of duplication." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "Documentation enhancement request versus spam placeholder; not duplicates." + "reason": "A vague documentation suggestion is not the same underlying report as a spam placeholder." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "Documentation enhancement request versus spam placeholder; no shared bug/change." + "reason": "No concrete overlap beyond low-quality metadata; not enough to treat as duplicates." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are spam-like placeholders, but there is no substantive evidence they describe the same underlying problem." + "reason": "Both are spam placeholders, but identical template text alone is insufficient to prove the same underlying bug or change." }, { "left": "issue:44572", "right": "issue:47036", "accept": false, - "reason": "Spam placeholder versus benchmark tip; different topics." + "reason": "Spam placeholder and benchmark tip are unrelated in title and implied intent." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus a Google LLC-related issue; unrelated." + "reason": "No shared concrete defect or code-path issue is visible." }, { "left": "issue:44573", "right": "issue:47036", "accept": false, - "reason": "Spam placeholder versus benchmark tip; not a duplicate pair." + "reason": "No evidence these refer to the same underlying problem; titles suggest different subjects." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus a Google LLC-related issue; no common underlying bug/request." + "reason": "Distinct issue titles with no supporting body or targets; not a safe duplicate merge." } ] }, @@ -22055,7 +22098,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "5412dff64a3db400b896d4cd15bed6345ee4611a4050a2215098a713c1b6a548", + "prepared_review_unit_hash": "02c2ed00cf0a0e06ef6048df6cddac76e154669b0dd78a241d0a9169d6042cc1", "component_budget": { "node_count": 2, "item_count": 2, @@ -22077,28 +22120,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two closed issues are highly similar and likely duplicates, but the packet only supports a soft match: same title and related wording, with no distinct evidence that they are unquestionably the same underlying issue. #45145 is the best canonical/best issue mainly because it was created first and already has an inbound reference.", - "confidence": 0.78, - "canonical_issue_reason": "#45145 is the best canonical issue because it was opened earlier and already has an inbound reference; #45151 does not add any clear unique evidence from the packet.", + "summary": "These two closed issues appear to be duplicates of the same report: identical titles, highly similar content size, and no distinguishing links or targets. The earlier issue is the better canonical choice.", + "confidence": 0.98, + "canonical_issue_reason": "Issue #45145 is earlier and has an inbound reference, so it is the better canonical representative of this duplicate pair.", "canonical_pr_reason": null, - "best_issue_reason": "#45145 is the best single issue to keep for triage because it is earlier and slightly better connected, while #45151 appears redundant but not provably identical from the available evidence.", + "best_issue_reason": "Issue #45145 is the strongest global issue candidate because it is the earliest instance of the duplicated report and already has some reference activity.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": false, - "reason": "The two issues have the same title and are plausibly duplicates, but the packet does not establish that they are the same underlying issue; with deterministic_accept=false and no distinct cross-links or content evidence, this should remain a likely/possible duplicate rather than an accepted merge." + "accept": true, + "reason": "Same title and evidently the same underlying energy-optimization report; no evidence of different bugs or scope." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The canonical choice of #45145 is plausible, but the writeup overstates the evidence. The packet only supports that the issues share the same title and have soft similarity; it does not establish that they are the same underlying issue or that #45151 adds no distinct information. The soft-edge verdict should be more conservative (e.g., likely duplicate / possible duplicate), especially since the packet\u2019s deterministic_accept is false." + "accept": true, + "feedback": "The assessment is grounded in the packet: both issues share the exact title, have very similar body lengths, and there are no explicit issue links or targets to distinguish them. Choosing the earlier issue with the inbound reference as canonical is conservative and reasonable. The duplicate/soft-edge verdict is also sufficiently cautious given the available evidence." }, "evaluator_used": true, - "retried": true, - "accepted_nontrivial_soft_edge": false, + "retried": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -22175,7 +22218,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "4517f338b112d383453f42a5c50d5a6e20d6d0b91ffeda96cf632fddf64acd92", + "prepared_review_unit_hash": "9013b8829ce7cd36308f9c021a22ad23ee36fe83098e0b63403438fbb7a16047", "component_budget": { "node_count": 2, "item_count": 2, @@ -22197,18 +22240,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated low-signal entries with different personal names and dates; there is no evidence they describe the same bug or request. This looks like accidental or spam-like duplication only in a very broad superficial sense.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "These two closed issues do not appear to be duplicates; they are separate name/date entries with no shared targets, files, or discussion context.", + "confidence": 0.96, + "canonical_issue_reason": "No canonical issue: the titles refer to different people and different dates, with no evidence they describe the same underlying problem.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong canonical candidate; both are empty, unrelated, and provide no actionable bug/context.", + "best_issue_reason": "Neither issue is a good global representative because they appear to be distinct records rather than duplicate reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Do not merge: the titles refer to different people and different dates, with no shared description, target, or code-related context indicating the same underlying issue." + "reason": "Only superficial title similarity; the issues name different people and contain no shared evidence of the same underlying bug or request." } ] }, @@ -22238,7 +22281,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "c72eca73fbdcaa670b9b92a393caf35609bb85c94296fb73d2c4c9b7087528de", + "prepared_review_unit_hash": "a2c7e544d9389d583affd9db25301989eaebd605bd7e34f7d776c19b5e16aded", "component_budget": { "node_count": 2, "item_count": 2, @@ -22260,18 +22303,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two closed issues only share a weak textual pattern (person name + date) and provide no evidence of the same bug or request. They should not be clustered as duplicates.", - "confidence": 0.97, - "canonical_issue_reason": "Neither issue is a strong canonical choice; both are title-only, unrelated, and lack any problem description or linkage.", + "summary": "The cluster contains two unrelated closed issues with only superficial similarity in title format (personal names and dates). They do not describe the same bug, feature, or failure mode, so this is not a duplicate cluster.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue: both entries are isolated, content-free, and unrelated beyond a similar name/date pattern.", "canonical_pr_reason": null, - "best_issue_reason": "No issue here is a good canonical representative because neither describes an actionable software problem or references the other.", + "best_issue_reason": "Neither issue is a strong canonical choice because neither contains problem details; if forced, no issue is meaningfully better than the other.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "The titles are different names with different dates and there is no body, linkage, or shared technical context indicating the same underlying issue." + "reason": "Reject: the titles refer to different people and different dates, with no shared bug report, code path, or explicit linkage." } ] }, @@ -22301,7 +22344,7 @@ "soft_pairs": [ "issue:45810|issue:46295" ], - "prepared_review_unit_hash": "c10e5c6378b46e65b78d1c6a177a98a9507375ac16a35438b9d0a15816ea8cae", + "prepared_review_unit_hash": "0f85cf7bfea147db706c444a40e2cd69934285d49e73204308cbdb6ca1c04477", "component_budget": { "node_count": 2, "item_count": 2, @@ -22323,18 +22366,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are not duplicates: one requests adding a Qwen3_5 token-classification model class for value-model use, while the other asks for FP8 training support. They concern different features and code paths.", - "confidence": 0.96, + "summary": "These two open issues are not duplicates: one requests a new Qwen3_5 token-classification/value-model class, while the other is about FP8 training support. They concern different code paths and different feature areas.", + "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 46295 is the broader infrastructure request and would be the more representative issue if this cluster needed a single anchor, but it is still not the same underlying request as 45810.", + "best_issue_reason": "No canonical issue emerges here because the cluster does not represent the same bug or request. If one must be chosen as the broader umbrella, issue #46295 is more general, but it is still unrelated to #45810.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45810", "right": "issue:46295", "accept": false, - "reason": "Different underlying feature requests: model-class addition for token classification vs FP8 training support. They do not describe the same bug or change." + "reason": "The titles and descriptions point to unrelated features: adding a Qwen3_5 token-classification model versus enabling FP8 training support. They do not describe the same underlying problem or change." } ] }, @@ -22374,12 +22417,12 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "b24d242089b5e1b49d69ddd285a6cb48d20b61a1be43bdf1d3703e518b9b60a7", + "prepared_review_unit_hash": "2bbb35c717746627a9072f3c52742765a965e529473deb9ad8717a6438b5e8e7", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5725, + "serialized_chars": 5727, "estimated_input_tokens": 1432, "estimated_eval_tokens": 3120 }, @@ -22387,7 +22430,7 @@ "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5725, + "serialized_chars": 5727, "estimated_input_tokens": 1432, "estimated_eval_tokens": 3120 }, @@ -22396,66 +22439,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This set contains two exact duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is a separate dataset-sales/sample post and does not match the DataForge example/service requests.", - "confidence": 0.96, - "canonical_issue_reason": "Best canonical candidate is issue 45872: it is a clear, concrete request and is exactly duplicated by 45873; the 45870/45871 pair is similar but more generic.", + "summary": "This cluster is actually two separate exact-duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is unrelated marketing/dataset content and should not be merged with the others.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue for the whole cluster: it mixes two distinct duplicate topics. For the first pair, 45870 is the cleanest representative because it is the earliest exact duplicate of 45871 and has the same title/body.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45872 is the strongest representative of the meaningful issue content here because it is specific, self-contained, and has an exact duplicate partner.", + "best_issue_reason": "45870 is the best single issue to keep if one representative is needed, but only for the 'Example: Financial time series prediction with DataForge' pair; the cluster should otherwise be split because 45872/45873 are a different duplicate pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Different request type: dataset bundle/sample promotion vs an example about financial time series prediction." + "reason": "Dataset bundle promotion vs. an example/tutorial issue; same financial theme but not the same underlying bug or change." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Different request type: dataset bundle/sample promotion vs an example about financial time series prediction." + "reason": "Dataset bundle promotion vs. an example/tutorial issue; not the same underlying request." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Same title and same apparent issue description; these look like the same underlying issue." + "reason": "Identical title and matching body size/content pattern; these are the same issue." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both are DataForge examples, but they describe different concrete requests: prediction example vs scraping-service + time-series request." + "reason": "Both mention financial time series, but one is an example prediction workflow and the other is a scraping-service feature request; different underlying issues." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Both are DataForge examples, but they describe different concrete requests: prediction example vs scraping-service + time-series request." + "reason": "Same as above: related domain, but different feature/problem statement." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Both are DataForge examples, but they describe different concrete requests: prediction example vs scraping-service + time-series request." + "reason": "Different requests: example prediction workflow vs. financial time series plus scraping service." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Both are DataForge examples, but they describe different concrete requests: prediction example vs scraping-service + time-series request." + "reason": "Different requests: example prediction workflow vs. financial time series plus scraping service." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Same title and same apparent issue description; these look like the same underlying issue." + "reason": "Identical title and matching body size/content pattern; these are the same issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 have identical titles, identical body lengths, and the soft-pair evidence shows jaccard 1.0, so treating them as duplicate pairs is reasonable. The distinction between 45869 and the DataForge example/service requests is also conservative and supported by the titles. No soft-edge verdict appears overstated." + "feedback": "The assessment is grounded in the packet: 45870/45871 and 45872/45873 have identical titles and matching body-length/Jaccard signals, while 45869 is a different dataset-promo issue. The cluster-splitting conclusion and the conservative cross-pair non-duplicates are reasonable." }, "evaluator_used": true, "retried": false, @@ -22482,7 +22525,7 @@ "soft_pairs": [ "issue:46439|issue:46441" ], - "prepared_review_unit_hash": "673cb9a1e7d8bf03d994256981498ff7ac4cf20a6dde4cb30b270daa209334cc", + "prepared_review_unit_hash": "029eb312bcb93d95de774f0ddb0667905ebd6bcc9d050b33a6b17153b18f5b31", "component_budget": { "node_count": 2, "item_count": 2, @@ -22504,24 +22547,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both issues concern static-cache handling around linear-attention, but they describe different failure modes: one is shape corruption during early initialization, the other is a Qwen3.5 KeyError for a missing `linear_attention` cache entry. They are related, but not clearly the same underlying bug.", - "confidence": 0.81, - "canonical_issue_reason": "Issue 46439 is the better canonical issue because it states the concrete mechanism (`early_initialization`) and the resulting corruption of linear-attention shapes, making it the broader and more diagnostic report.", + "summary": "The two issues are related in theme (`static` cache + linear-attention), but the evidence here is only titles/metadata, so they should not be treated as the same bug. The safer triage is to keep them separate and use the broader issue as the representative one.", + "confidence": 0.68, + "canonical_issue_reason": "Issue 46439 is the broader, more general title: it describes a `static` cache problem affecting linear-attention layers, while 46441 is a narrower Qwen3.5-specific crash report with a different surfaced exception.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46439 is the strongest standalone tracker: it is more general, more actionable, and likely covers the underlying static-cache initialization problem beyond a single model symptom.", + "best_issue_reason": "Issue 46439 is the best standalone issue because its title captures the underlying subsystem-level failure more generally than the model-specific crash in 46441.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46439", "right": "issue:46441", "accept": false, - "reason": "Related static-cache/linear-attention failures, but the reported symptoms and likely code paths differ (`RuntimeError` from wrong shapes vs `KeyError: 'linear_attention'` in Qwen3.5), so they are not safe to merge as the same bug." + "reason": "The titles suggest a shared area (`static` cache / linear-attention), but there is not enough evidence here to conclude they are the same underlying bug; the surfaced failures are different and the packet lacks bodies or diffs to connect them." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The result overstates what is supported by the packet. We only have titles and metadata here, not the issue bodies, so claims like `early_initialization`, shape corruption details, and a specific root-cause cache initialization bug are not grounded. The pair may be related by theme (`static` cache + linear-attention), but the soft-edge acceptance should stay conservative and not assert they are the same bug on this evidence alone." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -22545,12 +22591,12 @@ "soft_pairs": [ "issue:46551|issue:46552" ], - "prepared_review_unit_hash": "8715d8404c59bf07f9e63ad534796e46ec508c0da74646be833ea72b912ecef2", + "prepared_review_unit_hash": "412cce0fb379a567bef368b4b66d73a0121b71d305ac32237936286ce1cf7082", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1706, + "serialized_chars": 1707, "estimated_input_tokens": 427, "estimated_eval_tokens": 1110 }, @@ -22558,7 +22604,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1706, + "serialized_chars": 1707, "estimated_input_tokens": 427, "estimated_eval_tokens": 1110 }, @@ -22567,24 +22613,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are effectively identical duplicates describing the same Wav2Vec2CTCTokenizer v5 special-token regression. The open issue has the stronger triage value as the canonical issue.", + "summary": "The two issues are effectively identical reports of the same Wav2Vec2CTCTokenizer v5 bug; the newer open issue is the better canonical thread.", "confidence": 0.99, - "canonical_issue_reason": "issue:46552 is the open duplicate with inbound references, so it is the better canonical tracker for the shared bug report.", + "canonical_issue_reason": "Issue 46552 is the newer, still-open report with inbound references, so it is the best place to track and consolidate discussion for this bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46552 is best suited as the global issue because it is open, has references already pointing to it, and matches the same exact problem statement.", + "best_issue_reason": "Issue 46552 best represents the cluster because it is active and duplicates the exact same title/problem statement as 46551.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46551", "right": "issue:46552", "accept": true, - "reason": "Same title and essentially the same report/body; both describe the exact same tokenizer special-token leak bug, so this is a true duplicate pair." + "reason": "Exact same title and same underlying tokenizer bug report; these are duplicates of the same issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The conclusion is grounded in the packet: the two issues have identical titles and matching body length, and issue:46552 is open with inbound references, so using it as the canonical tracker is reasonable. The duplicate-pair acceptance is also conservative enough given the exact text match evidence." + "feedback": "The summary is grounded in the packet: both issues have the same title, the same body length, and the soft-similarity signal is maximal. Choosing the newer open issue (46552) as the canonical thread is a conservative inference supported by its open state and inbound references. The soft-edge duplicate verdict is also appropriately cautious." }, "evaluator_used": true, "retried": false, @@ -22611,40 +22657,40 @@ "soft_pairs": [ "issue:46657|issue:46692" ], - "prepared_review_unit_hash": "5f19ceb810ca47e05e3fd1de44733dece708c6b0307affdd2e6cf1a5c8df99c9", + "prepared_review_unit_hash": "dcc35bcf6b4b00ae895e7f33006ac3b2b442c9860e4a77c72983b5658bd4e884", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1567, - "estimated_input_tokens": 392, - "estimated_eval_tokens": 1040 + "serialized_chars": 1569, + "estimated_input_tokens": 393, + "estimated_eval_tokens": 1042 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1567, - "estimated_input_tokens": 392, - "estimated_eval_tokens": 1040 + "serialized_chars": 1569, + "estimated_input_tokens": 393, + "estimated_eval_tokens": 1042 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are related to training failures in object detection workflows, but they describe different underlying problems: one is a tensor shape mismatch, the other is GPU memory exhaustion after several epochs. They should not be treated as duplicates.", - "confidence": 0.96, - "canonical_issue_reason": "If one issue must serve as the cluster anchor, #46657 is the broader, more general training-error report; however, the two reports are not the same bug.", + "summary": "These two issues are not duplicates: one reports a tensor shape mismatch during object-detection training, while the other reports an out-of-memory failure after several epochs in RT-DETR v2. They describe different failure modes with no clear shared code-path evidence.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #46657 is the better canonical representative because it describes a concrete, specific training error rather than a broad resource issue.", "canonical_pr_reason": null, - "best_issue_reason": "#46657 is the better representative only because it is the more generic failure mode. #46692 is a separate OOM issue with a different symptom and likely different fix.", + "best_issue_reason": "#46657 is the most representative issue in the cluster due to its clearer error signature and tighter problem statement.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46657", "right": "issue:46692", "accept": false, - "reason": "Same broad training area, but different concrete failures: tensor size mismatch vs out-of-memory after epochs. The symptom, trigger, and likely fix are not the same underlying bug." + "reason": "Different underlying bugs: a tensor-size mismatch versus delayed OOM during training. The titles and problem descriptions point to separate failure modes, not the same code-path defect." } ] }, @@ -22674,12 +22720,12 @@ "soft_pairs": [ "issue:46822|issue:46823" ], - "prepared_review_unit_hash": "623576a8417ddfae1b8c6c700f474b65cdd18d0c355b5bd27bc6daac279b01d7", + "prepared_review_unit_hash": "c3e3cdcdadc5bdb1047b71953eb38c2df0a8ef7f5015fd798f744f34b10170c4", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1674, + "serialized_chars": 1673, "estimated_input_tokens": 419, "estimated_eval_tokens": 1094 }, @@ -22687,7 +22733,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1674, + "serialized_chars": 1673, "estimated_input_tokens": 419, "estimated_eval_tokens": 1094 }, @@ -22696,18 +22742,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated bug reports: one about ZeRO-3 zero.Init failing to partition a composite submodule and causing OOM, the other about 4-bit + device_map='auto' + max_memory incorrectly dispatching to CPU and raising ValueError. They should not be clustered as duplicates.", - "confidence": 0.96, - "canonical_issue_reason": "No true canonical issue for the pair because the reports describe different failure modes and code paths. If forced to anchor the cluster, issue 46822 is the better representative only because it is the clearer open report with a concrete repro.", + "summary": "These two issues are about different bugs in different loading paths: ZeRO-3/zero.Init partitioning for a multimodal submodule vs. 4-bit device_map/max_memory dispatch behavior. They should not be deduplicated.", + "confidence": 0.98, + "canonical_issue_reason": "No true canonical duplicate exists here; the issues describe unrelated failure modes and code paths.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46822 is the better standalone representative: it is open, specific, and describes a distinct ZeRO-3 partitioning/OOM problem more directly than 46823.", + "best_issue_reason": "Issue 46822 is the more concrete and narrowly scoped report, but it is not a duplicate of 46823.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46822", "right": "issue:46823", "accept": false, - "reason": "Different underlying bugs: ZeRO-3 composite-submodule partitioning/OOM versus 4-bit device_map/max_memory CPU dispatch ValueError. Similar broad theme of loading failures, but not the same code-path problem." + "reason": "Different underlying problems: one is ZeRO-3 zero.Init failing to partition a composite submodule causing OOM, the other is 4-bit loading with device_map='auto' and max_memory incorrectly dispatching to CPU. Same general ML loading area, but not the same bug." } ] }, @@ -22729,7 +22775,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-46906-20", + "cluster_id": "cluster-46906-21", "nodes": [ "issue:46906", "issue:46940", @@ -22750,7 +22796,8 @@ "issue:47279", "issue:47309", "issue:47377", - "issue:47390" + "issue:47390", + "issue:47406" ], "soft_pairs": [ "issue:46906|issue:46940", @@ -22812,418 +22859,452 @@ "issue:47309|issue:47377", "issue:47182|issue:47309", "issue:47227|issue:47309", + "issue:47390|issue:47406", "issue:47377|issue:47390", "issue:47279|issue:47390", "issue:47227|issue:47390", "issue:47256|issue:47390", - "issue:47182|issue:47390" + "issue:47377|issue:47406", + "issue:47279|issue:47406", + "issue:47120|issue:47406", + "issue:47256|issue:47406" ], - "prepared_review_unit_hash": "fec238ebdd937283fa95bdd340878be9d91ed246096cad666034502c9b763a3b", + "prepared_review_unit_hash": "1dbf57ba8b456c7a8de6e9596403b7c5c962de70a06a8711c7238eba1439d363", "component_budget": { - "node_count": 20, - "item_count": 20, - "soft_pair_count": 64, - "serialized_chars": 34210, - "estimated_input_tokens": 8553, - "estimated_eval_tokens": 17362 + "node_count": 21, + "item_count": 21, + "soft_pair_count": 68, + "serialized_chars": 36197, + "estimated_input_tokens": 9050, + "estimated_eval_tokens": 18356 }, "budget": { - "node_count": 20, - "item_count": 20, - "soft_pair_count": 64, - "serialized_chars": 34210, - "estimated_input_tokens": 8553, - "estimated_eval_tokens": 17362 + "node_count": 21, + "item_count": 21, + "soft_pair_count": 68, + "serialized_chars": 36197, + "estimated_input_tokens": 9050, + "estimated_eval_tokens": 18356 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All items are daily \"[serge] integration failure triage\" issues for different dates, so this cluster looks like a recurring report series rather than duplicate bug reports. No PRs are present.", - "confidence": 0.98, - "canonical_issue_reason": "issue:47279 is the most central dated triage item in the set, with the highest inbound reference count and substantial discussion, making it the best canonical anchor for this recurring issue series.", + "summary": "This cluster is a repeated series of daily '[serge] integration failure triage' issues, but they appear to be separate day-by-day reports rather than duplicates of one underlying bug. If one issue has to represent the set, issue 47279 is the best hub because it has the strongest inbound-reference/activity signal.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 47279 is the most central report in the series by inbound references and discussion activity, so it is the best canonical anchor if one must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": "issue:47279 is the most representative single issue among the triage reports: it is highly referenced, active, and sits in the middle of the sequence, so it best anchors the cluster.", + "best_issue_reason": "Issue 47279 has the strongest linkage/attention in the cluster, but the rest look like distinct daily triage reports rather than exact duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46906", "right": "issue:46940", "accept": false, - "reason": "Same triage-series title, but different dates; these are separate daily reports, not the same underlying issue." + "reason": "Same triage template, but different daily reports; no evidence they describe the same underlying failure." }, { "left": "issue:46906", "right": "issue:46948", "accept": false, - "reason": "Different daily triage report; no evidence they describe the exact same bug instance." + "reason": "Different dates in the triage series; similar wording alone is not enough to merge as duplicates." }, { "left": "issue:46906", "right": "issue:46969", "accept": false, - "reason": "These are sequential daily triage issues, not duplicate reports of one concrete problem." + "reason": "These look like separate daily failure reports, not the same bug." }, { "left": "issue:46906", "right": "issue:46985", "accept": false, - "reason": "Same recurring template, but date-specific triage indicates distinct reports." + "reason": "Same broad series, but distinct day-specific triage issues." }, { "left": "issue:46906", "right": "issue:47037", "accept": false, - "reason": "Both are dated triage reports; similarity is from the template, not a shared unique bug." + "reason": "Different daily triage reports; no concrete shared bug/change evidence." }, { "left": "issue:46940", "right": "issue:46948", "accept": false, - "reason": "Daily triage reports for different dates; too generic to treat as the same issue." + "reason": "Adjacent triage days, but still separate reports rather than one duplicate issue." }, { "left": "issue:46940", "right": "issue:46969", "accept": false, - "reason": "These look like separate day-by-day reports in a series, not duplicates." + "reason": "No sign of the same underlying failure; only the same report template." }, { "left": "issue:46940", "right": "issue:46985", "accept": false, - "reason": "Same broad subsystem and template, but different day\u2019s triage report." + "reason": "Different daily triage entries, not the same issue." }, { "left": "issue:46940", "right": "issue:46998", "accept": false, - "reason": "Not the same concrete failure; the dates show separate triage entries." + "reason": "Similar title series only; these are distinct day-by-day reports." }, { "left": "issue:46948", "right": "issue:46969", "accept": false, - "reason": "Recurring integration-triage issue, but each date is a distinct report." + "reason": "Separate daily triage issues with the same pattern, but not the same bug." }, { "left": "issue:46948", "right": "issue:46985", "accept": false, - "reason": "Different dated triage items; not enough to call them duplicates." + "reason": "No concrete evidence of one underlying problem shared by both reports." }, { "left": "issue:46948", "right": "issue:46998", "accept": false, - "reason": "These appear to be separate daily tracking issues, not one underlying bug." + "reason": "Different triage dates; similar wording is not enough to merge." }, { "left": "issue:46969", "right": "issue:46985", "accept": false, - "reason": "Adjacent daily triage reports, but still distinct issues." + "reason": "Distinct day-specific integration-failure triage reports." }, { "left": "issue:46969", "right": "issue:47037", "accept": false, - "reason": "Same recurring report format; the date difference indicates a new triage issue." + "reason": "Separate daily reports; no shared concrete failure description." }, { "left": "issue:46969", "right": "issue:47091", "accept": false, - "reason": "No concrete shared bug evidence beyond the generic triage series." + "reason": "Same series, but not the same underlying bug/change." }, { "left": "issue:46985", "right": "issue:46998", "accept": false, - "reason": "These are separate dates in the triage sequence, not one duplicate issue." + "reason": "These are consecutive triage reports, not duplicates." }, { "left": "issue:46985", "right": "issue:47037", "accept": false, - "reason": "Similarity comes from the template, not from identical underlying failure details." + "reason": "Different dated triage entries; no merge-worthy same bug evidence." }, { "left": "issue:46985", "right": "issue:47049", "accept": false, - "reason": "Different daily integration triage entries; do not merge as duplicates." + "reason": "Same template, different day and likely different failure instance." }, { "left": "issue:46985", "right": "issue:47091", "accept": false, - "reason": "Same recurring series, but no sign of the same concrete bug report." + "reason": "No indication these are the same concrete problem." }, { "left": "issue:46985", "right": "issue:47102", "accept": false, - "reason": "These are date-stamped triage reports; that is not sufficient for duplicate status." + "reason": "Separate daily integration-failure reports." }, { "left": "issue:46998", "right": "issue:47037", "accept": false, - "reason": "Separate daily reports in a recurring integration-failure series." + "reason": "Different daily triage reports; similar title is not enough." }, { "left": "issue:46998", "right": "issue:47049", "accept": false, - "reason": "Different report dates mean distinct issue instances." + "reason": "They belong to the same triage series but are distinct reports." }, { "left": "issue:46998", "right": "issue:47053", "accept": false, - "reason": "Same template, but no evidence of a shared concrete code-path problem." + "reason": "No shared concrete bug/change, just the same triage template." }, { "left": "issue:46998", "right": "issue:47091", "accept": false, - "reason": "Recurring triage item, not a duplicate bug report." + "reason": "Separate day-by-day reports, not duplicates." }, { "left": "issue:47037", "right": "issue:47049", "accept": false, - "reason": "Adjacent daily triage issues, but still separate reports." + "reason": "Same broad series, but different daily failure reports." }, { "left": "issue:47037", "right": "issue:47053", "accept": false, - "reason": "The titles are the same series with different dates; not duplicates." + "reason": "Different dates and no concrete same-code-path evidence." }, { "left": "issue:47037", "right": "issue:47091", "accept": false, - "reason": "No evidence these are the same failure beyond generic subsystem overlap." + "reason": "These are separate triage issues, not one underlying bug." }, { "left": "issue:47037", "right": "issue:47102", "accept": false, - "reason": "Different dated triage reports, so they should remain separate." + "reason": "Different daily reports in the same integration-failure sequence." }, { "left": "issue:47049", "right": "issue:47053", "accept": false, - "reason": "Same recurring triage pattern, but distinct daily issue entries." + "reason": "Similar title series only; not the same underlying failure." }, { "left": "issue:47049", "right": "issue:47091", "accept": false, - "reason": "No shared unique failure details; these are separate triage reports." + "reason": "Distinct triage days, no duplicate evidence." }, { "left": "issue:47049", "right": "issue:47102", "accept": false, - "reason": "Date-specific integration triage entries, not one duplicate issue." + "reason": "Separate reports with no concrete shared bug." }, { "left": "issue:47053", "right": "issue:47091", "accept": false, - "reason": "Both are daily reports, not the same underlying bug." + "reason": "Same report pattern, but different issue instances." }, { "left": "issue:47053", "right": "issue:47102", "accept": false, - "reason": "Same template and subsystem, but different triage days." + "reason": "No sign they are the same underlying failure." }, { "left": "issue:47091", "right": "issue:47102", "accept": false, - "reason": "These are separate dated triage issues rather than duplicate reports." + "reason": "Different daily triage issues, not duplicates." }, { "left": "issue:47102", "right": "issue:47256", "accept": false, - "reason": "Same recurring integration triage series, but different dates and likely different failures." + "reason": "Same series, but these are separate day-specific reports." }, { "left": "issue:47120", "right": "issue:47182", "accept": false, - "reason": "Both are daily triage reports; similarity is structural, not evidence of the same bug." + "reason": "Both are triage reports with similar shape, but no evidence of the same bug." }, { "left": "issue:47120", "right": "issue:47256", "accept": false, - "reason": "Different daily reports in a sequence, so not a duplicate pair." + "reason": "Different daily failures; same template only." }, { "left": "issue:47120", "right": "issue:47265", "accept": false, - "reason": "Recurring triage items for different days; do not merge." + "reason": "Separate reports in the series, not one duplicate issue." }, { "left": "issue:47120", "right": "issue:47279", "accept": false, - "reason": "Same report template, but the date difference indicates a separate issue entry." + "reason": "No concrete indication these describe the same underlying problem." }, { "left": "issue:47120", "right": "issue:47377", "accept": false, - "reason": "These are separate daily integration-failure triage reports." + "reason": "Distinct daily triage entries, not the same change/bug." + }, + { + "left": "issue:47120", + "right": "issue:47406", + "accept": false, + "reason": "Open/latest triage issue, but still a separate report from the earlier ones." }, { "left": "issue:47182", "right": "issue:47227", "accept": false, - "reason": "Adjacent date-stamped reports, but still distinct triage issues." + "reason": "Adjacent triage dates, but not the same underlying failure." }, { "left": "issue:47182", "right": "issue:47256", "accept": false, - "reason": "No evidence of one underlying bug; this is a recurring daily report series." + "reason": "Similar report series only; distinct reports." }, { "left": "issue:47182", "right": "issue:47265", "accept": false, - "reason": "Different triage date, so not a duplicate of the same report." + "reason": "No concrete shared bug/change evidence." }, { "left": "issue:47182", "right": "issue:47279", "accept": false, - "reason": "These look like separate daily triage logs, not one bug report." + "reason": "Different day in the same triage sequence; not a duplicate." }, { "left": "issue:47182", "right": "issue:47309", "accept": false, - "reason": "Same generic title, but different day and likely different integration failures." + "reason": "Separate daily integration-failure reports." }, { "left": "issue:47182", "right": "issue:47377", "accept": false, - "reason": "Recurring daily triage issue; not the same concrete change or bug." - }, - { - "left": "issue:47182", - "right": "issue:47390", - "accept": false, - "reason": "Open vs. closed daily triage entries; separate reports, not duplicates." + "reason": "Same broad triage topic, but not the same issue." }, { "left": "issue:47227", "right": "issue:47256", "accept": false, - "reason": "Different dated triage issues in the same series." + "reason": "Distinct daily reports; the titles are similar because they are part of the same series." }, { "left": "issue:47227", "right": "issue:47265", "accept": false, - "reason": "No concrete shared bug details beyond the repeated triage title." + "reason": "No evidence of one shared underlying bug." }, { "left": "issue:47227", "right": "issue:47279", "accept": false, - "reason": "These are distinct daily reports, not one underlying issue." + "reason": "Separate triage issues, not a duplicate pair." }, { "left": "issue:47227", "right": "issue:47309", "accept": false, - "reason": "Same recurring integration triage pattern, but separate day entries." + "reason": "Different daily failure reports with the same triage template." }, { "left": "issue:47227", "right": "issue:47377", "accept": false, - "reason": "Date-specific triage reports should not be merged as duplicates." + "reason": "Same series, but not the same concrete code-path problem." }, { "left": "issue:47227", "right": "issue:47390", "accept": false, - "reason": "Different daily report; no evidence they describe the same failure instance." + "reason": "Distinct reports; no sign they should be merged." }, { "left": "issue:47256", "right": "issue:47265", "accept": false, - "reason": "Both are daily integration triage reports, but not the same bug report." + "reason": "Adjacent daily triage reports, but still separate issues." }, { "left": "issue:47256", "right": "issue:47279", "accept": false, - "reason": "Recurring report series only; the date difference points to separate issues." + "reason": "No concrete evidence they are duplicates." }, { "left": "issue:47256", "right": "issue:47390", "accept": false, - "reason": "These are separate daily triage issues, not duplicate reports." + "reason": "Different day and likely different failure instance." + }, + { + "left": "issue:47256", + "right": "issue:47406", + "accept": false, + "reason": "Same triage series only; this is not the same underlying bug." }, { "left": "issue:47265", "right": "issue:47279", "accept": false, - "reason": "Different dates in the same triage series, so not duplicates." + "reason": "Separate daily triage reports, not one duplicate issue." }, { "left": "issue:47265", "right": "issue:47309", "accept": false, - "reason": "No shared concrete failure description; keep as separate reports." + "reason": "No shared concrete failure beyond the template." }, { "left": "issue:47265", "right": "issue:47377", "accept": false, - "reason": "These appear to be sequential daily triage entries rather than one bug." + "reason": "Different dated reports in the same series." }, { "left": "issue:47279", "right": "issue:47309", "accept": false, - "reason": "Same recurring triage template, but different dates and likely different failures." + "reason": "Same generic triage pattern, but distinct reports." }, { "left": "issue:47279", "right": "issue:47377", "accept": false, - "reason": "Distinct daily integration-failure reports, not the same underlying problem." + "reason": "No evidence of the same underlying bug/change." + }, + { + "left": "issue:47279", + "right": "issue:47390", + "accept": false, + "reason": "Separate triage issues; titles are similar because the series is repetitive." + }, + { + "left": "issue:47279", + "right": "issue:47406", + "accept": false, + "reason": "Latest issue in the series, but still a distinct daily report rather than a duplicate." }, { "left": "issue:47309", "right": "issue:47377", "accept": false, - "reason": "Different day\u2019s triage reports; insufficient evidence for duplication." + "reason": "Different day-specific integration-failure reports." }, { "left": "issue:47377", "right": "issue:47390", "accept": false, - "reason": "Open and closed daily reports in the same series, but not the same issue instance." + "reason": "Same series, but not the same concrete problem." + }, + { + "left": "issue:47377", + "right": "issue:47406", + "accept": false, + "reason": "Separate triage reports; no duplicate evidence." + }, + { + "left": "issue:47390", + "right": "issue:47406", + "accept": false, + "reason": "Different dated reports, not one underlying bug." } ] }, @@ -23253,7 +23334,7 @@ "soft_pairs": [ "issue:46983|issue:46984" ], - "prepared_review_unit_hash": "f54a03fbace1d25d77b1171b824857ea7d42309cfef8153b9dd00e2433de2e82", + "prepared_review_unit_hash": "b60aea8c07af99bce316c3f5bd6ebaa5d88eefa8162e2ba7039863c3c755f34b", "component_budget": { "node_count": 2, "item_count": 2, @@ -23275,27 +23356,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two closed issues with the exact same title and highly similar content; they appear to be duplicate requests for RWKV7 model support.", - "confidence": 0.98, - "canonical_issue_reason": "Issue 46984 is slightly better as the representative because it has more discussion activity, but both issues describe the same feature request.", + "summary": "The two closed issues share the exact same title, and the packet provides a strong soft-similarity signal between them. With no body text or explicit linkage exposed, they look closely related and possibly duplicate, but the evidence does not prove that conclusively.", + "confidence": 0.81, + "canonical_issue_reason": "Issue #46983 is a reasonable canonical choice because it has the exact same title as #46984 and there is no differentiating metadata in the packet; the duplicate call is plausible, but limited by the lack of body text.", "canonical_pr_reason": null, - "best_issue_reason": "46984 is the best issue to keep as the cluster representative since it appears to have slightly more engagement while matching the same request.", + "best_issue_reason": "#46983 is a suitable representative for the issue cluster because it is an exact title match and there are no visible signals that distinguish it from #46984.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46983", "right": "issue:46984", "accept": true, - "reason": "Same title and near-identical request for RWKV7 model support; these are duplicates of the same underlying issue." + "reason": "Accept: the issues have the exact same title and a strong soft-similarity signal, which makes them look like the same feature request, though the missing body text means the match cannot be proven definitively." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: both issues share the same title, have high soft similarity, and 46984 has slightly more discussion activity, so choosing it as the representative is reasonable. The duplicate-issue verdict is phrased conservatively enough given the available evidence." + "accept": false, + "feedback": "The analysis overstates the evidence. The packet supports an exact title match and a soft similarity signal between the two issues, but it does not expose the body text, so claims like \u201chighly similar body content,\u201d \u201csame problem statement,\u201d or a definitive duplicate conclusion are too strong. A more conservative summary would note that the two closed issues share the same title and show a strong soft-similarity signal, suggesting they may be related or duplicate, but not prove it." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -23319,7 +23400,7 @@ "soft_pairs": [ "issue:47007|issue:47015" ], - "prepared_review_unit_hash": "b1288e3c788a7fd26b124dde75cdda2fe15946988c4a70d27df9c4d968a3bd22", + "prepared_review_unit_hash": "9576b9d27f30e50f96dcc6e52edcadc60c39742547f5a3a1b8788c347b549cbd", "component_budget": { "node_count": 2, "item_count": 2, @@ -23341,18 +23422,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are thematically related AI/cybersecurity writeups, but they do not describe the same underlying problem or request; this is not a duplicate pair.", - "confidence": 0.89, - "canonical_issue_reason": null, + "summary": "Both items are closed issue-style posts about AI-assisted vulnerability discovery, but they read like related discussion/proposal titles rather than the same concrete bug or change. The overlap is thematic, not a clear duplicate report.", + "confidence": 0.64, + "canonical_issue_reason": "Issue 47015 is the broader, more general framing of the topic, so it is the better representative if this cluster must have a canonical issue.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 47015 is the best single issue representative because its title is the most umbrella-like and least specific to one angle of the topic.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:47007", "right": "issue:47015", "accept": false, - "reason": "Both mention AI-driven vulnerability discovery, but one is about lessons from advanced LLM use and the other is a broader transformer-based cybersecurity application; they read as distinct topics rather than the same issue." + "reason": "The titles overlap on AI-based vulnerability discovery, but they describe different angles ('lessons from using advanced LLMs' vs 'novel application of transformers') and do not identify the same underlying issue or concrete change." } ] }, @@ -23382,12 +23463,12 @@ "soft_pairs": [ "issue:47138|issue:47139" ], - "prepared_review_unit_hash": "6683d26a610601c64b981f98b19b0e76a54a6a15bd4f91ed86959f43122aff45", + "prepared_review_unit_hash": "018885568b514bff565dc29daa5dc1a441a1779cdebe828334c412c260e1c6bd", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1591, + "serialized_chars": 1590, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -23395,7 +23476,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1591, + "serialized_chars": 1590, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -23404,24 +23485,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues appear to be exact duplicates: same title, same apparent content, and strong similarity evidence. Issue #47138 is the better canonical representative because it has slightly more inbound references and discussion activity.", + "summary": "These two open issues are effectively duplicates: they have identical titles, similar metadata, and the same reported routing bug involving chat management endpoints under /v1 when base_url includes /v1.", "confidence": 0.99, - "canonical_issue_reason": "#47138 is the stronger canonical issue: it has more inbound references and at least some discussion activity, while #47139 is effectively identical.", + "canonical_issue_reason": "Issue #47138 is as good a canonical choice as #47139; with identical titles and no distinguishing content, either could serve as the primary tracker, but #47138 can be used as the representative by convention.", "canonical_pr_reason": null, - "best_issue_reason": "#47138 best represents the cluster because it is the most referenced and engaged of the two identical issues.", + "best_issue_reason": "#47138 is the best issue candidate because it is indistinguishable from #47139 on the available evidence and can serve as the single tracker for the duplicated report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:47138", "right": "issue:47139", "accept": true, - "reason": "Same title and no differentiating evidence; they look like duplicate reports of the same bug." + "reason": "Same title, same described bug, and no evidence of different scope; these are duplicate reports of the same underlying routing issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: the issues have identical titles, identical body lengths, and a soft-similarity score of 1.0, with no conflicting evidence. Choosing #47138 as canonical is also conservative and supported by its slightly higher inbound references and discussion activity. No overstatement beyond the available evidence is apparent." + "feedback": "The conclusion is grounded in the packet: the two issues have identical titles, identical body length, no conflicting metadata, and the only pair evidence is soft_similarity with Jaccard 1.0. The duplicate/one-canonical-tracker reading is conservative enough here." }, "evaluator_used": true, "retried": false, @@ -23440,7 +23521,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-82", + "cluster_id": "cluster-40546-83", "nodes": [ "pull_request:43251", "pull_request:43254", @@ -23482,16 +23563,17 @@ "pull_request:47284", "pull_request:47291", "pull_request:47292", - "pull_request:47295", "pull_request:47301", - "pull_request:47336", "pull_request:47360", "pull_request:47374", "pull_request:47383", "pull_request:47384", - "pull_request:47395" + "pull_request:47395", + "pull_request:47398", + "pull_request:47409" ], "soft_pairs": [ + "pull_request:47398|pull_request:47409", "pull_request:43251|pull_request:43254", "pull_request:46903|pull_request:47090", "pull_request:47090|pull_request:47395", @@ -23522,455 +23604,307 @@ "pull_request:44775|pull_request:47275", "pull_request:47275|pull_request:47301", "pull_request:47216|pull_request:47275", - "pull_request:46196|pull_request:47202", "pull_request:46912|pull_request:47179", - "pull_request:46912|pull_request:47284", + "pull_request:46196|pull_request:47202", "pull_request:47163|pull_request:47284", + "pull_request:46912|pull_request:47284", "pull_request:46836|pull_request:47202", "pull_request:46196|pull_request:47210", "pull_request:47003|pull_request:47204", "pull_request:47093|pull_request:47275", "pull_request:45157|pull_request:47093", "pull_request:46995|pull_request:47163", - "pull_request:47163|pull_request:47173", "pull_request:45157|pull_request:46395", + "pull_request:47163|pull_request:47173", "pull_request:46836|pull_request:47275", - "pull_request:46912|pull_request:47082", - "pull_request:47204|pull_request:47295", - "pull_request:47163|pull_request:47384", - "pull_request:44775|pull_request:47301", - "pull_request:46196|pull_request:46995", - "pull_request:47301|pull_request:47384", - "pull_request:47216|pull_request:47384", - "pull_request:47077|pull_request:47301", - "pull_request:47336|pull_request:47384", - "pull_request:44775|pull_request:47077", - "pull_request:47166|pull_request:47292" + "pull_request:46912|pull_request:47082" ], - "prepared_review_unit_hash": "1d03a853070ef843b7d40692086e66398c00b0c2c60dd2b38566a6540234eea3", + "prepared_review_unit_hash": "76844adcde1a0b8496e9fcd30bac122a6969f581cf488cc05ef3794bb6a2f3a4", "component_budget": { - "node_count": 82, - "item_count": 82, + "node_count": 83, + "item_count": 83, "soft_pair_count": 227, - "serialized_chars": 281443, - "estimated_input_tokens": 70361, - "estimated_eval_tokens": 140978 + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 54, - "serialized_chars": 126532, - "estimated_input_tokens": 31633, - "estimated_eval_tokens": 63522 + "soft_pair_count": 45, + "serialized_chars": 124004, + "estimated_input_tokens": 31001, + "estimated_eval_tokens": 62258 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a mix of unrelated PRs, with one clear duplicate cluster around the training-loss double-shift fix for CohereASR/PPFormulaNet (and its later extension). A second duplicate pair exists in fixed_cross_entropy kwargs support. Most other soft edges are broad subsystem overlaps only and should stay separate.", - "confidence": 0.94, + "summary": "The cluster is mostly a set of unrelated PRs, but there are a few clear duplicate-like subgroups around the same underlying fixes: `fixed_cross_entropy` kwargs forwarding, and several training-loss double-shift fixes. Most soft edges should be rejected because they only share a broad subsystem or test/doc surface area.", + "confidence": 0.78, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47090 is the strongest canonical duplicate candidate: it addresses the same double-shift training-loss bug across multiple model code paths, includes the aligned fix in both modeling files and tests, and is the most complete version of that change among the close duplicates.", + "canonical_pr_reason": "No single PR covers the whole cluster; among the duplicate-like candidates, 47090 is the strongest representative because it is issue-linked and fixes the same training-loss double-shift root cause across multiple affected models, not just one narrow case.", "best_issue_reason": null, - "best_pr_reason": "PR 47090 is the best global representative because it most fully captures the shared loss-shift bug and its broader model coverage, while the other near-duplicates are narrower or partial variants of the same fix.", + "best_pr_reason": "47090 is the best global representative of the meaningful duplicate cluster: it addresses the shared loss-shift bug with the broadest concrete code coverage and is more complete than the narrower CohereASR-only variants.", "soft_edge_verdicts": [ + { + "left": "pull_request:47398", + "right": "pull_request:47409", + "accept": true, + "reason": "Same `num_experts` back-compat fallback bug in the GLM MoE config path; 47409 is a broader superset and could plausibly absorb 47398." + }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Same fixed_cross_entropy enhancement: both add cross_entropy kwargs support in the same file and could be merged into one patch." + "reason": "Same `fixed_cross_entropy` signature/forwarding change to pass supported kwargs into `nn.functional.cross_entropy`; these are effectively the same fix." }, { "left": "pull_request:46903", "right": "pull_request:47090", "accept": true, - "reason": "Same underlying training-loss double-shift bug in the same family of model forward paths; 47090 is an expanded version of the same fix." + "reason": "Same training-loss double-shift bug class in the same model family; 47090 broadens the same fix across additional affected models and is compatible as one combined PR." }, { "left": "pull_request:47090", "right": "pull_request:47395", "accept": false, - "reason": "Same symptom class, but different concrete model/code path (CohereASR/Moonshine vs Git); not a single mergeable fix." + "reason": "Both involve loss shifting, but they are different concrete code paths and models (`cohere_asr`/`moonshine` vs `git`), so this is too broad to merge as duplicates." }, { "left": "pull_request:46903", "right": "pull_request:46958", "accept": true, - "reason": "Both fix the CohereASR double-shift loss bug; the implementations differ, but the underlying code-path problem is the same." + "reason": "Same CohereASR double-shift training-loss bug; the two PRs use different implementation styles, but they target the same concrete failure mode." }, { "left": "pull_request:46903", "right": "pull_request:47395", "accept": false, - "reason": "Different models and loss paths; sharing the generic double-shift theme is not enough for duplicate triage." + "reason": "Shared theme only; `GitForCausalLM` has a separate model-specific loss path, so this is not the same underlying change." }, { "left": "pull_request:47268", "right": "pull_request:47284", "accept": false, - "reason": "Both are failing tests, but for different models and different fixes; only a broad test-failure overlap." + "reason": "Unrelated test fixes in different model suites; no common bug or change." }, { "left": "pull_request:46958", "right": "pull_request:46974", "accept": false, - "reason": "CohereASR loss bug vs Musicgen generate typo are unrelated code-path problems." + "reason": "Different bugs in different files: CohereASR loss shifting vs Musicgen cache-length typo." }, { "left": "pull_request:46954", "right": "pull_request:46958", "accept": false, - "reason": "One is a Pytest v9 compatibility tweak; the other is a model loss fix. Same area, different bugs." + "reason": "One is a Pytest fp32-strictness cleanup, the other is a training-loss fix; they are not the same change." }, { "left": "pull_request:47173", "right": "pull_request:47284", "accept": false, - "reason": "Both touch tests, but they target different models and different XPU issues." + "reason": "Separate test-suite expectation updates for different models; not the same underlying issue." }, { "left": "pull_request:47202", "right": "pull_request:47204", "accept": false, - "reason": "Both are infra/Docker related, but one removes a builder workflow while the other adds exporter dependencies; not the same change." + "reason": "One removes a Docker workflow, the other adds exporter test dependencies to a Dockerfile; related to CI, but not the same fix." }, { "left": "pull_request:47204", "right": "pull_request:47374", "accept": false, - "reason": "Exporter/runtime docs vs exporter workflow/docs rework; related theme, different change sets." + "reason": "Dockerfile/runtime-test setup vs exporter documentation rewrite; these are unrelated changes." }, { "left": "pull_request:47284", "right": "pull_request:47292", "accept": false, - "reason": "Distinct failing test fixes for different model families." + "reason": "Different model test failures and different model codepaths; no duplicate bug here." }, { "left": "pull_request:47108", "right": "pull_request:47202", "accept": false, - "reason": "Different CI/docs-builder changes, not one underlying bug or feature." + "reason": "Workflow comment/pin cleanup is unrelated to the Docker image builder removal." }, { "left": "pull_request:47077", "right": "pull_request:47078", "accept": false, - "reason": "Both are Japanese docs updates, but one is a translation pass and the other is wording cleanup plus a toctree addition." + "reason": "Both touch Japanese docs, but one is a translation PR and the other is wording cleanup; not the same artifact change." }, { "left": "pull_request:47204", "right": "pull_request:47243", "accept": false, - "reason": "Exporter dependency setup vs Executorch exporter implementation fixes are related but not the same concrete change." + "reason": "Exporter test-dependency setup is different from executorch exporter code fixes; same area, different bug/change." }, { "left": "pull_request:47202", "right": "pull_request:47360", "accept": false, - "reason": "Both touch CI/workflows, but they address different pipeline changes and different files." + "reason": "Docker image builder removal and OTEL CI pilot workflow are distinct CI changes." }, { "left": "pull_request:47108", "right": "pull_request:47360", "accept": false, - "reason": "Different workflow changes; only a high-level CI overlap." + "reason": "Dependabot-compatible comment cleanup is unrelated to the OTEL workflow pilot." }, { "left": "pull_request:47105", "right": "pull_request:47129", "accept": false, - "reason": "Independent test re-enable vs converter bugfix; no shared underlying issue." + "reason": "VITS test re-enable and RT-DETR converter key fixes are unrelated." }, { "left": "pull_request:47077", "right": "pull_request:47212", "accept": false, - "reason": "Japanese docs wording/translation vs quantized cache docs update; unrelated work." + "reason": "Japanese docs edits vs KV cache documentation updates; different docs issues." }, { "left": "pull_request:46266", "right": "pull_request:46957", "accept": false, - "reason": "Different subsystems and goals: model conversion/mapping vs generation optimization for finegrained FP8." + "reason": "These are broad subsystem refactors, but they fix different concrete problems in different areas." }, { "left": "pull_request:47301", "right": "pull_request:47360", "accept": false, - "reason": "Docs for MTP support vs CI OTEL workflow; no duplicate code-path problem." + "reason": "Docs for MTP support and an OTEL CI workflow change are unrelated." }, { "left": "pull_request:47360", "right": "pull_request:47383", "accept": false, - "reason": "CI workflow instrumentation vs Kimi integration tests; unrelated changes." - }, - { - "left": "pull_request:47173", - "right": "pull_request:47184", - "accept": false, - "reason": "Both are test-only PRs, but they target different models and different accelerator decorators." + "reason": "CI workflow pilot vs Kimi integration tests; no shared underlying bug or change." }, { - "left": "pull_request:47216", - "right": "pull_request:47275", + "left": "pull_request:47275", + "right": "pull_request:47360", "accept": false, - "reason": "Compressed-tensors docs/code vs adding a new model architecture; only superficial overlap in repo area." + "reason": "Model addition PR vs CI workflow PR; unrelated." }, { - "left": "pull_request:47291", - "right": "pull_request:47384", + "left": "pull_request:47173", + "right": "pull_request:47184", "accept": false, - "reason": "Generation-config deprecation vs Gemma 4 heterogeneous attention config; different code paths." + "reason": "Different test fixes for different models/devices; not duplicates." }, { "left": "pull_request:47166", "right": "pull_request:47173", "accept": false, - "reason": "Separate test maintenance PRs for different frameworks and models." + "reason": "Quantization test expansion vs XPU expectation updates; different changes." }, { "left": "pull_request:47077", "right": "pull_request:47275", "accept": false, - "reason": "Docs translation updates vs adding a new model; not the same artifact or bug." + "reason": "Japanese doc translation vs new model code/tests; unrelated." }, { "left": "pull_request:44775", "right": "pull_request:47275", "accept": false, - "reason": "Docs parallelism update vs new model support; no duplicate relationship." + "reason": "Docs parallelism navigation update and EfficientViT-SAM model addition are unrelated." }, { "left": "pull_request:47275", "right": "pull_request:47301", "accept": false, - "reason": "New model support vs MTP docs; unrelated changes." + "reason": "New model addition vs assisted-decoding documentation; no duplicate change." }, { "left": "pull_request:47216", - "right": "pull_request:47384", - "accept": false, - "reason": "Compressed-tensors quantization work vs Gemma 4 config refactor; different features." - }, - { - "left": "pull_request:47163", - "right": "pull_request:47384", - "accept": false, - "reason": "Paged-attention XPU tests vs Gemma 4 per-layer config; no shared underlying bug." - }, - { - "left": "pull_request:47336", - "right": "pull_request:47384", - "accept": false, - "reason": "Offloading save bugfix vs Gemma 4 config validation; unrelated." - }, - { - "left": "pull_request:44775", - "right": "pull_request:47077", + "right": "pull_request:47275", "accept": false, - "reason": "Parallelism docs vs Japanese translation docs; same documentation domain only at a very broad level." + "reason": "Compressed-tensors quantization support vs a new model architecture; different concrete work." }, { - "left": "pull_request:47166", - "right": "pull_request:47292", + "left": "pull_request:46912", + "right": "pull_request:47179", "accept": false, - "reason": "Different test suites and different accelerator-specific failures." + "reason": "RoPE test/model refactors and Granite-SWA model support are different changes." }, { - "left": "pull_request:46995", - "right": "pull_request:47163", + "left": "pull_request:46196", + "right": "pull_request:47202", "accept": false, - "reason": "Flash-linear-attention availability check vs paged-attention XPU tests; different functionality." + "reason": "SDPA dispatch support and Docker image builder removal are unrelated." }, { "left": "pull_request:47163", "right": "pull_request:47284", "accept": false, - "reason": "Paged-attention tests and mimo_v2_flash fixes are unrelated test/model issues." + "reason": "Paged-attention XPU test coverage vs Mimo V2 Flash fixes; different issues." }, { - "left": "pull_request:45157", - "right": "pull_request:47093", - "accept": false, - "reason": "Prism GGUF support vs OpenPangu-V2 model addition; both model-adjacent, but different features and code paths." - }, - { - "left": "pull_request:45157", - "right": "pull_request:46395", + "left": "pull_request:46912", + "right": "pull_request:47284", "accept": false, - "reason": "GGUF quantization support vs Param2MoE model support are unrelated additions." + "reason": "RoPE-related test/model work is unrelated to Mimo V2 Flash test fixes." }, { "left": "pull_request:46836", "right": "pull_request:47202", "accept": false, - "reason": "Unlimited OCR model add vs Docker builder removal; no duplicate bug/change." - }, - { - "left": "pull_request:46836", - "right": "pull_request:47275", - "accept": false, - "reason": "New OCR model support vs new vision model support; both model additions, but not the same underlying change." - }, - { - "left": "pull_request:46912", - "right": "pull_request:47082", - "accept": false, - "reason": "RoPE test improvements vs RoMa model addition; unrelated." - }, - { - "left": "pull_request:46912", - "right": "pull_request:47284", - "accept": false, - "reason": "RoPE test work vs mimo_v2_flash test/model fixes; different targets." + "reason": "New Unlimited OCR model support is unrelated to Docker workflow cleanup." }, { - "left": "pull_request:47163", - "right": "pull_request:47284", + "left": "pull_request:46196", + "right": "pull_request:47210", "accept": false, - "reason": "XPU paged-attention tests vs mimo_v2_flash fixes; not the same issue." + "reason": "Both touch attention internals, but one is model dispatch support and the other is packed SDPA varlen routing; different concrete fixes." }, { "left": "pull_request:47003", "right": "pull_request:47204", "accept": false, - "reason": "OpenVINO exporter implementation vs torch-light exporter test environment setup; related area, different change." + "reason": "OpenVINO exporter addition is not the same change as adding ONNX/ExecuTorch deps to torch-light." }, { "left": "pull_request:47093", "right": "pull_request:47275", "accept": false, - "reason": "OpenPangu-V2 model support vs EfficientViT-SAM model support; different model integrations." - }, - { - "left": "pull_request:46196", - "right": "pull_request:46995", - "accept": false, - "reason": "SDPA attention dispatch support vs flash-linear-attention availability check; different integration work." - }, - { - "left": "pull_request:46912", - "right": "pull_request:47179", - "accept": false, - "reason": "RoPE tests vs Granite SWA model support; unrelated." - }, - { - "left": "pull_request:46196", - "right": "pull_request:47210", - "accept": false, - "reason": "Broad SDPA dispatch support vs packed-row varlen attention routing; same subsystem, but different concrete functionality." + "reason": "OpenPangu-V2 model support and EfficientViT-SAM model support are different model integrations." }, { - "left": "pull_request:47204", - "right": "pull_request:47295", + "left": "pull_request:45157", + "right": "pull_request:47093", "accept": false, - "reason": "Exporter environment setup vs DiffusionGemma mask casting fix; unrelated." + "reason": "PrismML GGUF dequantization support and OpenPangu-V2 model addition are unrelated." }, { - "left": "pull_request:46995", - "right": "pull_request:47163", + "left": "pull_request:45157", + "right": "pull_request:46395", "accept": false, - "reason": "Availability check change vs paged-attention XPU tests; different problem classes." + "reason": "GGUF dequantization support and Param2MoE model support are unrelated." }, { "left": "pull_request:47163", - "right": "pull_request:47384", - "accept": false, - "reason": "Paged-attention tests vs Gemma 4 heterogeneous config refactor; not duplicates." - }, - { - "left": "pull_request:47301", - "right": "pull_request:47384", - "accept": false, - "reason": "Assisted-decoding docs vs Gemma 4 config refactor; unrelated." - }, - { - "left": "pull_request:47077", - "right": "pull_request:47301", - "accept": false, - "reason": "Docs translation updates vs assisted-decoding docs; both docs, but not the same content change." - }, - { - "left": "pull_request:44775", - "right": "pull_request:47077", - "accept": false, - "reason": "Parallelism docs vs Japanese translation docs; broad documentation overlap only." - }, - { - "left": "pull_request:47166", - "right": "pull_request:47292", - "accept": false, - "reason": "Different test updates for different accelerator backends and models." - }, - { - "left": "pull_request:47166", "right": "pull_request:47173", "accept": false, - "reason": "Both are test maintenance PRs, but they target different test modules and failures." - }, - { - "left": "pull_request:47291", - "right": "pull_request:47384", - "accept": false, - "reason": "Generation config deprecation vs Gemma 4 config refactor; different code paths." - }, - { - "left": "pull_request:47216", - "right": "pull_request:47384", - "accept": false, - "reason": "Compressed-tensors support vs Gemma 4 config work; unrelated." - }, - { - "left": "pull_request:47163", - "right": "pull_request:47384", - "accept": false, - "reason": "Paged-attention XPU tests vs Gemma 4 config refactor; unrelated." - }, - { - "left": "pull_request:47336", - "right": "pull_request:47384", - "accept": false, - "reason": "Model offloading save bugfix vs Gemma 4 config changes; unrelated." + "reason": "XPU paged-attention tests and XPU expectation updates in model tests are separate test changes." }, { - "left": "pull_request:44775", - "right": "pull_request:47077", - "accept": false, - "reason": "Parallelism docs vs Japanese docs translation; not the same artifact." - }, - { - "left": "pull_request:47166", - "right": "pull_request:47292", + "left": "pull_request:46836", + "right": "pull_request:47275", "accept": false, - "reason": "Different test-failure fixes for different models/backends." + "reason": "Unlimited OCR model addition and EfficientViT-SAM model addition are unrelated model integrations." }, { - "left": "pull_request:47204", - "right": "pull_request:47243", + "left": "pull_request:46912", + "right": "pull_request:47082", "accept": false, - "reason": "Environment prep for exporter tests vs exporter code fixes; related but not duplicate." - }, - { - "left": "pull_request:47077", - "right": "pull_request:47212", - "accept": false, - "reason": "Japanese docs wording vs kv-cache docs update; unrelated." - }, - { - "left": "pull_request:47108", - "right": "pull_request:47360", - "accept": false, - "reason": "Docs-builder pin comment vs OTEL workflow switch; different CI changes." - }, - { - "left": "pull_request:46995", - "right": "pull_request:47163", - "accept": false, - "reason": "Feature availability check vs paged-attention tests; not the same issue." + "reason": "RoPE test/model work and RoMa keypoint matcher model addition are unrelated." } ] }, @@ -23992,16 +23926,14 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-82", + "cluster_id": "cluster-40546-83", "nodes": [ - "pull_request:43451", "pull_request:44517", "pull_request:44775", "pull_request:45157", "pull_request:46196", "pull_request:46266", "pull_request:46267", - "pull_request:46381", "pull_request:46509", "pull_request:46604", "pull_request:46658", @@ -24011,6 +23943,7 @@ "pull_request:46957", "pull_request:46995", "pull_request:47003", + "pull_request:47077", "pull_request:47078", "pull_request:47082", "pull_request:47093", @@ -24023,12 +23956,14 @@ "pull_request:47175", "pull_request:47179", "pull_request:47184", + "pull_request:47204", "pull_request:47210", "pull_request:47216", "pull_request:47243", "pull_request:47275", - "pull_request:47291", + "pull_request:47292", "pull_request:47294", + "pull_request:47295", "pull_request:47298", "pull_request:47301", "pull_request:47304", @@ -24039,17 +23974,27 @@ "pull_request:47352", "pull_request:47360", "pull_request:47383", - "pull_request:47398", - "pull_request:47399" + "pull_request:47384", + "pull_request:47398" ], "soft_pairs": [ + "pull_request:47204|pull_request:47295", + "pull_request:47163|pull_request:47384", + "pull_request:44775|pull_request:47301", + "pull_request:46196|pull_request:46995", + "pull_request:47301|pull_request:47384", + "pull_request:47216|pull_request:47384", + "pull_request:47077|pull_request:47301", + "pull_request:47336|pull_request:47384", + "pull_request:44775|pull_request:47077", "pull_request:47294|pull_request:47320", + "pull_request:47166|pull_request:47292", "pull_request:46267|pull_request:47383", - "pull_request:47093|pull_request:47143", "pull_request:46509|pull_request:46995", + "pull_request:47093|pull_request:47143", "pull_request:46995|pull_request:47143", - "pull_request:46266|pull_request:47301", "pull_request:46196|pull_request:47179", + "pull_request:46266|pull_request:47301", "pull_request:46266|pull_request:47383", "pull_request:44775|pull_request:47304", "pull_request:47082|pull_request:47210", @@ -24067,8 +24012,8 @@ "pull_request:44775|pull_request:46836", "pull_request:46604|pull_request:46658", "pull_request:47163|pull_request:47179", - "pull_request:47082|pull_request:47216", "pull_request:44517|pull_request:45157", + "pull_request:47082|pull_request:47216", "pull_request:47275|pull_request:47336", "pull_request:47210|pull_request:47304", "pull_request:47108|pull_request:47383", @@ -24084,364 +24029,355 @@ "pull_request:46895|pull_request:47306", "pull_request:46196|pull_request:47301", "pull_request:47105|pull_request:47298", - "pull_request:45157|pull_request:46658", "pull_request:47275|pull_request:47304", + "pull_request:45157|pull_request:46658", "pull_request:47078|pull_request:47322", - "pull_request:44517|pull_request:47301", "pull_request:47078|pull_request:47243", + "pull_request:44517|pull_request:47301", "pull_request:46895|pull_request:47398", "pull_request:47294|pull_request:47360", - "pull_request:47113|pull_request:47304", - "pull_request:46912|pull_request:47291", - "pull_request:46895|pull_request:47399", - "pull_request:47163|pull_request:47352", - "pull_request:43451|pull_request:44517", - "pull_request:47352|pull_request:47383", - "pull_request:44517|pull_request:47383", - "pull_request:44517|pull_request:47143", - "pull_request:43451|pull_request:47352", - "pull_request:47243|pull_request:47294", - "pull_request:47210|pull_request:47291", - "pull_request:46604|pull_request:47163", - "pull_request:46957|pull_request:47383", - "pull_request:46604|pull_request:47383", - "pull_request:46957|pull_request:47360", - "pull_request:47113|pull_request:47175", - "pull_request:46381|pull_request:47322", - "pull_request:43451|pull_request:46604", - "pull_request:46957|pull_request:47301", - "pull_request:47113|pull_request:47275", - "pull_request:47113|pull_request:47243", - "pull_request:46912|pull_request:47113" + "pull_request:47113|pull_request:47304" ], - "prepared_review_unit_hash": "ee0553c2d65109f6b8d1c347e2c5bf2d535bef7d6b0e73ad94480ea533504ed6", + "prepared_review_unit_hash": "942a9a9b224ac2407dea998023d0f0bfd27946cc1858392b892f17614977abd0", "component_budget": { - "node_count": 82, - "item_count": 82, + "node_count": 83, + "item_count": 83, "soft_pair_count": 227, - "serialized_chars": 281443, - "estimated_input_tokens": 70361, - "estimated_eval_tokens": 140978 + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 70, - "serialized_chars": 135502, - "estimated_input_tokens": 33876, - "estimated_eval_tokens": 68008 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 59, + "serialized_chars": 134391, + "estimated_input_tokens": 33598, + "estimated_eval_tokens": 67452 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab bag of unrelated pull requests: model additions, docs-only edits, CI/test updates, and isolated bugfixes. None of the soft edges look like the same underlying change or bug, so I rejected them all.", - "confidence": 0.98, + "summary": "This cluster is mostly a broad set of unrelated PRs that share only superficial keyword similarity (docs/model additions/tests/infrastructure), so none of the soft edges look like true duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR stands out: the items span unrelated features and fixes, so there is no one PR that subsumes the cluster.", + "canonical_pr_reason": "No clear canonical PR: the cluster is heterogeneous and does not center on one underlying bug or change.", "best_issue_reason": null, - "best_pr_reason": "No clear best PR for duplicate triage; the set does not form one cohesive change family, so any representative would be arbitrary.", + "best_pr_reason": "PR 47352 is the strongest representative only because it is a substantive core-runtime change with an explicit issue target, but it is still not a duplicate hub for the cluster.", "soft_edge_verdicts": [ + { + "left": "pull_request:47204", + "right": "pull_request:47295", + "accept": false, + "reason": "Unrelated: exporter test-environment setup vs a DiffusionGemma mask bugfix." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47384", + "accept": false, + "reason": "Unrelated tests: XPU paged-attention coverage vs Gemma 4 heterogeneous-attention config changes." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47301", + "accept": false, + "reason": "Different docs topics: distributed-training navigation vs assisted-decoding/MTP docs." + }, + { + "left": "pull_request:46196", + "right": "pull_request:46995", + "accept": false, + "reason": "Different code paths: SDPA dispatch for wav2vec2 models vs flash-linear-attention availability check." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47384", + "accept": false, + "reason": "Docs vs model config: MTP documentation change vs Gemma 4 per-layer config validation." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47384", + "accept": false, + "reason": "Quantization docs/code vs model architecture config; no shared underlying bug." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47301", + "accept": false, + "reason": "Translation/docs update for DistilBERT vs docs for multi-token prediction." + }, + { + "left": "pull_request:47336", + "right": "pull_request:47384", + "accept": false, + "reason": "Offloaded-save bugfix vs Gemma 4 config refactor; unrelated problems." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47077", + "accept": false, + "reason": "Both are documentation edits, but for different content and languages; not the same change." + }, { "left": "pull_request:47294", "right": "pull_request:47320", "accept": false, - "reason": "Kernel-native causal-conv1d utilities and response-schema deprecation are unrelated changes." + "reason": "Kernel integration work vs chat response-schema deprecation; unrelated subsystems." }, { - "left": "pull_request:46267", - "right": "pull_request:47383", + "left": "pull_request:47166", + "right": "pull_request:47292", "accept": false, - "reason": "Serve chat parsing changes and Kimi integration tests do not target the same bug or code path." + "reason": "Different test suites for different quantization/model areas; not a shared bug." }, { - "left": "pull_request:47093", - "right": "pull_request:47143", + "left": "pull_request:46267", + "right": "pull_request:47383", "accept": false, - "reason": "These are separate model additions (OpenPangu-V2 vs NeuCodec), not duplicates." + "reason": "Serve/chat parsing refactor vs Kimi integration tests; unrelated." }, { "left": "pull_request:46509", "right": "pull_request:46995", "accept": false, - "reason": "Nemotron Omni model support and MLU flash-linear-attention availability are unrelated." + "reason": "New Nemotron Omni model support vs flash-linear-attention availability; unrelated." }, { - "left": "pull_request:46995", + "left": "pull_request:47093", "right": "pull_request:47143", "accept": false, - "reason": "A backend availability tweak and a model addition are different changes." + "reason": "OpenPangu-V2 model addition vs NeuCodec model addition; different models and code paths." }, { - "left": "pull_request:46266", - "right": "pull_request:47301", + "left": "pull_request:46995", + "right": "pull_request:47143", "accept": false, - "reason": "Dinov2 modernization/conversion work is unrelated to MTP documentation." + "reason": "Utility/backend availability change vs new model support; no overlap in bug or fix." }, { "left": "pull_request:46196", "right": "pull_request:47179", "accept": false, - "reason": "SDPA dispatch support for wav2vec2 and Granite model support are different code paths." + "reason": "Audio SDPA dispatch change vs Granite-SWA model support; unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:47301", + "accept": false, + "reason": "DINO/DINOv3 conversion/model modernization vs MTP docs; unrelated." }, { "left": "pull_request:46266", "right": "pull_request:47383", "accept": false, - "reason": "Dinov2 conversion/model updates do not overlap with Kimi integration tests." + "reason": "Model conversion/architecture work vs Kimi integration tests; not the same underlying issue." }, { "left": "pull_request:44775", "right": "pull_request:47304", "accept": false, - "reason": "Two separate documentation-only PRs for different topics." + "reason": "Different docs maintenance changes; one is distributed-training nav, the other is MPS graph-cache docs cleanup." }, { "left": "pull_request:47082", "right": "pull_request:47210", "accept": false, - "reason": "A keypoint matcher model addition and packed SDPA varlen routing are unrelated." + "reason": "New RoMa model support vs SDPA packed-sequence optimization; unrelated." }, { "left": "pull_request:47093", "right": "pull_request:47163", "accept": false, - "reason": "OpenPangu-V2 model support and XPU paged-attention test coverage are not the same change." + "reason": "OpenPangu-V2 model support vs XPU paged-attention tests; unrelated." }, { "left": "pull_request:47179", "right": "pull_request:47210", "accept": false, - "reason": "Granite model support and packed SDPA varlen attention are unrelated changes." + "reason": "Model support vs attention-kernel optimization; different concrete changes." }, { "left": "pull_request:46604", "right": "pull_request:47003", "accept": false, - "reason": "Mistral tekken tokenizer utilities and the OpenVINO exporter are different features." + "reason": "Mistral tekken tokenizer utilities vs OpenVINO exporter implementation; different feature areas." }, { "left": "pull_request:47082", "right": "pull_request:47108", "accept": false, - "reason": "RoMa model support and a doc-builder comment tweak are clearly unrelated." + "reason": "RoMa model support vs a docs-build comment tweak; unrelated." }, { "left": "pull_request:47078", "right": "pull_request:47275", "accept": false, - "reason": "Japanese docs wording edits and EfficientViTSAM model support do not match." + "reason": "Japanese docs wording updates vs new EfficientViTSam model support; unrelated." }, { - "left": "pull_request:46266", - "right": "pull_request:47216", - "accept": false, - "reason": "Dinov2 conversion/model work and compressed-tensors fp8 changes are separate efforts." - }, - { - "left": "pull_request:47108", + "left": "pull_request:47082", "right": "pull_request:47301", "accept": false, - "reason": "A documentation comment change and MTP docs are not the same underlying change." + "reason": "Keypoint matcher model support vs MTP documentation; unrelated." }, { - "left": "pull_request:47082", - "right": "pull_request:47275", + "left": "pull_request:46995", + "right": "pull_request:47301", "accept": false, - "reason": "RoMa model support is unrelated to EfficientViTSAM support." + "reason": "MLU support for flash-linear-attention availability vs MTP docs; unrelated." }, { - "left": "pull_request:46995", + "left": "pull_request:47179", "right": "pull_request:47301", "accept": false, - "reason": "Backend availability logic and MTP documentation are unrelated." + "reason": "Granite-SWA model support vs MTP docs; unrelated." }, { "left": "pull_request:46266", "right": "pull_request:47336", "accept": false, - "reason": "Dinov2 conversion support and fully offloaded save fixes do not share a code path." + "reason": "DINO modernization/conversion changes vs fully-offloaded save bugfix; unrelated." }, { "left": "pull_request:47175", "right": "pull_request:47179", "accept": false, - "reason": "Olmo3 CI expectation updates and Granite model support are not duplicates." + "reason": "Olmo3 CI expectation updates vs Granite-SWA model support; unrelated." }, { "left": "pull_request:44775", "right": "pull_request:46836", "accept": false, - "reason": "Parallelism docs and Unlimited OCR model support are unrelated." + "reason": "Docs navigation edits vs Unlimited OCR model support; unrelated." }, { "left": "pull_request:46604", "right": "pull_request:46658", "accept": false, - "reason": "Mistral tokenizer utilities and Step-3.7 model tooling are different implementations." + "reason": "Mistral tokenizer utilities vs Step-3.7 Flash model/scripts; unrelated." }, { "left": "pull_request:47163", "right": "pull_request:47179", "accept": false, - "reason": "XPU paged-attention tests and Granite model support are unrelated." + "reason": "Paged-attention XPU test coverage vs Granite-SWA model addition; unrelated." }, { "left": "pull_request:44517", "right": "pull_request:45157", "accept": false, - "reason": "Qwen3-TTS model support and PrismML Bonsai support are different model integrations." + "reason": "Qwen3 TTS support vs Prism GGUF dequantization/import fixes; unrelated." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47216", + "accept": false, + "reason": "RoMa model addition vs compressed-tensors fp8 quantization work; unrelated." }, { "left": "pull_request:47275", "right": "pull_request:47336", "accept": false, - "reason": "EfficientViTSAM support and offloaded model saving fixes are unrelated changes." + "reason": "EfficientViTSam model support vs offloaded-save bugfix; unrelated." }, { "left": "pull_request:47210", "right": "pull_request:47304", "accept": false, - "reason": "Packed SDPA runtime changes and an MPS docs edit do not match." + "reason": "Packed SDPA runtime optimization vs docs cleanup for MPS graph cache; unrelated." }, { "left": "pull_request:47108", "right": "pull_request:47383", "accept": false, - "reason": "Doc-builder comment changes and Kimi integration tests are unrelated." + "reason": "Doc-builder pin comment tweak vs Kimi integration tests; unrelated." }, { "left": "pull_request:46957", "right": "pull_request:47216", "accept": false, - "reason": "Finegrained FP8 decode behavior and compressed-tensors fp8 docs/code are different quantization efforts." + "reason": "Finegrained FP8 decode optimization vs compressed-tensors quantization/docs; not the same issue." }, { - "left": "pull_request:46912", - "right": "pull_request:47078", + "left": "pull_request:47216", + "right": "pull_request:47336", "accept": false, - "reason": "RoPE testing changes and Japanese documentation wording are unrelated." + "reason": "Quantization format support vs fully-offloaded save bugfix; unrelated." }, { - "left": "pull_request:47179", - "right": "pull_request:47301", + "left": "pull_request:46912", + "right": "pull_request:47078", "accept": false, - "reason": "Granite model support and MTP docs are not the same PR family." + "reason": "RoPE test expansion vs Japanese docs wording; unrelated." }, { "left": "pull_request:47166", "right": "pull_request:47352", "accept": false, - "reason": "Quantization test expansions and distributed runtime utilities are unrelated." - }, - { - "left": "pull_request:43451", - "right": "pull_request:44517", - "accept": false, - "reason": "Molmo2 and Qwen3-TTS are separate model additions." + "reason": "Quantization test updates vs distributed runtime/mixin refactor; unrelated." }, { - "left": "pull_request:47352", - "right": "pull_request:47383", - "accept": false, - "reason": "Distributed runtime utilities and Kimi integration tests do not share a concrete bug or change." - }, - { - "left": "pull_request:44517", - "right": "pull_request:47383", - "accept": false, - "reason": "Qwen3-TTS model support and Kimi integration tests are unrelated." - }, - { - "left": "pull_request:43451", - "right": "pull_request:47143", - "accept": false, - "reason": "Molmo2 and NeuCodec are different model addition PRs." - }, - { - "left": "pull_request:43451", - "right": "pull_request:47352", - "accept": false, - "reason": "Model support for Molmo2 and distributed runtime utilities are unrelated." - }, - { - "left": "pull_request:47243", - "right": "pull_request:47294", - "accept": false, - "reason": "Executorch exporter fixes and kernel-native causal-conv1d functions are different subsystems." - }, - { - "left": "pull_request:47210", - "right": "pull_request:47291", - "accept": false, - "reason": "Packed SDPA attention routing and continuous-batching deprecation/config cleanup are unrelated." - }, - { - "left": "pull_request:46604", - "right": "pull_request:47163", - "accept": false, - "reason": "Mistral tokenizer utilities and XPU paged-attention tests do not fix the same issue." - }, - { - "left": "pull_request:46957", - "right": "pull_request:47383", + "left": "pull_request:46196", + "right": "pull_request:47301", "accept": false, - "reason": "Finegrained FP8/generation kernel work and Kimi integration tests are unrelated." + "reason": "SDPA backend dispatch change vs MTP docs; unrelated." }, { - "left": "pull_request:46604", - "right": "pull_request:47383", + "left": "pull_request:47105", + "right": "pull_request:47298", "accept": false, - "reason": "Mistral tekken utilities and Kimi integration tests are different changes." + "reason": "VITS test re-enable vs MI450 runtime/model-failure fixes; different scopes." }, { - "left": "pull_request:46957", - "right": "pull_request:47360", + "left": "pull_request:47275", + "right": "pull_request:47304", "accept": false, - "reason": "Kernel/quantization work and the OTEL CI workflow change are unrelated." + "reason": "Model support vs docs cleanup; unrelated." }, { - "left": "pull_request:47113", - "right": "pull_request:47175", + "left": "pull_request:45157", + "right": "pull_request:46658", "accept": false, - "reason": "Audio feature/proc fixes and Olmo3 CI expectation updates are not the same bug." + "reason": "Prism GGUF dequantization/import fixes vs Step-3.7 Flash model build/conversion scripts; unrelated." }, { - "left": "pull_request:46381", + "left": "pull_request:47078", "right": "pull_request:47322", "accept": false, - "reason": "Tokenizers dependency bump and Quark config-class changes are separate maintenance fixes." + "reason": "Japanese docs wording vs AMD Quark quantization config changes; unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:46604", + "left": "pull_request:47078", + "right": "pull_request:47243", "accept": false, - "reason": "Molmo2 model support and Mistral tekken utilities are unrelated PRs." + "reason": "Japanese docs wording vs Executorch exporter fixes; unrelated." }, { - "left": "pull_request:46957", + "left": "pull_request:44517", "right": "pull_request:47301", "accept": false, - "reason": "Finegrained FP8 runtime changes and MTP docs are unrelated." + "reason": "Qwen3 TTS model support vs assisted-decoding/MTP docs; unrelated." }, { - "left": "pull_request:47113", - "right": "pull_request:47275", + "left": "pull_request:46895", + "right": "pull_request:47398", "accept": false, - "reason": "Audio requirements fixes and EfficientViTSAM model support are unrelated." + "reason": "CohereASR training-loss bugfix vs GLM MoE config fallback fix; unrelated." }, { - "left": "pull_request:47113", - "right": "pull_request:47243", + "left": "pull_request:47294", + "right": "pull_request:47360", "accept": false, - "reason": "Audio feature/proc fixes and Executorch exporter fixes target different code paths." + "reason": "Kernel function integration work vs CI workflow instrumentation change; unrelated." }, { - "left": "pull_request:46912", - "right": "pull_request:47113", + "left": "pull_request:47113", + "right": "pull_request:47304", "accept": false, - "reason": "RoPE testing updates and audio requirements fixes are unrelated changes." + "reason": "Audio feature/proc requirement fixes vs docs cleanup; unrelated." } ] }, @@ -24463,7 +24399,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-82", + "cluster_id": "cluster-40546-83", "nodes": [ "pull_request:41855", "pull_request:42668", @@ -24472,24 +24408,26 @@ "pull_request:43451", "pull_request:43823", "pull_request:43995", + "pull_request:44517", "pull_request:44827", - "pull_request:45157", "pull_request:45170", "pull_request:45213", "pull_request:45655", "pull_request:46048", - "pull_request:46196", "pull_request:46381", "pull_request:46509", + "pull_request:46604", "pull_request:46658", "pull_request:46836", "pull_request:46895", "pull_request:46912", + "pull_request:46957", "pull_request:46958", "pull_request:47003", "pull_request:47077", "pull_request:47090", "pull_request:47096", + "pull_request:47113", "pull_request:47143", "pull_request:47163", "pull_request:47166", @@ -24497,23 +24435,43 @@ "pull_request:47175", "pull_request:47184", "pull_request:47210", - "pull_request:47216", "pull_request:47243", "pull_request:47275", "pull_request:47291", "pull_request:47292", "pull_request:47294", - "pull_request:47298", "pull_request:47301", "pull_request:47304", "pull_request:47306", "pull_request:47322", - "pull_request:47336", + "pull_request:47352", + "pull_request:47360", "pull_request:47367", "pull_request:47383", - "pull_request:47387" + "pull_request:47399" ], "soft_pairs": [ + "pull_request:46912|pull_request:47291", + "pull_request:47163|pull_request:47352", + "pull_request:46895|pull_request:47399", + "pull_request:43451|pull_request:44517", + "pull_request:47352|pull_request:47383", + "pull_request:44517|pull_request:47383", + "pull_request:47243|pull_request:47294", + "pull_request:44517|pull_request:47143", + "pull_request:43451|pull_request:47352", + "pull_request:47210|pull_request:47291", + "pull_request:46957|pull_request:47383", + "pull_request:46604|pull_request:47163", + "pull_request:46604|pull_request:47383", + "pull_request:46957|pull_request:47360", + "pull_request:47113|pull_request:47175", + "pull_request:46381|pull_request:47322", + "pull_request:43451|pull_request:46604", + "pull_request:46957|pull_request:47301", + "pull_request:47113|pull_request:47275", + "pull_request:47113|pull_request:47243", + "pull_request:46912|pull_request:47113", "pull_request:47077|pull_request:47367", "pull_request:47294|pull_request:47306", "pull_request:47275|pull_request:47367", @@ -24555,372 +24513,420 @@ "pull_request:46509|pull_request:47143", "pull_request:47163|pull_request:47184", "pull_request:47163|pull_request:47166", - "pull_request:46509|pull_request:47163", - "pull_request:45157|pull_request:46836", - "pull_request:46836|pull_request:47387", - "pull_request:47163|pull_request:47292", - "pull_request:47291|pull_request:47322", - "pull_request:45157|pull_request:46509", - "pull_request:47301|pull_request:47336", - "pull_request:47216|pull_request:47383", - "pull_request:47336|pull_request:47383", - "pull_request:46509|pull_request:47387", - "pull_request:47304|pull_request:47367", - "pull_request:46912|pull_request:47298", - "pull_request:47301|pull_request:47383", - "pull_request:47216|pull_request:47301", - "pull_request:46509|pull_request:47301", - "pull_request:47175|pull_request:47298", - "pull_request:43451|pull_request:46836", - "pull_request:47163|pull_request:47210", - "pull_request:47294|pull_request:47298", - "pull_request:45157|pull_request:47387", - "pull_request:46658|pull_request:47387", - "pull_request:46196|pull_request:47143", - "pull_request:47143|pull_request:47163", - "pull_request:46658|pull_request:47143", - "pull_request:46658|pull_request:47322", - "pull_request:47143|pull_request:47387", - "pull_request:47143|pull_request:47322", - "pull_request:45157|pull_request:47143", - "pull_request:47301|pull_request:47367", - "pull_request:43451|pull_request:47243", - "pull_request:47096|pull_request:47243" + "pull_request:46509|pull_request:47163" ], - "prepared_review_unit_hash": "0cc3056d9a0d6feacebca2d23d1c03ae35919c78099899fa18f1953501ca41ed", + "prepared_review_unit_hash": "54a016e2814bf3b7d6b0458ffd8cc667e6d5c990d0e8932c582d27fdfa9deafb", "component_budget": { - "node_count": 82, - "item_count": 82, + "node_count": 83, + "item_count": 83, "soft_pair_count": 227, - "serialized_chars": 281443, - "estimated_input_tokens": 70361, - "estimated_eval_tokens": 140978 + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 72, - "serialized_chars": 131881, - "estimated_input_tokens": 32971, - "estimated_eval_tokens": 66198 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 63, + "serialized_chars": 132890, + "estimated_input_tokens": 33223, + "estimated_eval_tokens": 66702 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Only one true duplicate cluster is present: the CohereASR training-loss double-shift fixes (46895/46958/47090). The rest are mostly unrelated model additions, docs, exporter changes, or test-only updates that share broad themes but not the same concrete bug/change.", + "summary": "Most links are false positives across unrelated model-add, docs, test, and exporter PRs. The only real duplicate cluster is CohereASR training-loss double-shift fixes: 46895, 46958, and 47090 all target the same bug, with 47090 being the best representative.", "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47090 is the clearest canonical representative of the CohereASR double-shift fix: it targets the same code path as the earlier variants, has the most explicit explanation of the bug, and broadens coverage to the related ASR model variants/tests.", + "canonical_pr_reason": "PR 47090 is the best canonical choice: it fixes the same CohereASR double-shift loss bug as 46895/46958, has the explicit issue target, and uses the clearest final approach (`shift_labels` to preserve loss reduction behavior).", "best_issue_reason": null, - "best_pr_reason": "PR 47090 is the strongest standalone fix in the only real duplicate cluster: it is the most complete and readable version of the same loss-shift correction and is the best merge candidate for that underlying change.", + "best_pr_reason": "PR 47090 is the strongest overall representative of the actual duplicate cluster because it is the latest, most complete fix for the same concrete training-loss bug and subsumes the earlier variants.", "soft_edge_verdicts": [ + { + "left": "pull_request:46912", + "right": "pull_request:47291", + "accept": false, + "reason": "RoPE test-value updates vs generation-config deprecation; different code paths and problems." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47352", + "accept": false, + "reason": "Paged-attention XPU test coverage vs distributed runtime utilities; unrelated changes." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47399", + "accept": false, + "reason": "CohereASR loss bug fix vs Cosmos 3 Edge patch packing order; different models and bugs." + }, + { + "left": "pull_request:43451", + "right": "pull_request:44517", + "accept": false, + "reason": "Separate new-model additions (Molmo2 vs Qwen3 TTS); not the same model or change." + }, + { + "left": "pull_request:47352", + "right": "pull_request:47383", + "accept": false, + "reason": "Distributed runtime utilities vs Kimi integration tests; only broad framework similarity." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47383", + "accept": false, + "reason": "Qwen3 TTS model add vs Kimi integration tests; unrelated artifacts." + }, + { + "left": "pull_request:47243", + "right": "pull_request:47294", + "accept": false, + "reason": "ExecuTorch exporter fixes vs kernel-native causal_conv1d support; different concrete functionality." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47143", + "accept": false, + "reason": "Different model additions (Qwen3 TTS vs NeuCodec), not the same underlying change." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47352", + "accept": false, + "reason": "Molmo2 model addition and distributed runtime utilities are unrelated." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47291", + "accept": false, + "reason": "Packed SDPA varlen-attention routing vs continuous-batching config deprecation; different bugs." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47383", + "accept": false, + "reason": "Finegrained FP8 decode optimization vs Kimi integration tests; not the same fix." + }, + { + "left": "pull_request:46604", + "right": "pull_request:47163", + "accept": false, + "reason": "Mistral tekken tokenizer utilities vs paged-attention XPU tests; unrelated." + }, + { + "left": "pull_request:46604", + "right": "pull_request:47383", + "accept": false, + "reason": "Mistral tekken utilities vs Kimi integration tests; different code and intent." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47360", + "accept": false, + "reason": "FP8 grouped/batched kernel optimization vs CI OTEL workflow change; unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47175", + "accept": false, + "reason": "Audio feature/proc requirements fix vs Olmo3 CI expectation refresh; different issues." + }, + { + "left": "pull_request:46381", + "right": "pull_request:47322", + "accept": false, + "reason": "Tokenizer version bump vs AMD Quark config-class rename; unrelated maintenance." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46604", + "accept": false, + "reason": "Molmo2 model add vs Mistral tekken tokenizer utilities; not the same artifact." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47301", + "accept": false, + "reason": "FP8 decode optimization vs MTP documentation; different subsystems." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47275", + "accept": false, + "reason": "Audio requirements fix vs EfficientViTSam model addition; unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47243", + "accept": false, + "reason": "Audio requirements fix vs ExecuTorch exporter fixes; different bugs." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47175", + "accept": false, + "reason": "RoPE-related tests vs Olmo3 CI expectation updates; different test concerns." + }, { "left": "pull_request:47077", "right": "pull_request:47367", "accept": false, - "reason": "Different areas: docs translation for DistilBERT vs pipeline device-placement behavior." + "reason": "Japanese DistilBERT docs translation vs pipeline device-placement clarification; unrelated docs." }, { "left": "pull_request:47294", "right": "pull_request:47306", "accept": false, - "reason": "Unrelated changes: kernel-native causal_conv1d plumbing vs AWS Neuron experiment scripts." + "reason": "Kernel-native causal_conv1d functions vs Neuron experiment scripts; different features." }, { "left": "pull_request:47275", "right": "pull_request:47367", "accept": false, - "reason": "Different underlying work: adding EfficientViT-SAM vs clarifying pipeline device placement." + "reason": "EfficientViTSam model add vs pipeline device-placement docs/code; unrelated." }, { "left": "pull_request:47210", "right": "pull_request:47367", "accept": false, - "reason": "Different code paths: packed SDPA varlen routing vs pipeline device-selection docs/code." + "reason": "Packed SDPA varlen attention vs pipeline device-placement clarification; different changes." }, { "left": "pull_request:43451", "right": "pull_request:46381", "accept": false, - "reason": "Unrelated: new Molmo2 model support vs a tokenizers dependency/parser compatibility update." + "reason": "Molmo2 model add vs tokenizer dependency bump; not the same work." }, { "left": "pull_request:46381", "right": "pull_request:47096", "accept": false, - "reason": "Different changes: tokenizers version support vs exporter fixes for Kimi/XCodec-related models." + "reason": "Tokenizer dependency changes vs ExecuTorch/Kimi exporter work; unrelated." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Different targets: Mistral tokenizer API additions vs ExecuTorch dynamic-shape export tests." + "reason": "Mistral tokenizer missing methods vs ExecuTorch dynamic-shape tests; different artifacts." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Different targets: tokenizer method additions vs adding MobileLLM model support." + "reason": "Mistral tokenizer methods vs MobileLLM model addition; unrelated." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Same broad subsystem (model loading) but not the same bug/change: processor auto-loading robustness vs a new model implementation." + "reason": "Processor auto-loading robustness vs MobileLLM model addition; different changes." }, { "left": "pull_request:43424", "right": "pull_request:47301", "accept": false, - "reason": "ExecuTorch export dynamic-shape tests are unrelated to assisted-decoding docs for MTP." - }, - { - "left": "pull_request:43424", - "right": "pull_request:47383", - "accept": false, - "reason": "Different work: ExecuTorch export tests vs Kimi integration tests." - }, - { - "left": "pull_request:43823", - "right": "pull_request:45655", - "accept": false, - "reason": "Unrelated: adding MobileLLM vs fixing config-class resolution order." + "reason": "ExecuTorch dynamic-shape test vs MTP docs; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:47383", "accept": false, - "reason": "Different fixes: Mistral4 test/debug work vs Kimi integration tests." + "reason": "Mistral4 test fixes vs Kimi integration tests; different models and bugs." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Different code paths: Mistral4 tests vs CLIP-like weight renaming for layrnorm/layernorm." + "reason": "Mistral4 tests vs CLIP-like `layrnorm` rename mapping; unrelated." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Different subsystems: ExecuTorch export tests vs Mistral4 test fixes." + "reason": "ExecuTorch export tests vs Mistral4 tests; only broad testing similarity." }, { "left": "pull_request:43823", - "right": "pull_request:43995", + "right": "pull_request:45213", "accept": false, - "reason": "Unrelated: new MobileLLM model vs Falcon output-interface refactor." + "reason": "MobileLLM model addition vs model-creation skill draft; not the same change." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Different areas: cross-entropy kwargs fix vs MobileLLM model addition." + "reason": "Cross-entropy kwargs fix vs MobileLLM model addition; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:46048", "accept": false, - "reason": "Unrelated: Mistral4 test fixes vs TDT loss kernel integration." + "reason": "Mistral4 test fixes vs TDT loss kernel addition; different code paths." }, { - "left": "pull_request:43424", - "right": "pull_request:43823", + "left": "pull_request:43995", + "right": "pull_request:45213", "accept": false, - "reason": "Different changes: ExecuTorch export behavior vs MobileLLM addition." + "reason": "Falcon output-interface refactor vs model-creation skill draft; unrelated." }, { - "left": "pull_request:43251", - "right": "pull_request:44827", + "left": "pull_request:44827", + "right": "pull_request:45213", "accept": false, - "reason": "Different code paths: cross-entropy kwargs handling vs Mistral4 test fixes." + "reason": "Mistral4 tests vs model-creation skill draft; not the same bug or feature." }, { - "left": "pull_request:44827", - "right": "pull_request:45655", + "left": "pull_request:43995", + "right": "pull_request:44827", "accept": false, - "reason": "Unrelated: Mistral4 test fixes vs config-resolution order fix." + "reason": "Falcon output-interface refactor vs Mistral4 tests; different models/problems." }, { - "left": "pull_request:43995", + "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "Falcon model refactor is unrelated to the model-creation skill scaffolding PR." + "reason": "MobileLLM model add vs model-creation skill draft; unrelated artifacts." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs fix vs MobileLLM model add; different concern." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "No shared bug/change: Mistral4 test fixes vs a model-creation skill/meta PR." + "reason": "Mistral4 tests vs model-creation skill draft; not a duplicate." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor vs model-creation skill draft; unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "Unrelated: MobileLLM model code vs model-creation skill/meta tooling." + "reason": "MobileLLM model add vs model-creation skill draft; different scope." }, { "left": "pull_request:46895", "right": "pull_request:47090", "accept": true, - "reason": "Same CohereASR training-loss double-shift bug and same forward/loss code path." + "reason": "Same CohereASR training-loss double-shift bug; same files and same issue target, with 47090 refining the fix." }, { "left": "pull_request:46958", "right": "pull_request:47090", "accept": true, - "reason": "Same underlying CohereASR loss-shift fix; 47090 is a refined variant of the same change." + "reason": "Same CohereASR loss bug in the same code path; 47090 is the more complete final variant." }, { "left": "pull_request:46895", "right": "pull_request:46958", "accept": true, - "reason": "Both fix the same CohereASR double-shift loss bug in the same model code." + "reason": "Both address the same CohereASR double-shift training-loss issue in the same files." }, { "left": "pull_request:47173", "right": "pull_request:47292", "accept": false, - "reason": "Both are XPU-related test updates, but they fix different model tests and are not the same change." + "reason": "XPU expectation updates in different test suites for different models; not the same bug." }, { "left": "pull_request:47166", "right": "pull_request:47184", "accept": false, - "reason": "Different test suites and different issues; same theme is not enough to merge as one PR." + "reason": "Quantization test tweaks vs Qwen3 Next accelerator test rename; unrelated." }, { "left": "pull_request:46912", "right": "pull_request:47175", "accept": false, - "reason": "Both touch test expectations, but they target different models and different failure causes." + "reason": "RoPE tests vs Olmo3 CI numeric adjustments; different test goals." }, { "left": "pull_request:47096", "right": "pull_request:47383", "accept": false, - "reason": "Different work: exporter/model-loading fixes for Kimi/XCodec vs Kimi integration tests." + "reason": "ExecuTorch exporter fixes vs Kimi integration tests; same model family but different concrete changes." }, { "left": "pull_request:47184", "right": "pull_request:47292", "accept": false, - "reason": "Different XPU test maintenance PRs for different models; not the same underlying bug." + "reason": "Qwen3 Next accelerator test rename vs TipsV2-DPT XPU expectation updates; unrelated." }, { "left": "pull_request:47301", "right": "pull_request:47304", "accept": false, - "reason": "Unrelated docs-only changes: MTP assisted decoding vs MPS graph-cache docs." + "reason": "MTP support docs vs MPS graph-cache docs; different documentation topics." }, { "left": "pull_request:47077", "right": "pull_request:47304", "accept": false, - "reason": "Different documentation topics and unrelated model areas." + "reason": "Japanese DistilBERT docs translation vs MPS graph-cache docs; unrelated." }, { "left": "pull_request:46658", "right": "pull_request:46836", "accept": false, - "reason": "Both are large model additions, but they are completely different models and code paths." + "reason": "Step3.7 Flash model scaffolding vs Unlimited OCR model addition; different models." }, { "left": "pull_request:47210", "right": "pull_request:47301", "accept": false, - "reason": "Different features: packed SDPA routing vs multi-token prediction docs." + "reason": "Packed SDPA varlen-attention code vs MTP docs; unrelated." }, { "left": "pull_request:47003", "right": "pull_request:47243", "accept": false, - "reason": "Both are exporter-related, but they address different exporters and different concrete issues." + "reason": "OpenVINO exporter work vs ExecuTorch exporter fixes; different exporters and code paths." }, { "left": "pull_request:46509", "right": "pull_request:47143", "accept": false, - "reason": "Different model additions: Nemotron Omni vs NeuCodec." + "reason": "Nemotron Omni model add vs NeuCodec model add; different models." }, { "left": "pull_request:47163", "right": "pull_request:47184", "accept": false, - "reason": "Different test fixes for different models; not the same code-path problem." + "reason": "Paged-attention XPU tests vs Qwen3 Next accelerator tests; different test suites." }, { "left": "pull_request:47163", "right": "pull_request:47166", "accept": false, - "reason": "Unrelated test updates across different quantization/generation suites." + "reason": "Paged-attention XPU tests vs quantization test updates; unrelated." }, { "left": "pull_request:46509", "right": "pull_request:47163", "accept": false, - "reason": "Different kinds of changes: new model support vs XPU paged-attention tests." - }, - { - "left": "pull_request:45157", - "right": "pull_request:46836", - "accept": false, - "reason": "Unrelated: PrismML GGUF support vs Unlimited OCR model support." - }, - { - "left": "pull_request:46836", - "right": "pull_request:47387", - "accept": false, - "reason": "Different features: OCR model addition vs one-to-many conversion support." + "reason": "Nemotron Omni model add vs paged-attention XPU tests; unrelated." }, { - "left": "pull_request:47163", + "left": "pull_request:47166", "right": "pull_request:47292", "accept": false, - "reason": "Different model-specific XPU test fixes; not the same underlying bug." - }, - { - "left": "pull_request:47291", - "right": "pull_request:47322", - "accept": false, - "reason": "Unrelated: generation config deprecation vs AMD Quark config-class changes." + "reason": "Quantization test updates vs TipsV2-DPT XPU failures; different models and fixes." }, { - "left": "pull_request:45157", - "right": "pull_request:46509", - "accept": false, - "reason": "Unrelated: PrismML GGUF dequantization support vs Nemotron Omni model support." - }, - { - "left": "pull_request:47301", - "right": "pull_request:47336", - "accept": false, - "reason": "Docs for MTP support are unrelated to fully-offloaded model saving." - }, - { - "left": "pull_request:47216", - "right": "pull_request:47383", - "accept": false, - "reason": "Compressed-tensors fp8 support is unrelated to Kimi integration tests." - }, - { - "left": "pull_request:47304", - "right": "pull_request:47367", - "accept": false, - "reason": "MPS graph-cache docs do not overlap with pipeline device-placement changes." - }, - { - "left": "pull_request:46658", - "right": "pull_request:47387", - "accept": false, - "reason": "Step-3.7 Flash model work is unrelated to one-to-many conversion support." - }, - { - "left": "pull_request:47096", - "right": "pull_request:47243", + "left": "pull_request:47173", + "right": "pull_request:47292", "accept": false, - "reason": "Both mention exporters, but they fix different exporter paths and different concrete bugs." + "reason": "Generic XPU expectation fixes vs TipsV2-DPT-specific test values; not the same change." } ] }, @@ -24942,12 +24948,16 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-82", + "cluster_id": "cluster-40546-83", "nodes": [ "pull_request:40546", "pull_request:43451", + "pull_request:45157", + "pull_request:46196", "pull_request:46509", "pull_request:46658", + "pull_request:46836", + "pull_request:46912", "pull_request:47003", "pull_request:47096", "pull_request:47143", @@ -24955,10 +24965,18 @@ "pull_request:47166", "pull_request:47175", "pull_request:47184", + "pull_request:47210", + "pull_request:47216", + "pull_request:47243", "pull_request:47291", + "pull_request:47292", + "pull_request:47294", "pull_request:47298", + "pull_request:47301", + "pull_request:47304", "pull_request:47306", "pull_request:47322", + "pull_request:47336", "pull_request:47367", "pull_request:47383", "pull_request:47387", @@ -24966,6 +24984,35 @@ "pull_request:47399" ], "soft_pairs": [ + "pull_request:45157|pull_request:46836", + "pull_request:46836|pull_request:47387", + "pull_request:47163|pull_request:47292", + "pull_request:47291|pull_request:47322", + "pull_request:45157|pull_request:46509", + "pull_request:47301|pull_request:47336", + "pull_request:47216|pull_request:47383", + "pull_request:47336|pull_request:47383", + "pull_request:46509|pull_request:47387", + "pull_request:47304|pull_request:47367", + "pull_request:46912|pull_request:47298", + "pull_request:47301|pull_request:47383", + "pull_request:47216|pull_request:47301", + "pull_request:46509|pull_request:47301", + "pull_request:47175|pull_request:47298", + "pull_request:47163|pull_request:47210", + "pull_request:47294|pull_request:47298", + "pull_request:45157|pull_request:47387", + "pull_request:46658|pull_request:47387", + "pull_request:46196|pull_request:47143", + "pull_request:47143|pull_request:47163", + "pull_request:46658|pull_request:47143", + "pull_request:46658|pull_request:47322", + "pull_request:47143|pull_request:47387", + "pull_request:47143|pull_request:47322", + "pull_request:45157|pull_request:47143", + "pull_request:47301|pull_request:47367", + "pull_request:43451|pull_request:47243", + "pull_request:47096|pull_request:47243", "pull_request:43451|pull_request:47291", "pull_request:47166|pull_request:47383", "pull_request:47184|pull_request:47383", @@ -24998,214 +25045,340 @@ "pull_request:47003|pull_request:47387", "pull_request:47096|pull_request:47143" ], - "prepared_review_unit_hash": "7d8172124327d2497462c64f5d066822a0d0cc2645c08ed3a733c6716c5d1d8d", + "prepared_review_unit_hash": "c5ba8e54f93ea214a77f939e52e2e6935f6b09324ddb1bdecdbb17dcba723aaa", "component_budget": { - "node_count": 82, - "item_count": 82, + "node_count": 83, + "item_count": 83, "soft_pair_count": 227, - "serialized_chars": 281443, - "estimated_input_tokens": 70361, - "estimated_eval_tokens": 140978 + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 }, "budget": { - "node_count": 20, - "item_count": 20, - "soft_pair_count": 31, - "serialized_chars": 55844, - "estimated_input_tokens": 13961, - "estimated_eval_tokens": 28178 + "node_count": 32, + "item_count": 32, + "soft_pair_count": 60, + "serialized_chars": 92588, + "estimated_input_tokens": 23147, + "estimated_eval_tokens": 46550 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: it mixes unrelated model-addition PRs, exporter/infrastructure changes, test-only updates, and a few isolated bug fixes. The soft similarity edges are false positives; none look like the same underlying change.", - "confidence": 0.97, + "summary": "These soft links are mostly false positives: the items span unrelated PRs for model additions, docs, tests, exporters, quantization, and bug fixes. I found no pair that clearly shares the same concrete underlying change or bug, so no duplicate cluster emerges.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "No true canonical duplicate exists. If forced to pick a representative, PR 43451 is the broadest model-addition PR, but it is not actually duplicate of the others.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 47399 is the cleanest narrowly scoped fix in the set (Cosmos 3 Edge patch packing order), but it is still unrelated to the other PRs.", + "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "pull_request:43451", - "right": "pull_request:47291", + "left": "pull_request:45157", + "right": "pull_request:46836", "accept": false, - "reason": "Molmo2 model integration vs a ContinuousBatchingConfig deprecation change; different code paths and goals." + "reason": "Different model-support PRs for unrelated architectures (PrismML Bonsai quantization vs Unlimited OCR)." }, { - "left": "pull_request:47166", - "right": "pull_request:47383", + "left": "pull_request:46836", + "right": "pull_request:47387", "accept": false, - "reason": "Quantization test adjustments vs Kimi integration tests; both are test-related but not the same bug or change." + "reason": "Unlimited OCR model support is unrelated to one-to-many conversion support in core loading." }, { - "left": "pull_request:47184", + "left": "pull_request:47163", + "right": "pull_request:47292", + "accept": false, + "reason": "Both mention XPU, but one adds paged-attention test coverage and the other fixes a specific model test; different issues." + }, + { + "left": "pull_request:47291", + "right": "pull_request:47322", + "accept": false, + "reason": "Different config changes for different subsystems (gen deprecation vs AMD Quark config class rename)." + }, + { + "left": "pull_request:45157", + "right": "pull_request:46509", + "accept": false, + "reason": "Unrelated model-support additions for PrismML Bonsai and Nemotron Omni." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47336", + "accept": false, + "reason": "Docs for MTP support and offloaded-model saving bugfix are unrelated changes." + }, + { + "left": "pull_request:47216", "right": "pull_request:47383", "accept": false, - "reason": "Qwen3 Next/VideMT test maintenance vs Kimi integration tests; unrelated models and failures." + "reason": "Compressed-tensors FP8 support and Kimi integration tests do not target the same problem." }, { - "left": "pull_request:43451", - "right": "pull_request:47175", + "left": "pull_request:47336", + "right": "pull_request:47383", "accept": false, - "reason": "Molmo2 model support vs OLMo3 CI expectation updates; different models and different fixes." + "reason": "Offloaded-model saving bugfix and Kimi integration tests are unrelated." }, { - "left": "pull_request:40546", - "right": "pull_request:43451", + "left": "pull_request:46509", + "right": "pull_request:47387", "accept": false, - "reason": "VibeVoice vs Molmo2 are separate model additions; same broad pattern, not the same change." + "reason": "Nemotron Omni support is unrelated to one-to-many conversion support." }, { - "left": "pull_request:47306", + "left": "pull_request:47304", "right": "pull_request:47367", "accept": false, - "reason": "Neuron TP experiments vs pipeline device-placement docs/code clarification; unrelated functionality." + "reason": "Both are docs-oriented, but they address different topics: MPS graph cache vs pipeline device placement." }, { - "left": "pull_request:47298", - "right": "pull_request:47399", + "left": "pull_request:46912", + "right": "pull_request:47298", "accept": false, - "reason": "MI450 model-failure fixes across many models vs a specific Cosmos 3 Edge patch-order bug; not the same issue." + "reason": "RoPE test expansion and MI450 image/model failure fixes are different issues." }, { - "left": "pull_request:47398", - "right": "pull_request:47399", + "left": "pull_request:47301", + "right": "pull_request:47383", "accept": false, - "reason": "GlmMoeDsa num_experts fallback fix vs Cosmos 3 Edge patch packing order; unrelated modules and bugs." + "reason": "MTP docs and Kimi integration tests are unrelated." }, { - "left": "pull_request:43451", + "left": "pull_request:47216", + "right": "pull_request:47301", + "accept": false, + "reason": "Compressed-tensors FP8 support is unrelated to MTP docs." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47301", + "accept": false, + "reason": "Nemotron Omni support and MTP docs do not share the same underlying change." + }, + { + "left": "pull_request:47175", + "right": "pull_request:47298", + "accept": false, + "reason": "Olmo3 CI test updates and MI450 failure fixes are not the same bug." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47210", + "accept": false, + "reason": "Paged-attention XPU tests and packed-SDPA varlen routing are different test/code-path changes." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47298", + "accept": false, + "reason": "Kernel-native causal_conv1d support is unrelated to MI450 image/model fixes." + }, + { + "left": "pull_request:45157", + "right": "pull_request:47387", + "accept": false, + "reason": "PrismML Bonsai support and one-to-many conversion support are unrelated." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47387", + "accept": false, + "reason": "Step-3.7 Flash model work and one-to-many conversion support are unrelated." + }, + { + "left": "pull_request:46196", "right": "pull_request:47143", "accept": false, - "reason": "Molmo2 model support vs NeuCodec support; both are model additions, but for different models and code paths." + "reason": "SDPA dispatch support across audio models and NeuCodec model addition are different changes." }, { - "left": "pull_request:43451", + "left": "pull_request:47143", "right": "pull_request:47163", "accept": false, - "reason": "Molmo2 model addition vs XPU paged-attention test support; not the same underlying change." + "reason": "NeuCodec model support and paged-attention XPU tests are unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:47383", + "left": "pull_request:46658", + "right": "pull_request:47143", "accept": false, - "reason": "Molmo2 model support vs Kimi integration tests; different artifacts and purposes." + "reason": "Step-3.7 Flash model support and NeuCodec support are different model additions." }, { - "left": "pull_request:43451", - "right": "pull_request:46509", + "left": "pull_request:46658", + "right": "pull_request:47322", "accept": false, - "reason": "Molmo2 support vs Nemotron Omni support; separate model integrations." + "reason": "Step-3.7 Flash work is unrelated to AMD Quark config changes." }, { - "left": "pull_request:47298", - "right": "pull_request:47398", + "left": "pull_request:47143", + "right": "pull_request:47387", "accept": false, - "reason": "MI450 platform fixes vs a GlmMoeDsa config bugfix; unrelated scope." + "reason": "NeuCodec support and one-to-many conversions are unrelated." }, { - "left": "pull_request:46658", - "right": "pull_request:47399", + "left": "pull_request:47143", + "right": "pull_request:47322", "accept": false, - "reason": "Step-3.7 Flash conversion/model build work vs Cosmos 3 Edge patch packing; different models and tasks." + "reason": "NeuCodec support and Quark config changes are unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:47096", + "left": "pull_request:45157", + "right": "pull_request:47143", + "accept": false, + "reason": "PrismML Bonsai support and NeuCodec support are unrelated model additions." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47367", "accept": false, - "reason": "Molmo2 integration vs exportability work for Kimi; unrelated model support changes." + "reason": "MTP docs and pipeline device-placement docs are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47003", + "right": "pull_request:47243", "accept": false, - "reason": "Molmo2 model addition vs OpenVINO exporter addition; one is model support, the other is export infrastructure." + "reason": "Molmo2 model addition and Executorch exporter fixes are unrelated." }, { - "left": "pull_request:47003", - "right": "pull_request:47096", + "left": "pull_request:47096", + "right": "pull_request:47243", "accept": false, - "reason": "OpenVINO exporter infrastructure vs Kimi exportability; related broadly to export, but not the same concrete change." + "reason": "Kimi exportability work and Executorch exporter fixes target different concrete code paths." }, { "left": "pull_request:43451", - "right": "pull_request:47322", + "right": "pull_request:47291", "accept": false, - "reason": "Molmo2 model support vs AMD Quark config-class compatibility fix; unrelated subsystems." + "reason": "Molmo2 support and continuous-batching config deprecation are unrelated." }, { - "left": "pull_request:47096", - "right": "pull_request:47322", + "left": "pull_request:47166", + "right": "pull_request:47383", + "accept": false, + "reason": "Quantization test updates and Kimi integration tests are unrelated." + }, + { + "left": "pull_request:47184", + "right": "pull_request:47383", "accept": false, - "reason": "Kimi exportability work vs Quark config-class changes; different tooling and code paths." + "reason": "Qwen3 Next multi-accelerator test changes and Kimi integration tests are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47298", + "right": "pull_request:47175", + "accept": false, + "reason": "Molmo2 model addition and Olmo3 CI test updates are unrelated." + }, + { + "left": "pull_request:40546", + "right": "pull_request:43451", + "accept": false, + "reason": "VibeVoice and Molmo2 are separate model-addition PRs with different code paths." + }, + { + "left": "pull_request:47306", + "right": "pull_request:47367", "accept": false, - "reason": "Molmo2 model addition vs MI450 failure fixes; no shared underlying bug." + "reason": "Neuron training experiments and pipeline device-placement docs are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47398", + "right": "pull_request:47298", "accept": false, - "reason": "Molmo2 model addition vs GlmMoeDsa config fallback fix; unrelated." + "reason": "Molmo2 support and MI450 image/model fixes are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47399", + "right": "pull_request:47398", "accept": false, - "reason": "Molmo2 model addition vs Cosmos patch packing fix; unrelated model and bug." + "reason": "Molmo2 support and GLM MoE DSA config fallback are unrelated." }, { "left": "pull_request:47096", "right": "pull_request:47398", "accept": false, - "reason": "Kimi exportability work vs GlmMoeDsa config fix; unrelated." + "reason": "Kimi exporter work and GLM MoE config fallback are unrelated." }, { "left": "pull_request:47096", "right": "pull_request:47399", "accept": false, - "reason": "Kimi exportability work vs Cosmos patch packing fix; unrelated." + "reason": "Kimi exporter work and Cosmos 3 Edge patch packing order fixes are unrelated." }, { "left": "pull_request:43451", "right": "pull_request:47306", "accept": false, - "reason": "Molmo2 model addition vs Neuron experiments; not the same change." + "reason": "Molmo2 support and Neuron experiments are unrelated." }, { "left": "pull_request:47096", - "right": "pull_request:47306", + "right": "pull_request:47387", + "accept": false, + "reason": "Kimi exportability work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47387", + "accept": false, + "reason": "OpenVINO exporter work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47143", "accept": false, - "reason": "Kimi exportability work vs Neuron training experiments; different stack and goal." + "reason": "Kimi exporter work and NeuCodec support are unrelated." }, { "left": "pull_request:47003", "right": "pull_request:47398", "accept": false, - "reason": "OpenVINO exporter addition vs GlmMoeDsa config fix; unrelated." + "reason": "OpenVINO exporter work and GLM MoE config fallback are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47306", + "accept": false, + "reason": "OpenVINO exporter work and Neuron experiments are unrelated." }, { "left": "pull_request:47003", "right": "pull_request:47387", "accept": false, - "reason": "OpenVINO exporter work vs one-to-many conversion support; both touch infra, but not the same concrete problem." + "reason": "OpenVINO exporter work and one-to-many conversions are unrelated." }, { "left": "pull_request:47096", "right": "pull_request:47387", "accept": false, - "reason": "Kimi exportability vs one-to-many conversion support; related through export tooling, but not a duplicate change." + "reason": "Kimi exportability work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47387", + "accept": false, + "reason": "OpenVINO exporter work and one-to-many conversions are unrelated." }, { "left": "pull_request:47096", "right": "pull_request:47143", "accept": false, - "reason": "Kimi exportability work vs NeuCodec model support; separate model integrations." + "reason": "Kimi exporter work and NeuCodec support are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47398", + "accept": false, + "reason": "OpenVINO exporter work and GLM MoE config fallback are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47306", + "accept": false, + "reason": "OpenVINO exporter work and Neuron experiments are unrelated." } ] }, @@ -25235,7 +25408,7 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "29eba4f72357f956b3c2220a220b35867676c2f9cb40b48f17ae352e7fcdc1a0", + "prepared_review_unit_hash": "0aaa3b2aa82331cfed11462fa4a70f7c07630bc65c4285cf1aa2c85354d4e381", "component_budget": { "node_count": 2, "item_count": 2, @@ -25257,7 +25430,7 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests are unrelated feature additions in different model families: TimesFM covariates support vs. DINOv3 image classification support. They do not describe the same underlying bug or change, so the soft duplicate edge should be rejected.", + "summary": "The two pull requests are unrelated: one adds TimesFM covariate forecasting, the other adds DINOv3 image classification support. They touch different model families, different code paths, and do not represent the same underlying change.", "confidence": 0.99, "canonical_issue_reason": null, "canonical_pr_reason": null, @@ -25268,7 +25441,7 @@ "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different subsystems and goals: one adds TimesFM forecasting with covariates, the other adds DINOv3 image classification support. No shared code path or mergeable single PR." + "reason": "Different models and features: TimesFM forecasting with covariates versus DINOv3 image classification. No shared bug or mergeable change scope." } ] }, @@ -25361,7 +25534,7 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "7008189458b5b4b5e066dea9b537c6b8ad3531d3b2ba62a0da3ee0e117019c0f", + "prepared_review_unit_hash": "27758471f7b85818ab4e27aa6a8c1c3e2f6737b666fa79c1e1bee65076332e04", "component_budget": { "node_count": 2, "item_count": 2, @@ -25383,18 +25556,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two workflow-maintenance PRs with overlapping GitHub Actions upgrades, but they target different action families and different underlying motivations; they are related, not duplicates.", - "confidence": 0.92, + "summary": "Two unrelated workflow-maintenance pull requests: one upgrades several GitHub Actions for Node 24 compatibility, the other bumps Docker build actions to newer versions. They overlap in some workflow files but do not fix the same underlying problem or represent a mergeable single change.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 42975 is the best representative because it has the clearest concrete goal\u2014Node 24 compatibility for workflow actions\u2014and a coherent set of related action-version bumps across many workflows.", + "canonical_pr_reason": "pull_request:42975 is the more coherent representative: it has a single explicit goal (Node 24 compatibility) and a consistent set of action-version upgrades across workflows.", "best_issue_reason": null, - "best_pr_reason": "PR 42975 is the strongest standalone candidate: its title matches the diff, the scope is consistent, and the change set is tied to a single compatibility concern rather than a generic upgrade sweep.", + "best_pr_reason": "pull_request:42975 is the best overall representative because its intent is clearer and broader, while 42976 is a narrower dependency bump focused on Docker build actions.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Both PRs update GitHub Actions in CI workflows, but 42975 focuses on Node 24-compatible action releases, while 42976 updates docker/build-push-action and other latest versions. They do not fix the same concrete code-path problem and would not naturally merge into one PR." + "reason": "Different maintenance updates: 42975 upgrades checkout/cache/upload-artifact for Node 24 compatibility; 42976 upgrades docker/build-push-action v5->v6. Shared workflow files are incidental, not the same concrete change." } ] }, @@ -25429,7 +25602,7 @@ "pull_request:43448|pull_request:45490", "pull_request:45490|pull_request:46562" ], - "prepared_review_unit_hash": "d4aa0e718ee7910dbc2292f6640b79b766aa02ab6e291c51515b79ac99488952", + "prepared_review_unit_hash": "3bf7fb48b380a2a02621681d15d297bc6f84abec588215ae7f3d407f6ddb5f91", "component_budget": { "node_count": 4, "item_count": 4, @@ -25451,36 +25624,36 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All four items are unrelated model-addition pull requests that happen to share Transformer integration boilerplate (docs, auto mappings, tests). They are not duplicates of the same underlying bug/change.", - "confidence": 0.96, + "summary": "These are four separate model-addition PRs that only overlap on shared infrastructure files like docs/auto mappings; they do not look like the same underlying change, so none should be clustered as duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "#46562 is the best single representative only because it is the most substantial and highest-activity feature PR in the set; the others are separate model additions, not duplicates.", + "canonical_pr_reason": "No canonical PR: each PR introduces a different model (ViT NEPA, Molmo, CTSM, Step3p7) with distinct code paths and tests, despite some shared registry/docs edits.", "best_issue_reason": null, - "best_pr_reason": "#46562 has the broadest, most complete model-integration footprint and the most review activity, so it is the best representative PR for this cluster.", + "best_pr_reason": "No single PR is a good duplicate representative for this cluster because the overlapping files are generic framework plumbing rather than evidence of one shared fix/change.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo. Shared auto/doc files are generic integration boilerplate, not the same code-path fix." + "reason": "Different model additions (ViT NEPA vs Molmo); shared auto/docs files are generic registration work, not the same concrete change." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different model additions: ViT NEPA vs CTSM. They touch similar framework plumbing but implement unrelated models." + "reason": "These add unrelated models (ViT NEPA vs CTSM) and touch different model implementations; overlap is only broad model registry/docs plumbing." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Different model additions: Molmo vs CTSM. Overlap is only in shared transformer registry/docs files, not one concrete bug or change." + "reason": "Molmo and CTSM are distinct model integrations with different code paths; shared files do not imply a single duplicate PR." }, { "left": "pull_request:45490", "right": "pull_request:46562", "accept": false, - "reason": "Different model additions: CTSM vs Step3p7. Both are model PRs with shared auto-mapping edits, but they are distinct features and not mergeable as one fix." + "reason": "CTSM and Step3p7 are different model additions; despite heavier overlap in auto-mapping files, they are not the same underlying change." } ] }, @@ -25531,7 +25704,7 @@ "pull_request:46539|pull_request:46922", "pull_request:46886|pull_request:46922" ], - "prepared_review_unit_hash": "8ea4c07ffc78b4c7c717c7ce8e5295cf62be750767f81989db184485604ae8b9", + "prepared_review_unit_hash": "602097d29f649eee4712ce34ce9ef23fc1bfe3bd34b778d44d308797e8529f27", "component_budget": { "node_count": 8, "item_count": 8, @@ -25553,114 +25726,114 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Mostly unrelated pull requests spanning whisper tests, dummy test edits, FP8 dequantization, generation validation, CI memory tweaks, and a README link change. The only likely duplicate pair is the two YOLOS device-mismatch fixes, which address the same `device_map=\"auto\"` bug in the same model file.", + "summary": "Most items are unrelated PRs bundled by soft similarity. The only strong duplicate-like pair is the YOLOS device-mismatch fixes (46886 and 46922), which target the same issue and code path; the rest should stay separate.", "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46922 is the best representative for the YOLOS bug: it fixes the concrete device mismatch in the forward pass by moving `mid_position_embeddings` to `hidden_states.device` and also adds the relevant no-split modules.", + "canonical_pr_reason": "PR 46922 is the best representative because it addresses the concrete YOLOS forward/device-mismatch fix directly in model code and also includes the split-module safeguard; it is the more complete version of the same change as 46886.", "best_issue_reason": null, - "best_pr_reason": "PR 46922 is the strongest cluster representative because it is the most complete fix for the only clear duplicate bug in the set; the other PRs are unrelated test, CI, or subsystem changes.", + "best_pr_reason": "PR 46922 is the strongest standalone PR in this set for triage purposes: it has an explicit issue target, touches the real production code path, and is narrowly about one bug fix rather than test churn or unrelated maintenance.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Both are test-oriented, but they fix different things: Whisper tokenizer/tests vs. placeholder asserts in CLIP/ViT tests. Not the same underlying bug." + "reason": "Both are test-oriented PRs, but they affect different models and do not fix the same underlying bug or code path." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper decode/test changes and FP8 dequant/loading fixes are unrelated code paths and problems." + "reason": "Whisper tokenizer/test fix and FP8 dequant logic are unrelated subsystems with different failure modes." }, { "left": "pull_request:43291", "right": "pull_request:46380", "accept": false, - "reason": "Whisper decoding vs. generation watermarking validation are different features with no shared bug." + "reason": "Whisper decoding tests and generation-watermarking validation are separate issues." }, { "left": "pull_request:43291", "right": "pull_request:46922", "accept": false, - "reason": "Whisper tokenizer/test work is unrelated to YOLOS device placement/device_map behavior." + "reason": "Whisper tests and YOLOS device-mismatch fixes are unrelated." }, { "left": "pull_request:43665", "right": "pull_request:46539", "accept": false, - "reason": "Dummy test edits for CLIP/ViT do not match the Phimoe CI memory-limit/integration-test fix." + "reason": "These are both test-heavy PRs, but they target different models and different behaviors; no shared bug is evident." }, { "left": "pull_request:43665", "right": "pull_request:47393", "accept": false, - "reason": "Placeholder test edits and a README URL update are unrelated." + "reason": "CLIP/ViT test edits and a README URL change are unrelated." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "One is fake test failures, the other is a quantization/dequantization fix; no common concrete bug." + "reason": "Model test failures in CLIP/ViT are not the same as the FP8 dequant fix." }, { "left": "pull_request:43665", "right": "pull_request:46380", "accept": false, - "reason": "Test placeholder assertions and generation-config watermark validation are not the same change." + "reason": "Different components and different bug classes; test changes do not indicate the same underlying fix." }, { "left": "pull_request:43665", "right": "pull_request:46922", "accept": false, - "reason": "Generic test edits do not match the YOLOS runtime device mismatch fix." + "reason": "CLIP/ViT test updates are unrelated to YOLOS runtime device placement." }, { "left": "pull_request:44815", "right": "pull_request:46539", "accept": false, - "reason": "FP8 dequantization/model-loading changes are unrelated to CI resource patching for Phimoe." + "reason": "FP8 quantization/dequantization logic and CI Phimoe integration-test infrastructure are different concerns." }, { "left": "pull_request:44815", "right": "pull_request:46380", "accept": false, - "reason": "Different subsystems and failure modes: FP8 quantization vs. generation config validation." + "reason": "The FP8 kernel/model-loading fix and watermarking/generation validation are unrelated code paths." }, { "left": "pull_request:44815", "right": "pull_request:46922", "accept": false, - "reason": "FP8 dequant/loading fixes and YOLOS device placement are distinct bugs." + "reason": "These PRs fix different model/runtime issues; no same bug or mergeable combined change is shown." }, { "left": "pull_request:46380", "right": "pull_request:46539", "accept": false, - "reason": "Generation watermark validation and Phimoe CI runner memory handling are unrelated." + "reason": "Generation watermarking validation and Phimoe CI/device-memory setup are unrelated." }, { "left": "pull_request:46380", "right": "pull_request:46922", "accept": false, - "reason": "Watermarking-config validation is not the same as fixing YOLOS device mismatch." + "reason": "A generation-config validation fix is not the same bug as YOLOS device mismatch." }, { "left": "pull_request:46539", "right": "pull_request:46922", "accept": false, - "reason": "CI CPU-memory patching for Phimoe integration tests is unrelated to the YOLOS model runtime bug." + "reason": "Phimoe CI memory/device-map handling and YOLOS model device placement are different issues." }, { "left": "pull_request:46886", "right": "pull_request:46922", "accept": true, - "reason": "Both target the same YOLOS `device_map=\"auto\"` mismatch bug in `modeling_yolos.py`; 46922 is a fuller version of the same fix and could plausibly be merged into one PR." + "reason": "Both PRs address the same YOLOS device-mismatch problem under device_map=\"auto\" in the same file, and 46922 plausibly subsumes 46886's fix." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet: it correctly treats most PRs as unrelated and identifies the only plausible duplicate pair as the two YOLOS fixes. The accepted soft-edge verdict is conservative because both PRs touch the same model file and share the same issue target, and 46922 plausibly subsumes 46886 rather than stretching to unrelated changes." + "feedback": "The analysis is grounded in the packet and stays conservative. The duplicate-like relationship between 46886 and 46922 is well supported: they share the same issue target and file, and 46922 clearly subsumes 46886 by adding the runtime device-placement fix plus the split-module safeguard. The other pairwise rejections are appropriately cautious given only soft similarity evidence." }, "evaluator_used": true, "retried": false, @@ -25690,7 +25863,7 @@ "pull_request:43488|pull_request:43532", "pull_request:43532|pull_request:46747" ], - "prepared_review_unit_hash": "cb99ded73cfbccafb452b2381730b069075800683a013ca84cbfa6149c5f2848", + "prepared_review_unit_hash": "d0cd80cd8783ddd95b6a182b6a787f047a9458af93948071d32182c68cd30a42", "component_budget": { "node_count": 3, "item_count": 3, @@ -25712,30 +25885,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three PRs are only superficially similar as '[don't merge]' or CI/test-related changes, but they modify different workflows and code paths, so this cluster does not look like a true duplicate set.", - "confidence": 0.95, + "summary": "All three items are synthetic '[don't merge]' pull requests, but they target different behaviors: version/setup bot checks, workflow diff/show tooling, and a security-gate workflow trigger change. The title similarity is superficial; there is no clear duplicate PR cluster here.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "If one PR must be treated as the cluster representative, 46747 is the most self-contained and concrete change: a single workflow file with a clear trigger shift.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "46747 is the best standalone PR candidate because it has the narrowest scope and the clearest code-path change, while the others are broader or mix multiple unrelated workflow/test changes.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:46747", "accept": false, - "reason": "Different changes: 43488 edits setup.py/dependency versioning and a ViT test, while 46747 changes the PR security-gate workflow trigger. They do not appear to fix the same underlying problem." + "reason": "Different purposes and files: 43488 is a setup/version-format and test bot probe, while 46747 changes a PR security-gate workflow trigger. Not the same underlying change." }, { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "43488 is a repo-bot/setup.py and test tweak; 43532 rewires several GitHub workflows plus testing utilities. Similar test-related intent is too broad to treat as the same PR." + "reason": "43488 checks repository formatting/versioning behavior; 43532 rewires several CI/notification workflows and testing helpers. Shared test-file overlap is not enough to call them the same bug/change." }, { "left": "pull_request:43532", "right": "pull_request:46747", "accept": false, - "reason": "Both touch workflows, but the actual behavior changes are different: 43532 updates self-comment/scheduled/slack automation and test helpers, whereas 46747 changes the security-gate workflow from pull_request_target to pull_request. Not the same concrete change." + "reason": "Both touch workflows, but they alter different CI paths for different goals: 43532 is a show-diff/commenting workflow change, while 46747 changes the security-gate trigger semantics. Not mergeable as one fix." } ] }, @@ -25765,7 +25938,7 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "119ee3cc98f4a85471e5b9a45f1988655f4fffdb627f8b1b77666a70e8ca2048", + "prepared_review_unit_hash": "e3de25ce051f50682a9e3eedabb4cd41cec242fe47e84f4499f00d0864cb70f9", "component_budget": { "node_count": 2, "item_count": 2, @@ -25787,18 +25960,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Related but distinct PRs: one adds a Promptable Concept Segmentation pipeline, the other adds a Promptable Visual Segmentation pipeline. They share scaffold/docs, but they target different tasks and model integrations, so they should not be deduplicated.", - "confidence": 0.94, + "summary": "These are related segmentation-feature PRs, but they implement different pipelines with different model integrations and task docs, so they should not be merged as duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR as a duplicate representative: the two PRs implement different segmentation pipelines with different concrete targets.", + "canonical_pr_reason": "No canonical PR: 43612 adds Promptable Concept Segmentation, while 43613 adds Promptable Visual Segmentation; they target different features and code paths.", "best_issue_reason": null, - "best_pr_reason": "PR 43613 is the broader/more complete segmentation pipeline addition, but it is still a separate change from 43612 rather than a duplicate.", + "best_pr_reason": "43613 is the better representative if one must be chosen because it is the broader and more fully integrated feature addition, but it is still distinct from 43612.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Different underlying features: concept segmentation vs visual segmentation. Shared docs/pipeline plumbing is not enough to treat them as the same code change, and they would not plausibly be merged into one PR without conflating two separate tasks." + "reason": "Reject: these are separate pipeline additions for different tasks, with different filenames, model hookups, and tests. They share scaffolding but not the same underlying change or fix." } ] }, @@ -25828,7 +26001,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "5e87c721a482522fe6be335509fb6ca76eae334bd2e8229e0a6fa9d4802d3916", + "prepared_review_unit_hash": "35fb69ad37cc0d5e4879915b38652f88706472648425f607270415895289615d", "component_budget": { "node_count": 2, "item_count": 2, @@ -25850,18 +26023,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs touch `trainer.py` and expose new overridable hooks for custom trainers, but they address different code paths: gradient-accumulation loss scaling vs. signature-column filtering. They are related by extensibility style, not by the same bug/change.", - "confidence": 0.94, + "summary": "These are two separate trainer customizability changes in the same file: one adds a hook for gradient-accumulation loss scaling, the other adds a hook for signature-column filtering. They are related at a high level but do not fix the same bug or change the same code path.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43651 is the more concrete behavioral change: it fixes how training loss is scaled during gradient accumulation and adds an explicit hook for custom trainers. PR #43663 is a separate refactor/hook for input column filtering, so it is not the best canonical representative of the cluster.", + "canonical_pr_reason": "PR #43651 is the better canonical representative because it introduces the more concrete behavioral change in the training step and clearly defines a new loss-scaling hook. PR #43663 is a separate refactor for input-column filtering.", "best_issue_reason": null, - "best_pr_reason": "PR #43651 is the best overall PR in this set because it makes a direct functional change to training behavior and introduces a narrowly scoped extensibility point. PR #43663 is useful but is a separate trainer customization concern.", + "best_pr_reason": "PR #43651 is the strongest standalone candidate: it targets a specific training-step bug/behavior around gradient accumulation and has a clear, self-contained code-path change. PR #43663 is also valid but addresses a different trainer extensibility point.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "They are both trainer extensibility changes, but they do not fix the same underlying bug or implement the same feature. One changes gradient-accumulation loss scaling; the other adds an override for signature-based column filtering." + "reason": "Both touch trainer.py and add customization hooks, but one is about loss scaling during gradient accumulation while the other is about signature-column filtering. They are different behaviors and would not plausibly be merged as one PR fixing the same underlying change." } ] }, @@ -25891,12 +26064,12 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "32ee13cd7304ef28ef8b16669d2496b6491705d4f55701720338760fcdf3662b", + "prepared_review_unit_hash": "19e95ebbd9da51e24862e164239093779e7e185e4f493ca817b129b130ead3ef", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3664, + "serialized_chars": 3663, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -25904,7 +26077,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3664, + "serialized_chars": 3663, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -25913,24 +26086,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are alternative fixes for the same CLI `TypeAdapter`/type-annotation NameError in `src/transformers/cli/serve.py`; they differ in implementation style but target the same runtime failure path.", - "confidence": 0.89, + "summary": "Two open PRs both address a TypeAdapter-related NameError in `src/transformers/cli/serve.py`, with one broader typing cleanup and one narrower pydantic-absence fix; they look closely related enough to treat as the same bug cluster.", + "confidence": 0.74, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43842 is the tighter, more direct fix for the `TypeAdapter` NameError: it keeps the change localized to the runtime/type-checking split and avoids broader annotation refactors.", + "canonical_pr_reason": "PR #43842 is the more direct and narrowly scoped fix for the reported runtime NameError when pydantic is unavailable, with explicit `TypeAdapter` handling under `TYPE_CHECKING`.", "best_issue_reason": null, - "best_pr_reason": "PR 43842 is the best representative of the cluster because it addresses the concrete failure with the smallest surface area and most directly matches the reported runtime bug.", + "best_pr_reason": "PR #43842 is the best representative because it most clearly targets the concrete failure mode and has the cleaner, more specific fix description.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", "accept": true, - "reason": "Both PRs fix the same underlying `TypeAdapter` NameError in the CLI serve code path; one uses postponed annotations, the other uses a `TYPE_CHECKING` import plus stringized annotation, so they are alternative implementations of the same bug fix." + "reason": "Both PRs target the same TypeAdapter NameError in the CLI serve code path and their changes overlap enough that they could plausibly be merged into one fix." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: both PRs touch the same file, target the same CLI `TypeAdapter`/annotation NameError class of failure, and their fixes are plausibly alternative implementations of one bugfix. The soft-edge acceptance is conservative enough here, and the pair looks mergeable into a single PR without requiring incompatible changes." + "feedback": "Grounded in the packet: both PRs target a TypeAdapter-related NameError in `src/transformers/cli/serve.py`, and the cluster summary stays appropriately cautious about their relationship. The soft-edge verdict is conservative enough because it only claims the changes could plausibly be merged into one fix." }, "evaluator_used": true, "retried": false, @@ -26120,382 +26293,460 @@ "pull_request:44044|pull_request:44154", "pull_request:44024|pull_request:44076" ], - "prepared_review_unit_hash": "fc96ad4b26ebca289fb0e23eee84181a3da2aee0428e1a868b00aeef7a0ab7df", + "prepared_review_unit_hash": "8db6007d91526401d4a8fdac61f61195bfd7c00ca0be8203f4170ec5c36a0927", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84514, - "estimated_input_tokens": 21129, - "estimated_eval_tokens": 42514 + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 }, "budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 72, - "serialized_chars": 76945, - "estimated_input_tokens": 19237, - "estimated_eval_tokens": 38730 + "serialized_chars": 76941, + "estimated_input_tokens": 19236, + "estimated_eval_tokens": 38728 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of independent, model-specific output-tracing refactors tied to the same umbrella tracking issue (#43979), not one duplicate bug. The only clear duplicates are within the same concrete GPT-J/GPT-Neo code paths.", - "confidence": 0.79, + "summary": "This cluster is mostly a batch of model-specific output-tracing refactors, not one duplicate set. Only a few same-family GPT-Neo/GPT-J PRs look like true rerolls of the same change; the rest touch different model code paths and should stay separate.", + "confidence": 0.61, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44066 is the clearest representative: it has the explicit #43979 link and a concrete GPT-J/CodeGen output-tracing refactor with overlapping follow-up work in 44722.", + "canonical_pr_reason": "PR 44722 is the strongest representative of the true duplicate subgroup: it is the latest and most complete GPT-J/CodeGen output-tracing refactor, with the same concrete file-level change as the earlier GPT-J PRs.", "best_issue_reason": null, - "best_pr_reason": "PR 44066 is the best single representative of the cluster theme because it is explicit, concrete, and central to the GPT-J output-tracing migration that appears to have follow-up duplicates.", + "best_pr_reason": "PR 44722 is the best global PR candidate because it most clearly captures the shared GPT-J/CodeGen output-tracing refactor and is broader/more complete than the earlier GPT-J variants.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different model families; same umbrella refactor theme and issue target are not enough to make these duplicates." + "reason": "Different model families and code paths; same tracing theme/issue tag is not enough." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "These touch unrelated model implementations; only the output-tracing initiative matches." + "reason": "CVT/FNet vs DeBERTa v2 are unrelated implementations; not the same underlying change." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Same GPT-J output-tracing refactor in the same code path; this looks like an alternate/overlapping version of the same change." + "reason": "Same GPT-J output-tracing refactor family and overlapping file-level change; this looks like a reroll/duplicate." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Both are output-tracing refactors, but they are for different model families and code paths." + "reason": "ResNet/RT-DETR ResNet vs EfficientNet are different model paths; not mergeable as one fix." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "Different model families; shared issue linkage does not make them the same change." + "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "GPT-J/CodeGen versus EfficientNet: unrelated concrete code paths." + "reason": "GPT-J/CodeGen output tracing vs EfficientNet output tracing are different model-specific changes." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa are unrelated implementations; only the refactor style matches." + "reason": "MobileNetV2 and DeBERTa v2 are unrelated; same broad initiative does not make them duplicates." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "Different model code paths; same output-tracing pattern only." + "reason": "GPT-J/CodeGen and MGP-STR are different models and concrete fixes." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen and MPT are separate model implementations." + "reason": "GPT-J/CodeGen vs MPT are separate model implementations." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Both are GPT-Neo output-tracing migrations in the same file/code path, so they look like the same underlying change." + "reason": "Both are GPT-Neo output-tracing refactors on the same file family; one is effectively a rerun of the same change." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are different model families." + "reason": "GPT-J/CodeGen vs GPT-Neo are different model-specific refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo and GPT-J are related but distinct code paths; not the same concrete fix." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet output tracing and GPT-J/CodeGen output tracing are unrelated patches." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are different model implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet vs GPT-J are different fixes, even if both mention output tracing." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are different model families and code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet and MPT are unrelated model-specific changes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are different models with different output-tracing code." }, { - "left": "pull_request:44018", - "right": "pull_request:44066", + "left": "pull_request:44068", + "right": "pull_request:44722", "accept": false, - "reason": "Different model families; only the general output-tracing refactor matches." + "reason": "GPT-Neo and GPT-J/CodeGen are not the same underlying patch." }, { - "left": "pull_request:44019", - "right": "pull_request:44722", + "left": "pull_request:44018", + "right": "pull_request:44116", "accept": false, - "reason": "ResNet-family versus GPT-J/CodeGen: unrelated concrete code paths." + "reason": "GPT-Neo and Flaubert are unrelated model refactors." }, { - "left": "pull_request:44086", + "left": "pull_request:44071", "right": "pull_request:44722", "accept": false, - "reason": "Different models and different files; same initiative only." + "reason": "MPT and GPT-J/CodeGen are different code paths." }, { - "left": "pull_request:44019", + "left": "pull_request:44018", "right": "pull_request:44085", "accept": false, - "reason": "ResNet-family versus GPT-J: not the same underlying bug or change." + "reason": "GPT-Neo and GPT-J are related but distinct model-specific refactors." }, { "left": "pull_request:44018", - "right": "pull_request:44019", + "right": "pull_request:44086", "accept": false, - "reason": "GPT-Neo and ResNet are different code paths despite similar refactor style." + "reason": "GPT-Neo and MGP-STR are different concrete changes." }, { - "left": "pull_request:44019", - "right": "pull_request:44071", + "left": "pull_request:44068", + "right": "pull_request:44086", "accept": false, - "reason": "ResNet-family and MPT are unrelated implementations." + "reason": "GPT-Neo and MGP-STR are unrelated model implementations." }, { - "left": "pull_request:44068", - "right": "pull_request:44116", + "left": "pull_request:44071", + "right": "pull_request:44085", "accept": false, - "reason": "GPT-Neo and Flaubert are unrelated model code paths." + "reason": "MPT and GPT-J are separate model code paths." }, { - "left": "pull_request:44073", - "right": "pull_request:44101", + "left": "pull_request:44085", + "right": "pull_request:44086", "accept": false, - "reason": "VisualBert and XLM are different model implementations; only the refactor pattern matches." + "reason": "GPT-J and MGP-STR are different architectures and fixes." }, { - "left": "pull_request:44056", - "right": "pull_request:44101", + "left": "pull_request:44059", + "right": "pull_request:44086", "accept": false, - "reason": "MPNet and XLM are unrelated code paths." + "reason": "GPT-2 and MGP-STR are unrelated concrete changes." }, { - "left": "pull_request:44074", + "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "TextNet and XLM are different models; not a duplicate change." + "reason": "ViLT and XLM are different model families; not a duplicate fix." }, { - "left": "pull_request:44056", + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are separate output-tracing refactors." + }, + { + "left": "pull_request:44017", "right": "pull_request:44161", "accept": false, - "reason": "MPNet and LongT5 are separate implementations." + "reason": "SegFormer and LongT5 are unrelated code paths." }, { - "left": "pull_request:43997", - "right": "pull_request:44017", + "left": "pull_request:44017", + "right": "pull_request:44073", "accept": false, - "reason": "RegNet/FNet-related work versus SegFormer: same tracking theme, different code paths." + "reason": "SegFormer and VisualBERT are different model-specific changes." }, { - "left": "pull_request:43997", - "right": "pull_request:44056", + "left": "pull_request:44017", + "right": "pull_request:44722", "accept": false, - "reason": "Different model families and different files." + "reason": "SegFormer and GPT-J/CodeGen are unrelated fixes." }, { - "left": "pull_request:44010", + "left": "pull_request:44068", "right": "pull_request:44071", "accept": false, - "reason": "SqueezeBERT and MPT are unrelated concrete implementations." + "reason": "GPT-Neo and MPT are different model families." }, { - "left": "pull_request:44010", + "left": "pull_request:43997", "right": "pull_request:44073", "accept": false, - "reason": "SqueezeBERT and VisualBert are different model code paths." + "reason": "RegNet and VisualBERT are unrelated implementations." }, { - "left": "pull_request:44010", - "right": "pull_request:44018", + "left": "pull_request:44017", + "right": "pull_request:44071", "accept": false, - "reason": "Different model families; same decorator migration only." + "reason": "SegFormer and MPT are different concrete changes." }, { - "left": "pull_request:44010", - "right": "pull_request:44044", + "left": "pull_request:44071", + "right": "pull_request:44086", "accept": false, - "reason": "SqueezeBERT and DeBERTaV2 are unrelated implementations." + "reason": "MPT and MGP-STR are different model code paths." }, { - "left": "pull_request:44044", - "right": "pull_request:44129", + "left": "pull_request:44073", + "right": "pull_request:44101", "accept": false, - "reason": "DeBERTaV2 and SpeechT5 are different model code paths." + "reason": "VisualBERT and XLM are distinct model families." }, { - "left": "pull_request:44044", - "right": "pull_request:44074", + "left": "pull_request:44056", + "right": "pull_request:44101", "accept": false, - "reason": "DeBERTaV2 and TextNet are unrelated models." + "reason": "MPNet and XLM are different concrete output-tracing refactors." }, { - "left": "pull_request:44024", - "right": "pull_request:44076", + "left": "pull_request:44059", + "right": "pull_request:44071", "accept": false, - "reason": "FocalNet and ImageGPT are different implementations; the similarity is only the standardized-output refactor." + "reason": "GPT-2 and MPT are unrelated model changes." }, { - "left": "pull_request:44066", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J/CodeGen output-tracing migration with overlapping files; 44722 looks like a near-superset/follow-up of the same change." + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT-2 are different model families and code paths." }, { - "left": "pull_request:44085", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J output-tracing refactor in the same file path; the second PR is an overlapping variant of the same change." + "left": "pull_request:44059", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are not the same underlying fix." }, { - "left": "pull_request:44085", - "right": "pull_request:44129", + "left": "pull_request:43997", + "right": "pull_request:44017", "accept": false, - "reason": "GPT-J versus SpeechT5: unrelated code paths." + "reason": "RegNet and SegFormer are different model-specific refactors." }, { - "left": "pull_request:43996", - "right": "pull_request:44007", + "left": "pull_request:44101", + "right": "pull_request:44161", "accept": false, - "reason": "Different model families; only the issue umbrella matches." + "reason": "XLM and LongT5 are unrelated code paths." }, { - "left": "pull_request:43996", - "right": "pull_request:44129", + "left": "pull_request:43997", + "right": "pull_request:44056", "accept": false, - "reason": "CVT/FNet-related work and SpeechT5 are unrelated code paths." + "reason": "RegNet and MPNet are different model implementations." }, { - "left": "pull_request:44044", + "left": "pull_request:44071", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTaV2 and SpeechT5 are different model implementations." + "reason": "MPT and SpeechT5 are unrelated model-specific changes." }, { - "left": "pull_request:44044", - "right": "pull_request:44154", + "left": "pull_request:44074", + "right": "pull_request:44101", "accept": false, - "reason": "DeBERTaV2 and VITS are unrelated model code paths." + "reason": "TextNet and XLM are different output-tracing refactors." }, { - "left": "pull_request:44017", - "right": "pull_request:44073", + "left": "pull_request:44056", + "right": "pull_request:44161", "accept": false, - "reason": "SegFormer and VisualBert are separate implementations." + "reason": "MPNet and LongT5 are unrelated concrete fixes." }, { "left": "pull_request:43997", - "right": "pull_request:44017", + "right": "pull_request:44018", "accept": false, - "reason": "RegNet/FNet-family versus SegFormer: same initiative, different code paths." + "reason": "RegNet and GPT-Neo are different model families." }, { "left": "pull_request:43997", - "right": "pull_request:44056", + "right": "pull_request:44161", "accept": false, - "reason": "Different model families; not the same underlying change." + "reason": "RegNet and LongT5 are unrelated patches." }, { - "left": "pull_request:44056", - "right": "pull_request:44073", + "left": "pull_request:44074", + "right": "pull_request:44161", "accept": false, - "reason": "MPNet and VisualBert are unrelated concrete code paths." + "reason": "TextNet and LongT5 are different model-specific changes." }, { - "left": "pull_request:44017", - "right": "pull_request:44071", + "left": "pull_request:44018", + "right": "pull_request:44161", "accept": false, - "reason": "SegFormer and MPT are different models." + "reason": "GPT-Neo and LongT5 are unrelated output-tracing refactors." }, { - "left": "pull_request:44068", - "right": "pull_request:44086", + "left": "pull_request:44010", + "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated implementations." + "reason": "SqueezeBERT and MPT are unrelated model code paths." }, { - "left": "pull_request:44068", - "right": "pull_request:44722", + "left": "pull_request:44018", + "right": "pull_request:44129", "accept": false, - "reason": "GPT-Neo versus GPT-J/CodeGen: not the same code path." + "reason": "GPT-Neo and SpeechT5 are different concrete fixes." }, { - "left": "pull_request:44068", - "right": "pull_request:44071", + "left": "pull_request:44010", + "right": "pull_request:44073", "accept": false, - "reason": "GPT-Neo and MPT are different model families." + "reason": "SqueezeBERT and VisualBERT are different model families." }, { - "left": "pull_request:44073", - "right": "pull_request:44161", + "left": "pull_request:43997", + "right": "pull_request:44074", "accept": false, - "reason": "VisualBert and LongT5 are unrelated code paths." + "reason": "RegNet and TextNet are unrelated model implementations." }, { - "left": "pull_request:44056", - "right": "pull_request:44161", + "left": "pull_request:44010", + "right": "pull_request:44018", "accept": false, - "reason": "MPNet and LongT5 are different model implementations." + "reason": "SqueezeBERT and GPT-Neo are different concrete changes." }, { "left": "pull_request:43997", - "right": "pull_request:44018", + "right": "pull_request:44129", "accept": false, - "reason": "RegNet and GPT-Neo are different concrete code paths." + "reason": "RegNet and SpeechT5 are unrelated model-specific refactors." }, { - "left": "pull_request:43997", - "right": "pull_request:44161", + "left": "pull_request:44071", + "right": "pull_request:44129", "accept": false, - "reason": "RegNet/FNet-related work and LongT5 are unrelated." + "reason": "MPT and SpeechT5 are different model code paths." }, { - "left": "pull_request:44074", + "left": "pull_request:43996", "right": "pull_request:44161", "accept": false, - "reason": "TextNet and LongT5 are different models." + "reason": "CVT/FNet and LongT5 are unrelated changes." }, { - "left": "pull_request:44018", - "right": "pull_request:44161", + "left": "pull_request:43996", + "right": "pull_request:44017", "accept": false, - "reason": "GPT-Neo and LongT5 are unrelated implementations." + "reason": "CVT/FNet and SegFormer are different model families." }, { "left": "pull_request:44010", "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBERT and DeBERTaV2 do not share a concrete code path." + "reason": "SqueezeBERT and DeBERTa v2 are unrelated refactors." }, { "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa v2 and TextNet are different code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing change with overlapping files; 44722 is the later, broader reroll." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both are GPT-J output-tracing refactors on the same file family, and 44722 expands the same change." + }, + { + "left": "pull_request:44085", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTaV2 and SpeechT5 are different code paths." + "reason": "GPT-J and SpeechT5 are unrelated model-specific patches." }, { - "left": "pull_request:44010", - "right": "pull_request:44018", + "left": "pull_request:43996", + "right": "pull_request:44007", "accept": false, - "reason": "SqueezeBERT and GPT-Neo are unrelated model implementations." + "reason": "CVT/FNet and ResNet/RT-DETR ResNet are different model families." }, { - "left": "pull_request:44098", - "right": "pull_request:44101", + "left": "pull_request:43996", + "right": "pull_request:44129", "accept": false, - "reason": "ViLT and XLM are different model families; same output-tracing theme only." + "reason": "CVT/FNet and SpeechT5 are unrelated code paths." }, { - "left": "pull_request:44059", - "right": "pull_request:44071", + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa v2 and SpeechT5 are different concrete fixes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", "accept": false, - "reason": "GPT-2 and MPT are separate code paths." + "reason": "DeBERTa v2 and ViTS are unrelated model refactors." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet standardized outputs and ImageGPT hook capture are different model-specific changes." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is conservative overall and grounded in the packet: most PRs are separate model-specific output-tracing refactors under the shared umbrella issue, with only a few same-file/code-path overlaps. The accepted soft pairs are plausibly mergeable into single bugfix PRs, especially the GPT-J/GPT-Neo overlaps. One minor wording issue is that the summary glosses over GPT-J/CodeGen as well as GPT-Neo overlaps, but that does not materially overstate the evidence." + "feedback": "The summary is well grounded in the packet: most PRs are separate model-specific output-tracing refactors, and the GPT-Neo/GPT-J/CodeGen subgroup is the only clearly duplicate-looking cluster. The conservative false verdicts are supported by distinct model families/code paths, and the true verdicts are limited to closely overlapping same-family changes with shared files or the same issue target. Best representative PR 44722 is also a reasonable choice because it is the later, broader GPT-J/CodeGen reroll." }, "evaluator_used": true, "retried": false, @@ -26530,14 +26781,14 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "5a0a9b92a3415af61c71f43418449c61f2108f2869b4d39292f4ebb202644e11", + "prepared_review_unit_hash": "36d27060a1eb0fff7a5bfb737a60e51fcd7b19a67031c701f557fc7d31b1bcd1", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84514, - "estimated_input_tokens": 21129, - "estimated_eval_tokens": 42514 + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 }, "budget": { "node_count": 6, @@ -26552,42 +26803,42 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These PRs are related by theme\u2014each migrates a different model to standardized output tracing\u2014but they are not duplicates of the same concrete change. They touch different model code paths and should stay separate.", - "confidence": 0.87, + "summary": "These are all model-specific PRs to migrate output tracing/capture to a standardized decorator pattern. They are similar in theme, but they do not fix the same concrete code path, so they should not be merged as duplicates.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "44129 is the strongest canonical representative because it is a self-contained output-tracing migration with an explicit issue target and a clearly scoped model-specific implementation.", + "canonical_pr_reason": "PR 44129 is the strongest representative: it has the clearest explicit issue link, a substantial tracing refactor, and the diff shows a concrete output-capture migration in a specific model path.", "best_issue_reason": null, - "best_pr_reason": "44129 is the best overall representative of the cluster: it is clearly framed, tied to a tracking issue, and implements the same general refactor pattern in a concrete model path.", + "best_pr_reason": "PR 44129 is the best single PR for the cluster because it is the most explicit and complete example of the shared refactor pattern; the others are narrower, model-specific variants of the same migration theme.", "soft_edge_verdicts": [ { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "Both are output-tracing refactors, but one is for VisualBert and the other for TextNet; they do not fix the same code path and would not be merged as one PR." + "reason": "Both are output-tracing refactors, but they touch different model code paths (VisualBert vs TextNet) and are not the same underlying change." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "These are separate model migrations (MPNet vs VisualBert). The change pattern is similar, but the underlying code and fix are different." + "reason": "These are separate model migrations (MPNet vs VisualBert). Same refactor theme, but not the same bug or change and not plausibly one PR." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "MPNet and TextNet are distinct model implementations; this is the same refactor style, not the same bug or change." + "reason": "MPNet and TextNet are unrelated model implementations; the similar decorator migration does not make them duplicates." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model code paths. They share a migration theme, but not a single duplicate fix." + "reason": "RegNet and SqueezeBert are different model paths with different output-handling code; this is a shared refactor pattern, not one concrete fix." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 are different targets with different implementation details; these are separate PRs, not one mergeable fix." + "reason": "SqueezeBert and SpeechT5 change different forward/output paths, so they should remain separate PRs despite similar standardized tracing goals." } ] }, @@ -26666,274 +26917,268 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "48a9a6f8db6174639c0cd970e00275b7e4cd5672c2a5498751941ab0d3c5665c", + "prepared_review_unit_hash": "0df79e68744390b1ddfa4e868b8b44ac32662d66d6842d683bc2fdd8ff9415dc", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 39, - "serialized_chars": 39802, - "estimated_input_tokens": 9951, - "estimated_eval_tokens": 20158 + "serialized_chars": 39789, + "estimated_input_tokens": 9948, + "estimated_eval_tokens": 20152 }, "budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 39, - "serialized_chars": 39802, - "estimated_input_tokens": 9951, - "estimated_eval_tokens": 20158 + "serialized_chars": 39789, + "estimated_input_tokens": 9948, + "estimated_eval_tokens": 20152 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All items are similar output-tracing refactors, but they target different model implementations and code paths, so they are not duplicates of one another.", - "confidence": 0.95, + "summary": "These PRs share a common output-tracing/refactor theme, but they target different model implementations and code paths, so they are not true duplicates of one another.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44003 is the clearest representative of the cluster: it applies the output-tracing refactor across two closely related Mamba-family files and shows the shared helper integration pattern.", + "canonical_pr_reason": "Pick PR 44025 as the representative anchor: it applies the shared output-tracing refactor to two closely related model variants and is one of the clearest, self-contained examples in the set.", "best_issue_reason": null, - "best_pr_reason": "PR 44003 is the strongest canonical example because it is the broadest and most infrastructure-oriented of the set, touching multiple related model files rather than a single isolated wrapper.", + "best_pr_reason": "PR 44025 is the best cluster representative because it is broad enough to reflect the refactor pattern without being a catch-all, and it touches a closely related pair of model files.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Same refactor theme, but different model code paths (`timm_backbone` vs `upernet`)." + "reason": "Same refactor theme, but different model families and forward paths; not the same concrete change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Different model families (`timm_backbone` vs `vision_text_dual_encoder`); not the same bug or change." + "reason": "Both are output-tracing refactors, but they affect unrelated model implementations and APIs." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Both are output-tracing refactors, but for unrelated models (`timm_backbone` vs `univnet`)." + "reason": "Different models and code paths; similarity is only at the boilerplate refactor level." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "`timm_backbone` and `mamba/falcon_mamba` are separate code paths; similarity is only at the refactor pattern level." + "reason": "Timm backbone vs. Mamba/Falcon Mamba are unrelated fixes, despite shared refactor wording." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Different architectures (`timm_backbone` vs `rwkv`); no shared concrete defect." + "reason": "Different architectures and output plumbing; not mergeable as one concrete bug fix." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "Separate model implementations (`mobilenet_v1` vs `univnet`), despite similar tracing cleanup." + "reason": "Different model code paths; shared output-tracing mechanics are too broad to treat as duplicate." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "`mobilenet_v1` and `upernet` have different forward paths and outputs." + "reason": "MobileNetV1 and UperNet are separate implementations with distinct forward behavior." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "Different model families (`mobilenet_v1` vs `depth_anything`); not mergeable as one fix." + "reason": "Unrelated model families; same refactor pattern does not imply the same underlying change." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "`mobilenet_v1` and `vision_encoder_decoder` are unrelated code paths." + "reason": "Different model architectures and files; only the tracing refactor motif is shared." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "`mobilenet_v1` and `speech_encoder_decoder` are different implementations, not one underlying bug." + "reason": "Separate encoder-decoder stack versus MobileNetV1; not the same concrete code-path problem." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Both refactor output handling, but in unrelated models (`vision_text_dual_encoder` vs `univnet`)." + "reason": "Vision-text dual encoder and UnivNet have unrelated forward paths; too broad to dedupe." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different concrete changes in `vision_text_dual_encoder` and `upernet`." + "reason": "Same pattern, different model subsystems; no shared bug or identical change." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "`vision_text_dual_encoder` does not share the same code path as `mamba/falcon_mamba`." + "reason": "These are distinct model implementations with different output contracts." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different model families; same broad refactor idea is not enough." + "reason": "Different model families; the shared refactor theme is not enough for a duplicate match." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "`univnet` and `upernet` are separate model implementations with different output semantics." + "reason": "UnivNet and UperNet are unrelated code paths; this is only a similar refactor style." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Unrelated code paths (`univnet` vs `mamba/falcon_mamba`)." + "reason": "No shared concrete bug or change; just a common tracing refactor pattern." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "Different models and different forward signatures; not the same underlying change." - }, - { - "left": "pull_request:44001", - "right": "pull_request:44004", - "accept": false, - "reason": "`univnet` and `codegen` are unrelated model code paths." + "reason": "Different model families and output behaviors; not a plausible single merged PR." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "`upernet` and `rwkv` are different architectures; no shared concrete bug." + "reason": "UperNet and RWKV are unrelated implementations; same refactor theme is insufficient." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "Different model families (`upernet` vs `mamba/falcon_mamba`)." + "reason": "These touch different subsystems and do not fix the same underlying issue." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "`mamba/falcon_mamba` and `codegen` are unrelated code paths." + "reason": "Mamba/Falcon Mamba output tracing and CodeGen cache/output handling are different concrete changes." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Different model implementations (`timm_backbone` vs `codegen`)." + "reason": "Different models and different forward semantics; only superficially similar refactor language." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "`univnet` and `codegen` are different subsystems and forward paths." + "reason": "UnivNet and CodeGen are separate code paths with distinct outputs and kwargs handling." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "`upernet` and `codegen` are unrelated model changes." + "reason": "UperNet and CodeGen are unrelated; not the same bug or fix." }, { "left": "pull_request:44000", "right": "pull_request:44004", "accept": false, - "reason": "`vision_text_dual_encoder` and `codegen` are separate model paths." + "reason": "Different model code paths and output signatures; too broad to merge as duplicates." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "Both are output-tracing refactors, but in different encoder-decoder/depth model families." + "reason": "Depth Anything family versus vision-encoder-decoder are different implementations, not one shared fix." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "`depth_anything`/`prompt_depth_anything` is unrelated to `speech_encoder_decoder`." + "reason": "Different model families and code paths; shared refactor wording is not enough." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "`depth_anything` and `superpoint` are separate vision models with different forward logic." + "reason": "Depth Anything and SuperPoint are unrelated forward implementations." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Different model code paths (`depth_anything` vs `rwkv`)." + "reason": "Different architectures and output handling; not the same underlying bug/change." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "`depth_anything` and `dpr` are unrelated model implementations." + "reason": "Depth Anything and DPR are separate model changes with no concrete overlap." }, { "left": "pull_request:44026", - "right": "pull_request:44027", + "right": "pull_request:44029", "accept": false, - "reason": "`vision_encoder_decoder` and `speech_encoder_decoder` are analogous wrappers, but not the same concrete bug or change." + "reason": "Vision-encoder-decoder and RWKV do not share the same code path; only the refactor theme matches." }, { "left": "pull_request:44026", - "right": "pull_request:44028", + "right": "pull_request:44027", "accept": false, - "reason": "`vision_encoder_decoder` and `superpoint` differ in code path and output handling." + "reason": "Encoder-decoder vision and speech wrappers are distinct implementations, not duplicates." }, { "left": "pull_request:44026", - "right": "pull_request:44029", + "right": "pull_request:44028", "accept": false, - "reason": "`vision_encoder_decoder` and `rwkv` are unrelated architectures." + "reason": "Different model families and output APIs; no single merged PR would make sense." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Different model families (`vision_encoder_decoder` vs `dpr`)." + "reason": "Separate encoder-decoder wrapper versus DPR encoder; not the same concrete fix." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "`speech_encoder_decoder` and `superpoint` are separate code paths." + "reason": "Speech encoder-decoder and SuperPoint are unrelated code paths." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "`speech_encoder_decoder` and `rwkv` do not share the same underlying fix." + "reason": "Different model families and output contracts; no duplicate relationship." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "`speech_encoder_decoder` and `dpr` are different model implementations." + "reason": "Speech encoder-decoder and DPR are separate changes with no shared bug." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "`superpoint` and `rwkv` are unrelated models." + "reason": "SuperPoint and RWKV are unrelated implementations; similarity is only in the refactor pattern." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "`superpoint` and `dpr` are different concrete code paths." + "reason": "Different model subsystems and output behavior; not mergeable as one change." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "`rwkv` and `dpr` are separate model families; same refactor style is not enough." + "reason": "RWKV and DPR touch different code paths and fix different output-tracing details." } ] }, @@ -26963,12 +27208,12 @@ "soft_pairs": [ "pull_request:44662|pull_request:45401" ], - "prepared_review_unit_hash": "1e0cfae13df67f208379396ba10c00bf152cb1bfc90979581f24666fac3fb54f", + "prepared_review_unit_hash": "dbe1daf59cd115c68f572826b93fde0591ee457d6abd6ffb8d87cdfac269dd79", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6552, + "serialized_chars": 6551, "estimated_input_tokens": 1638, "estimated_eval_tokens": 3532 }, @@ -26976,7 +27221,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6552, + "serialized_chars": 6551, "estimated_input_tokens": 1638, "estimated_eval_tokens": 3532 }, @@ -26985,18 +27230,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are unrelated model additions: one implements PenguinVL, the other adds Voxtral TTS support. They share only boilerplate registry/docs files, so the soft similarity is not enough to treat them as the same change.", - "confidence": 0.98, + "summary": "These are two separate model-addition PRs for different architectures (PenguinVL vs Voxtral TTS). They share common Transformers scaffolding files, but the underlying codepaths, model types, and goals are distinct, so they are not duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the two PRs fix/add different model families and do not share a concrete code-path problem.", + "canonical_pr_reason": "PR 45401 is the better canonical PR only in the sense that it has an explicit tracking issue and a narrower, well-scoped target (Voxtral TTS support). PR 44662 is a different model implementation entirely.", "best_issue_reason": null, - "best_pr_reason": "PR 45401 is the more self-contained candidate only because it has an explicit issue target and a narrower, clearly scoped feature, but it is still unrelated to 44662.", + "best_pr_reason": "PR 45401 is the strongest representative of its own change set because it is tied to a specific issue and focused on one model family. PR 44662 is not a better cluster representative since it targets a different model with different files and behavior.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Reject: PenguinVL implementation and Voxtral TTS support are distinct features with different models, configs, and modeling code. The overlap is limited to shared auto-registration/docs plumbing, which is not enough to indicate the same underlying change." + "reason": "Reject: they add support for different models (PenguinVL vs Voxtral TTS). Shared auto-registration/docs files are generic boilerplate for new model integrations and do not indicate the same underlying bug or change." } ] }, @@ -27026,12 +27271,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "bfb6763910582c6d9b3217a9832078693a3ce9ce0c6d9fcd5df2a8a3efab6ca4", + "prepared_review_unit_hash": "e6ee7e54dbfa9103a6b7855fc6d2441a68cdeb2503c74b0cb62abaac983a4f78", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, + "serialized_chars": 4407, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -27039,7 +27284,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, + "serialized_chars": 4407, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -27048,18 +27293,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are thematically related (both involve fusion), but they are not duplicates: one introduces a generic Module Fusion API, while the other adds n-to-1 kernel fusion via KernelConfig and model-loading integration. They target different abstractions and code paths.", - "confidence": 0.9, + "summary": "These two PRs overlap on fusion-related infrastructure, but they target different changes: #44979 introduces a generic Module Fusion API, while #45363 implements a specific KernelConfig-based n-to-1 kernel fusion path plus integration updates. They are related work, not duplicates.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "44979 is the cleaner foundational change: it introduces the new module_fusion API and focused tests for the core abstraction.", + "canonical_pr_reason": "No single canonical PR: #44979 is a general module-fusion API scaffold, while #45363 is a separate concrete kernel-fusion feature with different files and scope.", "best_issue_reason": null, - "best_pr_reason": "44979 is the best representative of this cluster because it is the central API-level fusion proposal, whereas 45363 is a separate downstream integration built on kernel configuration.", + "best_pr_reason": "#45363 is the better representative PR because it is the more complete, user-facing implementation of a concrete fusion mechanism rather than just the foundational API layer.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Reject: these are related fusion features, but not the same underlying change. 44979 adds a generic Module Fusion API in src/transformers/module_fusion.py, while 45363 implements KernelConfig-based n-to-1 kernel fusion across modeling/integration files and a demo script. Different abstractions and code paths." + "reason": "Reject: they are related but not the same underlying change. #44979 adds a generic Module Fusion API, whereas #45363 implements KernelConfig-based n-to-1 fusion and integration plumbing. They would not plausibly be merged as one PR." } ] }, @@ -27089,7 +27334,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "5cbbeff4849d10c3bc16b10ecc75335b162609f745096a368f890f4c8edd306a", + "prepared_review_unit_hash": "3a132a0c3c394290e317652dcc6e0170be3740bd078458f019cf884689857920", "component_budget": { "node_count": 2, "item_count": 2, @@ -27111,18 +27356,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests address the same Mistral regex patch AttributeError in `src/transformers/tokenization_utils_tokenizers.py` for issue #45081; PR 45317 is the fuller version with added tests, so it is the better canonical representative.", + "summary": "Both PRs fix the same AttributeError in `_patch_mistral_regex` by switching `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer` in the same file; 45317 is the more complete version because it adds tests.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45317 is the stronger canonical PR because it implements the same fix as 45086 but also adds test coverage and has more review activity, making it the more complete representative of the change.", + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same code fix as 45086 plus test coverage, making it the more complete change set.", "best_issue_reason": null, - "best_pr_reason": "PR 45317 is the best overall PR because it is the more complete and validated version of the same underlying fix, with tests included.", + "best_pr_reason": "PR 45317 is the best representative: same concrete fix as 45086, same target issue, and it includes validation tests.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same concrete bug fix in the same code path, same target issue, and the diffs show the same `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer` correction; 45317 is essentially the tested superset." + "reason": "Same underlying bug and same fix in the same code path (`_patch_mistral_regex`); 45317 simply extends 45086 with tests, so they are duplicates/overlapping changes." } ] }, @@ -27152,7 +27397,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "a7303f0a462a1a579795322bbda674435ecf0b33c58c873eea8e65da799d3764", + "prepared_review_unit_hash": "060ac854cebefb825878af0ae7b05b81a1c7539c46dd06a2674d52211bc0f0f7", "component_budget": { "node_count": 2, "item_count": 2, @@ -27174,18 +27419,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two NemotronH PRs, but they target different changes: one is a refactor toward GraniteMoeHybrid-compatible inheritance/conversion mapping, while the other introduces proper NemotronH dense/sparse model support and docs. They are related by subsystem, not the same underlying fix.", - "confidence": 0.92, + "summary": "These are related NemotronH-area pull requests, but they are not duplicates: one is a focused inheritance/conversion refactor for NemotronH, while the other splits the architecture into dedicated dense/sparse model variants and updates docs/tests.", + "confidence": 0.9, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45569 is the better canonical representative because it is the broader, more user-visible NemotronH change, adding the dense/sparse model split, auto mappings, docs, and tests. PR #45115 is a narrower internal refactor/conversion-mapping update.", + "canonical_pr_reason": "PR 45569 is the broader, more representative NemotronH change set because it introduces the new dense/sparse model structure and associated public-facing docs/model wiring.", "best_issue_reason": null, - "best_pr_reason": "PR #45569 is the strongest standalone PR in the cluster: it appears to deliver the actual model support and documentation for NemotronH variants. PR #45115 is a limited refactor and not a substitute for the full architecture work.", + "best_pr_reason": "PR 45569 is the best overall candidate: it contains the substantive model API/architecture work, not just internal conversion-mapping refactors.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "Related to NemotronH, but not the same concrete code-path problem. #45115 focuses on inheritance/refactor and conversion mapping; #45569 adds new dense/sparse model implementations and docs. They would not plausibly be merged as one duplicate PR." + "reason": "They share NemotronH core files, but the underlying changes differ: 45115 is about conversion-mapping and inheritance refactoring, while 45569 adds separate NemotronHDense/NemotronHSparse implementations plus docs/tests. They are related, not the same mergeable fix." } ] }, @@ -27215,7 +27460,7 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "a59fcef6df7d0e01fd93c5be0d9a504a13b444b6ad1635994d7db66f3f727f73", + "prepared_review_unit_hash": "4fd001073c0d5241f02cd0724950eaedebb2d91c2c28d4becbc588ca39e828e9", "component_budget": { "node_count": 2, "item_count": 2, @@ -27237,18 +27482,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are both typing-related PRs, but they address different changes: one is a broad mypy/typing pass across many Transformers modules, while the other adds shared typing protocols and updates a smaller set of typing-sensitive call sites. They are not the same underlying fix/change, so the soft duplicate edge should be rejected.", - "confidence": 0.92, + "summary": "These two PRs are related by typing work in transformers, but they target different changes: one is a broad type-checking sweep across many files, while the other adds shared typing protocols/aliases and fixes specific typing integration points. They are not the same underlying change.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR for deduplication: the PRs are related only at a broad typing/workflow level, but they implement different code changes and would not plausibly be merged as one duplicate change.", + "canonical_pr_reason": "PR #45425 is the more focused, self-contained typing-infrastructure change; #45415 is a broader type-checking sweep and does not subsume it.", "best_issue_reason": null, - "best_pr_reason": "PR 45425 is the better representative of a focused, coherent typing change set because it introduces reusable typing abstractions and targeted call-site updates, whereas PR 45415 is a broad repository-wide type-checking sweep.", + "best_pr_reason": "PR #45425 is the better representative of the cluster because it has a clearer, narrower purpose and concrete typing API additions, but it is still not a duplicate of #45415.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Shared typing scope and a few overlapping files are not enough: 45415 is a broad type-checking pass, while 45425 adds specific typing protocols and fixes. They do not appear to be the same concrete bug/change or merge cleanly into one PR." + "reason": "Different change sets: #45415 broadly enables type checking across many source files and adjusts pyproject ignores, while #45425 adds typing aliases/Protocols in _typing.py and related helpers. They do not appear to fix the same concrete code-path problem." } ] }, @@ -27278,12 +27523,12 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "c19aab39e2495d08e4a174fb7b0065aa76a288cd6a8513e834c8958f4a822b91", + "prepared_review_unit_hash": "6aed6f583f811074649c481b405a82a20c6c2882ab2cfd1fb800da52fced3ad0", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4508, + "serialized_chars": 4507, "estimated_input_tokens": 1127, "estimated_eval_tokens": 2510 }, @@ -27291,7 +27536,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4508, + "serialized_chars": 4507, "estimated_input_tokens": 1127, "estimated_eval_tokens": 2510 }, @@ -27300,18 +27545,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related only at a high level (expert parallelism / model loading) but fix different bugs on different code paths: one is about EP + DeepSpeed ZeRO-3 loading via accelerate, the other about EP + FSDP2 rank-0 broadcast overwriting experts. They should not be deduplicated.", - "confidence": 0.96, + "summary": "Both PRs are about expert-parallel loading, but they fix different failure modes: one addresses EP + ZeRO-3/accelerate config handling, the other fixes EP + FSDP2 expert tensors being overwritten during shard loading. They are related, not duplicates.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "No issue present. If choosing a representative PR, #45662 is the better canonical candidate because it addresses a concrete, clearly reproduced EP parameter corruption path with targeted code changes.", + "canonical_pr_reason": "No clear duplicate canonical PR: the two PRs target different concrete code-path bugs, despite overlapping in EP/loading infrastructure.", "best_issue_reason": null, - "best_pr_reason": "#45662 is the strongest standalone PR in this cluster: it fixes a specific expert-parameter overwrite bug in loading/sharding logic, whereas #45548 is a separate serialization/loading compatibility fix.", + "best_pr_reason": "PR 45662 is the stronger representative of the cluster because it fixes a concrete, reproducible overwrite bug in the loading path and spans the broader EP/FSDP2 remediation.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "Related theme, but not the same bug: #45548 removes distributed_config from config serialization and adds has_ep for EP/ZeRO-3 loading, while #45662 fixes FSDP2 post-shard wrapping to prevent expert parameters being overwritten. Different failure modes and code paths." + "reason": "Different underlying bugs: 45548 focuses on EP + DeepSpeed ZeRO-3/accelerate config serialization, while 45662 fixes experts being overwritten during distributed shard loading for FSDP2. Shared EP context is not enough to merge them as duplicates." } ] }, @@ -27341,40 +27586,40 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "976a7b0ee0896e361bb91e0c7ac7330e488dcff42dc78f600409cad05fac16b2", + "prepared_review_unit_hash": "24fb3b1b490be06bc0f1b3b500da0a59116f6e25e3b7a0a61a4d59978e69cb63", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4673, - "estimated_input_tokens": 1169, - "estimated_eval_tokens": 2594 + "serialized_chars": 4672, + "estimated_input_tokens": 1168, + "estimated_eval_tokens": 2592 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4673, - "estimated_input_tokens": 1169, - "estimated_eval_tokens": 2594 + "serialized_chars": 4672, + "estimated_input_tokens": 1168, + "estimated_eval_tokens": 2592 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same xdist-related captured_info logging collision and share the same core file in `testing_utils.py`; PR #45645 appears to be the broader superset that also preserves CI log collection, so this cluster is best represented by #45645.", - "confidence": 0.87, + "summary": "Both pull requests address the same xdist collision around captured_info output, with 45645 extending the code fix in 45639 and adding the CI-side log collection needed for the new per-worker files.", + "confidence": 0.86, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45645 is the more complete fix: it addresses the same xdist-safe `captured_info` output-path problem as #45639 and also updates the workflow to consume multiple per-worker log files.", + "canonical_pr_reason": "PR 45645 is the broader, more complete fix: it includes the xdist-safe testing_utils change plus the workflow update to collect all captured_info files, while 45639 only covers the library-side logging change.", "best_issue_reason": null, - "best_pr_reason": "PR #45645 is the best representative because it covers the same underlying bug plus the CI-side log aggregation needed for the new output format.", + "best_pr_reason": "PR 45645 is the best representative because it solves the underlying xdist debug-log collision end to end, not just the file-writing half of it.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying xdist collision in captured_info logging; #45645 extends the same fix with CI log handling, so they could plausibly be merged into one PR." + "reason": "Same underlying xdist captured_info collision fix; 45645 builds on the same testing_utils change and adds the necessary CI artifact handling, so they could plausibly be merged together." } ] }, @@ -27470,12 +27715,12 @@ "soft_pairs": [ "pull_request:46126|pull_request:46269" ], - "prepared_review_unit_hash": "881bc76557c94788bde2fdbca3f79f74e2de66fce98ff3a232e0960b22028ee5", + "prepared_review_unit_hash": "bf0592aeb9df3a25bb1879c6722ecc7f081a0ee47c01e59c9ccb27b194689c44", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5525, + "serialized_chars": 5526, "estimated_input_tokens": 1382, "estimated_eval_tokens": 3020 }, @@ -27483,7 +27728,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5525, + "serialized_chars": 5526, "estimated_input_tokens": 1382, "estimated_eval_tokens": 3020 }, @@ -27492,18 +27737,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two pull requests are not duplicates. #46126 is a focused EP + Trainer integration change, while #46269 is a broad distributed branch base that spans many unrelated distributed features and infra updates. Their overlap is only superficial/shared subsystem work.", + "summary": "Both PRs touch distributed training code, but they are not the same change: #46126 is a focused EP/Trainer integration, while #46269 is a broad distributed branch base spanning FSDP, integrations, model loading, CI, and multiple subsystems.", "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "#46126 is the best canonical PR because it represents a single, focused change set with a clear EP/Trainer integration purpose. #46269 is too broad and mixed to serve as a good duplicate-triage anchor.", + "canonical_pr_reason": "#46126 is the more focused, self-contained PR centered on expert-parallel routing and Trainer integration, making it the better canonical representative of a single concrete change.", "best_issue_reason": null, - "best_pr_reason": "#46126 is the best overall PR representative in this cluster: it is narrowly scoped and clearly describes one concrete integration path, unlike #46269 which is a large base branch touching many areas.", + "best_pr_reason": "#46126 is the best representative because it targets one specific feature path with a clear technical scope; #46269 is too broad and mixes many unrelated distributed changes.", "soft_edge_verdicts": [ { "left": "pull_request:46126", "right": "pull_request:46269", "accept": false, - "reason": "Reject: they do not appear to fix the same concrete bug or implement the same change. #46126 is a specific EP/Trainer integration PR; #46269 is a broad distributed branch base with many unrelated file changes, so the overlap is not enough to merge them as duplicates." + "reason": "Shared distributed filenames are not enough here: #46126 is a specific EP/Trainer feature, while #46269 is a large base branch touching many areas and is not plausibly mergeable as the same PR." } ] }, @@ -27571,130 +27816,130 @@ "pull_request:46504|pull_request:46716", "pull_request:46432|pull_request:46716" ], - "prepared_review_unit_hash": "0de91c4aeec928caec4b75ec56c00a6829262e44a1178be049d244e67790b94c", + "prepared_review_unit_hash": "8e8978aa7527355477e072582a5e02480bcd879954420f857ca74fa7bfef7027", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 31, - "serialized_chars": 26676, - "estimated_input_tokens": 6669, - "estimated_eval_tokens": 13594 + "serialized_chars": 26679, + "estimated_input_tokens": 6670, + "estimated_eval_tokens": 13596 }, "budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 31, - "serialized_chars": 26676, - "estimated_input_tokens": 6669, - "estimated_eval_tokens": 13594 + "serialized_chars": 26679, + "estimated_input_tokens": 6670, + "estimated_eval_tokens": 13596 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "No soft-duplicate PR pairs were convincing enough to accept. Most items are small, unrelated workflow maintenance changes; the closest group is the pr-ci-security-gate.yml edits, but they address different hardening steps rather than the same concrete fix.", - "confidence": 0.93, + "summary": "This cluster is mostly a set of unrelated workflow-maintenance PRs. The only tight subgroup is several edits to pr-ci-security-gate.yml, but they change different behaviors (checkout/diff checks, permissions, reusable-workflow comments, secrets handling), so they are not duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46554 is the strongest standalone representative: it makes the security gate materially stricter by checking out the PR and blocking workflow/CI-infra changes with explicit failure behavior.", + "canonical_pr_reason": "No single canonical PR: the cluster spans separate workflow-only changes across different files, and the pr-ci-security-gate PRs are about different safeguards rather than the same fix.", "best_issue_reason": null, - "best_pr_reason": "PR 46554 is the best overall PR candidate because it has the clearest, most substantive change and the most specific security intent among the set.", + "best_pr_reason": "If forced to pick one representative PR, 46554 is the most substantive change in the only coherent subgroup (pr-ci-security-gate), since it alters the core security-check logic.", "soft_edge_verdicts": [ { "left": "pull_request:46138", "right": "pull_request:46554", "accept": false, - "reason": "Different workflows and different fixes; both are shell/input handling changes, but not the same underlying bug or code path." + "reason": "Different workflow files and different changes; one updates self-comment CI stdin handling, the other changes pr-ci-security-gate logic." }, { "left": "pull_request:46138", "right": "pull_request:46558", "accept": false, - "reason": "Unrelated workflow changes: self-comment CI message piping vs PR CI security-gate permissions/approval logic." + "reason": "Unrelated workflow files and different behavior; no shared code-path or same bug fix." }, { "left": "pull_request:46138", "right": "pull_request:46563", "accept": false, - "reason": "Different workflow and purpose; one fixes stdin handling, the other documents a reusable security gate." + "reason": "Unrelated workflow files; self-comment CI input handling is not the same change as pr-ci-security-gate comments/guardrails." }, { "left": "pull_request:46138", "right": "pull_request:46716", "accept": false, - "reason": "Different workflow files and concerns; not the same concrete change." + "reason": "Different workflow file and different purpose; not the same underlying change." }, { "left": "pull_request:46138", "right": "pull_request:46504", "accept": false, - "reason": "Both are shell-safety tweaks, but they affect different workflows and different inputs, so not duplicates." + "reason": "Different workflow files and unrelated fixes; no duplicate relationship." }, { "left": "pull_request:46432", "right": "pull_request:46503", "accept": false, - "reason": "SSH runner hardening vs benchmark trigger expansion; separate workflow changes." + "reason": "Different workflow files and distinct changes; ssh-runner hardening is unrelated to benchmark dispatch behavior." }, { "left": "pull_request:46432", "right": "pull_request:46501", "accept": false, - "reason": "Different workflows and problems; no shared concrete fix." + "reason": "Different workflow files and different fixes; ssh-runner changes are not the same as anti-slop permission scoping." }, { "left": "pull_request:46432", "right": "pull_request:46502", "accept": false, - "reason": "SSH runner environment handling vs reviewer assignment checkout ref; unrelated." + "reason": "Different workflow files and unrelated behavior; no shared concrete bug." }, { "left": "pull_request:46432", "right": "pull_request:46554", "accept": false, - "reason": "Both touch CI/workflow security, but they fix different workflows and different issues." + "reason": "Different workflow files and different concerns; ssh-runner hardening is not the same as pr-ci-security-gate changes." }, { "left": "pull_request:46432", "right": "pull_request:46558", "accept": false, - "reason": "Different workflows and distinct hardening steps; not the same underlying bug/change." + "reason": "Different workflow files and unrelated fixes; not the same change." }, { "left": "pull_request:46501", "right": "pull_request:46503", "accept": false, - "reason": "Anti-slop permissions adjustment vs benchmark dispatch trigger; unrelated workflow changes." + "reason": "Both are workflow tweaks, but they address different workflows and different behavior, so they are not duplicates." }, { "left": "pull_request:46501", "right": "pull_request:46502", "accept": false, - "reason": "Different workflow files and different functionality; not duplicates." + "reason": "Same general area of CI workflows, but the files and fixes differ; not the same underlying issue." }, { "left": "pull_request:46501", "right": "pull_request:46504", "accept": false, - "reason": "Same broad CI area, but one changes anti-slop permissions and the other shell quoting in build-ci-docker-images; not the same fix." + "reason": "Different workflow files and unrelated changes; no common code-path problem." }, { "left": "pull_request:46501", "right": "pull_request:46554", "accept": false, - "reason": "Different workflows and different code paths; no duplicate evidence beyond broad CI/security similarity." + "reason": "anti-slop permission scoping is unrelated to pr-ci-security-gate workflow logic." }, { "left": "pull_request:46502", "right": "pull_request:46503", "accept": false, - "reason": "Reviewer assignment checkout ref vs benchmark dispatch support; unrelated." + "reason": "Different workflow files and distinct fixes; not mergeable as one duplicate change." }, { "left": "pull_request:46502", "right": "pull_request:46504", "accept": false, - "reason": "Different workflows and different changes; not the same bug or feature." + "reason": "Different workflow files and unrelated behavior changes." }, { "left": "pull_request:46502", @@ -27706,85 +27951,85 @@ "left": "pull_request:46503", "right": "pull_request:46504", "accept": false, - "reason": "Benchmark trigger condition vs build-ci-docker-images quoting fix; separate workflow updates." + "reason": "Different workflow files and different CI concerns; not the same bug or change." }, { "left": "pull_request:46503", "right": "pull_request:46554", "accept": false, - "reason": "Different workflows and different intents; not duplicates." + "reason": "Benchmark dispatch support is unrelated to pr-ci-security-gate workflow changes." }, { "left": "pull_request:46504", "right": "pull_request:46554", "accept": false, - "reason": "Both are workflow hardening, but one fixes shell input handling in build-ci-docker-images while the other changes PR CI security-gate behavior." + "reason": "Different workflow files and different behavior; not the same underlying fix." }, { "left": "pull_request:46504", "right": "pull_request:46558", "accept": false, - "reason": "Different workflow files and different fixes; not the same underlying issue." + "reason": "Different workflow files and unrelated CI logic changes." }, { "left": "pull_request:46554", "right": "pull_request:46558", "accept": false, - "reason": "Same file, but different hardening changes: one adds checkout-based checks, the other adjusts permissions/approval flow." + "reason": "Same workflow file, but different concrete changes: security-check/diff scanning versus permissions/approval gating." }, { "left": "pull_request:46554", "right": "pull_request:46563", "accept": false, - "reason": "Same workflow file, but one enforces security checks and the other only adds explanatory comments; not a duplicate change." + "reason": "Same workflow file, but one adds security checks and the other only adds safety comments about the reusable workflow." }, { "left": "pull_request:46554", "right": "pull_request:46716", "accept": false, - "reason": "Same workflow area, but one blocks risky PR changes while the other changes secrets inheritance; distinct fixes." + "reason": "Same workflow file, but different safeguards: checkout/diff-based checks versus removing inherited secrets." }, { "left": "pull_request:46558", "right": "pull_request:46563", "accept": false, - "reason": "Both touch pr-ci-security-gate.yml, but they address different aspects and are not the same concrete change." + "reason": "Same workflow file, but different changes: permissions/approval condition versus documentation comments." }, { "left": "pull_request:46558", "right": "pull_request:46716", "accept": false, - "reason": "Permissions/approval logic vs secrets handling; related security hardening, but not duplicate PRs." + "reason": "Same workflow file, but one adjusts job permissions/approval flow and the other changes secrets passing; not the same fix." }, { "left": "pull_request:46563", "right": "pull_request:46716", "accept": false, - "reason": "Documentation/comment update vs secrets inheritance change; not the same underlying fix." + "reason": "Same workflow file, but comments-only clarification is not the same as changing secrets inheritance." }, { "left": "pull_request:46504", "right": "pull_request:46563", "accept": false, - "reason": "Different workflow files and unrelated fixes; no evidence of a shared concrete bug." + "reason": "Different workflow files and unrelated fixes; no duplicate overlap beyond generic CI maintenance." }, { "left": "pull_request:46432", "right": "pull_request:46563", "accept": false, - "reason": "SSH runner workflow hardening vs security-gate comments; separate changes." + "reason": "Different workflow files and unrelated purposes; not the same underlying change." }, { "left": "pull_request:46504", "right": "pull_request:46716", "accept": false, - "reason": "Build CI shell quoting fix vs pr-ci-security-gate secrets change; unrelated." + "reason": "Different workflow files and different CI concerns; not a duplicate." }, { "left": "pull_request:46432", "right": "pull_request:46716", "accept": false, - "reason": "Different workflows and different code paths; not duplicates." + "reason": "Different workflow files and unrelated fixes; no shared concrete issue." } ] }, @@ -27877,7 +28122,7 @@ "soft_pairs": [ "pull_request:46310|pull_request:47162" ], - "prepared_review_unit_hash": "6e06ce90ddb10590397366b9532734fcdcd7b2627832d6c0662985a6512db56d", + "prepared_review_unit_hash": "c78619d4333d6a6498f73c61a48a777174a21ae9b55cc0bd783a287ab8fd2e6f", "component_budget": { "node_count": 2, "item_count": 2, @@ -27899,18 +28144,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs fix the same off-by-one in the sliding-window right mask (`-dist <= right_window_size`), and the later PR is the more complete version with test coverage and an additional model instance of the same fix.", - "confidence": 0.94, + "summary": "Both PRs fix the same off-by-one error in Moonshine Streaming\u2019s sliding-window right boundary; PR 47162 is the more complete superset because it also covers Gemma4 and adds tests.", + "confidence": 0.9, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47162 is the broader, more complete fix: it includes the MoonshineStreaming change from PR 46310, adds the same correction in Gemma4, and adds a test.", + "canonical_pr_reason": "PR 47162 is the better canonical PR because it covers the same Moonshine Streaming fix as 46310, extends the same boundary correction to Gemma4, and includes a test file, making it the more complete change set.", "best_issue_reason": null, - "best_pr_reason": "PR 47162 is the best representative because it fully covers the underlying right-window off-by-one bug and includes validation.", + "best_pr_reason": "PR 47162 is the best representative of the cluster because it contains the shared bug fix plus additional coverage, whereas 46310 is a narrower subset of the same fix.", "soft_edge_verdicts": [ { "left": "pull_request:46310", "right": "pull_request:47162", "accept": true, - "reason": "Both PRs address the same concrete off-by-one in `sliding_window_mask_function` right-window handling; PR 47162 subsumes PR 46310\u2019s MoonshineStreaming fix and could plausibly be merged as one change." + "reason": "Accept: both PRs correct the same off-by-one right-window mask bug (`-dist <= right_window_size`) in the same Moonshine Streaming code path, and 47162 is a superset that plausibly subsumes 46310." } ] }, @@ -27940,7 +28185,7 @@ "soft_pairs": [ "pull_request:46574|pull_request:46580" ], - "prepared_review_unit_hash": "d491199cbcf6f065cde88663163d1a554e9c29ce2b46112e3f379b76ff535d78", + "prepared_review_unit_hash": "0d69c2492508fc47d07ff807d649b6a5a2d4862d92d15554f9aed69fe18a4d49", "component_budget": { "node_count": 2, "item_count": 2, @@ -27962,18 +28207,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated tokenizer PRs: one fixes CodeLlama decode stripping, the other fixes DeepSeek-Coder/DeepSeek-LLM tokenizer-class resolution. They do not look like the same underlying bug or change.", - "confidence": 0.99, + "summary": "These are two unrelated pull requests: one fixes CodeLlama decoding of real leading spaces, and the other fixes DeepSeek-Coder/DeepSeek-LLM tokenizer class selection in auto-loading. They share only a vague tokenizer theme, not the same underlying bug or change.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the PRs address different models and different tokenizer failure modes.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "No single best PR for this cluster because the changes are unrelated.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:46574", "right": "pull_request:46580", "accept": false, - "reason": "Different code paths and different bugs: CodeLlama decoder setup/leading-space stripping vs DeepSeek checkpoint tokenizer-class overrides. They are not mergeable as one PR." + "reason": "Different code paths and failures: CodeLlama decoder stripping vs. auto tokenizer class overrides for DeepSeek checkpoints. Not the same fix and would not be merged as one PR." } ] }, @@ -28003,7 +28248,7 @@ "soft_pairs": [ "pull_request:46773|pull_request:47087" ], - "prepared_review_unit_hash": "0160305ecfa3c11158f634e0a41ae739607a53ad55d4c24acc2ebc43e844a8b3", + "prepared_review_unit_hash": "71c1148e891f219c27aefe1909893bdc3958e76bc032fb8aa955d1aa928992d0", "component_budget": { "node_count": 2, "item_count": 2, @@ -28025,18 +28270,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at the level of padding-mask logic in linear/recurrent attention models, but they fix different bugs: one is about batch-size-1 masking in linear-attention paths, the other about masking being skipped on continued forwards/chunked prefill/cache continuation. They should not be merged as duplicates.", - "confidence": 0.94, + "summary": "Two pull requests in the same masking area, but they fix different bugs: one restores padding masking for batch-size-1 linear-attention forwards, the other fixes recurrent masking being skipped on continued/cache forwards. They are related conceptually, but not duplicates.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47087 is the better canonical PR because it has an explicit issue target, broader and more central masking-utils changes, and the clearer/shared underlying regression being fixed across multiple recurrent-layer models.", + "canonical_pr_reason": "PR 47087 is the better canonical PR: it targets the shared masking utility at the source, has an explicit issue target, and covers the recurrent/cached-forward masking regression in a more centralized way.", "best_issue_reason": null, - "best_pr_reason": "PR 47087 is the best standalone PR to represent the cluster because it targets the common masking utility, has the explicit linked issue, and addresses a concrete, well-scoped bug across several affected models.", + "best_pr_reason": "PR 47087 is the strongest standalone change for the cluster because it fixes a concrete code-path bug in masking_utils and propagates the fix across affected recurrent models with tests.", "soft_edge_verdicts": [ { "left": "pull_request:46773", "right": "pull_request:47087", "accept": false, - "reason": "Do not merge: 46773 fixes padding-mask skipping for batch size 1 in linear-attention models, while 47087 fixes masking being skipped on continued forwards/cache continuation. Similar symptom area, but different code paths and different failure modes." + "reason": "Both concern skipped padding masks, but 46773 fixes a batch-size-1 condition in linear-attention model code paths while 47087 fixes masking on continued forwards/chunked prefill via masking_utils. Different triggers and different bugs, so not duplicates." } ] }, @@ -28068,12 +28313,12 @@ "pull_request:47074|pull_request:47076", "pull_request:47074|pull_request:47195" ], - "prepared_review_unit_hash": "274626889140a8b00aaad68aafb592e78bab24de5edea2ecb705a1755fca1534", + "prepared_review_unit_hash": "df63fc7596a26857ade3149c3470f9e111dec61cc9cef50cae19dde8fb9844f1", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 6434, + "serialized_chars": 6433, "estimated_input_tokens": 1609, "estimated_eval_tokens": 3474 }, @@ -28081,7 +28326,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 6434, + "serialized_chars": 6433, "estimated_input_tokens": 1609, "estimated_eval_tokens": 3474 }, @@ -28090,24 +28335,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three unrelated pull requests touching different subsystems: AnyToAnyPipeline input normalization, Gemma4 processor placeholder validation, and generation candidate length capping. None appear to fix the same concrete bug or change.", - "confidence": 0.96, + "summary": "No duplicates here: the three PRs fix different code paths in different subsystems, so none of the soft edges should be merged.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47074 is the best canonical anchor in this cluster because it has the clearest standalone bugfix and the highest review/discussion activity; the other PRs are separate fixes in unrelated code paths.", + "canonical_pr_reason": "No single canonical PR: 47074 fixes AnyToAnyPipeline input key normalization, 47076 fixes Gemma4 processor placeholder validation, and 47195 fixes candidate decoding length capping. They are separate bugs with no shared code path.", "best_issue_reason": null, - "best_pr_reason": "PR 47074 is the strongest representative PR here: it makes a narrowly scoped pipeline correction, has targeted test coverage, and is more mature than the other two open PRs.", + "best_pr_reason": "47074 is the best standalone representative only because it is the clearest user-facing bugfix, but it should not be treated as a duplicate of the others.", "soft_edge_verdicts": [ { "left": "pull_request:47074", "right": "pull_request:47076", "accept": false, - "reason": "Different code paths and problems: AnyToAnyPipeline key normalization vs Gemma4 audio placeholder validation. They are not the same underlying change." + "reason": "Different components and behaviors: multimodal pipeline input-key mapping vs Gemma4 audio/text placeholder construction. Shared multimodal context is too broad to indicate the same bug or mergeable PR." }, { "left": "pull_request:47074", "right": "pull_request:47195", "accept": false, - "reason": "Unrelated functionality: multimodal pipeline input keys vs generation candidate length budgeting. No shared concrete bug or mergeable common fix." + "reason": "Unrelated fixes in different code paths: pipeline input normalization vs generation candidate length capping. No evidence they address the same underlying defect." } ] }, @@ -28169,178 +28414,178 @@ "pull_request:47280|pull_request:47282", "pull_request:47310|pull_request:47380" ], - "prepared_review_unit_hash": "99bec095ca95040628fb9190f1d3c0026097bea4b3b2da934053d25a5c07598b", + "prepared_review_unit_hash": "8f61d10c20f34d9a3d576b9b437e8bd48034a78f9bdaeee49c3ccce654f5dab6", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 25, - "serialized_chars": 28950, - "estimated_input_tokens": 7238, - "estimated_eval_tokens": 14732 + "serialized_chars": 28954, + "estimated_input_tokens": 7239, + "estimated_eval_tokens": 14734 }, "budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 25, - "serialized_chars": 28950, - "estimated_input_tokens": 7238, - "estimated_eval_tokens": 14732 + "serialized_chars": 28954, + "estimated_input_tokens": 7239, + "estimated_eval_tokens": 14734 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly heterogeneous serge-generated fix PRs across unrelated models/subsystems. I only see two likely duplicate subgroups: MiniMax-M3-VL load_error fixes and Mamba2 OOM test fixes. Everything else is too model- or bug-specific to merge as duplicates.", - "confidence": 0.89, + "summary": "This cluster is mostly a set of unrelated model-specific integration-test fixes that only look superficially similar. I found two true duplicate pairs: the MiniMax-M3-VL load_error/offload-folder fixes and the Mamba2 OOM fixes. The EdgeTAM PRs and the rest are false positives or broader multi-bug umbrellas.", + "confidence": 0.84, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47280 is the clearest self-contained representative of the strongest duplicate subgroup here: a concrete MiniMax-M3-VL load_error workaround using an offload folder. The cluster as a whole is not one issue, but this is the best canonical PR among the near-duplicates.", + "canonical_pr_reason": "PR 47280 is the cleanest representative of a true duplicate fix: it directly addresses the MiniMax-M3-VL load_error by adding a temporary offload folder, with the same root cause as 47123 but a slightly more complete patch.", "best_issue_reason": null, - "best_pr_reason": "PR 47280 is the best overall representative because it fixes a concrete runtime failure with a targeted change, rather than mostly adjusting test expectations or bundling unrelated fixes.", + "best_pr_reason": "PR 47280 is the best overall representative among the actual duplicates because it is a direct, local fix for a concrete load_error, rather than a broad umbrella or a cross-model edit.", "soft_edge_verdicts": [ { "left": "pull_request:47122", "right": "pull_request:47281", "accept": false, - "reason": "Different models and different code paths; this is just generic similarity, not the same bug/change." + "reason": "Different models and different problems: EdgeTAM import_or_config kwarg filtering vs GLM46V modular/model refactor work." }, { "left": "pull_request:47122", "right": "pull_request:47123", "accept": false, - "reason": "Edgetam source fix vs MiniMax-M3-VL load_error workaround; unrelated failures." + "reason": "Different models and failure modes: EdgeTAM backbone kwarg handling vs MiniMax-M3-VL load_error/offload handling." }, { "left": "pull_request:47122", "right": "pull_request:47280", "accept": false, - "reason": "Different model and failure type; no shared concrete bug." + "reason": "Different models and unrelated code paths; this is EdgeTAM kwarg filtering vs MiniMax-M3-VL load_error mitigation." }, { "left": "pull_request:47122", "right": "pull_request:47224", "accept": false, - "reason": "47224 is a broader multi-fix bundle; the overlapping Edgetam hunk is not enough to treat the PRs as the same change." + "reason": "47224 is a multi-bug umbrella touching tokenizer/whisper/other fixes in addition to EdgeTAM, so it is not the same single change as 47122." }, { "left": "pull_request:47122", "right": "pull_request:47310", "accept": false, - "reason": "Unrelated Edgetam vs Gemma3 generation behavior fixes." + "reason": "Different models and unrelated fixes: EdgeTAM kwarg filtering vs Gemma3 generation cache behavior." }, { "left": "pull_request:47123", "right": "pull_request:47280", "accept": true, - "reason": "Same MiniMax-M3-VL integration load_error, same test file, and both add an offload-folder workaround; they look like the same underlying fix." + "reason": "Same underlying MiniMax-M3-VL load_error on model loading; both add an offload folder for accelerate/device_map behavior, with 47280 being a more complete version." }, { "left": "pull_request:47123", "right": "pull_request:47266", "accept": false, - "reason": "Different model and different failure mode." + "reason": "Both are load_error fixes, but for different models and different code paths, so they are not the same bug." }, { "left": "pull_request:47123", "right": "pull_request:47281", "accept": false, - "reason": "Different model and code path." + "reason": "MiniMax-M3-VL load_error fix vs GLM46V model refactor; different artifact and change type." }, { "left": "pull_request:47123", "right": "pull_request:47124", "accept": false, - "reason": "Different model and failure class; only the auto-generated title style is shared." + "reason": "Different models and failure modes: MiniMax-M3-VL load_error vs Falcon-Mamba output_mismatch." }, { "left": "pull_request:47124", "right": "pull_request:47310", "accept": false, - "reason": "Falcon-Mamba output expectation update vs core generation-config fix; not the same concrete bug." + "reason": "Both are output mismatches, but for different models and different generation code paths; too broad to be the same change." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47266", + "accept": false, + "reason": "Different model-specific failures and fixes: Falcon-Mamba output mismatch vs DeepSeek-V3.2 load_error." }, { "left": "pull_request:47124", "right": "pull_request:47282", "accept": false, - "reason": "Different model and different fix strategy." + "reason": "Different models and different root causes: Falcon-Mamba output mismatch vs Mamba2 OOM." }, { "left": "pull_request:47124", "right": "pull_request:47380", "accept": false, - "reason": "Different model and unrelated OOM/test-cleanup work." + "reason": "Different models and different failure modes; no shared concrete code-path problem." }, { "left": "pull_request:47124", "right": "pull_request:47281", "accept": false, - "reason": "Different model and code path." + "reason": "Different models and different kinds of changes; this is not the same bug." }, { "left": "pull_request:47266", "right": "pull_request:47280", "accept": false, - "reason": "Different models and failures; only both are load_error-style test fixes." - }, - { - "left": "pull_request:47266", - "right": "pull_request:47380", - "accept": false, - "reason": "Different model and different OOM workaround." + "reason": "Different models and different failure modes: DeepSeek-V3.2 load_error vs MiniMax-M3-VL load_error." }, { "left": "pull_request:47266", "right": "pull_request:47281", "accept": false, - "reason": "Different model and code path." + "reason": "Different model families and unrelated fixes; the similarity is only superficial integration-test cleanup." }, { "left": "pull_request:47280", "right": "pull_request:47281", "accept": false, - "reason": "Different model and code path." + "reason": "MiniMax-M3-VL load_error fix vs GLM46V model refactor; not the same underlying bug." }, { "left": "pull_request:47280", "right": "pull_request:47282", "accept": false, - "reason": "Different model and failure mode." + "reason": "Different models and different failures: MiniMax-M3-VL load_error vs Mamba2 OOM." }, { "left": "pull_request:47280", "right": "pull_request:47310", "accept": false, - "reason": "Different model and unrelated fix." + "reason": "Different models and unrelated fixes; no shared code path." }, { "left": "pull_request:47280", "right": "pull_request:47380", "accept": false, - "reason": "Different model and different OOM workaround; not the same concrete change." + "reason": "Different models and different failure modes; MiniMax-M3-VL load_error mitigation is unrelated to Mamba2 OOM." }, { "left": "pull_request:47281", "right": "pull_request:47282", "accept": false, - "reason": "Different model and fix surface." + "reason": "Different models and different fix types; GLM46V refactor/update vs Mamba2 OOM workaround." }, { "left": "pull_request:47282", "right": "pull_request:47380", "accept": true, - "reason": "Same Mamba2 integration OOM batch, same file, and both are memory-management workarounds for the same failing test suite; they could plausibly be merged." + "reason": "Same Mamba2 integration-test OOM problem in the same test file; both are plausible variants of the same memory-pressure fix and could be merged into one PR." }, { "left": "pull_request:47282", "right": "pull_request:47310", "accept": false, - "reason": "Mamba2 OOM workaround vs Gemma3 generation-config behavior fix; unrelated." + "reason": "Different models and unrelated failures: Mamba2 OOM vs Gemma3 output mismatch." }, { "left": "pull_request:47310", "right": "pull_request:47380", "accept": false, - "reason": "Core generation behavior fix vs Mamba2 test OOM workaround; different issues." + "reason": "Different models and different concrete issues; cache default handling in Gemma3 is unrelated to Mamba2 OOM." } ] }, @@ -28370,47 +28615,50 @@ "soft_pairs": [ "pull_request:47140|pull_request:47303" ], - "prepared_review_unit_hash": "8cbad5381f85743bc5699a0b85d02c7cefac9e96ae2b6a632cd37d5d040f29fa", + "prepared_review_unit_hash": "71bda56b95be761b35c09dd75f0d74633192a4e1af0250880651c2a08a5d4f26", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, - "estimated_input_tokens": 1102, - "estimated_eval_tokens": 2460 + "serialized_chars": 4409, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, - "estimated_input_tokens": 1102, - "estimated_eval_tokens": 2460 + "serialized_chars": 4409, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs modify the chat CLI\u2019s management URL construction in `src/transformers/cli/chat.py`, but they are not clearly the same fix: #47140 is a specific `/v1` base-url normalization change, while #47303 routes at least `load_model` requests to the service root with a different helper and a different tracking issue.", - "confidence": 0.67, + "summary": "Both PRs fix the same chat CLI management-endpoint URL bug: requests were being sent relative to the wrong base instead of the service root. The differences are mostly implementation style and scope, not a separate underlying change.", + "confidence": 0.82, "canonical_issue_reason": null, - "canonical_pr_reason": "No clear canonical PR for duplication: the two PRs overlap in subsystem and URL handling, but they target different issue numbers and implement different URL-routing strategies.", + "canonical_pr_reason": "PR #47303 is the better canonical PR: it directly frames the fix as routing chat management calls to the service root, has an explicit tracking issue, and uses a shared helper to normalize management URLs.", "best_issue_reason": null, - "best_pr_reason": "PR #47303 is the better representative of the cluster\u2019s broader intent because it explicitly frames the change as routing chat management calls to the service root, rather than a narrower `/v1`-only workaround.", + "best_pr_reason": "PR #47303 is the strongest representative of the cluster because it captures the general service-root routing problem more clearly than the /v1-specific variant and appears to cover the same chat management code path.", "soft_edge_verdicts": [ { "left": "pull_request:47140", "right": "pull_request:47303", - "accept": false, - "reason": "Reject: similar chat CLI URL logic, but #47140 fixes `/v1` management-base normalization for `RichInterface`, while #47303 uses a separate helper to route management calls to the service root and appears tied to a different concrete change." + "accept": true, + "reason": "Both changes address the same underlying chat CLI management URL bug in src/transformers/cli/chat.py, redirecting management requests from a /v1-style API base to the service root." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both PRs touch the same chat CLI management URL path in src/transformers/cli/chat.py and the titles/diffs support the shared service-root vs /v1-base correction. The soft-edge verdict is also conservative enough, since it says the changes appear to address the same underlying bug rather than claiming they are identical." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -28433,7 +28681,7 @@ "soft_pairs": [ "pull_request:47261|pull_request:47262" ], - "prepared_review_unit_hash": "d33f7265c2259f1c399cac647511111038688a29c7a4894192f0638635605282", + "prepared_review_unit_hash": "57a03110bcb78e2d550970d665196eaafb63311ba36411283d1dac107d8c9309", "component_budget": { "node_count": 2, "item_count": 2, @@ -28455,18 +28703,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are unrelated: one fixes loss scaling with label smoothing during gradient accumulation, the other adds/validates `min_tokens_to_keep` in `TopHLogitsWarper`. They should not be clustered as duplicates.", - "confidence": 0.99, + "summary": "The two PRs are unrelated: one fixes loss scaling with label smoothing during gradient accumulation in Trainer, and the other changes TopHLogitsWarper to honor min_tokens_to_keep. They touch different subsystems and different concrete code paths, so this is not a duplicate pair.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47261 is the more canonical representative of the cluster only in the sense that it is a concrete bug fix with an explicit issue link; however, it is not actually a duplicate of PR 47262.", + "canonical_pr_reason": "PR #47261 is the better canonical representative only in the sense that it is a concrete, self-contained bug fix with an explicit issue target; PR #47262 is a separate generation feature/validation change in a different path.", "best_issue_reason": null, - "best_pr_reason": "PR 47261 is the stronger standalone bug-fix PR: it targets a specific training-loss path, has an explicit issue reference, and includes the relevant trainer/test changes.", + "best_pr_reason": "PR #47261 is the most representative standalone PR here because it addresses a specific correctness bug with clear reproduction context, whereas PR #47262 is not the same underlying change and should not be clustered with it.", "soft_edge_verdicts": [ { "left": "pull_request:47261", "right": "pull_request:47262", "accept": false, - "reason": "Different code paths and problems: Trainer label-smoothing/gradient-accumulation loss scaling vs. generation TopH warper parameter handling. They are not the same underlying change and would not plausibly merge into one PR." + "reason": "Different bugs in different modules: Trainer label-smoothing/gradient-accumulation loss scaling vs TopHLogitsWarper token-retention handling. They are not the same underlying change and should not be merged." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 7b8795972b97df0e4c74017cde7f383178d11f4c..6f441fc6b9b2da0f1d0c994cbcde5cf13ab90e1c 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260719t000028z", + "analysis_id": "hybrid-model-20260719t060027z", "archived_artifacts": { - "hybrid": "snapshots/20260719T000028Z/analysis-runs/hybrid-model-20260719t000028z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260719T000028Z/analysis-runs/hybrid-model-20260719t000028z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-07-19T00:11:51Z", + "published_at": "2026-07-19T06:10:38Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260719T000028Z", + "snapshot_id": "20260719T060027Z", "variant": "hybrid" } diff --git a/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.json b/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..a4e7f015f22019402b30bd2e145a4a9b0a9b72dd --- /dev/null +++ b/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.json @@ -0,0 +1,5142 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260719T060027Z", + "generated_at": "2026-07-19T06:10:36Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "cluster_id": "cluster-43451-15", + "summary": "Cluster of 1 issues and 14 PRs centered on issue #47355.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 47355, + "canonical_pr_number": 47322, + "issue_numbers": [ + 47355 + ], + "pr_numbers": [ + 43451, + 46509, + 46658, + 47003, + 47096, + 47143, + 47163, + 47298, + 47306, + 47322, + 47383, + 47398, + 47399, + 47409 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43451, + "right_pr_number": 46509, + "code_similarity": 0.167, + "size_similarity": 0.378, + "file_overlap": 0.097, + "area_overlap": 0.074, + "patch_similarity": 0.245, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 1375, + 1382 + ] + ], + "right_ranges": [ + [ + 795, + 802 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 1097, + 1103 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 46658, + "code_similarity": 0.276, + "size_similarity": 0.77, + "file_overlap": 0.071, + "area_overlap": 0.163, + "patch_similarity": 0.286, + "shared_filenames": [ + "src/transformers/conversion_mapping.py", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/conversion_mapping.py", + "left_ranges": [ + [ + 737, + 769 + ] + ], + "right_ranges": [ + [ + 739, + 830 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 1085, + 1091 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47003, + "code_similarity": 0.202, + "size_similarity": 0.861, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.196, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47096, + "code_similarity": 0.036, + "size_similarity": 0.099, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.108, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47143, + "code_similarity": 0.154, + "size_similarity": 0.476, + "file_overlap": 0.097, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [ + "src/transformers/models/__init__.py", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/__init__.py", + "left_ranges": [ + [ + 290, + 296 + ] + ], + "right_ranges": [ + [ + 305, + 311 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.004, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.02, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47298, + "code_similarity": 0.02, + "size_similarity": 0.055, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.06, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47306, + "code_similarity": 0.044, + "size_similarity": 0.137, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.109, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47383, + "code_similarity": 0.011, + "size_similarity": 0.031, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.029, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47399, + "code_similarity": 0.003, + "size_similarity": 0.008, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.006, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47409, + "code_similarity": 0.011, + "size_similarity": 0.019, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.047, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 46658, + "code_similarity": 0.131, + "size_similarity": 0.291, + "file_overlap": 0.051, + "area_overlap": 0.081, + "patch_similarity": 0.196, + "shared_filenames": [ + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 1085, + 1091 + ] + ] + } + ] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47003, + "code_similarity": 0.115, + "size_similarity": 0.438, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.182, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47096, + "code_similarity": 0.071, + "size_similarity": 0.263, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.121, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47143, + "code_similarity": 0.254, + "size_similarity": 0.793, + "file_overlap": 0.071, + "area_overlap": 0.143, + "patch_similarity": 0.161, + "shared_filenames": [ + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47163, + "code_similarity": 0.006, + "size_similarity": 0.012, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.024, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47298, + "code_similarity": 0.039, + "size_similarity": 0.146, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.064, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47306, + "code_similarity": 0.092, + "size_similarity": 0.363, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.127, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47322, + "code_similarity": 0.002, + "size_similarity": 0.006, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47383, + "code_similarity": 0.022, + "size_similarity": 0.082, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.036, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47398, + "code_similarity": 0.002, + "size_similarity": 0.007, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.005, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47399, + "code_similarity": 0.006, + "size_similarity": 0.022, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.009, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47409, + "code_similarity": 0.021, + "size_similarity": 0.049, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.073, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47003, + "code_similarity": 0.166, + "size_similarity": 0.663, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.221, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47096, + "code_similarity": 0.03, + "size_similarity": 0.077, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.097, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47143, + "code_similarity": 0.116, + "size_similarity": 0.367, + "file_overlap": 0.051, + "area_overlap": 0.0, + "patch_similarity": 0.183, + "shared_filenames": [ + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1085, + 1091 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47163, + "code_similarity": 0.003, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.015, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47298, + "code_similarity": 0.015, + "size_similarity": 0.042, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.045, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47306, + "code_similarity": 0.035, + "size_similarity": 0.105, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.096, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47383, + "code_similarity": 0.007, + "size_similarity": 0.024, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.017, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47399, + "code_similarity": 0.002, + "size_similarity": 0.006, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.005, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47409, + "code_similarity": 0.008, + "size_similarity": 0.014, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.036, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47096, + "code_similarity": 0.21, + "size_similarity": 0.115, + "file_overlap": 0.113, + "area_overlap": 0.352, + "patch_similarity": 0.201, + "shared_filenames": [ + "src/transformers/exporters/exporter_dynamo.py", + "src/transformers/exporters/exporter_executorch.py", + "src/transformers/exporters/utils.py", + "src/transformers/models/kimi_k25/modeling_kimi_k25.py", + "src/transformers/models/kimi_k25/modular_kimi_k25.py", + "src/transformers/vision_utils.py", + "tests/exporters/test_export.py", + "tests/models/kimi_k25/test_modeling_kimi_k25.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/exporters/exporter_dynamo.py", + "left_ranges": [ + [ + 43, + 49 + ], + [ + 352, + 359 + ], + [ + 383, + 390 + ], + [ + 452, + 473 + ], + [ + 490, + 510 + ], + [ + 551, + 562 + ], + [ + 597, + 609 + ] + ], + "right_ranges": [ + [ + 350, + 357 + ], + [ + 381, + 388 + ] + ] + }, + { + "filename": "src/transformers/exporters/exporter_executorch.py", + "left_ranges": [ + [ + 70, + 88 + ], + [ + 196, + 204 + ] + ], + "right_ranges": [ + [ + 418, + 442 + ] + ] + }, + { + "filename": "src/transformers/exporters/utils.py", + "left_ranges": [ + [ + 232, + 238 + ], + [ + 432, + 441 + ], + [ + 444, + 467 + ], + [ + 472, + 498 + ], + [ + 514, + 536 + ], + [ + 662, + 706 + ], + [ + 747, + 759 + ], + [ + 808, + 815 + ] + ], + "right_ranges": [ + [ + 432, + 441 + ], + [ + 444, + 455 + ], + [ + 464, + 490 + ], + [ + 708, + 720 + ] + ] + }, + { + "filename": "src/transformers/models/kimi_k25/modeling_kimi_k25.py", + "left_ranges": [ + [ + 43, + 49 + ], + [ + 99, + 172 + ], + [ + 176, + 184 + ], + [ + 191, + 212 + ], + [ + 218, + 226 + ], + [ + 281, + 294 + ], + [ + 500, + 532 + ], + [ + 545, + 550 + ], + [ + 557, + 569 + ], + [ + 575, + 582 + ] + ], + "right_ranges": [ + [ + 43, + 49 + ], + [ + 99, + 172 + ], + [ + 176, + 184 + ], + [ + 189, + 210 + ], + [ + 216, + 224 + ], + [ + 279, + 292 + ], + [ + 498, + 530 + ], + [ + 543, + 548 + ], + [ + 555, + 567 + ], + [ + 573, + 580 + ] + ] + }, + { + "filename": "src/transformers/models/kimi_k25/modular_kimi_k25.py", + "left_ranges": [ + [ + 35, + 41 + ], + [ + 54, + 154 + ], + [ + 240, + 248 + ], + [ + 255, + 276 + ], + [ + 282, + 305 + ], + [ + 431, + 436 + ], + [ + 443, + 455 + ], + [ + 461, + 468 + ] + ], + "right_ranges": [ + [ + 35, + 41 + ], + [ + 54, + 154 + ], + [ + 240, + 248 + ], + [ + 253, + 274 + ], + [ + 280, + 303 + ], + [ + 429, + 434 + ], + [ + 441, + 453 + ], + [ + 459, + 466 + ] + ] + }, + { + "filename": "src/transformers/vision_utils.py", + "left_ranges": [ + [ + 32, + 61 + ] + ], + "right_ranges": [ + [ + 32, + 61 + ] + ] + }, + { + "filename": "tests/exporters/test_export.py", + "left_ranges": [ + [ + 13, + 21 + ], + [ + 26, + 32 + ], + [ + 37, + 43 + ], + [ + 49, + 132 + ], + [ + 239, + 279 + ], + [ + 307, + 341 + ], + [ + 354, + 461 + ], + [ + 493, + 517 + ], + [ + 532, + 539 + ], + [ + 564, + 570 + ], + [ + 573, + 582 + ], + [ + 599, + 605 + ], + [ + 608, + 617 + ], + [ + 621, + 656 + ], + [ + 659, + 677 + ], + [ + 701, + 708 + ], + [ + 711, + 717 + ], + [ + 720, + 726 + ], + [ + 732, + 738 + ], + [ + 755, + 761 + ], + [ + 764, + 773 + ], + [ + 778, + 813 + ], + [ + 816, + 834 + ] + ], + "right_ranges": [ + [ + 181, + 188 + ] + ] + }, + { + "filename": "tests/models/kimi_k25/test_modeling_kimi_k25.py", + "left_ranges": [ + [ + 127, + 132 + ] + ], + "right_ranges": [ + [ + 127, + 132 + ] + ] + } + ] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47143, + "code_similarity": 0.132, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.145, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.02, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47298, + "code_similarity": 0.027, + "size_similarity": 0.064, + "file_overlap": 0.023, + "area_overlap": 0.0, + "patch_similarity": 0.052, + "shared_filenames": [ + "src/transformers/utils/__init__.py", + "src/transformers/utils/import_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/utils/__init__.py", + "left_ranges": [ + [ + 182, + 188 + ] + ], + "right_ranges": [ + [ + 253, + 259 + ] + ] + }, + { + "filename": "src/transformers/utils/import_utils.py", + "left_ranges": [ + [ + 916, + 926 + ] + ], + "right_ranges": [ + [ + 756, + 785 + ] + ] + } + ] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47306, + "code_similarity": 0.046, + "size_similarity": 0.159, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.093, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47383, + "code_similarity": 0.015, + "size_similarity": 0.036, + "file_overlap": 0.014, + "area_overlap": 0.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "tests/models/kimi_k25/test_modeling_kimi_k25.py" + ], + "shared_file_areas": [ + { + "filename": "tests/models/kimi_k25/test_modeling_kimi_k25.py", + "left_ranges": [ + [ + 127, + 132 + ] + ], + "right_ranges": [ + [ + 18, + 24 + ], + [ + 27, + 36 + ], + [ + 216, + 401 + ] + ] + } + ] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47399, + "code_similarity": 0.003, + "size_similarity": 0.01, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.007, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47409, + "code_similarity": 0.011, + "size_similarity": 0.022, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.042, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47143, + "code_similarity": 0.052, + "size_similarity": 0.208, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.069, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47163, + "code_similarity": 0.012, + "size_similarity": 0.045, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.021, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47298, + "code_similarity": 0.12, + "size_similarity": 0.554, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.06, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47306, + "code_similarity": 0.158, + "size_similarity": 0.725, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.089, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47322, + "code_similarity": 0.006, + "size_similarity": 0.022, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.011, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47383, + "code_similarity": 0.103, + "size_similarity": 0.314, + "file_overlap": 0.125, + "area_overlap": 0.0, + "patch_similarity": 0.02, + "shared_filenames": [ + "tests/models/kimi_k25/test_modeling_kimi_k25.py" + ], + "shared_file_areas": [ + { + "filename": "tests/models/kimi_k25/test_modeling_kimi_k25.py", + "left_ranges": [ + [ + 127, + 132 + ] + ], + "right_ranges": [ + [ + 18, + 24 + ], + [ + 27, + 36 + ], + [ + 216, + 401 + ] + ] + } + ] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47398, + "code_similarity": 0.007, + "size_similarity": 0.025, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.011, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47399, + "code_similarity": 0.021, + "size_similarity": 0.083, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.028, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47096, + "right_pr_number": 47409, + "code_similarity": 0.045, + "size_similarity": 0.188, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.052, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.009, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.013, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47298, + "code_similarity": 0.028, + "size_similarity": 0.116, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.033, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47306, + "code_similarity": 0.069, + "size_similarity": 0.288, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.079, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47383, + "code_similarity": 0.015, + "size_similarity": 0.065, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.016, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47398, + "code_similarity": 0.001, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.002, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47399, + "code_similarity": 0.004, + "size_similarity": 0.017, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47409, + "code_similarity": 0.013, + "size_similarity": 0.039, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.034, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47298, + "code_similarity": 0.022, + "size_similarity": 0.08, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.04, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47306, + "code_similarity": 0.011, + "size_similarity": 0.032, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.029, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47322, + "code_similarity": 0.101, + "size_similarity": 0.5, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.008, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47383, + "code_similarity": 0.03, + "size_similarity": 0.142, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.013, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47398, + "code_similarity": 0.116, + "size_similarity": 0.571, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.009, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47399, + "code_similarity": 0.113, + "size_similarity": 0.538, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.033, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47409, + "code_similarity": 0.052, + "size_similarity": 0.237, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.032, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47298, + "right_pr_number": 47306, + "code_similarity": 0.089, + "size_similarity": 0.402, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.055, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47298, + "right_pr_number": 47322, + "code_similarity": 0.009, + "size_similarity": 0.04, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.007, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47298, + "right_pr_number": 47383, + "code_similarity": 0.118, + "size_similarity": 0.566, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.033, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47298, + "right_pr_number": 47398, + "code_similarity": 0.01, + "size_similarity": 0.046, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.007, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47298, + "right_pr_number": 47399, + "code_similarity": 0.034, + "size_similarity": 0.149, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.024, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47298, + "right_pr_number": 47409, + "code_similarity": 0.074, + "size_similarity": 0.339, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.042, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47306, + "right_pr_number": 47322, + "code_similarity": 0.004, + "size_similarity": 0.016, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.008, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47306, + "right_pr_number": 47383, + "code_similarity": 0.049, + "size_similarity": 0.227, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.026, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47306, + "right_pr_number": 47398, + "code_similarity": 0.005, + "size_similarity": 0.018, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.008, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47306, + "right_pr_number": 47399, + "code_similarity": 0.014, + "size_similarity": 0.06, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.013, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47306, + "right_pr_number": 47409, + "code_similarity": 0.045, + "size_similarity": 0.136, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.116, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47322, + "right_pr_number": 47383, + "code_similarity": 0.015, + "size_similarity": 0.071, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.005, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47322, + "right_pr_number": 47398, + "code_similarity": 0.195, + "size_similarity": 0.875, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.133, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47322, + "right_pr_number": 47399, + "code_similarity": 0.054, + "size_similarity": 0.269, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.0, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47322, + "right_pr_number": 47409, + "code_similarity": 0.029, + "size_similarity": 0.119, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.032, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47383, + "right_pr_number": 47398, + "code_similarity": 0.017, + "size_similarity": 0.081, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.006, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47383, + "right_pr_number": 47399, + "code_similarity": 0.053, + "size_similarity": 0.264, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47383, + "right_pr_number": 47409, + "code_similarity": 0.123, + "size_similarity": 0.599, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.022, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47398, + "right_pr_number": 47399, + "code_similarity": 0.062, + "size_similarity": 0.308, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.0, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47398, + "right_pr_number": 47409, + "code_similarity": 0.286, + "size_similarity": 0.136, + "file_overlap": 0.25, + "area_overlap": 0.5, + "patch_similarity": 0.059, + "shared_filenames": [ + "src/transformers/models/glm_moe_dsa/configuration_glm_moe_dsa.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/glm_moe_dsa/configuration_glm_moe_dsa.py", + "left_ranges": [ + [ + 159, + 173 + ] + ], + "right_ranges": [ + [ + 158, + 166 + ] + ] + } + ] + }, + { + "left_pr_number": 47399, + "right_pr_number": 47409, + "code_similarity": 0.093, + "size_similarity": 0.441, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.031, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-43979-11", + "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 43979, + "canonical_pr_number": 44007, + "issue_numbers": [ + 43979 + ], + "pr_numbers": [ + 43996, + 44007, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43996, + "right_pr_number": 44007, + "code_similarity": 0.179, + "size_similarity": 0.576, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.429, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44013, + "code_similarity": 0.122, + "size_similarity": 0.318, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.392, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44044, + "code_similarity": 0.245, + "size_similarity": 0.864, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.479, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44066, + "code_similarity": 0.225, + "size_similarity": 0.818, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.408, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44072, + "code_similarity": 0.14, + "size_similarity": 0.303, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.528, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44085, + "code_similarity": 0.216, + "size_similarity": 0.783, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44129, + "code_similarity": 0.163, + "size_similarity": 0.643, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.229, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44154, + "code_similarity": 0.153, + "size_similarity": 0.535, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.31, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44722, + "code_similarity": 0.225, + "size_similarity": 0.848, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.368, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44013, + "code_similarity": 0.19, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.531, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44044, + "code_similarity": 0.186, + "size_similarity": 0.667, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.354, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44066, + "code_similarity": 0.188, + "size_similarity": 0.704, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.315, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44072, + "code_similarity": 0.212, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.708, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44085, + "code_similarity": 0.195, + "size_similarity": 0.735, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.318, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44129, + "code_similarity": 0.103, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.191, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44154, + "code_similarity": 0.238, + "size_similarity": 0.93, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.344, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44722, + "code_similarity": 0.178, + "size_similarity": 0.679, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.28, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44044, + "code_similarity": 0.126, + "size_similarity": 0.368, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.351, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44066, + "code_similarity": 0.127, + "size_similarity": 0.389, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.325, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44072, + "code_similarity": 0.29, + "size_similarity": 0.952, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.667, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44085, + "code_similarity": 0.131, + "size_similarity": 0.406, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.329, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44129, + "code_similarity": 0.07, + "size_similarity": 0.205, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.192, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44154, + "code_similarity": 0.177, + "size_similarity": 0.594, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.389, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44722, + "code_similarity": 0.118, + "size_similarity": 0.375, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.287, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44066, + "code_similarity": 0.25, + "size_similarity": 0.947, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.404, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44072, + "code_similarity": 0.136, + "size_similarity": 0.351, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.442, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44085, + "code_similarity": 0.24, + "size_similarity": 0.906, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.394, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44129, + "code_similarity": 0.147, + "size_similarity": 0.555, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.243, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44154, + "code_similarity": 0.17, + "size_similarity": 0.62, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.306, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44722, + "code_similarity": 0.257, + "size_similarity": 0.982, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.402, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44072, + "code_similarity": 0.133, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.393, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44085, + "code_similarity": 0.763, + "size_similarity": 0.957, + "file_overlap": 0.5, + "area_overlap": 0.825, + "patch_similarity": 0.887, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ] + } + ] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44129, + "code_similarity": 0.145, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.263, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44154, + "code_similarity": 0.174, + "size_similarity": 0.654, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.286, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44722, + "code_similarity": 0.907, + "size_similarity": 0.964, + "file_overlap": 1.0, + "area_overlap": 0.808, + "patch_similarity": 0.874, + "shared_filenames": [ + "src/transformers/models/codegen/modeling_codegen.py", + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/codegen/modeling_codegen.py", + "left_ranges": [ + [ + 245, + 266 + ] + ], + "right_ranges": [ + [ + 228, + 234 + ] + ] + }, + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44085, + "code_similarity": 0.137, + "size_similarity": 0.387, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44129, + "code_similarity": 0.074, + "size_similarity": 0.195, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.231, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44154, + "code_similarity": 0.175, + "size_similarity": 0.566, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.414, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44722, + "code_similarity": 0.124, + "size_similarity": 0.357, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.347, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44129, + "code_similarity": 0.141, + "size_similarity": 0.503, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.272, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44154, + "code_similarity": 0.18, + "size_similarity": 0.684, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.289, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44722, + "code_similarity": 0.728, + "size_similarity": 0.923, + "file_overlap": 0.5, + "area_overlap": 0.791, + "patch_similarity": 0.78, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44154, + "code_similarity": 0.099, + "size_similarity": 0.344, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44722, + "code_similarity": 0.146, + "size_similarity": 0.545, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.247, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44154, + "right_pr_number": 44722, + "code_similarity": 0.164, + "size_similarity": 0.631, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.255, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-46894-9", + "summary": "Cluster of 4 issues and 5 PRs centered on issue #46894.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46894, + "canonical_pr_number": 46903, + "issue_numbers": [ + 46894, + 46897, + 46901, + 47394 + ], + "pr_numbers": [ + 46895, + 46903, + 46958, + 47090, + 47395 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46895, + "right_pr_number": 46903, + "code_similarity": 0.533, + "size_similarity": 0.404, + "file_overlap": 0.333, + "area_overlap": 0.933, + "patch_similarity": 0.17, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 637, + 650 + ] + ], + "right_ranges": [ + [ + 637, + 651 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 500, + 513 + ] + ], + "right_ranges": [ + [ + 500, + 514 + ] + ] + } + ] + }, + { + "left_pr_number": 46895, + "right_pr_number": 46958, + "code_similarity": 0.779, + "size_similarity": 0.915, + "file_overlap": 1.0, + "area_overlap": 0.5, + "patch_similarity": 0.81, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "tests/models/cohere_asr/test_modeling_cohere_asr.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 637, + 650 + ] + ], + "right_ranges": [ + [ + 22, + 28 + ], + [ + 638, + 645 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 500, + 513 + ] + ], + "right_ranges": [ + [ + 16, + 22 + ], + [ + 501, + 508 + ] + ] + }, + { + "filename": "tests/models/cohere_asr/test_modeling_cohere_asr.py", + "left_ranges": [ + [ + 35, + 41 + ], + [ + 176, + 214 + ] + ], + "right_ranges": [ + [ + 15, + 21 + ], + [ + 180, + 221 + ] + ] + } + ] + }, + { + "left_pr_number": 46895, + "right_pr_number": 47090, + "code_similarity": 0.459, + "size_similarity": 0.281, + "file_overlap": 0.273, + "area_overlap": 0.765, + "patch_similarity": 0.355, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "tests/models/cohere_asr/test_modeling_cohere_asr.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 637, + 650 + ] + ], + "right_ranges": [ + [ + 637, + 649 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 500, + 513 + ] + ], + "right_ranges": [ + [ + 500, + 512 + ] + ] + }, + { + "filename": "tests/models/cohere_asr/test_modeling_cohere_asr.py", + "left_ranges": [ + [ + 35, + 41 + ], + [ + 176, + 214 + ] + ], + "right_ranges": [ + [ + 175, + 224 + ] + ] + } + ] + }, + { + "left_pr_number": 46895, + "right_pr_number": 47395, + "code_similarity": 0.187, + "size_similarity": 0.683, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.333, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46903, + "right_pr_number": 46958, + "code_similarity": 0.325, + "size_similarity": 0.369, + "file_overlap": 0.333, + "area_overlap": 0.364, + "patch_similarity": 0.156, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 637, + 651 + ] + ], + "right_ranges": [ + [ + 22, + 28 + ], + [ + 638, + 645 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 500, + 514 + ] + ], + "right_ranges": [ + [ + 16, + 22 + ], + [ + 501, + 508 + ] + ] + } + ] + }, + { + "left_pr_number": 46903, + "right_pr_number": 47090, + "code_similarity": 0.576, + "size_similarity": 0.695, + "file_overlap": 0.333, + "area_overlap": 0.871, + "patch_similarity": 0.214, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "src/transformers/models/pp_formulanet/modeling_pp_formulanet.py", + "src/transformers/models/pp_formulanet/modular_pp_formulanet.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 637, + 651 + ] + ], + "right_ranges": [ + [ + 637, + 649 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 500, + 514 + ] + ], + "right_ranges": [ + [ + 500, + 512 + ] + ] + }, + { + "filename": "src/transformers/models/pp_formulanet/modeling_pp_formulanet.py", + "left_ranges": [ + [ + 1103, + 1116 + ] + ], + "right_ranges": [ + [ + 1103, + 1118 + ] + ] + }, + { + "filename": "src/transformers/models/pp_formulanet/modular_pp_formulanet.py", + "left_ranges": [ + [ + 488, + 501 + ] + ], + "right_ranges": [ + [ + 488, + 503 + ] + ] + } + ] + }, + { + "left_pr_number": 46903, + "right_pr_number": 47395, + "code_similarity": 0.083, + "size_similarity": 0.276, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.186, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46958, + "right_pr_number": 47090, + "code_similarity": 0.397, + "size_similarity": 0.257, + "file_overlap": 0.273, + "area_overlap": 0.598, + "patch_similarity": 0.362, + "shared_filenames": [ + "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "tests/models/cohere_asr/test_modeling_cohere_asr.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/cohere_asr/modeling_cohere_asr.py", + "left_ranges": [ + [ + 22, + 28 + ], + [ + 638, + 645 + ] + ], + "right_ranges": [ + [ + 637, + 649 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 16, + 22 + ], + [ + 501, + 508 + ] + ], + "right_ranges": [ + [ + 500, + 512 + ] + ] + }, + { + "filename": "tests/models/cohere_asr/test_modeling_cohere_asr.py", + "left_ranges": [ + [ + 15, + 21 + ], + [ + 180, + 221 + ] + ], + "right_ranges": [ + [ + 175, + 224 + ] + ] + } + ] + }, + { + "left_pr_number": 46958, + "right_pr_number": 47395, + "code_similarity": 0.201, + "size_similarity": 0.747, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.342, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47090, + "right_pr_number": 47395, + "code_similarity": 0.078, + "size_similarity": 0.192, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.262, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-41211-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41211.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41211, + "canonical_pr_number": 44339, + "issue_numbers": [ + 41211 + ], + "pr_numbers": [ + 41356, + 44339 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41356, + "right_pr_number": 44339, + "code_similarity": 0.155, + "size_similarity": 0.094, + "file_overlap": 0.273, + "area_overlap": 0.078, + "patch_similarity": 0.183, + "shared_filenames": [ + "docs/source/en/model_doc/deimv2.md", + "src/transformers/models/deimv2/__init__.py", + "src/transformers/models/deimv2/configuration_deimv2.py", + "src/transformers/models/deimv2/modeling_deimv2.py", + "tests/models/deimv2/__init__.py", + "tests/models/deimv2/test_modeling_deimv2.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/model_doc/deimv2.md", + "left_ranges": [ + [ + 1, + 132 + ] + ], + "right_ranges": [ + [ + 1, + 65 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/__init__.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/configuration_deimv2.py", + "left_ranges": [ + [ + 1, + 74 + ] + ], + "right_ranges": [ + [ + 1, + 266 + ] + ] + }, + { + "filename": "tests/models/deimv2/test_modeling_deimv2.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 1734 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-47138-4", + "summary": "Cluster of 2 issues and 2 PRs centered on issue #47138.", + "status": "open", + "confidence": 0.6, + "canonical_issue_number": 47138, + "canonical_pr_number": 47303, + "issue_numbers": [ + 47138, + 47139 + ], + "pr_numbers": [ + 47140, + 47303 + ], + "evidence_types": [ + "closing_reference", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 47140, + "right_pr_number": 47303, + "code_similarity": 0.656, + "size_similarity": 0.65, + "file_overlap": 1.0, + "area_overlap": 0.447, + "patch_similarity": 0.466, + "shared_filenames": [ + "src/transformers/cli/chat.py", + "tests/cli/test_chat.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/cli/chat.py", + "left_ranges": [ + [ + 20, + 26 + ], + [ + 103, + 122 + ], + [ + 346, + 356 + ], + [ + 383, + 389 + ], + [ + 473, + 479 + ] + ], + "right_ranges": [ + [ + 103, + 115 + ], + [ + 216, + 224 + ], + [ + 383, + 389 + ] + ] + }, + { + "filename": "tests/cli/test_chat.py", + "left_ranges": [ + [ + 15, + 28 + ], + [ + 51, + 105 + ] + ], + "right_ranges": [ + [ + 14, + 29 + ], + [ + 52, + 76 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43240-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43240, + "canonical_pr_number": 43251, + "issue_numbers": [ + 43240 + ], + "pr_numbers": [ + 43251, + 43254 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43251, + "right_pr_number": 43254, + "code_similarity": 0.794, + "size_similarity": 0.64, + "file_overlap": 1.0, + "area_overlap": 0.667, + "patch_similarity": 0.882, + "shared_filenames": [ + "src/transformers/loss/loss_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/loss/loss_utils.py", + "left_ranges": [ + [ + 30, + 50 + ] + ], + "right_ranges": [ + [ + 30, + 43 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45647-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45647.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 45647, + "canonical_pr_number": 45738, + "issue_numbers": [ + 45647 + ], + "pr_numbers": [ + 45737, + 45738 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 45737, + "right_pr_number": 45738, + "code_similarity": 0.526, + "size_similarity": 0.11, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py", + "left_ranges": [ + [ + 491, + 497 + ] + ], + "right_ranges": [ + [ + 491, + 497 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46159-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46159.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 46159, + "canonical_pr_number": 46174, + "issue_numbers": [ + 46159 + ], + "pr_numbers": [ + 46174, + 46780 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 46174, + "right_pr_number": 46780, + "code_similarity": 0.841, + "size_similarity": 0.87, + "file_overlap": 1.0, + "area_overlap": 0.767, + "patch_similarity": 0.657, + "shared_filenames": [ + "src/transformers/models/aimv2/modeling_aimv2.py", + "src/transformers/models/d_fine/modeling_d_fine.py", + "src/transformers/models/deimv2/modeling_deimv2.py", + "src/transformers/models/pp_doclayout_v2/modeling_pp_doclayout_v2.py", + "src/transformers/models/pp_doclayout_v3/modeling_pp_doclayout_v3.py", + "src/transformers/models/rt_detr/modeling_rt_detr.py", + "src/transformers/models/rt_detr_v2/modeling_rt_detr_v2.py", + "src/transformers/models/vit_mae/modeling_vit_mae.py", + "src/transformers/models/vit_mae/modular_vit_mae.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/aimv2/modeling_aimv2.py", + "left_ranges": [ + [ + 134, + 141 + ], + [ + 144, + 154 + ], + [ + 157, + 165 + ] + ], + "right_ranges": [ + [ + 134, + 140 + ], + [ + 143, + 153 + ], + [ + 156, + 162 + ] + ] + }, + { + "filename": "src/transformers/models/d_fine/modeling_d_fine.py", + "left_ranges": [ + [ + 631, + 638 + ], + [ + 641, + 651 + ], + [ + 654, + 662 + ] + ], + "right_ranges": [ + [ + 631, + 637 + ], + [ + 640, + 650 + ], + [ + 653, + 659 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/modeling_deimv2.py", + "left_ranges": [ + [ + 703, + 710 + ], + [ + 713, + 723 + ], + [ + 726, + 734 + ] + ], + "right_ranges": [ + [ + 703, + 709 + ], + [ + 712, + 722 + ], + [ + 725, + 731 + ] + ] + }, + { + "filename": "src/transformers/models/pp_doclayout_v2/modeling_pp_doclayout_v2.py", + "left_ranges": [ + [ + 1551, + 1558 + ], + [ + 1561, + 1571 + ], + [ + 1574, + 1582 + ] + ], + "right_ranges": [ + [ + 1551, + 1557 + ], + [ + 1560, + 1570 + ], + [ + 1573, + 1579 + ] + ] + }, + { + "filename": "src/transformers/models/pp_doclayout_v3/modeling_pp_doclayout_v3.py", + "left_ranges": [ + [ + 804, + 811 + ], + [ + 814, + 824 + ], + [ + 827, + 835 + ] + ], + "right_ranges": [ + [ + 804, + 810 + ], + [ + 813, + 823 + ], + [ + 826, + 832 + ] + ] + }, + { + "filename": "src/transformers/models/rt_detr/modeling_rt_detr.py", + "left_ranges": [ + [ + 847, + 854 + ], + [ + 857, + 867 + ], + [ + 870, + 878 + ] + ], + "right_ranges": [ + [ + 847, + 853 + ], + [ + 856, + 866 + ], + [ + 869, + 875 + ] + ] + }, + { + "filename": "src/transformers/models/rt_detr_v2/modeling_rt_detr_v2.py", + "left_ranges": [ + [ + 976, + 983 + ], + [ + 986, + 996 + ], + [ + 999, + 1007 + ] + ], + "right_ranges": [ + [ + 976, + 982 + ], + [ + 985, + 995 + ], + [ + 998, + 1004 + ] + ] + }, + { + "filename": "src/transformers/models/vit_mae/modeling_vit_mae.py", + "left_ranges": [ + [ + 152, + 159 + ], + [ + 162, + 172 + ], + [ + 175, + 183 + ] + ], + "right_ranges": [ + [ + 152, + 158 + ], + [ + 161, + 171 + ], + [ + 174, + 180 + ] + ] + }, + { + "filename": "src/transformers/models/vit_mae/modular_vit_mae.py", + "left_ranges": [ + [ + 62, + 69 + ], + [ + 72, + 82 + ], + [ + 85, + 93 + ] + ], + "right_ranges": [ + [ + 62, + 68 + ], + [ + 71, + 81 + ], + [ + 84, + 90 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46869-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46869.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 46869, + "canonical_pr_number": 46886, + "issue_numbers": [ + 46869 + ], + "pr_numbers": [ + 46886, + 46922 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 46886, + "right_pr_number": 46922, + "code_similarity": 0.711, + "size_similarity": 0.857, + "file_overlap": 1.0, + "area_overlap": 0.5, + "patch_similarity": 0.429, + "shared_filenames": [ + "src/transformers/models/yolos/modeling_yolos.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/yolos/modeling_yolos.py", + "left_ranges": [ + [ + 446, + 452 + ] + ], + "right_ranges": [ + [ + 434, + 440 + ], + [ + 446, + 452 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45561-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45561.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 45561, + "canonical_pr_number": 45645, + "issue_numbers": [ + 45561 + ], + "pr_numbers": [ + 45639, + 45645 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45639, + "right_pr_number": 45645, + "code_similarity": 0.721, + "size_similarity": 0.973, + "file_overlap": 0.75, + "area_overlap": 0.563, + "patch_similarity": 0.696, + "shared_filenames": [ + "src/transformers/testing_utils.py", + "tests/utils/test_testing_utils.py", + "utils/notification_service.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/testing_utils.py", + "left_ranges": [ + [ + 3611, + 3618 + ], + [ + 3835, + 3869 + ] + ], + "right_ranges": [ + [ + 3525, + 3558 + ], + [ + 3782, + 3788 + ] + ] + }, + { + "filename": "tests/utils/test_testing_utils.py", + "left_ranges": [ + [ + 1, + 144 + ] + ], + "right_ranges": [ + [ + 1, + 114 + ] + ] + }, + { + "filename": "utils/notification_service.py", + "left_ranges": [ + [ + 935, + 967 + ] + ], + "right_ranges": [ + [ + 935, + 967 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46305-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46305.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46305, + "canonical_pr_number": 47162, + "issue_numbers": [ + 46305 + ], + "pr_numbers": [ + 46310, + 47162 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46310, + "right_pr_number": 47162, + "code_similarity": 0.577, + "size_similarity": 0.338, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.062, + "shared_filenames": [ + "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py", + "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/moonshine_streaming/modeling_moonshine_streaming.py", + "left_ranges": [ + [ + 356, + 362 + ] + ], + "right_ranges": [ + [ + 356, + 362 + ] + ] + }, + { + "filename": "src/transformers/models/moonshine_streaming/modular_moonshine_streaming.py", + "left_ranges": [ + [ + 269, + 275 + ] + ], + "right_ranges": [ + [ + 269, + 275 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45081-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45081.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 45081, + "canonical_pr_number": 45317, + "issue_numbers": [ + 45081 + ], + "pr_numbers": [ + 45086, + 45317 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45086, + "right_pr_number": 45317, + "code_similarity": 0.589, + "size_similarity": 0.34, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.136, + "shared_filenames": [ + "src/transformers/tokenization_utils_tokenizers.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/tokenization_utils_tokenizers.py", + "left_ranges": [ + [ + 1360, + 1370 + ], + [ + 1374, + 1380 + ] + ], + "right_ranges": [ + [ + 1360, + 1370 + ], + [ + 1374, + 1380 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43698-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43698.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43698, + "canonical_pr_number": 43779, + "issue_numbers": [ + 43698 + ], + "pr_numbers": [ + 43779, + 43816 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43779, + "right_pr_number": 43816, + "code_similarity": 0.418, + "size_similarity": 0.538, + "file_overlap": 1.0, + "area_overlap": 0.02, + "patch_similarity": 0.02, + "shared_filenames": [ + "src/transformers/integrations/integration_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/integrations/integration_utils.py", + "left_ranges": [ + [ + 2227, + 2241 + ], + [ + 2303, + 2309 + ] + ], + "right_ranges": [ + [ + 2278, + 2291 + ], + [ + 2309, + 2322 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43656-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 43656, + "issue_numbers": [], + "pr_numbers": [ + 43656, + 43842 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43656, + "right_pr_number": 43842, + "code_similarity": 0.405, + "size_similarity": 0.5, + "file_overlap": 1.0, + "area_overlap": 0.0, + "patch_similarity": 0.036, + "shared_filenames": [ + "src/transformers/cli/serve.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/cli/serve.py", + "left_ranges": [ + [ + 11, + 18 + ], + [ + 30, + 36 + ], + [ + 315, + 323 + ], + [ + 665, + 671 + ], + [ + 931, + 937 + ], + [ + 1843, + 1849 + ], + [ + 1868, + 1874 + ] + ], + "right_ranges": [ + [ + 54, + 61 + ], + [ + 587, + 593 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-47123-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 47123, + "issue_numbers": [], + "pr_numbers": [ + 47123, + 47280 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 47123, + "right_pr_number": 47280, + "code_similarity": 0.547, + "size_similarity": 0.522, + "file_overlap": 1.0, + "area_overlap": 0.368, + "patch_similarity": 0.093, + "shared_filenames": [ + "tests/models/minimax_m3_vl/test_modeling_minimax_m3_vl.py" + ], + "shared_file_areas": [ + { + "filename": "tests/models/minimax_m3_vl/test_modeling_minimax_m3_vl.py", + "left_ranges": [ + [ + 14, + 20 + ], + [ + 487, + 493 + ] + ], + "right_ranges": [ + [ + 14, + 21 + ], + [ + 481, + 499 + ], + [ + 520, + 530 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-47282-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 47380, + "issue_numbers": [], + "pr_numbers": [ + 47282, + 47380 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 47282, + "right_pr_number": 47380, + "code_similarity": 0.476, + "size_similarity": 0.5, + "file_overlap": 1.0, + "area_overlap": 0.194, + "patch_similarity": 0.056, + "shared_filenames": [ + "tests/models/mamba2/test_modeling_mamba2.py" + ], + "shared_file_areas": [ + { + "filename": "tests/models/mamba2/test_modeling_mamba2.py", + "left_ranges": [ + [ + 420, + 438 + ], + [ + 450, + 474 + ], + [ + 482, + 524 + ] + ], + "right_ranges": [ + [ + 18, + 24 + ], + [ + 406, + 419 + ], + [ + 426, + 432 + ], + [ + 457, + 463 + ], + [ + 487, + 493 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-44018-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 44068, + "issue_numbers": [], + "pr_numbers": [ + 44018, + 44068 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 44018, + "right_pr_number": 44068, + "code_similarity": 0.766, + "size_similarity": 0.939, + "file_overlap": 1.0, + "area_overlap": 0.425, + "patch_similarity": 0.866, + "shared_filenames": [ + "src/transformers/models/gpt_neo/modeling_gpt_neo.py", + "tests/models/gpt_neo/test_modeling_gpt_neo.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gpt_neo/modeling_gpt_neo.py", + "left_ranges": [ + [ + 26, + 46 + ], + [ + 138, + 143 + ], + [ + 182, + 187 + ], + [ + 283, + 295 + ], + [ + 327, + 341 + ], + [ + 348, + 354 + ], + [ + 360, + 369 + ], + [ + 399, + 406 + ], + [ + 411, + 419 + ], + [ + 428, + 433 + ], + [ + 472, + 492 + ], + [ + 507, + 513 + ], + [ + 519, + 528 + ], + [ + 541, + 559 + ], + [ + 562, + 567 + ], + [ + 595, + 601 + ], + [ + 607, + 614 + ], + [ + 627, + 643 + ], + [ + 685, + 690 + ], + [ + 708, + 714 + ], + [ + 720, + 727 + ], + [ + 740, + 757 + ], + [ + 761, + 766 + ], + [ + 780, + 786 + ], + [ + 791, + 798 + ], + [ + 807, + 822 + ], + [ + 840, + 845 + ] + ], + "right_ranges": [ + [ + 26, + 31 + ], + [ + 34, + 44 + ], + [ + 136, + 141 + ], + [ + 180, + 185 + ], + [ + 281, + 293 + ], + [ + 325, + 339 + ], + [ + 346, + 352 + ], + [ + 358, + 367 + ], + [ + 397, + 404 + ], + [ + 409, + 417 + ], + [ + 426, + 434 + ], + [ + 463, + 483 + ], + [ + 498, + 504 + ], + [ + 510, + 519 + ], + [ + 532, + 550 + ], + [ + 553, + 564 + ], + [ + 586, + 592 + ], + [ + 598, + 605 + ], + [ + 618, + 634 + ], + [ + 676, + 688 + ], + [ + 699, + 705 + ], + [ + 711, + 718 + ], + [ + 731, + 748 + ], + [ + 752, + 762 + ], + [ + 771, + 777 + ], + [ + 782, + 789 + ], + [ + 798, + 813 + ], + [ + 831, + 836 + ] + ] + }, + { + "filename": "tests/models/gpt_neo/test_modeling_gpt_neo.py", + "left_ranges": [ + [ + 458, + 464 + ] + ], + "right_ranges": [ + [ + 458, + 464 + ] + ] + } + ] + } + ] + } + ], + "duplicate_issues": [ + { + "cluster_id": "cluster-46894-9", + "canonical_issue_number": 46894, + "duplicate_issue_numbers": [ + 46897, + 46901, + 47394 + ], + "reason": "Issues in cluster-46894-9 are treated as duplicates because they share closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-47138-4", + "canonical_issue_number": 47138, + "duplicate_issue_numbers": [ + 47139 + ], + "reason": "Issues in cluster-47138-4 are treated as duplicates because they share closing_reference, soft_similarity evidence." + } + ], + "duplicate_prs": [ + { + "cluster_id": "cluster-41211-3", + "canonical_pr_number": 44339, + "duplicate_pr_numbers": [ + 41356 + ], + "target_issue_number": 41211, + "reason": "PRs in cluster-41211-3 are treated as duplicates because they converge on issue #41211 with closing_reference evidence." + }, + { + "cluster_id": "cluster-43240-3", + "canonical_pr_number": 43251, + "duplicate_pr_numbers": [ + 43254 + ], + "target_issue_number": 43240, + "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43451-15", + "canonical_pr_number": 47322, + "duplicate_pr_numbers": [ + 43451, + 46509, + 46658, + 47003, + 47096, + 47143, + 47163, + 47298, + 47306, + 47383, + 47398, + 47399, + 47409 + ], + "target_issue_number": 47355, + "reason": "PRs in cluster-43451-15 are treated as duplicates because they converge on issue #47355 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-43656-2", + "canonical_pr_number": 43656, + "duplicate_pr_numbers": [ + 43842 + ], + "target_issue_number": null, + "reason": "PRs in cluster-43656-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-43698-3", + "canonical_pr_number": 43779, + "duplicate_pr_numbers": [ + 43816 + ], + "target_issue_number": 43698, + "reason": "PRs in cluster-43698-3 are treated as duplicates because they converge on issue #43698 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43979-11", + "canonical_pr_number": 44007, + "duplicate_pr_numbers": [ + 43996, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "target_issue_number": 43979, + "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-44018-2", + "canonical_pr_number": 44068, + "duplicate_pr_numbers": [ + 44018 + ], + "target_issue_number": null, + "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-45081-3", + "canonical_pr_number": 45317, + "duplicate_pr_numbers": [ + 45086 + ], + "target_issue_number": 45081, + "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45561-3", + "canonical_pr_number": 45645, + "duplicate_pr_numbers": [ + 45639 + ], + "target_issue_number": 45561, + "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45647-3", + "canonical_pr_number": 45738, + "duplicate_pr_numbers": [ + 45737 + ], + "target_issue_number": 45647, + "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46159-3", + "canonical_pr_number": 46174, + "duplicate_pr_numbers": [ + 46780 + ], + "target_issue_number": 46159, + "reason": "PRs in cluster-46159-3 are treated as duplicates because they converge on issue #46159 with closing_reference evidence." + }, + { + "cluster_id": "cluster-46305-3", + "canonical_pr_number": 47162, + "duplicate_pr_numbers": [ + 46310 + ], + "target_issue_number": 46305, + "reason": "PRs in cluster-46305-3 are treated as duplicates because they converge on issue #46305 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-46869-3", + "canonical_pr_number": 46886, + "duplicate_pr_numbers": [ + 46922 + ], + "target_issue_number": 46869, + "reason": "PRs in cluster-46869-3 are treated as duplicates because they converge on issue #46869 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46894-9", + "canonical_pr_number": 46903, + "duplicate_pr_numbers": [ + 46895, + 46958, + 47090, + 47395 + ], + "target_issue_number": 46894, + "reason": "PRs in cluster-46894-9 are treated as duplicates because they converge on issue #46894 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-47123-2", + "canonical_pr_number": 47123, + "duplicate_pr_numbers": [ + 47280 + ], + "target_issue_number": null, + "reason": "PRs in cluster-47123-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-47138-4", + "canonical_pr_number": 47303, + "duplicate_pr_numbers": [ + 47140 + ], + "target_issue_number": 47138, + "reason": "PRs in cluster-47138-4 are treated as duplicates because they converge on issue #47138 with closing_reference, soft_similarity evidence." + }, + { + "cluster_id": "cluster-47282-2", + "canonical_pr_number": 47380, + "duplicate_pr_numbers": [ + 47282 + ], + "target_issue_number": null, + "reason": "PRs in cluster-47282-2 are treated as duplicates because they share soft_similarity evidence." + } + ], + "best_issue": { + "cluster_id": "cluster-43451-15", + "issue_number": 47355, + "reason": "Issue #47355 is the strongest global issue candidate because it is open, belongs to a cluster with 15 artifacts, and carries 5 discussion comments plus 4 inbound references.", + "score": 264.0 + }, + "best_pr": { + "cluster_id": "cluster-43451-15", + "pr_number": 47322, + "reason": "PR #47322 is the strongest global PR candidate because it is open, belongs to a cluster with 15 artifacts, links to 1 issues, and carries 2 review/discussion events.", + "score": 278.986 + } +} diff --git a/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..55aa110cfca666d67dcbf83cfb0c03f11e1e19de --- /dev/null +++ b/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,28729 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260719T060027Z", + "generated_at": "2026-07-19T06:10:36Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35308", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43425", + "issue:43541", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43828", + "issue:43975", + "issue:44292", + "issue:44322", + "issue:44457", + "issue:44509", + "issue:44512", + "issue:44560", + "issue:44805", + "issue:44841", + "issue:44918", + "issue:45081", + "issue:45161", + "issue:45237", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45800", + "issue:45823", + "issue:45859", + "issue:45874", + "issue:45901", + "issue:45902", + "issue:46015", + "issue:46165", + "issue:46294", + "issue:46468", + "issue:46515", + "issue:46661", + "issue:46869" + ], + "soft_pairs": [ + "issue:44322|issue:45464", + "issue:45237|issue:45685", + "issue:43716|issue:45237", + "issue:43828|issue:45237", + "issue:45237|issue:46468", + "issue:45237|issue:45507", + "issue:30277|issue:35545", + "issue:35545|issue:42915", + "issue:35545|issue:41720", + "issue:45901|issue:46015", + "issue:35545|issue:36010", + "issue:44509|issue:44512", + "issue:44918|issue:46165", + "issue:46165|issue:46869", + "issue:44805|issue:45823", + "issue:43638|issue:44805", + "issue:43638|issue:46165", + "issue:43596|issue:44805", + "issue:44805|issue:46294", + "issue:46165|issue:46468", + "issue:44805|issue:45161", + "issue:44560|issue:46165", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:34928|issue:35308", + "issue:44841|issue:46661", + "issue:24643|issue:30277", + "issue:43716|issue:46869", + "issue:34928|issue:44805", + "issue:45685|issue:46869", + "issue:34928|issue:36331", + "issue:45859|issue:45902", + "issue:43638|issue:45161", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:44292|issue:45237", + "issue:45874|issue:46294", + "issue:39290|issue:45081", + "issue:45800|issue:46515", + "issue:45823|issue:45874", + "issue:43425|issue:44292", + "issue:34928|issue:41720", + "issue:43828|issue:45161", + "issue:43541|issue:43716", + "issue:43828|issue:44292", + "issue:43716|issue:45685", + "issue:45507|issue:45685", + "issue:44292|issue:44918", + "issue:43716|issue:44292", + "issue:44918|issue:45161", + "issue:43039|issue:44918", + "issue:41762|issue:43638" + ], + "prepared_review_unit_hash": "f65c9956318cfbe5c0418ff2bada5621451827258d140a051fa2039ce96170f5", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 56, + "serialized_chars": 44512, + "estimated_input_tokens": 11128, + "estimated_eval_tokens": 22512 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are broad, unrelated runtime/model-support issues; only the two BertModel ZeRO-3 IndexError reports look like a true duplicate pair.", + "confidence": 0.95, + "canonical_issue_reason": "issue:43596 is the broadest, clearest representative of the shared BertModel + DeepSpeed ZeRO-3 zero-sized-tensor IndexError; issue:43638 is the narrower non-pretrained Bert variant of the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43596 best captures the underlying bug and is the most general duplicate target among the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different failures: config attribute missing vs streaming chat/completions inference issue." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Different hardware/backend bugs (AMD GPU failure vs MPS histogram op not implemented)." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different bug classes: dtype mismatch in Mistral-3 vs GPT-OSS running on AMD GPUs." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to GPT-OSS AMD GPU failure." + }, + { + "left": "issue:45237", + "right": "issue:46468", + "accept": false, + "reason": "Different models and symptoms; one is AMD GPU runtime failure, the other a broken official example." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different model/backend issues; not the same concrete code path." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "ZeRO-3 collective mismatch is unrelated to ModernBERT ONNX export failure." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Both are export/model errors, but the model-specific causes are different." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export failure and Qwen3 auto device-map cuda assert are unrelated." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Table QA crash and outdated RAG examples are different docs/runtime issues." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export bug is unrelated to GenerationMixin import failure." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are stale docs, but they concern different removed content and likely different edits." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "SFT trainer embedding unpacking and training-resume/trust-remote-code are different problems." + }, + { + "left": "issue:46165", + "right": "issue:46869", + "accept": false, + "reason": "Resume-training issue is unrelated to YOLO multi-GPU device-map failure." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask-shape IndexError and Gemma4 PLE device mismatch are not the same bug." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Both are IndexErrors, but the model/setup and failure mechanism differ." + }, + { + "left": "issue:43638", + "right": "issue:46165", + "accept": false, + "reason": "Bert ZeRO-3 init bug is unrelated to resume-training trust-remote-code failure." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Different IndexError contexts; not the same Bert ZeRO-3 initialization bug." + }, + { + "left": "issue:44805", + "right": "issue:46294", + "accept": false, + "reason": "Mask/indexing error is unrelated to Qwen3.5 video token/template mismatch." + }, + { + "left": "issue:46165", + "right": "issue:46468", + "accept": false, + "reason": "Training resume problem and broken official example are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Shape-mismatch IndexError is not the same as GPT-OSS tensor-parallel failure." + }, + { + "left": "issue:44560", + "right": "issue:46165", + "accept": false, + "reason": "Video input StopIteration and training-resume issue are unrelated." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice preset bug is unrelated to ModernBERT ONNX export." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Both mention distributed/ZeRO-3, but the collective mismatch and Bert init IndexError are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same underlying BertModel + DeepSpeed ZeRO-3 zero-sized tensor IndexError; one is just the non-pretrained variant." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross-entropy regression and GPT-OSS tensor-parallel failure are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Different issues: detokenization regression vs LoRA merge/save/reload mismatch." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Qwen3 NVFP4 runtime failure is unrelated to LoRA merge/save behavior." + }, + { + "left": "issue:34928", + "right": "issue:35308", + "accept": false, + "reason": "Both are distributed-training crashes, but activation-checkpointing/FSDP mismatch is not the same as IterableDataset length handling." + }, + { + "left": "issue:44841", + "right": "issue:46661", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to remote-code model_class config_class AttributeError." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "DeepSpeed weight-shape error and collective mismatch are different failure modes." + }, + { + "left": "issue:43716", + "right": "issue:46869", + "accept": false, + "reason": "Mistral-3 dtype mismatch and YOLO multi-GPU device-map failure are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Different distributed/shape errors with different triggers and code paths." + }, + { + "left": "issue:45685", + "right": "issue:46869", + "accept": false, + "reason": "MPS histogram limitation is unrelated to YOLO multi-GPU device-map auto failure." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "Activation-checkpointing/FSDP error is unrelated to CustomTrainer compute_loss signature breakage." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Both touch Qwen3_5Moe rotary embedding, but CPU offload incompatibility and uninitialized-memory after to_empty are distinct bugs." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "Bert ZeRO-3 initialization IndexError is unrelated to GPT-OSS tensor-parallel failure." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "Different models and different distributed failure modes." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Mistral-3 dtype mismatch and GraniteMoEHybrid invalid method call are unrelated." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime error is unrelated to GPT-OSS on AMD GPUs." + }, + { + "left": "issue:45874", + "right": "issue:46294", + "accept": false, + "reason": "Reverse embedding lookup issue is unrelated to video feature/token mismatch." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Gemma3 vLLM config attribute error and Mistral tokenizer regex patch crash are unrelated." + }, + { + "left": "issue:45800", + "right": "issue:46515", + "accept": false, + "reason": "Torch-version incompatibility reports are similar in theme but appear to be different compatibility cases." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Device-map forward mismatch and expensive reverse embedding lookup are different Gemma4 bugs." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "General Torch incompatibility and Qwen-3 NVFP4 runtime error are not the same issue." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "FSDP/checkpointing mismatch and Qwen3 auto device-map cuda assert are unrelated distributed bugs." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Autocast dtype mismatch is unrelated to GPT-OSS tensor-parallel failure." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Mixtral grouped_mm torch-dynamo tracing error and Mistral-3 dtype mismatch are different bugs." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Different model-specific runtime errors with different root causes." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Mistral-3 dtype mismatch and MPS histogram limitation are unrelated." + }, + { + "left": "issue:45507", + "right": "issue:45685", + "accept": false, + "reason": "Invalid method call in GraniteMoEHybrid is unrelated to an MPS int histogram backend limitation." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime error and Qwen3.5 embedding unpacking with TRL are different failures." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Mistral-3 dtype mismatch and Qwen-3 NVFP4 error are unrelated." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL embedding unpacking failure and GPT-OSS tensor-parallel failure are not the same bug." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy regression is unrelated to Qwen3.5 SFT trainer unpacking." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Gemma3 ZeRO-3 empty-tensor IndexError is similar in shape, but the model and code path differ from the BertModel bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet, and the lone accepted soft-edge pair (issue:43596 vs issue:43638) is a conservative, title-supported duplicate judgment. No overstatement stands out." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 2, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:30277", + "issue:34928", + "issue:35308", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:42617", + "issue:43039", + "issue:43296", + "issue:43425", + "issue:43531", + "issue:43596", + "issue:43638", + "issue:43792", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44512", + "issue:44560", + "issue:44589", + "issue:44805", + "issue:44841", + "issue:44863", + "issue:45020", + "issue:45070", + "issue:45237", + "issue:45685", + "issue:45800", + "issue:45812", + "issue:45823", + "issue:45901", + "issue:45920", + "issue:46015", + "issue:46294", + "issue:46304", + "issue:46402", + "issue:46772", + "issue:46901", + "issue:47394" + ], + "soft_pairs": [ + "issue:43827|issue:44512", + "issue:43828|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:45237|issue:46015", + "issue:43039|issue:44805", + "issue:45800|issue:46304", + "issue:43296|issue:46304", + "issue:39290|issue:41720", + "issue:43866|issue:46402", + "issue:41762|issue:43596", + "issue:45237|issue:45901", + "issue:44589|issue:46304", + "issue:43901|issue:44512", + "issue:43039|issue:44560", + "issue:44841|issue:45823", + "issue:45823|issue:46294", + "issue:43975|issue:46015", + "issue:43866|issue:46772", + "issue:45020|issue:46304", + "issue:44863|issue:46015", + "issue:35308|issue:42617", + "issue:45812|issue:45920", + "issue:46901|issue:47394", + "issue:43872|issue:46304", + "issue:44292|issue:46015", + "issue:39290|issue:43531", + "issue:43866|issue:44863", + "issue:41093|issue:43638", + "issue:43792|issue:45901", + "issue:43854|issue:43866", + "issue:43425|issue:45070" + ], + "prepared_review_unit_hash": "1e9124a1400e676c59d1d11f6695c4dad2fcb7b4eca525c314fa73aa2104c988", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 36961, + "estimated_input_tokens": 9241, + "estimated_eval_tokens": 18738 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous, with two likely duplicate subgroups: an AutoTokenizer v5 regression affecting multiple models, and a causal-LM training-loss double-shift issue. Most other soft-similarity pairs are false positives and should stay separate.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45920 is the broadest and most explicit tokenizer regression report, covering multiple affected model families and framing it as a v5-wide silent behavior change.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the strongest anchor for the cluster\u2019s main duplicate theme: it is broad, concrete, and clearly describes the regression in a way that can subsume the narrower Granite-specific report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs-related v5 complaints, but they refer to different removed/deprecated docs content and do not look like the same concrete change." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Different failure modes: autocast dtype mismatch vs MPS histogram/int support." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "One is a Liger Kernel cross-entropy call issue; the other is an autocast dtype mismatch. No shared code path." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "Different distributed/training errors: collective mismatch with DeepSpeed ZeRO3 vs a trainer API signature mismatch." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation checkpointing/FSDP recompute mismatch is unrelated to a missing GenerationMixin import." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "AMD GPU compatibility and outdated RAG docs are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Cross-entropy/Liger Kernel behavior is unrelated to a mask-vs-tensor shape mismatch." + }, + { + "left": "issue:45800", + "right": "issue:46304", + "accept": false, + "reason": "Both mention version incompatibility, but one is torch/transformers generally and the other is USE_DISTRIBUTED=0 support; not the same bug." + }, + { + "left": "issue:43296", + "right": "issue:46304", + "accept": false, + "reason": "vLLM model-loading failure and distributed support breakage are separate issues." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Different model-specific failures: Gemma3 config attribute missing vs Qwen3 auto device-map cuda assert." + }, + { + "left": "issue:43866", + "right": "issue:46402", + "accept": false, + "reason": "Checkpoint corruption and missing text-encoder weight loading are not the same underlying problem." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Both are ZeRO-3 index errors, but they involve different models and the evidence is too thin to treat as one concrete bug." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "GPU compatibility for GPT-OSS is unrelated to table-question-answering crashes." + }, + { + "left": "issue:44589", + "right": "issue:46304", + "accept": false, + "reason": "Float8 storage lookup failure is unrelated to distributed build/runtime support." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "One is a pipeline docs mismatch; the other is a removed run command mention. Similar topic, different docs change." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Cross-entropy/Liger Kernel issue is unrelated to Qwen3-vl video StopIteration." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Different model families and different forward-path failures; not the same bug." + }, + { + "left": "issue:45823", + "right": "issue:46294", + "accept": false, + "reason": "Both involve multimodal model loading/input, but one is device-map forward mismatch and the other is video token/template mismatch." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization bug and outdated RAG docs are unrelated." + }, + { + "left": "issue:43866", + "right": "issue:46772", + "accept": false, + "reason": "Corrupted checkpoint and NVIDIA ModelOpt compatibility are separate problems." + }, + { + "left": "issue:45020", + "right": "issue:46304", + "accept": false, + "reason": "Remote-code regressions and distributed build support are different change areas." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading is unrelated to RAG example documentation." + }, + { + "left": "issue:35308", + "right": "issue:42617", + "accept": false, + "reason": "Both are multi-GPU/distributed failures, but the specific crash modes and entry points differ too much." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Both describe the same tokenizer-ID regression in v5; 45920 is the broader multi-model report and 45812 is a narrower Granite-specific instance." + }, + { + "left": "issue:46901", + "right": "issue:47394", + "accept": true, + "reason": "Both report the same training-loss double-shift bug in causal-LM loss computation; they could plausibly be fixed by one PR." + }, + { + "left": "issue:43872", + "right": "issue:46304", + "accept": false, + "reason": "bitsandbytes storage/type incompatibility is unrelated to distributed support changes." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime error is unrelated to RAG docs." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Gemma3 config attribute error and Qwen3-MoE sliding-window issue are separate model-specific regressions." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Different loading failures: corrupted Ovis2 checkpoint vs NemotronH checkpoint-loading implementation." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Mask/tensor shape mismatch and ZeRO-3 BertModel index error are different bugs." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper load failure and table QA crash are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Unit-test model-loading failure and corrupted checkpoint are not the same issue." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch/transformers incompatibility and pydantic PretrainedConfig breakage are different changes." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is well grounded in the packet and stays conservative overall. The two accepted duplicate pairs are plausible from the issue titles, and the rejected soft matches mostly avoid overclaiming. The canonical issue choice of issue:45920 is also supported by the broader v5 tokenizer regression framing." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 3, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36010", + "issue:39186", + "issue:39290", + "issue:41093", + "issue:41762", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43716", + "issue:43792", + "issue:43828", + "issue:43866", + "issue:43872", + "issue:43975", + "issue:43994", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44479", + "issue:44805", + "issue:44841", + "issue:44918", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45362", + "issue:45399", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45901", + "issue:45902", + "issue:46165", + "issue:46294", + "issue:46468", + "issue:46640", + "issue:46661", + "issue:46869", + "issue:47228" + ], + "soft_pairs": [ + "issue:43541|issue:47228", + "issue:43792|issue:45070", + "issue:43596|issue:45800", + "issue:43975|issue:45901", + "issue:45362|issue:45464", + "issue:43541|issue:46869", + "issue:44292|issue:45901", + "issue:39290|issue:45823", + "issue:36010|issue:39290", + "issue:45464|issue:45685", + "issue:43541|issue:45685", + "issue:43541|issue:43828", + "issue:44841|issue:45084", + "issue:45507|issue:46869", + "issue:43404|issue:43866", + "issue:41093|issue:41762", + "issue:45399|issue:46661", + "issue:43572|issue:45070", + "issue:45800|issue:45823", + "issue:43872|issue:47228", + "issue:44805|issue:44918", + "issue:43828|issue:45464", + "issue:43716|issue:47228", + "issue:43296|issue:43366", + "issue:43366|issue:45464", + "issue:44291|issue:45070", + "issue:43531|issue:45070", + "issue:46869|issue:47228", + "issue:44387|issue:45005", + "issue:45020|issue:45800", + "issue:45507|issue:47228", + "issue:43716|issue:46294", + "issue:41093|issue:43596", + "issue:44479|issue:46294", + "issue:43054|issue:43994", + "issue:46468|issue:46869", + "issue:43596|issue:46165", + "issue:46640|issue:46661", + "issue:45759|issue:45902", + "issue:39186|issue:44918", + "issue:43716|issue:44918" + ], + "prepared_review_unit_hash": "f0b8d71cf51242eda3ea6adda963ab7113bcab321635a51109fb5651794af361", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 41, + "serialized_chars": 39847, + "estimated_input_tokens": 9962, + "estimated_eval_tokens": 20180 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a loose collection of unrelated Transformers issues, with only a couple of clear duplicate-style pairs. The strongest matches are the repeated mask-shape IndexError reports and the Qwen video input/token mismatch regression.", + "confidence": 0.57, + "canonical_issue_reason": "Issue 41093 is the best representative because it is a generic, recurring error signature and has a near-verbatim duplicate (44805), making it the clearest canonical issue among the set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 41093 is the most suitable global issue anchor here: it is concise, generic, and maps cleanly to an exact duplicate report with the same failure mode.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43541", + "right": "issue:47228", + "accept": false, + "reason": "Both are regression/runtime failures, but one is a grouped_mm torch-dynamo issue and the other is a bf16 compilation regression; not the same bug." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Unrelated failures: Whisper model loading vs a PretrainedConfig/pydantic field breakage." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "Different root causes: DeepSpeed ZeRO-3 empty tensor indexing vs torch/transformers version incompatibility." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Different models and symptoms: detokenization bug vs table-question-answering crash." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both involve Qwen3.5, but one is a chat crash and the other is a streaming API failure; too different to treat as the same bug." + }, + { + "left": "issue:43541", + "right": "issue:46869", + "accept": false, + "reason": "Grouped_mm tracing error and YOLO multi-GPU device-map failure are different code paths." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Different failures: NVFP4 model load error vs table QA task crash." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3 sliding_window attribute error is unrelated to Gemma4 device mismatch." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to Gemma3 config attribute access." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Chat/completions streaming failure is unrelated to the MPS histogram kernel error." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Different bugs: torch-dynamo grouped_mm tracing vs MPS int histogram support." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Grouped_mm tracing failure is not the same as an autocast dtype mismatch." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to the template-node compilation error." + }, + { + "left": "issue:45507", + "right": "issue:46869", + "accept": false, + "reason": "GraniteMoE hybrid invalid method call is unrelated to YOLO multi-GPU device mapping." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "lm_head weight tying bug is unrelated to a corrupted checkpoint report." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are IndexError reports, but the shapes and loading paths differ; not enough to assume the same underlying bug." + }, + { + "left": "issue:45399", + "right": "issue:46661", + "accept": false, + "reason": "Flash-Attn fallback gating and remote-code config_class lookup are different issues." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing pad_token_idx and pydantic PretrainedConfig breakage are separate regressions." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Version incompatibility vs Gemma4 device mismatch are not the same bug." + }, + { + "left": "issue:43872", + "right": "issue:47228", + "accept": false, + "reason": "bitsandbytes constructor incompatibility is unrelated to the Sam3 bf16 compilation regression." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "They both mention shape/embedding problems, but one is a mask-index mismatch and the other is a TRL SFT embedding unpacking failure." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Autocast dtype mismatch is unrelated to streaming chat/completions failure." + }, + { + "left": "issue:43716", + "right": "issue:47228", + "accept": false, + "reason": "Mistral-3 image preprocessor dtype mismatch is unrelated to Sam3 bf16 compilation." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "PaddleOCR-VL loading failure and GGUF gpt-oss support are different model/loader issues." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF architecture support is unrelated to Qwen3.5 streaming inference failures." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized TypeError is unrelated to the PretrainedConfig pydantic field issue." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and pydantic config serialization are different defects." + }, + { + "left": "issue:46869", + "right": "issue:47228", + "accept": false, + "reason": "YOLO multi-GPU device-map failure is unrelated to Sam3 bf16 compilation regression." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 memory/OOM regression and tied-weights translation-model issues are different problems." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Remote-code model breakage is broader than, and unrelated to, the torch 2.4.1 / transformers 5.8.0 incompatibility." + }, + { + "left": "issue:45507", + "right": "issue:47228", + "accept": false, + "reason": "GraniteMoE hybrid method error and Sam3 compilation regression are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:46294", + "accept": false, + "reason": "Image preprocessor dtype mismatch and video token/feature mismatch are different multimodal bugs." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are IndexErrors, but they occur in different contexts and don\u2019t share the same concrete failure mode." + }, + { + "left": "issue:44479", + "right": "issue:46294", + "accept": true, + "reason": "Both describe the same Qwen video-input regression: video features/tokens become mismatched in the chat-template path." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Siglip2 text-embedding quality complaint is not the same as the AutoModel/pipeline nonsensical-output report." + }, + { + "left": "issue:46468", + "right": "issue:46869", + "accept": false, + "reason": "Broken official Nemotron example and YOLO multi-GPU device-map failure are different issues." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "DeepSpeed ZeRO-3 index error is unrelated to resuming training with local models and remote code." + }, + { + "left": "issue:46640", + "right": "issue:46661", + "accept": false, + "reason": "Flash Attention 2 issue and remote-code config_class error are different failure modes." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Composite-config unwrapping bug and uninitialized rotary embedding memory are distinct Qwen3.5-MoE issues." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP 2-D weight requirement is unrelated to TRL SFT embedding unpacking." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Mistral-3 image preprocessor dtype mismatch is unrelated to Qwen3.5 embedding unpacking in TRL." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": true, + "reason": "Same mask-shape IndexError; these look like the same underlying indexing bug reported twice." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet and stays conservative overall. The canonical anchor choice for issue 41093 is defensible because 44805 is a near-verbatim duplicate of the same mask-shape IndexError, and the other accepted pair (44479/46294) is also a plausible same-bug match from the issue titles. The remaining soft-edge rejections are appropriately cautious and do not overclaim duplicate status." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 4, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:33666", + "issue:35308", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:41762", + "issue:42915", + "issue:43054", + "issue:43404", + "issue:43644", + "issue:43645", + "issue:43828", + "issue:43872", + "issue:43873", + "issue:43950", + "issue:43975", + "issue:44292", + "issue:44360", + "issue:44492", + "issue:44512", + "issue:44530", + "issue:44534", + "issue:44589", + "issue:44863", + "issue:44918", + "issue:44928", + "issue:44960", + "issue:45005", + "issue:45183", + "issue:45399", + "issue:45507", + "issue:45632", + "issue:45698", + "issue:45701", + "issue:45902", + "issue:45920", + "issue:45987", + "issue:46032", + "issue:46463", + "issue:46489", + "issue:46491", + "issue:46858", + "issue:46899", + "issue:46932", + "issue:47246", + "issue:47332", + "issue:47333" + ], + "soft_pairs": [ + "issue:43645|issue:45399", + "issue:40990|issue:44960", + "issue:42915|issue:45399", + "issue:45632|issue:45698", + "issue:43404|issue:45005", + "issue:44360|issue:44512", + "issue:43873|issue:45005", + "issue:35308|issue:36331", + "issue:44492|issue:44512", + "issue:44918|issue:45507", + "issue:44530|issue:45005", + "issue:43975|issue:44292", + "issue:45183|issue:46463", + "issue:44863|issue:45005", + "issue:35308|issue:42915", + "issue:39692|issue:43054", + "issue:46489|issue:46491", + "issue:43950|issue:44534", + "issue:36331|issue:41762", + "issue:43644|issue:43950", + "issue:33666|issue:35308", + "issue:46032|issue:47246", + "issue:43872|issue:44589", + "issue:46899|issue:46932", + "issue:43054|issue:45701", + "issue:45920|issue:46489", + "issue:38175|issue:43054", + "issue:47332|issue:47333", + "issue:43828|issue:44928", + "issue:45987|issue:46858", + "issue:43950|issue:45902" + ], + "prepared_review_unit_hash": "1b474c22f9554ad15ce5d0bc75f8922d90838d276f72de4f64cea11998cea507", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37123, + "estimated_input_tokens": 9281, + "estimated_eval_tokens": 18818 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is still a heterogeneous issue cluster. The strongest duplicate evidence is around the non-persistent-buffer regression in v5, while most other pairs are only thematically related. I removed the over-strong acceptance for the custom-module loading pair: they may be adjacent reports, but the packet does not prove a single underlying bug.", + "confidence": 0.81, + "canonical_issue_reason": "#43950 is the clearest representative regression report here: it is specific, actionable, and matches the strongest confirmed duplicate set around non-persistent buffer corruption in v5.", + "canonical_pr_reason": null, + "best_issue_reason": "#43950 is the best anchor for the cluster because it is concrete, broadly phrased, and directly covers one of the strongest same-bug pairs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Different problems: notebook/custom-model initialization vs flash-attn2 fallback gating." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "Different model/perplexity reports with no clear shared code path." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Qwen3Moe FP8 failure is unrelated to flash-attn fallback checks." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Related custom-module loading symptoms, but the packet only shows titles and does not establish the same bug or a mergeable single fix." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but they concern different model families and different failure modes." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "Unrelated: DSA indexer activation bug vs documentation cleanup." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to tied-weight translation-model issues." + }, + { + "left": "issue:35308", + "right": "issue:36331", + "accept": false, + "reason": "Different training bugs: iterable dataset length mismatch vs custom compute_loss signature breakage." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "A typo in cache strategies is unrelated to a removed CLI docs reference." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Different model/trainer issues with no shared failing code path." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied weights." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Different tokenizer/runtime failures on different models." + }, + { + "left": "issue:45183", + "right": "issue:46463", + "accept": false, + "reason": "Related area, but one is broad input_ids/inputs_embeds validation and the other is a T5-specific decoder-input error message." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading is unrelated to tied-weight translation-model behavior." + }, + { + "left": "issue:35308", + "right": "issue:42915", + "accept": false, + "reason": "Multi-GPU IterableDataset crash and FP8 model failure are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Both are SigLIP2-related, but one is docs/example mismatch and the other is an actual embedding quality issue." + }, + { + "left": "issue:46489", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different incorrect outputs." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same non-persistent-buffer regression in v5, just phrased as the broader symptom versus the specific from_pretrained corruption." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Different training API breakages: custom trainer loss signature vs ZeRO-3 model loading failure." + }, + { + "left": "issue:43644", + "right": "issue:43950", + "accept": true, + "reason": "Clear duplicate pair: both report v5 corrupting non-persistent buffers with junk values during loading/materialization." + }, + { + "left": "issue:33666", + "right": "issue:35308", + "accept": false, + "reason": "Both are training-related, but they describe different multi-GPU/data-loader issues." + }, + { + "left": "issue:46032", + "right": "issue:47246", + "accept": false, + "reason": "Related Mamba-family code, but not the same concrete bug: seq_len>1 cache corruption vs NemotronH slow-path recurrence propagation." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Different quantization/storage incompatibilities with different error signatures." + }, + { + "left": "issue:46899", + "right": "issue:46932", + "accept": true, + "reason": "Same Gemma4 quantized vision/audio embedder dtype-casting bug; #46932 is explicitly the fix for the same issue." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "Text-embedding quality regression and tokenization-version change are different bugs." + }, + { + "left": "issue:45920", + "right": "issue:46489", + "accept": false, + "reason": "Broad AutoTokenizer token-id regression across models is distinct from a DeepSeek-Coder v1 encode/decode issue." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Different SigLIP2 symptoms: zero probabilities vs degraded text embeddings." + }, + { + "left": "issue:47332", + "right": "issue:47333", + "accept": true, + "reason": "Same offloading-save failure: both report inability to save disk-offloaded Qwen3 models, differing only by model variant." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Different dtype-related failures on different models and workloads." + }, + { + "left": "issue:45987", + "right": "issue:46858", + "accept": true, + "reason": "Same cache contract bug: get_seq_length returns a tensor instead of a plain int, breaking downstream arithmetic." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Non-persistent buffer loading corruption is unrelated to uninitialized rotary-embedding memory after materialization." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Most of the summary is grounded, but one soft-edge acceptance is too strong: issue:45632 vs issue:45698 look related in custom-module loading, yet the packet only gives titles and does not establish that they are the same bug or a mergeable single fix. The rest of the accepted pairs are more defensible, but this one overstates the evidence, so I can\u2019t accept the cluster evaluation." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 5, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:24643", + "issue:28530", + "issue:41553", + "issue:41762", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43404", + "issue:43493", + "issue:43643", + "issue:43782", + "issue:43824", + "issue:43827", + "issue:43828", + "issue:43856", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44368", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44509", + "issue:44661", + "issue:44910", + "issue:44960", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45237", + "issue:45356", + "issue:45399", + "issue:45488", + "issue:45663", + "issue:45685", + "issue:45715", + "issue:45759", + "issue:45823", + "issue:45859", + "issue:45902", + "issue:46315", + "issue:46489", + "issue:46661", + "issue:46869" + ], + "soft_pairs": [ + "issue:43054|issue:43493", + "issue:44910|issue:45859", + "issue:45663|issue:45902", + "issue:44403|issue:44451", + "issue:42947|issue:44387", + "issue:24643|issue:28530", + "issue:43975|issue:45237", + "issue:44403|issue:44488", + "issue:41762|issue:42915", + "issue:43643|issue:45020", + "issue:44991|issue:45399", + "issue:42947|issue:43856", + "issue:45081|issue:45399", + "issue:43824|issue:43872", + "issue:45685|issue:45823", + "issue:41553|issue:43883", + "issue:45020|issue:45823", + "issue:43782|issue:44960", + "issue:44661|issue:45020", + "issue:43828|issue:45823", + "issue:45020|issue:45356", + "issue:41553|issue:43881", + "issue:42491|issue:44368", + "issue:43872|issue:46869", + "issue:45759|issue:46661", + "issue:45715|issue:46315", + "issue:45020|issue:46661", + "issue:45488|issue:46489", + "issue:43404|issue:43950", + "issue:43827|issue:44509", + "issue:43881|issue:43883" + ], + "prepared_review_unit_hash": "04c22ca897bb290180acee7727bb3137645cdfe6678ea1efdbf0bb2edd13c85a", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37489, + "estimated_input_tokens": 9373, + "estimated_eval_tokens": 19002 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are only loosely related and should stay separate. The only clear duplicate-style match is the v5 docs cleanup around removed pipeline tasks. No pull requests were present.", + "confidence": 0.83, + "canonical_issue_reason": "Issue 44509 is the clearest and most complete statement of the v5 pipeline-docs cleanup problem; 43827 is the same issue in a narrower form.", + "canonical_pr_reason": null, + "best_issue_reason": "44509 is the best representative issue for the duplicate docs problem because it explicitly covers the removed pipeline task references and has the broadest scope.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Same model family, but one is a quality complaint and the other is a JAX/HF discrepancy report; not enough to treat as the same concrete bug." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Both involve Qwen3.5/MoE-related runtime behavior, but the flash-attention position_ids crash and CPU offload rotary embedding issue are different code paths." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Gemma4 FSDP2 KeyError and Qwen3_5MoeRotaryEmbedding uninitialized-memory bug are unrelated failures." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic loading noise vs a specific model load failure; different symptoms and likely different fixes." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Both mention memory, but gradient checkpointing/LoRA inefficacy and int4 reserved-memory OOM are distinct regressions." + }, + { + "left": "issue:24643", + "right": "issue:28530", + "accept": false, + "reason": "DeepSpeed weight-dimension training error and early-stopping metric configuration are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45237", + "accept": false, + "reason": "Tokenizer detokenization bug and AMD GPU model support failure are different issues." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "These are both loading-related, but they concern different models and no shared underlying failure is evident." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Different model families and different load-time errors; no sign of the same bug." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "The remote-code regression theme is broad, but the specific AutoConfig missing-fields bug is not clearly the same concrete failure as the general report." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading failure and flash-attn fallback gating are unrelated." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Both mention training/memory, but one is about gradient checkpointing with PEFT LoRA and the other about Qwen3 MoE memory usage." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Mistral tokenizer regex patch crash and flash-attn fallback blocking are separate subsystems." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Qwen2_5_VL import failure and bitsandbytes Int8Params incompatibility are different bugs." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "MPS histogram-int error and Gemma4 device-mismatch forward error are unrelated." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Voxtral tokenizer loading error and Molmo tied-weights attribute error are different model-specific bugs." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "A broad remote-code regression does not match the specific Gemma4 device-map forward bug." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL weight_only loading error and generic GLM5 loading failure are not the same failure mode." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "add-new-model-like registry logic and remote-code loading regressions are different code paths." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch and Gemma4 device mismatch are separate runtime problems." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Remote-code breakage and Kimi-K2.5 tokenizer codec/fix_mistral_regex regression are unrelated." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Different models and different load errors; no duplicate evidence." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "LoRA model incompatibility across HF versions and a tie_word_embeddings warning during Qwen3.5 fine-tuning are not the same bug." + }, + { + "left": "issue:43872", + "right": "issue:46869", + "accept": false, + "reason": "bitsandbytes Int8Params incompatibility and YOLO multi-GPU device-map failure are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:46661", + "accept": false, + "reason": "Composite-config unwrapping for Qwen multimodal models and remote-code model_class config_class handling are different issues." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "Both are tokenizer performance regressions, but convert_ids_to_tokens and convert_tokens_to_ids have different hot paths and fixes." + }, + { + "left": "issue:45020", + "right": "issue:46661", + "accept": false, + "reason": "The general remote-code regression is too broad to equate with this specific config_class AttributeError without more evidence." + }, + { + "left": "issue:45488", + "right": "issue:46489", + "accept": false, + "reason": "Both concern tokenizer correctness, but they involve different model families and different tokenizer mechanisms." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Tied lm_head weights and non-persistent buffer corruption are separate from one another." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both report the same v5 docs problem: stale references to removed pipeline tasks/models in the summarization/translation docs." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "GLM-4v-9b loading failure and Molmo missing all_tied_weights_keys are different model-specific bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The evaluation is grounded in the packet. The only accepted duplicate-style pair, issue:43827 \u2194 issue:44509, is well supported by the titles and fits the v5 docs cleanup theme. The rejected soft-edge pairs are phrased conservatively and do not overclaim duplication or mergeability. No PR pairs were present." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 6, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42617", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43366", + "issue:43531", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43792", + "issue:43824", + "issue:43828", + "issue:43854", + "issue:43940", + "issue:43950", + "issue:43957", + "issue:44155", + "issue:44220", + "issue:44479", + "issue:44530", + "issue:44560", + "issue:44589", + "issue:44805", + "issue:44863", + "issue:44918", + "issue:44928", + "issue:44991", + "issue:45305", + "issue:45313", + "issue:45325", + "issue:45584", + "issue:45588", + "issue:45663", + "issue:45685", + "issue:45823", + "issue:45834", + "issue:45902", + "issue:45923", + "issue:46402", + "issue:46468", + "issue:46515", + "issue:46894", + "issue:46897" + ], + "soft_pairs": [ + "issue:43792|issue:44220", + "issue:36331|issue:41093", + "issue:43257|issue:45305", + "issue:46894|issue:46897", + "issue:42947|issue:45923", + "issue:43940|issue:45313", + "issue:43854|issue:44863", + "issue:24643|issue:30064", + "issue:44479|issue:44560", + "issue:41093|issue:44805", + "issue:43366|issue:43828", + "issue:44928|issue:45588", + "issue:43716|issue:46468", + "issue:43366|issue:43531", + "issue:44530|issue:44863", + "issue:43596|issue:46515", + "issue:44863|issue:46402", + "issue:41553|issue:44991", + "issue:45325|issue:45902", + "issue:43366|issue:44589", + "issue:43638|issue:46515", + "issue:43716|issue:44560", + "issue:45663|issue:45823", + "issue:42915|issue:43957", + "issue:44155|issue:45584", + "issue:43824|issue:43957", + "issue:42617|issue:43366", + "issue:43950|issue:45834", + "issue:44560|issue:45685", + "issue:44560|issue:44918" + ], + "prepared_review_unit_hash": "b3f11ead635fc4ea842b396ce0cab22a05caa78f9db6849bff229069670c163c", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36762, + "estimated_input_tokens": 9191, + "estimated_eval_tokens": 18638 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly heterogeneous and does not form one coherent duplicate cluster. The only clear duplicate-looking edge is the pair of training-loss double-shift reports; the rest are similar only at the subsystem or symptom level and should stay separate.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 46894 is the best representative because it states the underlying training-loss double-shift bug directly and is the earlier, clearer report; 46897 appears to be the same bug on another model.", + "canonical_pr_reason": null, + "best_issue_reason": "46894 is the strongest standalone issue in the set and the only solid anchor for deduplication within this mostly unrelated collection.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Both are Whisper/audio-related, but one is a generic run failure and the other is a specific fbank feature extraction problem; not the same concrete bug." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Different failures: Trainer signature mismatch versus a tensor mask shape IndexError." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Both mention Qwen3 and DeepSpeed, but one is about weight conversion/loading and the other is about gradient averaging; different bugs." + }, + { + "left": "issue:46894", + "right": "issue:46897", + "accept": true, + "reason": "Same training-loss double-shift bug pattern, likely fixed by one shared change." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing effectiveness and a missing supports_gradient_checkpointing flag are related only loosely; not the same bug." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both are ZeRO-3 loading issues on Qwen variants, but the models and failure modes differ enough that they do not look like one bug." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different model families and different loading failures; not a duplicate." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "Unrelated issues: a DeepSpeed training runtime error versus an image processor segmentation-map failure." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-related Qwen regressions, but one is a broad v5.3.0 regression and the other is a specific StopIteration path; not enough to merge." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "Same generic mask-shape error wording, but no evidence they hit the same code path or model." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "One is GGUF architecture support, the other is a dtype mismatch under autocast; unrelated." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Different code paths and symptoms: RLHF/SDPA fallback versus flash-attention handling of s_aux=None." + }, + { + "left": "issue:43716", + "right": "issue:46468", + "accept": false, + "reason": "Different models and different failures: image-preprocessor dtype mismatch versus a broken Nemotron example." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "GGUF support for gpt-oss and a sliding_window issue in Qwen3-MoE are unrelated." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache group-type crash and NemotronH checkpoint loading are separate problems." + }, + { + "left": "issue:43596", + "right": "issue:46515", + "accept": false, + "reason": "Zero-3/Bert index error versus a broad torch compatibility complaint; not the same bug." + }, + { + "left": "issue:44863", + "right": "issue:46402", + "accept": false, + "reason": "Different Nemotron/Sam3 loading problems with different symptoms and code paths." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/loading issues, but for different models and different error modes." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Different bugs: temporal position scaling in Qwen2.5-VL versus uninitialized inv_freq after meta-to-empty materialization." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "Architecture support for gpt-oss and a Float8 storage lookup error are unrelated." + }, + { + "left": "issue:43638", + "right": "issue:46515", + "accept": false, + "reason": "Different root causes and symptoms; no duplicate evidence." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Mistral-3 image dtype mismatch and Qwen3-vl video StopIteration are separate multimodal bugs." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "FSDP2 shared_kv_states logic and a PLE device-map mismatch are different forward-pass issues." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "FineGrainedFP8Config failure and meta-device loading breakage are different model-loading problems." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription handling are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Import exposure for a missing class versus model-loading breakage on meta device; not the same issue." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "A failed 3d_parallel.py run and GGUF gpt-oss support are unrelated." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption and Kosmos2.5 long-ocr index errors are different regressions." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3-vl video StopIteration and an MPS histogram limitation on Int are unrelated." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Video embedding StopIteration and TRL SFT unpacking failure are different code paths and bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The single accepted soft edge (46894/46897) is supported by both titles describing the same training-loss double-shift pattern, while the other pairs are reasonably left as non-duplicates based on differing models, symptoms, or code paths." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 7, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:39692", + "issue:41553", + "issue:41720", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43317", + "issue:43550", + "issue:43596", + "issue:43632", + "issue:43638", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43720", + "issue:43994", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44560", + "issue:44661", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44877", + "issue:44910", + "issue:44960", + "issue:44977", + "issue:45030", + "issue:45276", + "issue:45313", + "issue:45357", + "issue:45507", + "issue:45672", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:46468", + "issue:46489", + "issue:46515", + "issue:46846", + "issue:46858", + "issue:46899", + "issue:47254" + ], + "soft_pairs": [ + "issue:41553|issue:44488", + "issue:43645|issue:43646", + "issue:44683|issue:46515", + "issue:36010|issue:42915", + "issue:44910|issue:46858", + "issue:43550|issue:44910", + "issue:45823|issue:46515", + "issue:42947|issue:47254", + "issue:45812|issue:46489", + "issue:44661|issue:45823", + "issue:44560|issue:44805", + "issue:44877|issue:45030", + "issue:38175|issue:43994", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:44661|issue:44805", + "issue:42491|issue:43720", + "issue:45672|issue:46899", + "issue:43632|issue:45834", + "issue:43317|issue:44756", + "issue:41720|issue:42915", + "issue:38175|issue:39692", + "issue:42915|issue:44756", + "issue:43644|issue:44534", + "issue:42491|issue:45276", + "issue:44960|issue:44977", + "issue:44451|issue:44488", + "issue:45507|issue:46468", + "issue:44863|issue:46846", + "issue:45313|issue:45357" + ], + "prepared_review_unit_hash": "b4ab555a743f338c11d96f0bd4095eb1431176052bf2384ba7fb27992468732f", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35962, + "estimated_input_tokens": 8991, + "estimated_eval_tokens": 18238 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose set of Transformers v5 regressions with only a few true duplicate pairs. The clearest duplicates are the custom-model initialization break and the non-persistent buffer junk regression; the SigLIP2 checkpoint output reports also look like the same underlying model bug.", + "confidence": 0.64, + "canonical_issue_reason": "issue:43646 is the broadest and most representative issue among the true duplicate candidates: it describes the core custom model initialization regression, while issue:43645 is just the notebook-specific manifestation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best cluster anchor because it is the most general, clearly stated v5 regression and likely subsumes related custom-model initialization reports better than the narrower variants.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Both are load failures, but they concern different models and different reported failure modes; not enough evidence of the same code-path bug." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Same underlying custom model initialization regression in Transformers v5; one report is notebook-specific, the other is the general breakage." + }, + { + "left": "issue:44683", + "right": "issue:46515", + "accept": false, + "reason": "Different torch version compatibility problems in different contexts; no indication they are the same fixable bug." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "Completely different failures: an import error versus a Qwen3Moe FP8 loading problem." + }, + { + "left": "issue:44910", + "right": "issue:46858", + "accept": false, + "reason": "Both involve attention/cache internals, but the failure mechanisms are different: illegal memory access from misread position_ids versus slice arithmetic on a device tensor." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Different models, different attention implementations, and different crash modes; too broad to be the same bug." + }, + { + "left": "issue:45823", + "right": "issue:46515", + "accept": false, + "reason": "A Gemma4 device-mismatch issue is unrelated to a generic torch version incompatibility report." + }, + { + "left": "issue:42947", + "right": "issue:47254", + "accept": false, + "reason": "Both mention PEFT/gradient checkpointing, but one is a Transformers v5 regression and the other is a PEFT + DeepSpeed ZeRO-3 checkpoint error; not clearly the same issue." + }, + { + "left": "issue:45812", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions, but one is a silent Granite-wide token-id bug while the other is a DeepSeek-Coder tokenizer gap; different model families and likely different fixes." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping edge case and Gemma4 device placement failure are unrelated." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Both are runtime errors, but the model/task and failure mechanics are different; not the same underlying bug." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both mention config validation, but they affect different model families and are not clearly the same validation regression." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Same specific SigLIP2 checkpoint and same broken inference behavior; one reports zero probabilities, the other nonsensical outputs through AutoModel/pipeline." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "A ZeRO-3/BertModel index error is unrelated to the tokenizer-mapping failure in add-new-model-like." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Different subsystems and symptoms: DeepSpeed zero3 initialization versus tokenizer mapping logic." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer generation tooling and a tensor mask shape error are not the same bug." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Both involve loading/compatibility, but one is a qwen3_moe LoRA version mismatch and the other is a BitNet packed-weights accelerate-loading bug." + }, + { + "left": "issue:45672", + "right": "issue:46899", + "accept": false, + "reason": "Related Gemma4 multimodal quantization area, but one is specifically an audio uint8 torch.finfo TypeError and the other is unsafe casting in multimodal embedders." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "Both are v5 regressions, but the issues are unrelated: hf initialization flag breakage versus long OCR index errors." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Device-map dequantized loading and Strix Halo mmap OOM are different loading problems with different fixes." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "A cudaErrorAssert during auto device mapping is unrelated to a qwen3_moe FineGrainedFP8Config failure." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same model family, but one is a direct inference regression and the other is a documentation example with multiple distinct errors; not the same underlying bug." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Different models and different failure classes; no shared code-path evidence." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "These are the same Transformers v5 non-persistent buffer regression stated with nearly identical wording." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "LoRA version incompatibility and gemma4 resize_token_embeddings behavior are different issues." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different models and different flash-attention failures; similar wording is not enough to merge them." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both fail to load specific BERT checkpoints, but the titles do not show the same checkpoint family or the same root cause, so this is too weak to merge." + }, + { + "left": "issue:45507", + "right": "issue:46468", + "accept": false, + "reason": "Invalid method calls in GraniteMoEHybrid and a broken official Nemotron example are unrelated." + }, + { + "left": "issue:44863", + "right": "issue:46846", + "accept": false, + "reason": "NemotronH checkpoint loading and Qwen3.5 tensor-parallel plan issues are different bugs." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5 regressions, but one is ZeRO-3 weight loading and the other is incorrect visual encoder key saving; different code paths and fixes." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall and conservative on the soft edges. The anchor choice is reasonable, the duplicate calls are supported by near-identical titles, and the rejections avoid over-merging on weak similarity." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 8, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43644", + "issue:43645", + "issue:43708", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44075", + "issue:44079", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44387", + "issue:44402", + "issue:44488", + "issue:44756", + "issue:44910", + "issue:44960", + "issue:44991", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:45859", + "issue:46050", + "issue:46459", + "issue:46512", + "issue:46829", + "issue:46897", + "issue:46901" + ], + "soft_pairs": [ + "issue:46459|issue:46512", + "issue:44488|issue:44991", + "issue:39692|issue:43994", + "issue:43644|issue:43645", + "issue:43257|issue:43854", + "issue:43856|issue:43873", + "issue:42491|issue:45335", + "issue:42491|issue:45292", + "issue:43854|issue:43931", + "issue:43708|issue:44075", + "issue:43856|issue:44387", + "issue:44164|issue:44291", + "issue:45488|issue:45812", + "issue:44960|issue:46050", + "issue:45276|issue:45335", + "issue:45362|issue:45406", + "issue:43425|issue:43957", + "issue:44291|issue:44756", + "issue:44402|issue:44488", + "issue:41720|issue:44155", + "issue:45663|issue:45759", + "issue:39692|issue:44960", + "issue:44910|issue:45588", + "issue:46897|issue:46901", + "issue:43404|issue:43454", + "issue:43873|issue:44387", + "issue:42947|issue:45704", + "issue:45663|issue:45859", + "issue:44155|issue:45823", + "issue:44079|issue:46829", + "issue:43957|issue:44756" + ], + "prepared_review_unit_hash": "4dabee40f272c1dcac34090a500c70287f69dc79c40806d6e47890ac3c19ea58", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37282, + "estimated_input_tokens": 9321, + "estimated_eval_tokens": 18898 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is thematically related around Transformers 5.x regressions, but the items mostly describe distinct model-specific bugs and API issues rather than one duplicate chain. I did not find any soft-edge pair that is clearly the same concrete bug/change.", + "confidence": 0.67, + "canonical_issue_reason": "issue:46459 is the best single representative: it is a concrete, high-impact composite-model loading regression with a clear repro and broad downstream impact, whereas most other issues are narrower model-specific failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46459 is the strongest global candidate because it captures the cluster\u2019s main loading/dtype regression theme and has broader applicability than the more model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46459", + "right": "issue:46512", + "accept": false, + "reason": "Both involve dtype handling during model construction/loading, but one is about ignored load dtype on composite checkpoints and the other about stale config.dtype after from_config; related area, different bug." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/model load regressions on transformers>=5, but they affect different models and appear to have different failure modes." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both mention SigLIP2 bad outputs, but one is a docs/example mismatch plus quantization failure while the other is an AutoModel/pipeline result issue; not clearly the same code path." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "One is about non-persistent buffers being junk-filled, the other about custom-model initialization in Jupyter; same release, different bugs." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different models and symptoms: Qwen3 MoE conversion with accelerate/deepspeed versus GLM-4.7 load failures in unit tests." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "One is Qwen3 MoE training memory usage, the other is quantization/offloading behavior; broad subsystem overlap only." + }, + { + "left": "issue:42491", + "right": "issue:45335", + "accept": false, + "reason": "LoRA checkpoint compatibility on hf5.x versus resize_token_embeddings not updating decoder embeddings; unrelated concrete failures." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Different model-loading/compatibility issue versus a resize_token_embeddings propagation bug." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "GLM-4.7 unit-test load failure and Qwen3-VL shape mismatch are separate model-specific loading bugs." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Trainer checkpoint resumption math and optimizer argument handling are unrelated." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Training memory efficiency and int4 quantization reserved-memory OOM are different resource-management bugs." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "save/from_pretrained extra_state handling and init_empty_weights argument compatibility are distinct serialization/loading regressions." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Different tokenizer regressions affecting different model families; no shared concrete code path is evident." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "GLM5 loading and sinq save/load breakage are unrelated issues." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": false, + "reason": "Both are resize_token_embeddings propagation bugs, but they target different architectures and different nested embedding fields." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Qwen3.5 chat crash and Gemma4Processor missing _tokenizer are different runtime failures." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Torch 2.10 compatibility and meta-device loading failures are separate regressions." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "Unexpected init_empty_weights argument and mmap-induced OOM are unrelated." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Different tokenizer/vocab mismatch reports for different models; not the same underlying bug." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "CUDA assert on Qwen3 auto device mapping and AudioFlamingo3 batched inference leak are unrelated." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Gemma4 FSDP2 attention state rebuild and Qwen 3.5/3.6 text_config unwrapping are different model-loading/training issues." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "SigLIP2 docs/example failure and GLM5 loading are different models and different bug classes." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Flash-attention illegal memory access from packed-sequence misinterpretation versus AttributeError on s_aux=None are different failures in different code paths." + }, + { + "left": "issue:46897", + "right": "issue:46901", + "accept": false, + "reason": "Both are double-shift training-loss bugs, but they affect different model families and there is no evidence of a shared implementation bug." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both mention untied lm_head weights, but they are separate multimodal model classes with no clear shared code path from the titles alone." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Quantization/offloading behavior and int4 reserved-memory OOM are related only at a high level, not the same bug." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing ineffectiveness in PEFT LoRA and a T5 RMSNorm memory leak are unrelated." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different Gemma4/FSDP2 state bug versus Qwen3_5Moe rotary embedding CPU-offload incompatibility." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo3 batched inference leak and Gemma4 PLE device mismatch are distinct runtime issues." + }, + { + "left": "issue:44079", + "right": "issue:46829", + "accept": false, + "reason": "ModelOutput key assignment bug and VideoPrism hidden_states mispopulation are different output-structure problems." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading failures and Strix Halo mmap OOM are different loading/memory issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 9, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:42796", + "issue:43335", + "issue:43381", + "issue:43425", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43854", + "issue:43957", + "issue:44164", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44530", + "issue:44589", + "issue:44683", + "issue:44756", + "issue:44863", + "issue:44877", + "issue:44912", + "issue:44928", + "issue:45084", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45812", + "issue:45823", + "issue:46315", + "issue:46326", + "issue:46468", + "issue:46491", + "issue:46869", + "issue:46894", + "issue:46901", + "issue:47228", + "issue:47332" + ], + "soft_pairs": [ + "issue:43335|issue:45663", + "issue:44863|issue:46468", + "issue:44479|issue:44877", + "issue:36010|issue:39404", + "issue:44387|issue:44756", + "issue:45276|issue:45292", + "issue:44292|issue:44912", + "issue:43381|issue:44928", + "issue:43957|issue:47332", + "issue:44164|issue:44756", + "issue:42796|issue:46326", + "issue:45588|issue:45663", + "issue:43645|issue:44164", + "issue:43425|issue:47228", + "issue:43854|issue:45216", + "issue:45684|issue:45698", + "issue:45292|issue:45335", + "issue:44589|issue:44756", + "issue:43425|issue:43828", + "issue:44402|issue:44451", + "issue:43646|issue:43708", + "issue:43425|issue:46869", + "issue:46894|issue:46901", + "issue:44683|issue:45823", + "issue:43425|issue:43716", + "issue:45812|issue:46315", + "issue:46315|issue:46491", + "issue:44530|issue:45084", + "issue:39404|issue:45701" + ], + "prepared_review_unit_hash": "7c49db93f71302c5f6affd5b78db9f1c7773e1034ac35524bcb4e88e5580d069", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36280, + "estimated_input_tokens": 9070, + "estimated_eval_tokens": 18396 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are unrelated despite superficial overlap. The only clear duplicate families here are the `resize_token_embeddings` regression reports and the training-loss double-shift reports; the `resize_token_embeddings` issue is the best canonical anchor.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45292 is the cleanest general statement of the shared `resize_token_embeddings` bug, and it matches the more model-specific reports in issue:45276 and issue:45335.", + "canonical_pr_reason": null, + "best_issue_reason": "It is the most model-agnostic, reusable duplicate target among the issues shown, with clear manifestations across multiple model variants.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Different bugs: sparse-layer construction vs Gemma4 FSDP2 shared_kv_states handling." + }, + { + "left": "issue:44863", + "right": "issue:46468", + "accept": false, + "reason": "Both are Nemotron-related, but one is checkpoint loading and the other is an example breakage; not the same concrete bug." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Different regressions in different model families and code paths." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to Whisper pipeline `return_language` behavior." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Both mention OOM, but the causes differ: int4 quantization memory growth vs mmap behavior on Strix Halo." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same `resize_token_embeddings` propagation bug, with 45276 being a model-specific manifestation." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Different quantization/loading problems with different symptoms and likely different fixes." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Gradient checkpointing eval-mode issue is unrelated to RLHF NaN from 3D position_ids / SDPA fallback." + }, + { + "left": "issue:43957", + "right": "issue:47332", + "accept": false, + "reason": "Different model-loading/offloading failures on different models and paths." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to mmap-triggered OOM." + }, + { + "left": "issue:42796", + "right": "issue:46326", + "accept": false, + "reason": "Tokenizer BPE marker handling and `continue_final_message` are distinct tokenizer behaviors." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Flash-attention `s_aux=None` crash is a different issue from Gemma4 FSDP2 `shared_kv_states` rebuilding." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Custom-model notebook init regression is unrelated to `extra_state` save/from_pretrained handling." + }, + { + "left": "issue:43425", + "right": "issue:47228", + "accept": false, + "reason": "Torch compatibility breakage and Sam3 bf16 compilation regression are different code paths." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Unit-test model loading failure and Qwen3.5 save_pretrained regression are different problems." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Permissions propagation from source files is not the same as loading the wrong custom module after save_pretrained." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Both report the same `resize_token_embeddings` bug affecting model embeddings/output embeddings." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Storage-object lookup failure is unrelated to mmap/OOM behavior." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version incompatibility is not the same as the autocast dtype-mismatch bug." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer vocab mismatch and model load failure are different issues." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization regression is unrelated to Trainer `max_steps` miscalculation on checkpoint resume." + }, + { + "left": "issue:43425", + "right": "issue:46869", + "accept": false, + "reason": "Torch incompatibility and YOLO multi-GPU device-map failure are different concrete bugs." + }, + { + "left": "issue:46894", + "right": "issue:46901", + "accept": true, + "reason": "Same training-loss double-shift pattern: both describe labels being shifted incorrectly during loss computation." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex_attention regression is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch version incompatibility is different from the image-preprocessor/model dtype mismatch." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Wrong token IDs and slower `convert_tokens_to_ids` are separate tokenizer regressions." + }, + { + "left": "issue:46315", + "right": "issue:46491", + "accept": false, + "reason": "Performance regression and CodeLlama round-trip space loss are unrelated tokenizer issues." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttentionCache group-type crash is unrelated to non-template-node compilation failure." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline regression is unrelated to tokenizer-version-dependent tokenization changes." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The canonical `resize_token_embeddings` anchor at issue:45292 is well-supported by the closely related model-specific reports in issue:45276 and issue:45335, and the training-loss double-shift pair is also a reasonable duplicate callout. The unrelated pairs are correctly rejected rather than over-merged." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 10, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36032", + "issue:39404", + "issue:40990", + "issue:42796", + "issue:42915", + "issue:43404", + "issue:43452", + "issue:43596", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43746", + "issue:43856", + "issue:43901", + "issue:43906", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "issue:44479", + "issue:44488", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44671", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44898", + "issue:44912", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45479", + "issue:45759", + "issue:45823", + "issue:46050", + "issue:46326", + "issue:46665", + "issue:46821", + "issue:46894", + "issue:46899", + "issue:47050", + "issue:47331", + "issue:47333" + ], + "soft_pairs": [ + "issue:43611|issue:43645", + "issue:44479|issue:44805", + "issue:44589|issue:45084", + "issue:39404|issue:46665", + "issue:44164|issue:45759", + "issue:36032|issue:43452", + "issue:43994|issue:44079", + "issue:40990|issue:47331", + "issue:44265|issue:44898", + "issue:43901|issue:44509", + "issue:44991|issue:45081", + "issue:46050|issue:46326", + "issue:43906|issue:46326", + "issue:44671|issue:44991", + "issue:43906|issue:44530", + "issue:39404|issue:44912", + "issue:45823|issue:46899", + "issue:43596|issue:44683", + "issue:44164|issue:46821", + "issue:44756|issue:47333", + "issue:44488|issue:44987", + "issue:43856|issue:44756", + "issue:44488|issue:45081", + "issue:36032|issue:42915", + "issue:43632|issue:47050", + "issue:43716|issue:45084", + "issue:43746|issue:47331", + "issue:42796|issue:43404", + "issue:45479|issue:46894", + "issue:43611|issue:43644" + ], + "prepared_review_unit_hash": "9b55eded07e48c47e6e78c96350750869e3c36bb2d3ca4c9fdab9a8d3966fdf8", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37016, + "estimated_input_tokens": 9254, + "estimated_eval_tokens": 18764 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous. The only clear thematic thread is a loose set of Transformers v5 loading/initialization regressions, but none of the soft pairs look like the same concrete bug or change, so all soft edges should be rejected.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43611 is the broadest and most central of the v5 regression reports here: it targets model loading breakage in Transformers 5.0.0 and best represents the shared initialization/loading theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best issue to anchor the cluster because it is a general, high-signal v5 model-loading regression; the other candidates are either narrower model-specific failures, tokenizer bugs, docs issues, or unrelated runtime errors.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both are v5 breakage reports, but one is about base_model_prefix loading and the other about custom model initialization in Jupyter; different failure modes and likely different fixes." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression on Qwen VL models is unrelated to a generic mask-shape IndexError; no evidence of the same code path." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "One is a float8 storage lookup error, the other is a template-node compilation error; different subsystems and root causes." + }, + { + "left": "issue:39404", + "right": "issue:46665", + "accept": false, + "reason": "Whisper pipeline return_language behavior and from_pretrained-inside-classes inconsistency are separate issues." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state save/load handling and composite Qwen text_config unwrapping are different loading problems." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5 tokenizer add_special_tokens conflict and gguf_file handling for AutoTokenizer/AutoModel are distinct loader failures." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Incorrect SigLIP2 outputs and ModelOutput key assignment bugs are different behaviors with different likely code paths." + }, + { + "left": "issue:40990", + "right": "issue:47331", + "accept": false, + "reason": "High perplexity on gpt-oss-20b and Metal quantization batch corruption are unrelated problems." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export failures from torch_compilable_check are not the same as Perceiver non-default-resolution interpolation failures." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs issues, but they concern different pipeline/task documentation changes and not the same underlying bug." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer-loading failures on different models; not the same concrete bug." + }, + { + "left": "issue:46050", + "right": "issue:46326", + "accept": false, + "reason": "Quantized save/load breakage via sinq is unrelated to continue_final_message on mistral_common tokenizers." + }, + { + "left": "issue:43906", + "right": "issue:46326", + "accept": false, + "reason": "An isolated reproduction of another issue and a tokenizer-specific continue_final_message bug are not the same bug." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT masked-LM prediction regression and est-roberta tokenizer loading failure are different issues." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "Generic reproduction report vs Qwen3.5 PagedAttentionCache crash; different code paths." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline return_language and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46899", + "accept": false, + "reason": "Both mention Gemma4, but one is a device-map forward mismatch and the other is a 4-bit dtype-cast bug in multimodal embedders; different fixes." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Deepspeed ZeRO-3/BertModel initialization and compiled flex_attention on torch>=2.9 are unrelated." + }, + { + "left": "issue:44164", + "right": "issue:46821", + "accept": false, + "reason": "extra_state save/load handling and TimesFM window_size AttributeError are different bugs." + }, + { + "left": "issue:44756", + "right": "issue:47333", + "accept": false, + "reason": "Strix Halo mmap OOM and disk-offloaded Qwen3-30B-A3B save failure are distinct storage/offloading problems." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Two different model-loading failures; same broad area but not the same underlying bug." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "MoE training memory inefficiency and Strix Halo mmap OOM are separate memory issues with different causes." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer-loading failures in different model families; no shared concrete bug." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "T5 tokenizer init conflict and Qwen3Moe fine-grained FP8 failure are unrelated." + }, + { + "left": "issue:43632", + "right": "issue:47050", + "accept": false, + "reason": "_is_hf_initialized flag breakage and Canine fp16 ONNX attention dtype mismatch are different problems." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Mistral-3 image-preprocessor dtype mismatch and template-node compilation failure are unrelated." + }, + { + "left": "issue:43746", + "right": "issue:47331", + "accept": false, + "reason": "PEFT local checkpoint loading for GraniteSpeech and Metal quantization batch corruption are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43404", + "accept": false, + "reason": "Tokenizer BPE-marker handling and lm_head weight tying in Mistral3ForConditionalGeneration are different bugs, even if both involve Mistral-family models." + }, + { + "left": "issue:45479", + "right": "issue:46894", + "accept": false, + "reason": "Zero-loss classification bug and ASR double-shift training-loss bug are different loss-shaping issues." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but one is about loading with base_model_prefix and the other about non-persistent buffer initialization; not the same defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 11, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42915", + "issue:43299", + "issue:43421", + "issue:43452", + "issue:43482", + "issue:43541", + "issue:43582", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43746", + "issue:43873", + "issue:43931", + "issue:44206", + "issue:44220", + "issue:44403", + "issue:44451", + "issue:44462", + "issue:44530", + "issue:44661", + "issue:44671", + "issue:44683", + "issue:44737", + "issue:44811", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45479", + "issue:45507", + "issue:45584", + "issue:45859", + "issue:46178", + "issue:46285", + "issue:46326", + "issue:46402", + "issue:46489", + "issue:46897", + "issue:47312", + "issue:47331" + ], + "soft_pairs": [ + "issue:44462|issue:46489", + "issue:44661|issue:44683", + "issue:43299|issue:43931", + "issue:46285|issue:46402", + "issue:43541|issue:45084", + "issue:45479|issue:46897", + "issue:44451|issue:44987", + "issue:42915|issue:44530", + "issue:43873|issue:47331", + "issue:40990|issue:43421", + "issue:43643|issue:44530", + "issue:36032|issue:43645", + "issue:40444|issue:46326", + "issue:34567|issue:40444", + "issue:44737|issue:45859", + "issue:39404|issue:44671", + "issue:43644|issue:43646", + "issue:42175|issue:46178", + "issue:44987|issue:44991", + "issue:44206|issue:44220", + "issue:45084|issue:45507", + "issue:43582|issue:43632", + "issue:43452|issue:43482", + "issue:43482|issue:47312", + "issue:43746|issue:46326", + "issue:44811|issue:45584", + "issue:43582|issue:47331", + "issue:44403|issue:45081" + ], + "prepared_review_unit_hash": "3b379ffe87d35bbd0c62fc83ef513cd8fdabe2f9f01f34609e614f36bc15e36b", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36123, + "estimated_input_tokens": 9031, + "estimated_eval_tokens": 18318 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a duplicate set in the strict sense; it mixes many separate Transformers regressions across tokenizers, model loading, quantization, audio, and training behavior. I found no soft edge that clearly represents the same underlying bug/change.", + "confidence": 0.88, + "canonical_issue_reason": "No single canonical issue: the items cover many distinct failures and code paths (tokenizer loading, GGUF, model init, quantization, Whisper/audio, trainer state, etc.), so collapsing them would over-merge unrelated bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must be used as a representative anchor, `issue:43646` is the broadest user-facing Transformers v5 custom-model initialization regression, but it is still not a true canonical duplicate for the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44462", + "right": "issue:46489", + "accept": false, + "reason": "Both involve tokenizer behavior, but one is a repo/tokenizer.json loading regression and the other is a model-specific DeepSeek-Coder tokenizer output issue; not the same concrete bug." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Unrelated subsystems: add-new-model-like mapping failure vs compiled flex_attention on torch>=2.9." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Same general Qwen3-VL area, but one is a v5 dev loading regression and the other is a specific weight-shape mismatch for a concrete model; not clearly one bug." + }, + { + "left": "issue:46285", + "right": "issue:46402", + "accept": false, + "reason": "Different model families and failures: CLIP text-model structure inconsistency vs Sam3Model text-encoder weights not loading." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Different failures and code paths: torch dynamo grouped_mm tracing vs template compilation error." + }, + { + "left": "issue:45479", + "right": "issue:46897", + "accept": false, + "reason": "Both are loss-related, but they are distinct model-specific training bugs with different mechanisms." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are model loading regressions, but for different models and likely different causes." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Both touch Qwen/MoE-ish areas, but one is FineGrainedFP8Config failure and the other is PagedAttentionCache linear_attention crash; not the same issue." + }, + { + "left": "issue:43873", + "right": "issue:47331", + "accept": false, + "reason": "Both mention quantization/offload, but one is offloading behavior and the other is a Metal batched-generate corruption bug in a different code path." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Unrelated: perplexity regression on gpt-oss vs runtime special-token/post-processor update behavior." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "Different concerns: trust_remote_code missing fields vs cache crash on Qwen3.5 models." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer method-name conflict is unrelated to custom model initialization in notebooks." + }, + { + "left": "issue:40444", + "right": "issue:46326", + "accept": false, + "reason": "Different areas: multi-image IterableDataset fine-tuning failure vs mistral_common continue_final_message behavior." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counting bug is unrelated to Qwen2.5-VL multi-image IterableDataset failure." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both are device-related, but one is XLNet positional encoding CPU placement and the other is Qwen3.5 MoE CPU offload incompatibility." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline return_language regression is unrelated to CamemBERT masked-LM prediction changes." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 initialization regressions, but one is non-persistent buffers being filled incorrectly and the other is custom model initialization; not the same concrete bug." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "Both mention backend plumbing, but one is packaging/install extras and the other is a BACKENDS_MAPPING validation error for tensorflow_text." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/model loading failures, but for different models and likely different regressions." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "Both are audio feature-extraction issues, but one is an unsupported `center` arg and the other is `_torch_extract_fbank_features()`; not clearly the same bug." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Completely different failures: template compilation vs GraniteMoEHybrid invalid method call." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError is unrelated to the `_is_hf_initialized` flag regression." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both involve GGUF loading, but one is a general `gguf_file`/from_pretrained break and the other is a model-specific Qwen2.5-GGUF failure; too little evidence that they are the same bug." + }, + { + "left": "issue:43746", + "right": "issue:46326", + "accept": false, + "reason": "PEFT-adapter local checkpoint loading and mistral_common continue_final_message are separate problems." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both are Whisper-related, but one is `batch_decode` ignoring `skip_special_tokens` and the other is empty-transcription generation failure after `align_special_tokens`; distinct symptoms and likely distinct fixes." + }, + { + "left": "issue:43582", + "right": "issue:47331", + "accept": false, + "reason": "Apple Silicon warmup TypeError is unrelated to Metal quantization batched-generate corruption." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "General loading noise vs Mistral regex patch crash are different issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 12, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28530", + "issue:34567", + "issue:36032", + "issue:41720", + "issue:42915", + "issue:43208", + "issue:43262", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43425", + "issue:43482", + "issue:43531", + "issue:43540", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43792", + "issue:43819", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44849", + "issue:44877", + "issue:44987", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45923", + "issue:45987", + "issue:46661", + "issue:46869", + "issue:46899", + "issue:47013", + "issue:47254", + "issue:47331" + ], + "soft_pairs": [ + "issue:43262|issue:47331", + "issue:43645|issue:46661", + "issue:43208|issue:47013", + "issue:43792|issue:46869", + "issue:43317|issue:43482", + "issue:42915|issue:45084", + "issue:44987|issue:45081", + "issue:43646|issue:44534", + "issue:42915|issue:44291", + "issue:43262|issue:43377", + "issue:28530|issue:34567", + "issue:36032|issue:43742", + "issue:44062|issue:44987", + "issue:43377|issue:43819", + "issue:44451|issue:45081", + "issue:43404|issue:44466", + "issue:43994|issue:44610", + "issue:44849|issue:45325", + "issue:42915|issue:44589", + "issue:36032|issue:41720", + "issue:43531|issue:44877", + "issue:44265|issue:44610", + "issue:44008|issue:46899", + "issue:44479|issue:45325", + "issue:43322|issue:43540", + "issue:43425|issue:44403", + "issue:45923|issue:47254", + "issue:44464|issue:45987" + ], + "prepared_review_unit_hash": "864bc2afc3ae159e6615b72783ff24c1d50e69249fa06d77f371331b2ca3e853", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35814, + "estimated_input_tokens": 8954, + "estimated_eval_tokens": 18164 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose semantic grouping of many distinct Transformers bugs, mostly around model loading, tokenizers, multimodal preprocessing, quantization, and trainer/runtime regressions. The soft pairs are only superficially related; none look like the same underlying bug or a plausible merged PR.", + "confidence": 0.96, + "canonical_issue_reason": "No true canonical issue: the items span many different code paths and failure modes, so there isn't a single duplicate to anchor the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must represent the cluster, issue:43646 is the broadest umbrella because it covers a generic Transformers 5.0 custom-model initialization regression and is closest to several nearby v5 loading breakage reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43262", + "right": "issue:47331", + "accept": false, + "reason": "Audio chat-template sampling-rate bug vs Metal quantization batch corruption; unrelated components and fixes." + }, + { + "left": "issue:43645", + "right": "issue:46661", + "accept": false, + "reason": "Both involve custom/remote-code models, but one is Jupyter initialization and the other is missing config_class on AutoModel; different bugs." + }, + { + "left": "issue:43208", + "right": "issue:47013", + "accept": false, + "reason": "Both are xLSTM, but one reports multiple training bugs for smaller models while the other is a specific return_last_states unpacking crash." + }, + { + "left": "issue:43792", + "right": "issue:46869", + "accept": false, + "reason": "Both are load failures, but Whisper runtime failure and YOLO multi-GPU device-map failure are different code paths." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "GPU+CPU offload dequantized model loading vs Qwen2.5-GGUF v5 loading; different model families and failure modes." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3Moe fine-grained FP8 failure vs template compilation error; not the same underlying bug." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Model loading failure for physical-intelligence/fast vs Mistral tokenizer regex patch crash; separate tokenizer/model paths." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Custom model initialization regression vs non-persistent buffer serialization junk; unrelated defects." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3Moe FP8 config issue vs init_empty_weights unexpected argument; different initialization/quantization problems." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio sampling-rate default bug vs MIMI batch/single output mismatch from padding mask support; different modalities and logic." + }, + { + "left": "issue:28530", + "right": "issue:34567", + "accept": false, + "reason": "Early-stopping metric lookup bug vs TrainerState token counter not updating; unrelated trainer features." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "T5 tokenizer method-name conflict vs MobileLLM key error; different tokenizer/model loading issues." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken special-argument conflict vs a model loading failure; unrelated tokenizer internals." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "MIMI padding-mask batch discrepancy vs DAC from_latents/forward mismatch due to missing STE; different models and bugs." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "ScandiBERT load failure vs Mistral regex patch crash; not the same bug." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both concern tied lm_head weights, but one is a missing tie in Mistral3 and the other is device-dependent serialization; related theme, different defect." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Siglip2 nonsense outputs with AutoModel/pipeline vs OmDet processor image-size mismatch; unrelated vision bugs." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Qwen3.5 hidden-states bug vs Qwen2.5-VL rope-index scaling bug; different model versions and code paths." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3Moe FP8 config failure vs missing Float8 storage type; different low-level quantization/storage problems." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer attribute collision vs Qwen3 auto device-map CUDA assert; unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Sliding-window issue in Qwen3-MoE vs strict config blocking granite_speech; different subsystems." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export / torch_compilable_check incompatibility vs OmDet processor size mismatch; separate issues." + }, + { + "left": "issue:44008", + "right": "issue:46899", + "accept": false, + "reason": "Gemma3n forward variable collision vs Gemma4 multimodal quantization dtype cast; different model generations and bugs." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Video-input regression in Qwen multimodal models vs Qwen2.5-VL rope scaling for still images; not the same defect." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Llava Next segfault on load vs Qwen3OmniMoe video-processing ValueError; different models and failure points." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch version incompatibility vs noisy loading output; unrelated." + }, + { + "left": "issue:45923", + "right": "issue:47254", + "accept": false, + "reason": "Missing supports_gradient_checkpointing flag vs PEFT+DeepSpeed ZeRO-3 checkpoint error; both mention checkpointing but are different bugs." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Chunked generation inconsistency with compiled forward vs StaticCache seq-length return type bug; separate generation/cache paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 13, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:43089", + "issue:43296", + "issue:43381", + "issue:43421", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43611", + "issue:43632", + "issue:43716", + "issue:43746", + "issue:43792", + "issue:43872", + "issue:43883", + "issue:44079", + "issue:44220", + "issue:44291", + "issue:44387", + "issue:44403", + "issue:44488", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44683", + "issue:44756", + "issue:44811", + "issue:44991", + "issue:45081", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45663", + "issue:45812", + "issue:45923", + "issue:45987", + "issue:46315", + "issue:46491", + "issue:46710", + "issue:46821", + "issue:46899", + "issue:47404" + ], + "soft_pairs": [ + "issue:43632|issue:46821", + "issue:43089|issue:45987", + "issue:43296|issue:43531", + "issue:44220|issue:44683", + "issue:40444|issue:45325", + "issue:43872|issue:44291", + "issue:43883|issue:44488", + "issue:43611|issue:44534", + "issue:44991|issue:46315", + "issue:43716|issue:46899", + "issue:44403|issue:44991", + "issue:45216|issue:45357", + "issue:42796|issue:43746", + "issue:44291|issue:44589", + "issue:45812|issue:46491", + "issue:43381|issue:45923", + "issue:43792|issue:44589", + "issue:33666|issue:40444", + "issue:40990|issue:44811", + "issue:43632|issue:44756", + "issue:42796|issue:46710", + "issue:43531|issue:44291", + "issue:35707|issue:40444", + "issue:44079|issue:45663", + "issue:41720|issue:45081", + "issue:43883|issue:44991", + "issue:43482|issue:43525", + "issue:43421|issue:44568", + "issue:43296|issue:46821", + "issue:42175|issue:43296", + "issue:44387|issue:47404" + ], + "prepared_review_unit_hash": "d1884092157880ff10de9b7333483290619be529e5b73bbea7d9bade3aaf9b08", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37211, + "estimated_input_tokens": 9303, + "estimated_eval_tokens": 18862 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated issue reports, with only a couple of close duplicate-like pairs. The strongest duplicate signal is the `_is_hf_initialized` regression, and a second likely duplicate pair is the Qwen3.5 `save_pretrained` checkpoint-key regression.", + "confidence": 0.84, + "canonical_issue_reason": "issue:44291 is the broadest and most root-cause-oriented report in the `_is_hf_initialized` regression family, so it is the best canonical issue for that duplicate pair.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44291 best represents the shared underlying bug around the new `_is_hf_initialized` argument and model loading failures in Transformers v5.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43632", + "right": "issue:46821", + "accept": false, + "reason": "Both are v5-era regressions, but they describe different models and different failures (`_is_hf_initialized` vs `window_size` attribute access)." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation performance overhead and `StaticCache.get_seq_length()` contract mismatch are unrelated bugs." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "Different model families and different failure modes; no shared concrete code path." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction and compiled flex attention on torch 2.9 are separate subsystems and issues." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Both involve Qwen2.5-VL, but one is training with multiple images and the other is still-image position ID scaling; not the same bug." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Both report the same underlying `_is_hf_initialized` regression introduced by Transformers v5, just surfacing in different loading paths." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Missing `all_tied_weights_keys` and inability to load `cjvt/sleng-bert` are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "`base_model_prefix` loading failure and non-persistent buffer junk are different regressions." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Tokenizer load failure for a specific model and slow `convert_tokens_to_ids` are distinct issues." + }, + { + "left": "issue:43716", + "right": "issue:46899", + "accept": false, + "reason": "Different multimodal dtype/quantization bugs in different models; not one change." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Loading noise and tokenizer load failure do not share the same concrete bug." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Both describe the same Qwen3.5 `save_pretrained` checkpoint serialization regression, with 45357 looking like a later-specific manifestation." + }, + { + "left": "issue:42796", + "right": "issue:43746", + "accept": false, + "reason": "Tokenizer BPE-marker handling and GraniteSpeech PEFT local-checkpoint loading are unrelated." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected `_is_hf_initialized` and missing Float8 storage are different runtime failures." + }, + { + "left": "issue:45812", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different tokenization behaviors." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing in eval mode and a missing `supports_gradient_checkpointing` flag are related only at a high level, not the same bug." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure and Float8 storage lookup failure are separate problems." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Multi-GPU training and multi-image IterableDataset finetuning are different training issues." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "High perplexity and Whisper `batch_decode` special-token handling are unrelated." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "`_is_hf_initialized` regression and mmap/OOM behavior on Strix Halo are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer marker handling for Ministral and DeepSeek accuracy regression are different bugs." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3-MoE sliding window behavior and `_is_hf_initialized` loading errors are unrelated." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with embeds/past KV is not the same as multimodal IterableDataset training failure." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "`ModelOutput` key assignment and Gemma4 FSDP2 shared_kv_states rebuilding are different code paths." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Auto device mapping CUDA assert and Mistral regex patch crash are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Missing model attribute and tokenizer loading failure do not match." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Qwen2.5-GGUF loading failure and missing `pad_token_id` on Llama4Config are different incompatibilities." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Runtime special-token post-processor updates and `add_special_tokens=True` not adding BOS/EOS are related to tokenization but not the same bug." + }, + { + "left": "issue:43296", + "right": "issue:46821", + "accept": false, + "reason": "Different models and different attribute failures; no shared underlying issue." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "Packaging/backends omission and model loading failure are unrelated." + }, + { + "left": "issue:44387", + "right": "issue:47404", + "accept": false, + "reason": "Both are memory-related, but one is int4 quantization OOM and the other is device_map loading transients; not the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative on the soft-edge calls. The two accepted duplicate-like pairs are plausible from the titles alone, and the rejected pairs are generally described with appropriate caution rather than overstated as identical bugs." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 14, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42796", + "issue:43262", + "issue:43317", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43606", + "issue:43611", + "issue:43644", + "issue:43646", + "issue:43701", + "issue:43746", + "issue:43818", + "issue:43828", + "issue:43856", + "issue:43873", + "issue:43994", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44462", + "issue:44466", + "issue:44479", + "issue:44509", + "issue:44534", + "issue:44829", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45542", + "issue:45584", + "issue:45701", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:46402", + "issue:46821", + "issue:46909", + "issue:47217" + ], + "soft_pairs": [ + "issue:42175|issue:45542", + "issue:44462|issue:45812", + "issue:44079|issue:44534", + "issue:43482|issue:45538", + "issue:43611|issue:43646", + "issue:43317|issue:43856", + "issue:44265|issue:45072", + "issue:44509|issue:46909", + "issue:45325|issue:45834", + "issue:40444|issue:45381", + "issue:43377|issue:45491", + "issue:43644|issue:44079", + "issue:44291|issue:44877", + "issue:45325|issue:45381", + "issue:43994|issue:44898", + "issue:40444|issue:47217", + "issue:43454|issue:44466", + "issue:43818|issue:46402", + "issue:43818|issue:45357", + "issue:43873|issue:45701", + "issue:43701|issue:45325", + "issue:44829|issue:45479", + "issue:43606|issue:44265", + "issue:42796|issue:45381", + "issue:44479|issue:45823", + "issue:43828|issue:46821", + "issue:40990|issue:45584", + "issue:43262|issue:43746" + ], + "prepared_review_unit_hash": "ccdb9210bb0f05a707e5cfb5e59e25a2b8b79a2421c44692961f84f13fd6fb93", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36356, + "estimated_input_tokens": 9089, + "estimated_eval_tokens": 18434 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated, model-specific regression reports rather than a single duplicate cluster. Under a strict same-code-path standard, none of the soft candidate pairs should be merged. The strongest representative issue is the broad offloading/quantization regression (#43873), which is active and well-referenced.", + "confidence": 0.84, + "canonical_issue_reason": "#43873 is the broadest and most central issue here: it has the highest discussion/inbound activity and describes a general offloading+quantization failure rather than a narrow model-specific edge case.", + "canonical_pr_reason": null, + "best_issue_reason": "#43873 is the best anchor for this set because it is the most general, highest-activity issue and likely to attract related offloading regressions if they truly converge.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both involve TensorFlow backend packaging, but one is about `.[torch]` install extras omitting TF and the other is about tensorboard being present without TF; different failure modes." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "One is about ignoring `tokenizer.json`, the other about wrong Granite token IDs; same area, but not the same underlying bug." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "`ModelOutput` key assignment and non-persistent buffer serialization are unrelated internals with different code paths." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "GGUF loading in v5 and CLIPTokenizer `model_max_length` are unrelated bugs." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 loading regressions, but one is tied to `base_model_prefix` handling and the other to custom initialization; not the same concrete bug." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Device-map offload/dequantization and Qwen3 MoE memory usage are distinct problems." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "`torch.export` with `torch_compilable_check` and bfloat16 dtype mismatches are different inference/export failures." + }, + { + "left": "issue:44509", + "right": "issue:46909", + "accept": false, + "reason": "Both are documentation-related, but they concern different docs/pages and different fixes." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Qwen2.5-VL rope-index scaling and Kosmos2.5 long-OCR index errors are different model-specific bugs." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Multi-image finetuning failure in an IterableDataset is not the same as the video-position-id regression in Qwen2.5-VL inference." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Missing padding-mask support in MIMI and Gemma3 NaNs from sliding-window all-padding cases are different batching/masking bugs." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Non-persistent buffer junk and `ModelOutput` key assignment are unrelated v5 regressions." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "`init_empty_weights` argument mismatch and strict config loading are different load-time failures." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "They both mention Qwen2.5-VL video/vision positions, but one is still-image rope scaling and the other is video input position IDs; not clearly the same bug." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Bad SigLIP2 outputs and Perceiver interpolation failures are separate vision-model issues." + }, + { + "left": "issue:40444", + "right": "issue:47217", + "accept": false, + "reason": "Both involve VL image placeholders, but one is a finetuning/dataset issue and the other is a runtime failure when the placeholder appears in user text." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Weight-tying failure in AyaVision and device-dependent `lm_head.weight` serialization inconsistency are related only at a very high level." + }, + { + "left": "issue:43818", + "right": "issue:46402", + "accept": false, + "reason": "Video-LLaVA temporal-attention/weight-sharing issues are unrelated to Sam3 text-encoder loading." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Video-LLaVA architecture bugs and Qwen3.5 visual-encoder save_pretrained keys are different subsystems and fixes." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "The latter is a broad tokenization-version regression, not the same offloading/quantization bug." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "`resume_from_checkpoint` key mismatch and Qwen2.5-VL rope-index scaling are unrelated." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both can cause degenerate training, but one is tied to Flash Attention 3 behavior and the other to `problem_type`/`num_labels=1`; different root causes." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU-offload device mismatch and `torch.export`/compilable-check failures are distinct code paths." + }, + { + "left": "issue:42796", + "right": "issue:45381", + "accept": false, + "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL video position IDs are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Qwen VL video regression and Gemma4 PLE device mismatch are different model families and different failure modes." + }, + { + "left": "issue:43828", + "right": "issue:46821", + "accept": false, + "reason": "Autocast dtype mismatch and TimesFM `window_size` attribute errors are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "High perplexity on GPT-OSS and Whisper empty-transcription failure are unrelated model behaviors." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio processor sampling-rate defaulting and GraniteSpeech PEFT checkpoint loading are different bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 15, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:36331", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42617", + "issue:42796", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43450", + "issue:43454", + "issue:43550", + "issue:43572", + "issue:43582", + "issue:43606", + "issue:43720", + "issue:43749", + "issue:43819", + "issue:43873", + "issue:44186", + "issue:44265", + "issue:44387", + "issue:44534", + "issue:44568", + "issue:44610", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45920", + "issue:45923", + "issue:46050", + "issue:46209", + "issue:46305", + "issue:46315", + "issue:46821", + "issue:47404", + "issue:47405" + ], + "soft_pairs": [ + "issue:34634|issue:35707", + "issue:35707|issue:41720", + "issue:44186|issue:44265", + "issue:41720|issue:45823", + "issue:43606|issue:43819", + "issue:43454|issue:43550", + "issue:42617|issue:42915", + "issue:45823|issue:46821", + "issue:42796|issue:46209", + "issue:43257|issue:45923", + "issue:40990|issue:42947", + "issue:44568|issue:45920", + "issue:47404|issue:47405", + "issue:43454|issue:45072", + "issue:43377|issue:44610", + "issue:36331|issue:42617", + "issue:41553|issue:43582", + "issue:41720|issue:43322", + "issue:43572|issue:45759", + "issue:36331|issue:42915", + "issue:44877|issue:44991", + "issue:43749|issue:44534", + "issue:43454|issue:43720", + "issue:43317|issue:43873", + "issue:43377|issue:44898", + "issue:46305|issue:46315", + "issue:41720|issue:44977", + "issue:43322|issue:46050", + "issue:41720|issue:42617", + "issue:40990|issue:45701", + "issue:44387|issue:45704", + "issue:43450|issue:45701" + ], + "prepared_review_unit_hash": "1ea24f9cc6b50939f9fe68dfe7ce92c105e5db4b68bbb567ad792ee277f4dbd2", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37356, + "estimated_input_tokens": 9339, + "estimated_eval_tokens": 18934 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose set of Transformers bug reports with some broad thematic overlap around loading, offloading, quantization, and tokenization, but the paired items are not duplicates of the same concrete bug. I would not merge any of the soft-edge candidates.", + "confidence": 0.73, + "canonical_issue_reason": "issue:43873 is the broadest and most central issue in the set for the recurring offloading/quantization loading regressions, and it has the strongest discussion/inbound-reference signal. The other issues are either narrower model-specific failures or different code paths entirely.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best representative issue for this cluster because it captures the general offloading + quantization failure theme better than the more specialized reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "Different features and failure modes: Bark voice presets vs generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Unrelated bugs: progressive generation state handling vs device-mapping/CUDA assert on Qwen3." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Both involve batched behavior, but one is a tokenizer crash on NER padding/truncation and the other is torch.export with torch_compilable_check." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both are device-map related, but the concrete failures differ: CUDA assert on Qwen3 vs PLE forward device mismatch on Gemma4." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Different bugs and code paths: CPU-offload device mismatch vs DAC latent reconstruction/STE mismatch." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "AyaVision lm_head tying and garbage text generation is unrelated to Bamba torch.compile + SDPA failure." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "No shared underlying bug; one is a 3d_parallel.py runtime issue, the other is FineGrainedFP8Config failure on Qwen3Moe." + }, + { + "left": "issue:45823", + "right": "issue:46821", + "accept": false, + "reason": "Device-mapping mismatch and TimesFM window_size AttributeError are different issues." + }, + { + "left": "issue:42796", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer BPE marker handling is unrelated to FP8 inference on B200." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "Accelerate+deepspeed MOE weight conversion is different from a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "Perplexity regression on gpt-oss is not the same bug as ineffective gradient checkpointing with LoRA." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "Both are tokenizer regressions, but one is missing add_special_tokens behavior for a specific tokenizer while the other is wrong token IDs across several models." + }, + { + "left": "issue:47404", + "right": "issue:47405", + "accept": false, + "reason": "Loading OOM from conversion-op transients is unrelated to silent conversion skipping on model_type mismatch." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "Weight tying/generation corruption is unrelated to dtype mismatches in bfloat16 inference." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "Both mention batched inputs, but they affect different components and modalities: MIMI encoder padding masks vs OmDet processor image sizing." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "Trainer compute_loss signature breakage is unrelated to the 3d_parallel.py runtime failure." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "AutoTokenizer loading Voxtral error messaging is unrelated to an Apple Silicon TypeError in caching_allocator_warmup." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Different loading failures: Qwen3 auto device mapping CUDA assert vs Llava Next segmentation fault." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Trainer loss-signature incompatibility is not the same as Qwen3Moe FP8 configuration failure." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading for granite_speech and tokenizer loading for est-roberta are different failures." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "FSDP CPU RAM efficient loading breakage is not the same as non-persistent buffers being filled with junk." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "AyaVision lm_head tying is unrelated to BitNet packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Related theme, but not the same concrete bug: dequantized model offload loading failure vs broader offloading-with-quantization behavior." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "MIMI batching/padding-mask mismatch is unrelated to Perceiver non-default resolution failures." + }, + { + "left": "issue:46305", + "right": "issue:46315", + "accept": false, + "reason": "Streaming mask off-by-one and tokenizer convert_tokens_to_ids slowdown are different subsystems and symptoms." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3 device-map CUDA assert is unrelated to Qwen3.5 flash-attention generation issues." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Llava Next loading segfault is not the same bug as sinq quantized save/load breakage." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Device-mapping CUDA assert and 3d_parallel.py runtime failure are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "High perplexity on a model and tokenization changes across versions are not the same underlying bug." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 CUDA reserved-memory regression is unrelated to T5 using apex.FusedRMSNorm with a memory leak." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor batched-shape bug is unrelated to version-dependent tokenization changes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 16, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:26413", + "issue:28530", + "issue:33666", + "issue:41720", + "issue:42915", + "issue:43065", + "issue:43295", + "issue:43317", + "issue:43404", + "issue:43531", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43708", + "issue:43720", + "issue:43873", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44423", + "issue:44451", + "issue:44466", + "issue:44488", + "issue:44610", + "issue:44756", + "issue:44898", + "issue:45072", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45685", + "issue:45701", + "issue:45704", + "issue:45823", + "issue:46291", + "issue:46402", + "issue:46612", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:46869", + "issue:46897", + "issue:47020", + "issue:47317", + "issue:47404" + ], + "soft_pairs": [ + "issue:43873|issue:44756", + "issue:45685|issue:46821", + "issue:44466|issue:45357", + "issue:26413|issue:28530", + "issue:44186|issue:45072", + "issue:44451|issue:47020", + "issue:43295|issue:45701", + "issue:43065|issue:46402", + "issue:46612|issue:46858", + "issue:43550|issue:43720", + "issue:44186|issue:44610", + "issue:43646|issue:46665", + "issue:43550|issue:43606", + "issue:43577|issue:43708", + "issue:46897|issue:47317", + "issue:46821|issue:46869", + "issue:42915|issue:43317", + "issue:41720|issue:45161", + "issue:44423|issue:45406", + "issue:44610|issue:44898", + "issue:28530|issue:33666", + "issue:43404|issue:45357", + "issue:44008|issue:45823", + "issue:42915|issue:46291", + "issue:43881|issue:44488", + "issue:43531|issue:43572", + "issue:45704|issue:47404" + ], + "prepared_review_unit_hash": "e4ba8acee3ace3f55f96e49656f794832b0d88d3ffa4955cabe7553d5f3dfe01", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35336, + "estimated_input_tokens": 8834, + "estimated_eval_tokens": 17924 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: the soft-similarity pairs mostly share broad themes (training, loading, quantization, multimodal, serving) but not the same concrete root cause. I would not merge any of the proposed soft edges.", + "confidence": 0.78, + "canonical_issue_reason": "Issue #43708 is the most concrete, well-scoped, and reproducible Trainer bug in the set, with a clear failure mode around resuming checkpoints and changing per-device batch size.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43708 is the strongest representative issue overall: specific symptom, clear repro path, and direct impact on training correctness.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Both involve offloading/memory behavior, but one is about quantization offload while the other is Strix Halo mmap OOM; different bugs and fixes." + }, + { + "left": "issue:45685", + "right": "issue:46821", + "accept": false, + "reason": "Completely different failures: MPS histogram support vs TimesFM window_size handling." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both are save/serialization regressions, but one is about tied lm_head weights and the other about incorrect visual encoder keys." + }, + { + "left": "issue:26413", + "right": "issue:28530", + "accept": false, + "reason": "Different Trainer issues: checkpoint resumption epoch_iterator failure vs early stopping/metric_for_best_model handling." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Unrelated subsystems: LayoutLMv2 tokenizer padding/NER crash vs dtype mismatches in inference." + }, + { + "left": "issue:44451", + "right": "issue:47020", + "accept": false, + "reason": "Both affect tokenizer/model loading, but the concrete bugs are different model families and different code paths." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Both mention version-related tokenization behavior, but one is a processor/tokenizer API regression and the other is a broader tokenization change report." + }, + { + "left": "issue:43065", + "right": "issue:46402", + "accept": false, + "reason": "Same model family, but one is a dummy Conv2d in Sam3PixelDecoder and the other is missing text encoder weight loading." + }, + { + "left": "issue:46612", + "right": "issue:46858", + "accept": false, + "reason": "Both touch cache logic, but one is beam-search cache reorder and the other is static-cache seq length arithmetic in prefill." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Different model/runtime bugs: torch.compile+SDPA for Bamba vs packed-weight unpacking during accelerate loading for BitNet." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Different vision preprocessing bugs with different models and failure modes." + }, + { + "left": "issue:43646", + "right": "issue:46665", + "accept": false, + "reason": "One is a Transformers 5.0 custom model init regression; the other is a from_pretrained behavior inconsistency inside Python classes." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Both are runtime/device issues, but they affect different models and have different root causes." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Different problems: dtype propagation in Blip2 vs Trainer max_steps recomputation on resume." + }, + { + "left": "issue:46897", + "right": "issue:47317", + "accept": false, + "reason": "Both are training-loss bugs, but they concern different models and no shared concrete code path is established." + }, + { + "left": "issue:46821", + "right": "issue:46869", + "accept": false, + "reason": "TimesFM window_size attribute handling is unrelated to YOLO multi-GPU device_map loading." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Both involve quantization/loading, but one is FineGrainedFP8Config and the other is gpu+cpu offload of a dequantized model." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Different distributed/model-parallel issues: device-mapping cudaErrorAssert vs tensor-parallel support for GPT-OSS MoE." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both are serve crashes on multimodal models, but the immediate exceptions and offending code paths differ." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Different model-specific image size/preprocessing bugs." + }, + { + "left": "issue:28530", + "right": "issue:33666", + "accept": false, + "reason": "Early stopping metric lookup and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "One is tied lm_head serialization in Mistral3ForConditionalGeneration; the other is incorrect visual encoder keys in Qwen3.5 save_pretrained." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Different Gemma-family bugs: variable name collision in forward vs device mismatch under device_map=auto." + }, + { + "left": "issue:42915", + "right": "issue:46291", + "accept": false, + "reason": "FineGrainedFP8Config loading and kernels package revision/version requirements are unrelated." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Both are model loading failures, but for different models and likely different failure mechanisms." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Unrelated configuration bugs: Qwen3-MoE sliding window vs StableLmConfig pad_token_idx." + }, + { + "left": "issue:45704", + "right": "issue:47404", + "accept": false, + "reason": "Both relate to memory, but one is an apex RMSNorm leak and the other is device-map loading OOM from conversion transients." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 17, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28530", + "issue:29159", + "issue:30064", + "issue:33357", + "issue:34634", + "issue:35707", + "issue:36032", + "issue:41720", + "issue:42915", + "issue:43257", + "issue:43316", + "issue:43322", + "issue:43441", + "issue:43452", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43653", + "issue:43716", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:43881", + "issue:44186", + "issue:44361", + "issue:44464", + "issue:44534", + "issue:44671", + "issue:44734", + "issue:44743", + "issue:44913", + "issue:44991", + "issue:45479", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:46894", + "issue:47020", + "issue:47145", + "issue:47254", + "issue:47317", + "issue:47337" + ], + "soft_pairs": [ + "issue:43441|issue:47337", + "issue:44186|issue:44361", + "issue:28530|issue:30064", + "issue:43716|issue:46821", + "issue:44734|issue:46858", + "issue:44464|issue:46858", + "issue:44671|issue:44743", + "issue:29159|issue:36032", + "issue:43316|issue:44913", + "issue:30064|issue:34634", + "issue:43452|issue:47020", + "issue:45479|issue:47317", + "issue:45632|issue:45684", + "issue:45701|issue:47145", + "issue:42915|issue:43322", + "issue:33357|issue:35707", + "issue:43540|issue:43782", + "issue:43873|issue:45704", + "issue:43531|issue:45759", + "issue:41720|issue:43746", + "issue:43881|issue:44991", + "issue:43550|issue:43653", + "issue:44534|issue:46665", + "issue:46894|issue:47317", + "issue:43257|issue:47254", + "issue:43531|issue:46821" + ], + "prepared_review_unit_hash": "638da0eab1a22ebe92c843437efca133209b30c0b08ec8aab49bbe53c1d8f1c2", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34821, + "estimated_input_tokens": 8706, + "estimated_eval_tokens": 17668 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous and does not contain any clear duplicate PRs; the soft pairs mostly share broad subsystem keywords (tokenizers, cache, Qwen, generation) but describe different concrete failures. The best issue anchor is the most broadly discussed active bug report, not a true duplicate hub.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 43873 is the most central/open report in the set, with the strongest inbound discussion and a broad, actionable bug description. There is no strong duplicate cluster here, so it is the best available anchor rather than a true canonical duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "43873 has the most evidence of being a live, impactful problem and is the most suitable representative issue among otherwise unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43441", + "right": "issue:47337", + "accept": false, + "reason": "Different domains: FlashAttention/model loading on macOS vs a flaky bbox-range test." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Both tokenizer bugs, but different models and failure modes; not the same underlying defect." + }, + { + "left": "issue:28530", + "right": "issue:30064", + "accept": false, + "reason": "Early-stopping metric selection is unrelated to image processor void-segmentation handling." + }, + { + "left": "issue:43716", + "right": "issue:46821", + "accept": false, + "reason": "Different components and errors: multimodal dtype mismatch vs TimesFM window_size AttributeError." + }, + { + "left": "issue:44734", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache logic, but one is serve/KV-continuation indexing and the other is static-cache seq_length on non-CUDA compile paths." + }, + { + "left": "issue:44464", + "right": "issue:46858", + "accept": false, + "reason": "Chunked generation inconsistency is a different code path from static-cache slice arithmetic with torch.compile." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "Different models and bugs: CamemBERT masked LM predictions vs Qwen3 recurrent-state reset with cache." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Tokenizer inconsistency is too broad; T5Tokenizer load failure is a specific special-token name collision." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Both are config/API-ish, but one is a Gemma3TextConfig discrepancy and the other is GPTNeoX rotary_pct persistence." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Unrelated processor bugs affecting segmentation maps vs BarkProcessor voice_preset." + }, + { + "left": "issue:43452", + "right": "issue:47020", + "accept": false, + "reason": "Both touch tokenizer/model loading, but they are distinct bugs with different root causes and affected classes." + }, + { + "left": "issue:45479", + "right": "issue:47317", + "accept": false, + "reason": "Both mention training loss, but the model families and likely failure mechanisms are different; insufficient evidence of one underlying bug." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Cache-path collision for trust_remote_code is unrelated to file permission propagation in save_pretrained." + }, + { + "left": "issue:45701", + "right": "issue:47145", + "accept": false, + "reason": "Tokenizer-version differences and FP8 weight_scale corruption are different problems." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "FineGrainedFP8Config failure and Llava Next segfault are unrelated model-loading issues." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "MacOS bus error in a CLIP model is unrelated to progressive generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Different multimodal bugs: video input processing vs from_pretrained weight_only loading." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to T5's apex.FusedRMSNorm memory leak." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Both involve Qwen3, but one is sliding_window behavior in MoE and the other is config unwrapping for multimodal models." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Auto device mapping CUDA assert and PEFT checkpoint loading are distinct failures." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Different model families and different loading failures; no shared code path." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile/SDPA model execution bug is unrelated to a tokenizer special-token registration issue." + }, + { + "left": "issue:44534", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffers junk in v5 is not the same as inconsistent from_pretrained behavior inside Python classes." + }, + { + "left": "issue:46894", + "right": "issue:47317", + "accept": false, + "reason": "Both are training-loss complaints, but they are model-specific and the descriptions do not establish the same underlying bug." + }, + { + "left": "issue:43257", + "right": "issue:47254", + "accept": false, + "reason": "MoE weight conversion under accelerate/deepspeed and ZeRO-3 checkpointing errors are different distributed-training problems." + }, + { + "left": "issue:43531", + "right": "issue:46821", + "accept": false, + "reason": "Qwen3-MoE sliding-window bug is unrelated to TimesFM's window_size AttributeError." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 18, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:34567", + "issue:34634", + "issue:35707", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:43329", + "issue:43388", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43606", + "issue:43644", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43883", + "issue:44062", + "issue:44077", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44589", + "issue:44610", + "issue:44625", + "issue:44898", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45059", + "issue:45072", + "issue:45357", + "issue:45406", + "issue:45464", + "issue:45479", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:45923", + "issue:46050", + "issue:46402", + "issue:46665", + "issue:46858", + "issue:47333", + "issue:47405" + ], + "soft_pairs": [ + "issue:43526|issue:43883", + "issue:43606|issue:43720", + "issue:42490|issue:43329", + "issue:44492|issue:44509", + "issue:43531|issue:44589", + "issue:45406|issue:45464", + "issue:34634|issue:41720", + "issue:34567|issue:35707", + "issue:43606|issue:43653", + "issue:44451|issue:44991", + "issue:45759|issue:45859", + "issue:43540|issue:46858", + "issue:43388|issue:45059", + "issue:45704|issue:47405", + "issue:44987|issue:45042", + "issue:43644|issue:46665", + "issue:44077|issue:44625", + "issue:45357|issue:46402", + "issue:45923|issue:47333", + "issue:44062|issue:44451", + "issue:44918|issue:44936", + "issue:44898|issue:45072", + "issue:39692|issue:44610", + "issue:44625|issue:45479", + "issue:43746|issue:46050" + ], + "prepared_review_unit_hash": "06e0aafb4f46818d708d60c763a2b0b43a5beba3ea5193160d043e82427dd3f6", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 25, + "serialized_chars": 34958, + "estimated_input_tokens": 8740, + "estimated_eval_tokens": 17736 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a grab-bag of unrelated bug reports across tokenizers, multimodal processors, loading/offloading, training, and docs. None of the soft-edge pairs look like the same underlying bug or change, so all are rejected. I picked the broad loading-correctness regression as the canonical issue.", + "confidence": 0.96, + "canonical_issue_reason": "issue:47405 is the most broadly applicable and highest-impact bug in the set: loading a converted checkpoint through the wrong Auto class silently yields wrong weights. It is more canonical than the narrowly scoped, model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:47405 is the strongest standalone issue for triage because it describes a generic, severe correctness failure in model loading rather than a model-specific edge case or documentation mismatch.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "Different subsystems and failure modes: BeitImageProcessorFast label reduction vs Molmo missing all_tied_weights_keys." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "Both mention offload/loading, but one is a Bark CPU device mismatch and the other is BitNet packed-weight unpacking; not the same bug." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Local path import warning vs multimodal video token counting undefined variables; unrelated." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A typo in cache-strategy docs is unrelated to removed pipeline tasks in docs." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and Float8 storage lookup failure are different code paths." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Qwen3.5 streaming API failures." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice_preset bug and Qwen3 auto device mapping cuda assert are different model-specific problems." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token counter not updating is unrelated to progressive generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Bark CPU offload mismatch vs BigBirdTokenizer special-token registration; no shared underlying defect." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-load failures, but for different models and likely different causes." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Same Qwen family, but one is config unwrapping for from_config and the other is CPU-offload rotary embedding compatibility." + }, + { + "left": "issue:43540", + "right": "issue:46858", + "accept": false, + "reason": "Video processing ValueError in Qwen3OmniMoe is unrelated to cache length tensor arithmetic." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Metric gathering label truncation is unrelated to SAM3 text/bounding-box behavior." + }, + { + "left": "issue:45704", + "right": "issue:47405", + "accept": false, + "reason": "T5 Apex RMSNorm memory leak and wrong Auto-class loading of converted checkpoints are different failures." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Model loading failure for physical-intelligence/fast is unrelated to PIL backend torchvision dependency regressions." + }, + { + "left": "issue:43644", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffer junk in v5.0.0 and from_pretrained behavior inside Python classes are different issues." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "PatchTSMixer post_init policy and Qwen3.5 num_labels propagation are unrelated." + }, + { + "left": "issue:45357", + "right": "issue:46402", + "accept": false, + "reason": "Both involve multimodal weights, but one is Qwen3.5 save_pretrained key naming and the other is Sam3Model text encoder loading." + }, + { + "left": "issue:45923", + "right": "issue:47333", + "accept": false, + "reason": "Missing supports_gradient_checkpointing flag is unrelated to saving a disk-offloaded model." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken argument duplication and ScandiBERT load failure are not the same bug." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "TRL SFT input-embedding unpacking and trainer.evaluate after trainer.train are distinct training-path issues." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver image-size handling and dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Docs example errors for SigLIP2 do not match OmDet-Turbo processor/model resolution mismatch." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "num_labels propagation in Qwen3.5 is not the same as the single-label zero-loss classification bug." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "PEFT adapter loading from local checkpoints and sinq quantized save/load breakage are different save/load mechanisms." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 19, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:42886", + "issue:43295", + "issue:43381", + "issue:43421", + "issue:43493", + "issue:43526", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43653", + "issue:43673", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43819", + "issue:43844", + "issue:43873", + "issue:43881", + "issue:43927", + "issue:43994", + "issue:44186", + "issue:44220", + "issue:44368", + "issue:44464", + "issue:44514", + "issue:44545", + "issue:44671", + "issue:44843", + "issue:44857", + "issue:44995", + "issue:45081", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45834", + "issue:45907", + "issue:46032", + "issue:46294", + "issue:46463", + "issue:46655", + "issue:46693", + "issue:46858", + "issue:47050", + "issue:47337" + ], + "soft_pairs": [ + "issue:44514|issue:44545", + "issue:30064|issue:36010", + "issue:42886|issue:44843", + "issue:43421|issue:43927", + "issue:43746|issue:43873", + "issue:46655|issue:46693", + "issue:43381|issue:43844", + "issue:43421|issue:43653", + "issue:39692|issue:43493", + "issue:45081|issue:45356", + "issue:44464|issue:47050", + "issue:44995|issue:46858", + "issue:43673|issue:47337", + "issue:43782|issue:45381", + "issue:43994|issue:44671", + "issue:43421|issue:44186", + "issue:44368|issue:45357", + "issue:45356|issue:45907", + "issue:46032|issue:46463", + "issue:44857|issue:47050", + "issue:39692|issue:43873", + "issue:39692|issue:47050", + "issue:43596|issue:44220", + "issue:43720|issue:43819", + "issue:43540|issue:46294", + "issue:43295|issue:45834", + "issue:43550|issue:43927", + "issue:43550|issue:43819", + "issue:43526|issue:43881" + ], + "prepared_review_unit_hash": "47101b6772ebee240a749a48c6759a15ee93b3f2fee12d27709bc14c9dd8a323", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36228, + "estimated_input_tokens": 9057, + "estimated_eval_tokens": 18370 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous transformer bugs/regressions, with only one clear duplicate pair: the two Qwen2_5_VLProcessor apply_chat_template reports. The rest are related only at a broad subsystem level (tokenizers, generation, multimodal, cache/offload) and should stay separate.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43873 is the broadest and most central report in the set, with the highest discussion and multiple inbound references; the other issues are narrower, model-specific regressions or tokenizer bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best cluster anchor because it is general enough to represent the most central theme here (offloading/quantization behavior) and has the strongest engagement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same concrete bug: Qwen2_5_VLProcessor.apply_chat_template crashes on batched input when padding=False; titles differ only in wording." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Unrelated subsystems and failures: image processor segmentation maps vs a GenerationMixin import error." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both involve offline/tokenizer loading, but 44843 is specifically the mistral regex patch calling model_info; 42886 is a broader cache-loading failure." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Different bugs: runtime special-token/post-processor updates vs save/load losing custom token IDs and causing generation IndexError." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Different concrete problems: PEFT local checkpoint loading vs offloading behavior with quantization." + }, + { + "left": "issue:46655", + "right": "issue:46693", + "accept": false, + "reason": "Different code paths: video processor padding calculation vs attention-performance regression in long-context generation." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Different training/runtime issues: eval-mode gradient checkpointing vs DeepSpeed ZeRO-3 gradient growth." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Both tokenizer-related, but one is post-processor updates after runtime special-token changes and the other is BigBird mask-token registration; not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Same model family, but one is a broken documentation example and the other is an implementation discrepancy with JAX; not one underlying issue." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both mention fix_mistral_regex, but 45081 is a tokenizer attribute crash while 45356 is a codec-handling regression; close, but not the same concrete bug." + }, + { + "left": "issue:44464", + "right": "issue:47050", + "accept": false, + "reason": "Different failures: compiled-forward chunked generation inconsistency vs ONNX fp16 attention dtype mismatch." + }, + { + "left": "issue:44995", + "right": "issue:46858", + "accept": false, + "reason": "Both mention cache behavior, but one is a stale indexer cache in GlmMoeDsa and the other is device-tensor seq length arithmetic in static cache/torch.compile." + }, + { + "left": "issue:43673", + "right": "issue:47337", + "accept": false, + "reason": "GenerationMixin chunked_prefill cache issue is unrelated to a flaky bbox-range test." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen VL issues, but one is from_pretrained weight_only loading and the other is video vision_position_ids; different bugs." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Completely different models and symptoms: SigLIP2 inference drift vs CamemBERT masked LM predictions." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Both tokenizer-adjacent, but one is about special-token mutation at runtime and the other about LayoutLMv2 NER/batching crashes." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "The Qwen3.5 warning about tie_word_embeddings is unrelated to a save_pretrained visual-encoder-key regression." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Mistral/Kimi tokenizer regression is unrelated to list_repo_files forgetting to forward revision." + }, + { + "left": "issue:46032", + "right": "issue:46463", + "accept": false, + "reason": "Mamba2Mixer use_cache correctness bug is unrelated to a T5 decoder-input validation error message." + }, + { + "left": "issue:44857", + "right": "issue:47050", + "accept": false, + "reason": "Different model/tooling paths: LwDetr float16 AMP crash vs CanineModel ONNX export dtype mismatch." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "Same broad ML area, but SigLIP2 docs/example problems are not the same bug as quantization offloading behavior." + }, + { + "left": "issue:39692", + "right": "issue:47050", + "accept": false, + "reason": "Unrelated model-specific issues: SigLIP2 example/model mismatch vs CanineModel ONNX dtype mismatch." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "Different failure surfaces: DeepSpeed ZeRO-3/BertModel index error vs audio feature extraction issue." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "Different model internals and bugs: BitNet packed-weight loading vs DAC from_latents missing STE consistency." + }, + { + "left": "issue:43540", + "right": "issue:46294", + "accept": false, + "reason": "Both involve Qwen video input, but one is a ValueError in video processing for Qwen3OmniMoe and the other is chat-template token/feature mismatch in Qwen3.5-9B." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Different bugs: processor.tokenizer regression with images vs Kosmos2.5 OCR long-input index error." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "Unrelated problems: torch.compile/SDPA failure in Bamba vs DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Different model and failure mode: Bamba torch.compile/SDPA bug vs DAC forward/from_latents mismatch." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "Different issues: BeitImageProcessorFast reduce_labels returns one label vs glm-4v-9b loading failure." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The canonical-anchor choice is reasonable given 43873\u2019s higher discussion/inbound activity, and the only accepted soft edge (44514/44545) is a clear duplicate by title and bug description. The other pairwise rejects are appropriately cautious and do not overstate evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 20, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:30064", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43335", + "issue:43531", + "issue:43540", + "issue:43611", + "issue:43643", + "issue:43653", + "issue:43720", + "issue:43756", + "issue:43819", + "issue:44008", + "issue:44062", + "issue:44186", + "issue:44220", + "issue:44442", + "issue:44448", + "issue:44488", + "issue:44561", + "issue:44661", + "issue:44671", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45381", + "issue:45834", + "issue:45987", + "issue:46650", + "issue:46665", + "issue:46693", + "issue:46829", + "issue:46832", + "issue:46858", + "issue:46932", + "issue:47020", + "issue:47050", + "issue:47405" + ], + "soft_pairs": [ + "issue:44488|issue:47020", + "issue:44062|issue:44488", + "issue:43531|issue:43643", + "issue:44220|issue:44661", + "issue:43653|issue:43720", + "issue:45381|issue:46693", + "issue:45072|issue:47050", + "issue:41720|issue:44991", + "issue:30064|issue:36331", + "issue:44821|issue:45084", + "issue:43611|issue:46665", + "issue:46829|issue:46832", + "issue:44186|issue:44898", + "issue:46650|issue:47405", + "issue:43540|issue:44008", + "issue:44991|issue:47020", + "issue:43653|issue:43819", + "issue:34634|issue:36010", + "issue:44843|issue:45081", + "issue:43643|issue:44877", + "issue:43335|issue:43756", + "issue:45987|issue:46829", + "issue:44008|issue:46858", + "issue:39692|issue:45834", + "issue:44561|issue:45020", + "issue:44186|issue:44442", + "issue:44448|issue:44671", + "issue:45987|issue:46932" + ], + "prepared_review_unit_hash": "98eeae7dfff603f22719d20dd5515c670fa1e6a8061f90b389517952176ac5b0", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35882, + "estimated_input_tokens": 8971, + "estimated_eval_tokens": 18198 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a collection of distinct Transformers bugs, many around loading/tokenizer/modeling regressions, but they do not collapse into a single underlying issue. The soft-linked pairs share broad theme similarity only, not the same concrete code path, so none should be merged as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45020 is the broadest and most central-sounding report in the set, covering recent-version regressions with remote_code models and overlapping with several other loading/config/tokenizer complaints. It is still not an exact duplicate of the others, but it is the best umbrella candidate among these issues.", + "canonical_pr_reason": null, + "best_issue_reason": "#45020 is the most suitable representative because it describes a high-level regression category rather than a narrowly model-specific failure, making it the closest thing to a cluster anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44488", + "right": "issue:47020", + "accept": false, + "reason": "Both are tokenizer-loading failures, but one is a specific model load failure and the other is a constructor/API mismatch in XLMRobertaTokenizer; different root causes and code paths." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "These are unrelated tokenizer bugs: one is an AddedToken keyword conflict, the other is a model/tokenizer load failure for a specific checkpoint." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "One is a sliding_window handling bug in Qwen3-MoE; the other is missing fields from trust_remote_code AutoConfig loading. Similar category, different defect." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Different failure modes: one is an audio feature extraction issue, the other is an add-new-model-like/tokenizer-mapping problem." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Both are CI/loading related, but one is BigBirdTokenizer special-token decoding and the other is BitNet packed-weight unpacking during accelerate loading; not the same bug." + }, + { + "left": "issue:45381", + "right": "issue:46693", + "accept": false, + "reason": "Video input position-id mismatch for Qwen2.5-VL is unrelated to the FA2 long-context performance regression." + }, + { + "left": "issue:45072", + "right": "issue:47050", + "accept": false, + "reason": "Both mention dtype mismatches, but they occur in different models and code paths: SwitchTransformers/TimmWrapperModel versus Canine attention ONNX export." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Both involve model loading/version regressions, but one is a CUDA assert with auto device mapping and the other is tokenizer loading for a specific checkpoint." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image processor segmentation-map handling and Trainer.compute_loss signature mismatch are completely separate issues." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "One is loading an image processor from URL; the other is a template compilation error. No shared root cause." + }, + { + "left": "issue:43611", + "right": "issue:46665", + "accept": false, + "reason": "Both concern loading behavior, but one is base_model_prefix handling in Transformers 5.0 and the other is inconsistent from_pretrained behavior inside Python classes." + }, + { + "left": "issue:46829", + "right": "issue:46832", + "accept": false, + "reason": "Different model families and failures: wrong hidden_states output versus RTDetr feature-level mismatch." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Both are image/LayoutLM-family bugs, but one is tokenizer NER/padding truncation crashes and the other is Perceiver resolution/interpolation failure." + }, + { + "left": "issue:46650", + "right": "issue:47405", + "accept": false, + "reason": "Both involve loading/conversion correctness, but one is save_pretrained key renaming corruption while the other is silent model_type mismatch during Auto class loading." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Video-input processing in Qwen3OmniMoe and Gemma3n forward variable collision are unrelated model-specific bugs." + }, + { + "left": "issue:44991", + "right": "issue:47020", + "accept": false, + "reason": "Both are tokenizer loading issues, but one is a checkpoint-specific tokenizer breakage and the other is a constructor bug in XLMRobertaTokenizer." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Special-token decoding in BigBirdTokenizer and DAC.from_latents STE mismatch are unrelated defects." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice_preset and GenerationMixin import failure are different processor/import issues with no shared implementation bug." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Offline HF_hub model_info calls and mistral_regex backend_tokenizer attribute access are different failures in different helper paths." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both touch config loading, but one is missing fields under trust_remote_code and the other is strict config rejecting granite_speech; not the same bug." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are unrelated architectural/configuration bugs." + }, + { + "left": "issue:45987", + "right": "issue:46829", + "accept": false, + "reason": "StaticCache sequence length return type and VideoPrism hidden_states output mismatch are different API-contract issues." + }, + { + "left": "issue:44008", + "right": "issue:46858", + "accept": false, + "reason": "Gemma3n variable collision in forward is unrelated to static cache device tensor arithmetic under torch.compile." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 documentation/model-processor mismatch and Kosmos2.5 long OCR index error are distinct issues in different examples/models." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Both are broad v5/remote_code regressions, but one is specifically about removal of is_torch_fx_available and the other is a general remote_code breakage report; not the same concrete defect." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are unrelated tokenizer problems." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Pegasus v4/v5 output differences and CamemBERT masked LM prediction errors are model-specific behavior regressions, not one bug." + }, + { + "left": "issue:45987", + "right": "issue:46932", + "accept": false, + "reason": "StaticCache sequence length contract issue and Gemma4 quantized embedder dtype casting are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 21, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:34634", + "issue:36331", + "issue:43322", + "issue:43421", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43618", + "issue:43644", + "issue:43645", + "issue:43697", + "issue:43716", + "issue:43749", + "issue:43782", + "issue:43818", + "issue:43825", + "issue:43931", + "issue:43994", + "issue:44008", + "issue:44291", + "issue:44479", + "issue:44610", + "issue:44805", + "issue:44811", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45183", + "issue:45381", + "issue:45479", + "issue:45563", + "issue:45584", + "issue:45685", + "issue:45834", + "issue:46291", + "issue:46294", + "issue:46463", + "issue:46512", + "issue:46665", + "issue:46752", + "issue:46846", + "issue:47276", + "issue:47337" + ], + "soft_pairs": [ + "issue:44805|issue:44936", + "issue:43618|issue:43697", + "issue:44991|issue:46291", + "issue:43322|issue:45685", + "issue:45584|issue:45834", + "issue:43421|issue:44811", + "issue:46294|issue:46463", + "issue:43606|issue:45072", + "issue:46752|issue:47337", + "issue:43550|issue:45072", + "issue:44479|issue:45834", + "issue:43716|issue:44008", + "issue:43782|issue:44977", + "issue:45183|issue:45479", + "issue:43577|issue:43818", + "issue:28218|issue:29159", + "issue:43644|issue:43749", + "issue:45563|issue:47276", + "issue:43582|issue:46512", + "issue:43540|issue:45381", + "issue:43825|issue:43994", + "issue:44610|issue:45072", + "issue:43645|issue:46665", + "issue:43931|issue:46846", + "issue:34634|issue:36331", + "issue:43749|issue:44291" + ], + "prepared_review_unit_hash": "75c2c8909e383a84065343baad5af71b10c2f08d95e18686169a8f13b9b397bd", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35719, + "estimated_input_tokens": 8930, + "estimated_eval_tokens": 18116 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No pairs are strong enough to confirm as duplicates from titles alone. The tokenizer pair (28218, 29159) is the closest match, but it should be treated as related tokenizer behavior rather than the same bug. No pull requests are present.", + "confidence": 0.56, + "canonical_issue_reason": "Issue 29159 is the broadest tokenizer report and the best anchor for this sub-area, but the evidence is insufficient to collapse 28218 into it as a duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "29159 is the best representative of the tokenizer-related issues because it frames the behavior at the broader slow-vs-fast inconsistency level, even though the duplicate link is not strong enough to accept.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Different failure modes: mask/index shape mismatch versus trainer train/evaluate state behavior." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "CLIP attentions being dropped and RTDetr output drift are unrelated model behaviors." + }, + { + "left": "issue:44991", + "right": "issue:46291", + "accept": false, + "reason": "Tokenizer loading failure and pipeline import/version requirement are different code paths." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Llava Next segfault and MPS histogram TypeError are unrelated runtime failures." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription handling and Kosmos2.5 OCR index error are different bugs." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Special-token post-processor updates and Whisper batch_decode skipping are separate tokenizer behaviors." + }, + { + "left": "issue:46294", + "right": "issue:46463", + "accept": false, + "reason": "Qwen3.5 video token/feature mismatch and T5 decoder-input validation messaging are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "CPU offload device mismatch and dtype mismatch are not the same defect." + }, + { + "left": "issue:46752", + "right": "issue:47337", + "accept": false, + "reason": "Empty chat-template input crash and bbox-range flaky test are different problems." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Torch.compile/SDPA failure is unrelated to generic dtype mismatch reports." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Qwen video regression and Kosmos OCR index error are not the same underlying issue." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Image-preprocessor dtype mismatch and Gemma3n variable collision are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Weight-only load error and flash-attention generation issue differ in both symptom and path." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Input validation messaging and zero-loss classification are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "BLIP2 dtype retention and Video-LLaVA temporal-attention issues are different model problems." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": false, + "reason": "Both are tokenizer-related, but the packet only shows a broad inconsistency report versus an added-token spacing symptom; not enough evidence that they are the same bug." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Non-persistent buffer junk and FSDP CPU-RAM-efficient loading are different loading regressions." + }, + { + "left": "issue:45563", + "right": "issue:47276", + "accept": false, + "reason": "Both are warnings, but they refer to different generation features and code paths." + }, + { + "left": "issue:43582", + "right": "issue:46512", + "accept": false, + "reason": "Apple Silicon warmup TypeError and stale config dtype after from_config are distinct dtype/loading bugs." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "These are different Qwen video regressions with no clear shared root cause from titles alone." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "Pipeline translation warning text and SigLIP2 output quality are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Processor image-size mismatch and dtype mismatch are unrelated issues." + }, + { + "left": "issue:43645", + "right": "issue:46665", + "accept": false, + "reason": "Notebook custom-model initialization and from_pretrained-in-class behavior are different lifecycle bugs." + }, + { + "left": "issue:43931", + "right": "issue:46846", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 TP-plan omissions are separate loading/parallelism problems." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice_preset failure and CustomTrainer signature mismatch are unrelated." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and init_empty_weights unexpected argument are both loading-related but not the same defect." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The overall summary is directionally fine, but the accepted soft edge overstates the evidence. From the packet we only have titles and soft similarity, so saying 28218 and 29159 are the \"same tokenizer added-token spacing inconsistency\" is stronger than the data supports. A more conservative verdict would be that they are plausibly related tokenizer issues, not clearly duplicates. The rest of the soft-edge rejections look appropriately cautious, and the no-PR note is grounded." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 22, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36010", + "issue:36331", + "issue:42175", + "issue:42915", + "issue:43262", + "issue:43388", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43638", + "issue:43645", + "issue:43653", + "issue:43720", + "issue:43742", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:44589", + "issue:44671", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:45003", + "issue:45020", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45372", + "issue:45701", + "issue:46032", + "issue:46175", + "issue:46285", + "issue:46291", + "issue:46326", + "issue:46697", + "issue:46728", + "issue:46851", + "issue:46932", + "issue:47217", + "issue:47254", + "issue:47276", + "issue:47332", + "issue:47333", + "issue:47337" + ], + "soft_pairs": [ + "issue:43720|issue:45072", + "issue:47276|issue:47337", + "issue:42175|issue:45081", + "issue:47254|issue:47333", + "issue:43388|issue:46032", + "issue:44671|issue:45701", + "issue:43742|issue:45020", + "issue:47254|issue:47332", + "issue:43577|issue:43582", + "issue:46697|issue:46728", + "issue:42915|issue:47333", + "issue:44589|issue:45042", + "issue:43782|issue:45325", + "issue:43262|issue:47217", + "issue:46851|issue:47337", + "issue:36010|issue:43645", + "issue:43262|issue:43819", + "issue:46175|issue:46932", + "issue:43653|issue:44912", + "issue:45003|issue:46285", + "issue:43582|issue:44857", + "issue:36010|issue:36331", + "issue:43638|issue:44936", + "issue:43531|issue:45042", + "issue:45372|issue:46326", + "issue:43550|issue:44912", + "issue:43825|issue:43827", + "issue:42175|issue:46291" + ], + "prepared_review_unit_hash": "5f764323ad4e141caa1d97fa4b2236a274334bcd28c62bf472bd246184ccd19a", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35862, + "estimated_input_tokens": 8966, + "estimated_eval_tokens": 18188 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issue reports grouped by loose lexical similarity. I reject every soft edge: they describe different bugs, models, or subsystems. There is no real PR candidate here; among the issues, #45020 is the broadest regression-style report and the least bad canonical representative, but the cluster is not a true duplicate set.", + "confidence": 0.72, + "canonical_issue_reason": "#45020 is the broadest, most cross-cutting report in the set: a recent-version regression affecting model loading with `remote_code`, which is closer to a cluster-level umbrella than the many model-specific or subsystem-specific issues here.", + "canonical_pr_reason": null, + "best_issue_reason": "Best overall issue representative for this noisy cluster because it has the widest applicability and the most plausible overlap with other loading/regression reports, even though the set is not genuinely duplicate-heavy.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Different failures: BitNet packed-weight unpacking during accelerate loading vs dtype mismatch in bfloat16 inference for other models." + }, + { + "left": "issue:47276", + "right": "issue:47337", + "accept": false, + "reason": "A misleading warning about continuous batching is unrelated to a flaky bbox-range test." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Backend installation/package-dependency issue vs a Mistral tokenizer regex crash; different code paths and symptoms." + }, + { + "left": "issue:47254", + "right": "issue:47333", + "accept": false, + "reason": "PEFT+ZeRO-3 gradient checkpointing error vs disk-offloaded Qwen3 model save failure; not the same bug." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Metric gathering tuple-label truncation and Mamba2 cache correctness are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT masked-LM prediction regression vs general tokenization changes; same area loosely, but not the same underlying defect." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "MobileLLM key error on load and remote_code breakage are both loading-related, but they are different concrete failure modes." + }, + { + "left": "issue:47254", + "right": "issue:47332", + "accept": false, + "reason": "Gradient checkpointing/ZeRO-3 training error vs inability to save a disk-offloaded VL model." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "BLIP2 dtype handling vs an Apple Silicon allocator warmup TypeError; unrelated subsystems." + }, + { + "left": "issue:46697", + "right": "issue:46728", + "accept": false, + "reason": "Both are image-processor bugs, but one over-pads and the other miscounts patches; different algorithms and fixes." + }, + { + "left": "issue:42915", + "right": "issue:47333", + "accept": false, + "reason": "Qwen3MoE FineGrainedFP8 loading failure vs a disk-offloading save problem for a different Qwen3 model." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Float8 storage-object lookup error is unrelated to PIL backend unexpectedly requiring torchvision." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Qwen3-VL `weight_only=True` load error and Qwen2.5-VL rope index scaling are distinct model-specific bugs." + }, + { + "left": "issue:43262", + "right": "issue:47217", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and VL image-placeholder handling are different template/processor issues." + }, + { + "left": "issue:46851", + "right": "issue:47337", + "accept": false, + "reason": "Variable-length position_id behavior in Qwen3.5 is unrelated to a flaky bbox-coordinate test." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "GenerationMixin import failure and Jupyter custom-model initialization breakage are different compatibility issues." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio processor chat-template defaulting vs DAC latent conversion mismatch; no shared bug." + }, + { + "left": "issue:46175", + "right": "issue:46932", + "accept": false, + "reason": "Gemma4 vision patch-embedder one-hot memory blowup vs quantized embedders dtype casting; same model family but not the same defect." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug vs MXFP4 quantization fallback on a different model." + }, + { + "left": "issue:45003", + "right": "issue:46285", + "accept": false, + "reason": "Unsafe `sys.modules` access in modeling_utils and CLIP text-model structure changes are unrelated." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon M4 TypeError vs a CUDA float16 AMP crash in a loss module; different environments and causes." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "Importing `GenerationMixin` and a custom trainer kwarg mismatch are unrelated API breakages." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Zero-3/Bert index error during training and `trainer.evaluate()` failing after `train()` do not describe the same bug." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior is unrelated to PIL backend dependency requirements." + }, + { + "left": "issue:45372", + "right": "issue:46326", + "accept": false, + "reason": "Both mention mistral_common, but one is a missing import during Gemma4 processor loading and the other is a `continue_final_message` tokenizer bug." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "Torch.compile SDPA failure in Bamba-9B-v2 and MXFP4 fallback in git-oss-20b are different model/runtime problems." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "One is an incorrect runtime error message about translation tasks; the other is docs still referencing removed pipeline APIs. Related theme, not the same issue." + }, + { + "left": "issue:42175", + "right": "issue:46291", + "accept": false, + "reason": "Missing TensorFlow backend in a torch extra install and kernels-v0.15 import/version requirements are separate dependency issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 23, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29159", + "issue:34567", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42490", + "issue:42915", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43526", + "issue:43643", + "issue:43696", + "issue:43697", + "issue:43742", + "issue:43746", + "issue:43819", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:44060", + "issue:44291", + "issue:44368", + "issue:44387", + "issue:44479", + "issue:44554", + "issue:44589", + "issue:44610", + "issue:44617", + "issue:44811", + "issue:44821", + "issue:44912", + "issue:44977", + "issue:44991", + "issue:45042", + "issue:45081", + "issue:45084", + "issue:45406", + "issue:45538", + "issue:45600", + "issue:45759", + "issue:45859", + "issue:46209", + "issue:47020", + "issue:47050", + "issue:47333" + ], + "soft_pairs": [ + "issue:43452|issue:44991", + "issue:45081|issue:45406", + "issue:43257|issue:43329", + "issue:43819|issue:43927", + "issue:43906|issue:46209", + "issue:43526|issue:43697", + "issue:44060|issue:44368", + "issue:43854|issue:44977", + "issue:42915|issue:44821", + "issue:43746|issue:44811", + "issue:43257|issue:43696", + "issue:45084|issue:45538", + "issue:38175|issue:44610", + "issue:43262|issue:44811", + "issue:43643|issue:45759", + "issue:42490|issue:43257", + "issue:45859|issue:47050", + "issue:42915|issue:43742", + "issue:43329|issue:45600", + "issue:43329|issue:44479", + "issue:44291|issue:45042", + "issue:43526|issue:44554", + "issue:43329|issue:43450", + "issue:34567|issue:40990", + "issue:39692|issue:44912", + "issue:29159|issue:34567", + "issue:44387|issue:44617", + "issue:44589|issue:47333", + "issue:45081|issue:47020" + ], + "prepared_review_unit_hash": "06e69fae6510e7370f35037df03e3c850fcb6129b25c9ce107700e24713e83e1", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36670, + "estimated_input_tokens": 9168, + "estimated_eval_tokens": 18592 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft duplicate pairs should be merged here. The cluster is a grab-bag of unrelated tokenizer, multimodal, quantization, offloading, and model-loading bugs; several pairs share broad subsystem language but not the same concrete failure mode.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Both involve tokenizer loading, but they are different failure modes: gguf_file handling for AutoTokenizer/AutoModel vs a specific EMBEDDIA/est-roberta tokenizer regression." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Different code paths and symptoms: Mistral regex patching crash in tokenizer setup vs Gemma4Processor missing _tokenizer in transformers serve." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MOE weight conversion with accelerate/deepspeed is unrelated to a VLM video-token counting bug in _get_num_multimodal_tokens." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC latent reconstruction mismatch is unrelated to DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "The first is an isolated reproduction of another issue; the second is a B200 FP8 inference problem. Not the same bug." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels truncation and RTDetrV2 output differences are separate model/processor issues." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "One is an incorrect tied-weights warning in Qwen3-Next internals; the other is a user-facing warning during ms-swift LoRA fine-tuning. Same family, different problem." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "GLM-4.7-Flash load failure in tests is not the same as Qwen3.5 flash-attention generation issues." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "FineGrainedFP8Config failure for Qwen3Moe is unrelated to AutoImageProcessor URL loading." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "PEFT adapter checkpoint loading and Whisper batch_decode skip_special_tokens are different processor/model-loading bugs." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Both mention Qwen3 and loading/runtime issues, but one is MOE weight conversion with accelerate/deepspeed and the other is an outright CUDA OOM on GPT-oss-20b." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Template compilation error and CLIPTokenizer model_max_length misuse are unrelated." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "SigLIP2 zero-probabilities and OmDet-Turbo 640px-vs-224px preprocessing mismatch are different bugs." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and Whisper batch_decode token skipping are separate audio-processing issues." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Missing fields from trust_remote_code AutoConfig and not unwrapping text_config in composite multimodal configs are distinct config-loading problems." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Trailing-slash path import warning is unrelated to Qwen3 MOE weight conversion." + }, + { + "left": "issue:45859", + "right": "issue:47050", + "accept": false, + "reason": "CPU offload incompatibility in Qwen3_5MoeTextRotaryEmbedding is unrelated to a CanineModel fp16 ONNX dtype mismatch." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "FineGrainedFP8Config failure and MobileLLM-125M key error do not describe the same defect." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Video-token counting bug in multimodal processors is unrelated to auto_mappings.py referencing removed Sam3LiteText configs." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "The first is a missing/undefined helper in video token counting; the second is a broader v5.3.0 video-input regression across several Qwen models." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized TypeError and PIL backend requiring torchvision are separate compatibility issues." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "reduce_labels returning one label and an MPS attention correctness bug with differing head dims are unrelated." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Video token counting undefined names and batched video processor output shape are different bugs in adjacent areas, not one concrete failure." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "TrainerState token-count tracking and GPT-oss perplexity are unrelated training/runtime issues." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "SigLIP2 documentation errors do not match a GPT-oss MXFP4 quantization fallback problem." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer inconsistency and TrainerState num_input_tokens_seen not updating are different subsystems and symptoms." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both mention OOM, but one is int4 quantization reserved-memory growth and the other is Sam3Video running out of memory; no shared concrete root cause." + }, + { + "left": "issue:44589", + "right": "issue:47333", + "accept": false, + "reason": "Float8 storage lookup failure is not the same as save failures for disk-offloaded Qwen3-30B-A3B." + }, + { + "left": "issue:45081", + "right": "issue:47020", + "accept": false, + "reason": "Mistral regex patching crash and XLMRobertaTokenizer Unigram initialization are distinct tokenizer bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 24, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29127", + "issue:29159", + "issue:29870", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:42175", + "issue:42915", + "issue:43257", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43525", + "issue:43540", + "issue:43577", + "issue:43650", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43824", + "issue:43906", + "issue:43931", + "issue:44291", + "issue:44479", + "issue:44568", + "issue:44589", + "issue:44756", + "issue:44821", + "issue:44912", + "issue:44936", + "issue:44987", + "issue:45127", + "issue:45356", + "issue:45357", + "issue:45459", + "issue:45538", + "issue:45803", + "issue:46291", + "issue:46421", + "issue:46424", + "issue:46693", + "issue:46701", + "issue:46821", + "issue:47145", + "issue:47332" + ], + "soft_pairs": [ + "issue:43257|issue:43931", + "issue:44987|issue:45538", + "issue:43425|issue:46291", + "issue:43525|issue:46291", + "issue:45459|issue:45803", + "issue:44912|issue:47145", + "issue:28282|issue:29870", + "issue:43650|issue:43906", + "issue:28282|issue:34689", + "issue:33357|issue:34567", + "issue:43540|issue:46701", + "issue:44568|issue:45356", + "issue:43454|issue:43577", + "issue:44291|issue:44821", + "issue:46421|issue:46424", + "issue:43742|issue:44291", + "issue:43749|issue:44756", + "issue:44936|issue:46821", + "issue:42175|issue:43824", + "issue:24643|issue:29159", + "issue:44589|issue:47332", + "issue:28282|issue:29127", + "issue:44479|issue:46693", + "issue:43404|issue:45127", + "issue:42915|issue:45538", + "issue:43697|issue:45357" + ], + "prepared_review_unit_hash": "19cce58d12b678c05b46c1b14eb749ae30488625a1cd980522347229ac290c68", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34982, + "estimated_input_tokens": 8746, + "estimated_eval_tokens": 17748 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: most pairs are only loosely similar by theme (loading, tokenizers, quantization, multimodal, cache, trainer issues) but do not look like the same underlying bug. I would not merge any of the soft-edge pairs.", + "confidence": 0.63, + "canonical_issue_reason": "No single issue cleanly represents a duplicate set here; if a representative must be chosen, issue 43749 is the broadest/highest-activity infrastructure regression and most plausible cluster anchor.", + "canonical_pr_reason": "No pull requests are present in the cluster.", + "best_issue_reason": "Issue 43749 is the strongest representative candidate because it has the most discussion/inbound references and describes a broad runtime regression, making it a better cluster anchor than the highly model-specific reports.", + "best_pr_reason": "No pull requests are present, so there is no best PR candidate.", + "soft_edge_verdicts": [ + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Different models and different failure modes: Qwen3 MOE conversion vs Qwen3-VL weight-shape loading mismatch." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated bugs: one is a model-loading failure on a specific repo, the other is tokenizer max-length handling." + }, + { + "left": "issue:43425", + "right": "issue:46291", + "accept": false, + "reason": "Torch compatibility and pipeline import/version requirements are separate issues with different root causes." + }, + { + "left": "issue:43525", + "right": "issue:46291", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config is unrelated to kernels dependency validation on pipeline import." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both mention broad exception handling, but they are distinct code paths and distinct bugs; not the same change." + }, + { + "left": "issue:44912", + "right": "issue:47145", + "accept": false, + "reason": "Both concern quantized checkpoint loading, but one is MXFP4 fallback behavior and the other is FP8 weight_scale omission." + }, + { + "left": "issue:28282", + "right": "issue:29870", + "accept": false, + "reason": "PyTorch-not-found import error is unrelated to Bart generate() decoder_attention_mask assignment." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "A vague issue placeholder is not the same as a reproduction of a specific prior bug." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Generic PyTorch import error is not the same as a model-loading regression for Llama 3.2 Vision." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "MacOS CLIP bus error and TrainerState token-count tracking are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:46701", + "accept": false, + "reason": "Different multimodal failures in different models and pipelines; no shared concrete code path." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different tokenizers and stem from different causes." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "One is broken lm_head weight tying; the other is dtype propagation for Blip2/QFormer." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "init_empty_weights argument handling and AutoImageProcessor URL loading are unrelated." + }, + { + "left": "issue:46421", + "right": "issue:46424", + "accept": false, + "reason": "Both are cache-related, but one is recompilation/lazy init and the other is cache_length configuration semantics." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "KeyError during model loading is not the same as init_empty_weights rejecting _is_hf_initialized." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU/RAM-efficient loading and Strix Halo mmap OOM are different loading/memory issues." + }, + { + "left": "issue:44936", + "right": "issue:46821", + "accept": false, + "reason": "Trainer train/evaluate sequencing bug is unrelated to TimesFM window_size AttributeError." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Backend packaging/dependency issue and missing Qwen2.5-VL export are unrelated." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed weight-dimension error is unrelated to tokenizer slow/fast inconsistency." + }, + { + "left": "issue:44589", + "right": "issue:47332", + "accept": false, + "reason": "FP8 storage lookup and disk-offloaded save failure are different bugs." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "Missing PyTorch import and LayoutLMv3 error messaging are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:46693", + "accept": false, + "reason": "Both are regressions, but one is video-input behavior across Qwen VL models and the other is FA2 long-context performance." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both involve tied embeddings, but one is a specific model weight-tying bug and the other is LoRA merge behavior with extended vocab." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "Qwen3Moe FP8 config failure is unrelated to CLIPTokenizer model_max_length." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetr output differences in v5 and Qwen3.5 visual encoder key saving are unrelated regressions." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 25, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:27957", + "issue:28282", + "issue:34567", + "issue:36331", + "issue:38175", + "issue:39379", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43452", + "issue:43493", + "issue:43531", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43742", + "issue:43749", + "issue:43994", + "issue:44062", + "issue:44479", + "issue:44488", + "issue:44554", + "issue:44560", + "issue:44568", + "issue:44589", + "issue:44792", + "issue:44821", + "issue:44849", + "issue:44857", + "issue:44898", + "issue:44964", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45081", + "issue:45137", + "issue:45381", + "issue:45405", + "issue:45488", + "issue:45538", + "issue:45701", + "issue:45907", + "issue:46512", + "issue:46832", + "issue:46869", + "issue:47254", + "issue:47332", + "issue:47333" + ], + "soft_pairs": [ + "issue:45137|issue:45405", + "issue:44062|issue:44991", + "issue:43582|issue:47254", + "issue:43531|issue:44821", + "issue:44568|issue:45488", + "issue:39379|issue:44554", + "issue:44857|issue:46832", + "issue:45538|issue:46869", + "issue:42915|issue:43749", + "issue:27957|issue:28282", + "issue:43540|issue:44479", + "issue:43540|issue:44560", + "issue:41720|issue:43540", + "issue:34567|issue:36331", + "issue:43452|issue:43582", + "issue:44062|issue:45081", + "issue:42175|issue:44488", + "issue:44479|issue:45381", + "issue:45042|issue:45081", + "issue:42915|issue:47332", + "issue:43577|issue:46512", + "issue:44987|issue:45907", + "issue:43493|issue:43994", + "issue:38175|issue:44898", + "issue:43742|issue:44589", + "issue:44792|issue:44964", + "issue:44849|issue:45701", + "issue:43742|issue:47333" + ], + "prepared_review_unit_hash": "a338573a5ea32be8091588e79bbf09b766fc87c1ab1d5fe41e4323ca7936dde8", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35923, + "estimated_input_tokens": 8981, + "estimated_eval_tokens": 18218 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: tokenizer, dtype, multimodal, training, and device-mapping bugs are all mixed together. I only see two plausible duplicate groups: the Qwen video-input regression pair and the SigLIP2 output-discrepancy pair.", + "confidence": 0.88, + "canonical_issue_reason": "issue:44479 is the broadest and clearest representative of a real duplicate set here: it frames the Qwen video-input regression across multiple variants, while 45381 is a narrower manifestation of the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the best overall issue representative because it describes the widest-impact regression in the set and cleanly subsumes a more specific duplicate report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "Different problems: DeepSpeed ZeRO3 deque handling vs a PEFT version bump." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "Both involve tokenizers, but one is an AddedToken constructor error and the other is a tokenizer load failure for a specific model." + }, + { + "left": "issue:43582", + "right": "issue:47254", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError is unrelated to PEFT + ZeRO-3 gradient-checkpointing checkpoint errors." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding-window handling and AutoImageProcessor-from-URL loading are different code paths and different subsystems." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different tokenizer implementations and distinct failure modes." + }, + { + "left": "issue:39379", + "right": "issue:44554", + "accept": false, + "reason": "mllama mask handling is not the same as the MPS attention correctness issue with mismatched value/query head dims." + }, + { + "left": "issue:44857", + "right": "issue:46832", + "accept": false, + "reason": "RT-DETR AMP crash and RTDetrModel feature-level mismatch are separate bugs." + }, + { + "left": "issue:45538", + "right": "issue:46869", + "accept": false, + "reason": "CLIPTokenizer max-length behavior is unrelated to YOLO auto device-mapping failures." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "FineGrainedFP8Config on Qwen3-MoE and FSDP CPU RAM efficient loading are different loading/training paths." + }, + { + "left": "issue:27957", + "right": "issue:28282", + "accept": false, + "reason": "XLMRoberta Flash Attention 2 support is unrelated to missing-PyTorch ImportError." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are Qwen video-input issues, but they involve different model families and different concrete failures, so they are not clearly the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Qwen3OmniMoe video ValueError and Qwen3-vl-embedding StopIteration are different symptoms on different code paths." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Auto device mapping on Qwen3 and Qwen3OmniMoe video-processing errors are unrelated." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "TrainerState token-count tracking and a custom compute_loss kwarg mismatch are separate trainer bugs." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "gguf_file loading failures and Apple Silicon allocator TypeErrors do not share a concrete code path." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "AddedToken special-argument duplication is not the same as the Mistral regex patch crashing on a missing backend_tokenizer." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Packaging/backend dependency inclusion is unrelated to the sleng-bert loading failure." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Same underlying Qwen video-input regression in transformers 5.3.0; 45381 is a specific qwen2.5-vl manifestation of the broader bug in 44479." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Image-processor torchvision dependency regression and Mistral tokenizer regex patch failure are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:47332", + "accept": false, + "reason": "Qwen3-MoE FP8 configuration failure and disk-offloaded model save failure are distinct issues." + }, + { + "left": "issue:43577", + "right": "issue:46512", + "accept": false, + "reason": "Both mention dtype, but one is model/qformer weights not casting and the other is config.dtype going stale after from_config." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model-loading regression for physical-intelligence/fast is unrelated to forwarding revision in list_repo_files." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "Both describe the same SigLIP2 malfunction on google/siglip2-base-patch16-224: one as HF-vs-JAX discrepancy, the other as nonsensical AutoModel/pipeline outputs." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "SigLIP2 zero-probability behavior and Perceiver interpolate_pos_encoding failure are different model families and bugs." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "MobileLLM key error and Float8 storage lookup failure are unrelated loading/runtime errors." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Janus image-generation test failure and Phi-4 multimodal loading failure are different model-specific problems." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 output_hidden_states behavior is unrelated to version-dependent tokenization changes." + }, + { + "left": "issue:43742", + "right": "issue:47333", + "accept": false, + "reason": "MobileLLM key error and Qwen3 offloaded-save failure are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The two accepted duplicate pairs are the strongest matches by title/body theme, and the rejected soft pairs are mostly separated by model family or bug class. No clear overstatement of evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 26, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29870", + "issue:32090", + "issue:41762", + "issue:42796", + "issue:43257", + "issue:43452", + "issue:43540", + "issue:43577", + "issue:43653", + "issue:43697", + "issue:43716", + "issue:43742", + "issue:43756", + "issue:43906", + "issue:44361", + "issue:44466", + "issue:44488", + "issue:44568", + "issue:44625", + "issue:44737", + "issue:44843", + "issue:44877", + "issue:44913", + "issue:44936", + "issue:45005", + "issue:45072", + "issue:45325", + "issue:45405", + "issue:45588", + "issue:45600", + "issue:45663", + "issue:45672", + "issue:45674", + "issue:45823", + "issue:46121", + "issue:46399", + "issue:46463", + "issue:46491", + "issue:46693", + "issue:46701", + "issue:46829", + "issue:46846", + "issue:46899", + "issue:47276", + "issue:47332", + "issue:47405" + ], + "soft_pairs": [ + "issue:43756|issue:45325", + "issue:43257|issue:43756", + "issue:43756|issue:46399", + "issue:43716|issue:46701", + "issue:45823|issue:46463", + "issue:45672|issue:45674", + "issue:46463|issue:47276", + "issue:44737|issue:46463", + "issue:45405|issue:45600", + "issue:43906|issue:46693", + "issue:43257|issue:47405", + "issue:24643|issue:29870", + "issue:43577|issue:45072", + "issue:44568|issue:46491", + "issue:44625|issue:44877", + "issue:43540|issue:46693", + "issue:43452|issue:44488", + "issue:43742|issue:47332", + "issue:28282|issue:32090", + "issue:42796|issue:43906", + "issue:44466|issue:45005", + "issue:41762|issue:44936", + "issue:46463|issue:46899", + "issue:43452|issue:44843", + "issue:44913|issue:46121", + "issue:43697|issue:46829", + "issue:45663|issue:46846", + "issue:43653|issue:44361", + "issue:44737|issue:45588" + ], + "prepared_review_unit_hash": "c89520898d5d8ded27a3311c0765590f2e026b594ff11d3cff19f0ec61e2449c", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 37028, + "estimated_input_tokens": 9257, + "estimated_eval_tokens": 18770 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a heterogeneous set of unrelated issues. The only duplicate-like pair is the v5 tied-weights regression on translation models; the rest differ by model, subsystem, or failure mode and should stay separate.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 45005 is the broadest representative of the tied-weights v5 regression, and 44466 reads like a concrete device-dependent repro of that same problem.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45005 is the best overall issue candidate because it captures the underlying tied-weights regression more broadly than the specific lm_head.weight serialization symptom in 44466.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Both touch model internals for different families, but one is Smollm3 RoPE-layer selection and the other is Qwen2.5-VL temporal position scaling; different bugs." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MoE weight-conversion/loading issue vs Smollm3 RoPE-layer dropping; unrelated code paths and symptoms." + }, + { + "left": "issue:43756", + "right": "issue:46399", + "accept": false, + "reason": "Smollm3 RoPE configuration bug and Qwen3.5 kernel-registration crash are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:46701", + "accept": false, + "reason": "Different models and failures: Mistral-3 dtype mismatch in preprocessing vs DiffusionGemma pipeline generation breakage." + }, + { + "left": "issue:45823", + "right": "issue:46463", + "accept": false, + "reason": "Gemma4 device-map forward mismatch and T5 decoder-input validation are not the same bug." + }, + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Both involve 4-bit inference, but one is a uint8 finfo crash in Gemma4 audio modules and the other is a BitsAndBytesConfig exclusion regression; distinct fixes." + }, + { + "left": "issue:46463", + "right": "issue:47276", + "accept": false, + "reason": "Incorrect decoder-input validation/error message vs a synced_gpus warning in continuous batching; unrelated." + }, + { + "left": "issue:44737", + "right": "issue:46463", + "accept": false, + "reason": "XLNet CPU device placement bug vs T5 validation logic; no shared underlying defect." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "PEFT version bump problem and stale auto_mappings config references are unrelated maintenance issues." + }, + { + "left": "issue:43906", + "right": "issue:46693", + "accept": false, + "reason": "One is an isolated reproduction of an older issue; the other is an FA2 long-context performance regression. Not the same bug." + }, + { + "left": "issue:43257", + "right": "issue:47405", + "accept": false, + "reason": "Qwen3 MoE weight conversion problem vs silent wrong-Auto-class conversion skip; different loading failures." + }, + { + "left": "issue:24643", + "right": "issue:29870", + "accept": false, + "reason": "DeepSpeed training RuntimeError about weight dimensionality and Bart generate unbound decoder_attention_mask are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype, but they affect different model families and code paths; not one concrete bug." + }, + { + "left": "issue:44568", + "right": "issue:46491", + "accept": false, + "reason": "Different tokenizer regressions: missing BOS/EOS addition vs leading-space round-trip stripping." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and strict config loading for granite_speech are separate config issues." + }, + { + "left": "issue:43540", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3OmniMoe video-input value error vs FA2 performance regression; unrelated." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "Both are model/tokenizer loading complaints, but the gguf_file breakage and sleng-bert loading failure have different causes." + }, + { + "left": "issue:43742", + "right": "issue:47332", + "accept": false, + "reason": "MobileLLM loading KeyError vs inability to save a disk-offloaded Qwen3-VL model; different subsystems." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "Missing PyTorch import error and Trainer broadcast NoneType error are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43906", + "accept": false, + "reason": "Tokenizer BPE-marker handling and an isolated reproduction of another issue are not the same underlying bug." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": true, + "reason": "Same v5 tied-weights regression: 44466 is a concrete device-dependent serialization symptom, while 45005 describes the broader tied-weights problem on translation models." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "ZeRO-3 load-time index error for Gemma3 and trainer.evaluate() failing after training are unrelated." + }, + { + "left": "issue:46463", + "right": "issue:46899", + "accept": false, + "reason": "T5 decoder validation and Gemma4 4-bit multimodal dtype casting are different bugs." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "gguf_file loading breakage and offline model_info() patching are separate from_pretrained issues with different causes." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "Both involve RoPE config handling, but one is persistence of rotary_pct on reload and the other is a TypeError in validation input handling." + }, + { + "left": "issue:43697", + "right": "issue:46829", + "accept": false, + "reason": "Different vision-model output mismatches with different outputs being wrong; no shared code path." + }, + { + "left": "issue:45663", + "right": "issue:46846", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Qwen3.5 tensor-parallel plan omission are unrelated distributed-training bugs." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "Both are tokenizer-related, but one is special-token registration/empty decode and the other is an AttributeError in MLuke tasks." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU arange bug and flash_attention s_aux=None crash are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The judgment is grounded in the packet: the one accepted pair, 44466/45005, is a plausible shared v5 tied-weights regression, and the rest of the pairwise rejections stay conservative by distinguishing model, subsystem, or failure mode. The overall summary does not overclaim beyond the available titles and pair signals." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 27, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29159", + "issue:33357", + "issue:33666", + "issue:36010", + "issue:38175", + "issue:40444", + "issue:43295", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43749", + "issue:43761", + "issue:43927", + "issue:44361", + "issue:44442", + "issue:44493", + "issue:44568", + "issue:44756", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44857", + "issue:44898", + "issue:44991", + "issue:45216", + "issue:45356", + "issue:45357", + "issue:45372", + "issue:45375", + "issue:45542", + "issue:45701", + "issue:45704", + "issue:45920", + "issue:46050", + "issue:46170", + "issue:46178", + "issue:46701", + "issue:46821", + "issue:46846", + "issue:47000", + "issue:47013", + "issue:47332", + "issue:47404" + ], + "soft_pairs": [ + "issue:43720|issue:44898", + "issue:44821|issue:46821", + "issue:38175|issue:43295", + "issue:44493|issue:46050", + "issue:46846|issue:47404", + "issue:44361|issue:44442", + "issue:44756|issue:47404", + "issue:43697|issue:45216", + "issue:45357|issue:45920", + "issue:29159|issue:33357", + "issue:43688|issue:43749", + "issue:40444|issue:44811", + "issue:44756|issue:47332", + "issue:43643|issue:45375", + "issue:44843|issue:44991", + "issue:45372|issue:46170", + "issue:44857|issue:47000", + "issue:43295|issue:45356", + "issue:43688|issue:45704", + "issue:43688|issue:45701", + "issue:45542|issue:46178", + "issue:43525|issue:46701", + "issue:43653|issue:44568", + "issue:43646|issue:43749", + "issue:43653|issue:43927", + "issue:43761|issue:47013", + "issue:33666|issue:36010" + ], + "prepared_review_unit_hash": "5dcf718b4f4434fac96221ff50483b3829b1693a433c0770cdc8474cafd70267", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35493, + "estimated_input_tokens": 8874, + "estimated_eval_tokens": 18004 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issue reports that only share broad subsystem vocabulary (tokenizer, OOM, regression, processor). None of the soft pairs look like the same underlying bug or change, so I would not merge any of them as duplicates. The most representative issue overall is the broad tokenizer regression affecting multiple models.", + "confidence": 0.62, + "canonical_issue_reason": "issue:45920 is the broadest and clearest tokenizer regression here: it describes wrong token IDs across multiple models, not a narrow model-specific symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the best single issue to keep as the canonical reference because it captures a widespread tokenizer regression with multiple affected models and a concrete failure mode.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "Different model families and failures: BitNet loading/unpacking vs Perceiver interpolation/output mismatch." + }, + { + "left": "issue:44821", + "right": "issue:46821", + "accept": false, + "reason": "Unrelated bugs: AutoImageProcessor URL loading vs TimesFM window_size AttributeError." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Both mention processor/tokenizer, but one is SigLIP2 zero probabilities and the other is a regression in custom model code passing images to tokenizer." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Different root causes: unexpected key warnings with position ids vs quantized model save/load breakage." + }, + { + "left": "issue:46846", + "right": "issue:47404", + "accept": false, + "reason": "Both are memory/TP-related, but one is missing TP plan entries and the other is device_map loading OOM from conversion transients." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer failures for different tokenizers: MLuke AttributeError vs AutoTokenizer failing on FastSpeech2ConformerTokenizer." + }, + { + "left": "issue:44756", + "right": "issue:47404", + "accept": false, + "reason": "Both involve OOM, but one is mmap on Strix Halo and the other is device_map loading/conversion-op transients." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "Distinct regressions: RTDetrV2 output drift vs Qwen3.5 save_pretrained checkpoint corruption." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Related to Qwen/Tokenizer save/load behavior, but one is visual encoder key saving and the other is wrong token IDs across many models." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer consistency bug vs MacOS bus error on CLIP; no shared underlying failure." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Different subsystems: auxiliary loss normalization vs FSDP CPU RAM efficient loading." + }, + { + "left": "issue:40444", + "right": "issue:44811", + "accept": false, + "reason": "Different features: multi-image VL finetuning failure vs Whisper batch_decode skip_special_tokens bug." + }, + { + "left": "issue:44756", + "right": "issue:47332", + "accept": false, + "reason": "Both involve memory/offloading, but the causes differ: mmap OOM vs inability to save a disk-offloaded model." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both concern config fields, but one is generic trust_remote_code missing fields and the other is a specific Qwen3_5MoeVisionConfig field dropped by @strict." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline model_info patch bug vs a tokenizer load failure for EMBEDDIA/est-roberta; different code paths and symptoms." + }, + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Different dependency/import problems: mistral_common ReasoningEffort import vs Windows ROCm torch.distributed dependency issues." + }, + { + "left": "issue:44857", + "right": "issue:47000", + "accept": false, + "reason": "Different RTDetr issues: float16 AMP crash vs infeasible cost matrix in RTDetrLoss." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Different regression targets: processor.tokenizer/images handling vs Kimi-K2.5 codec handling and mistral regex warning." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Unrelated training/model bugs: auxiliary loss normalization vs T5 apex.FusedRMSNorm memory leak." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Different bug classes: auxiliary loss normalization vs tokenization changes across versions." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Both are backend/import errors, but they affect different missing backends and code paths." + }, + { + "left": "issue:43525", + "right": "issue:46701", + "accept": false, + "reason": "Llama4Config pad_token_id attribute error is unrelated to DiffusionGemma image-text pipeline failure." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both involve special tokens, but one is BigBird mask token registration and the other is mdeberta-v3 BOS/EOS insertion behavior." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Generic custom model initialization breakage vs FSDP CPU RAM efficient loading; different failure modes." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "Different token/config issues: BigBird special-token decode bug vs DiaConfig losing custom token IDs on save/load." + }, + { + "left": "issue:43761", + "right": "issue:47013", + "accept": false, + "reason": "Different model/runtime problems: CLIPVision hidden_states regression vs xLSTM train-mode state unpacking crash." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Multi-GPU Qwen2-VL training request vs GenerationMixin import error; not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 28, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28218", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:36331", + "issue:41720", + "issue:42898", + "issue:43278", + "issue:43452", + "issue:43504", + "issue:43525", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43643", + "issue:43688", + "issue:43720", + "issue:43756", + "issue:43761", + "issue:43782", + "issue:43927", + "issue:44361", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44484", + "issue:44554", + "issue:44561", + "issue:44610", + "issue:44625", + "issue:44704", + "issue:44734", + "issue:44743", + "issue:44977", + "issue:45405", + "issue:45440", + "issue:46050", + "issue:46402", + "issue:46491", + "issue:46612", + "issue:46693", + "issue:46829", + "issue:46832", + "issue:46932", + "issue:47254", + "issue:47276" + ], + "soft_pairs": [ + "issue:43756|issue:46402", + "issue:44464|issue:44743", + "issue:46832|issue:46932", + "issue:44479|issue:44625", + "issue:43761|issue:46829", + "issue:43550|issue:44361", + "issue:43452|issue:43577", + "issue:42898|issue:46491", + "issue:33666|issue:36331", + "issue:41720|issue:44464", + "issue:32090|issue:33666", + "issue:44466|issue:46050", + "issue:44977|issue:46693", + "issue:43504|issue:43550", + "issue:44464|issue:44977", + "issue:45405|issue:47254", + "issue:43278|issue:43688", + "issue:43782|issue:46693", + "issue:43643|issue:44561", + "issue:43688|issue:45440", + "issue:43504|issue:43720", + "issue:43720|issue:43927", + "issue:43606|issue:43927", + "issue:44734|issue:46612", + "issue:44625|issue:47276", + "issue:43504|issue:44610", + "issue:43720|issue:44361", + "issue:44484|issue:44704", + "issue:43525|issue:43572", + "issue:43577|issue:44554", + "issue:28218|issue:33357" + ], + "prepared_review_unit_hash": "938ca0e0d75da8d1ca4d16206152a5087f18c4b0d1df88199ea5aae7f3dd0de6", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36542, + "estimated_input_tokens": 9136, + "estimated_eval_tokens": 18528 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly noisy and spans unrelated bug reports. The only clearly on-theme item is the tokenizer spacing report, so I\u2019d anchor the cluster on issue #28218; no PRs are present.", + "confidence": 0.76, + "canonical_issue_reason": "#28218 is the broadest and most direct match for the tokenizer-space bug class, and its title matches the shared underlying symptom better than the more model-specific variants.", + "canonical_pr_reason": null, + "best_issue_reason": "#28218 is the best representative issue because it is the clearest, most general statement of the tokenizer spacing problem and can subsume related whitespace-regression reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43756", + "right": "issue:46402", + "accept": false, + "reason": "Unrelated model/blog RoPE-layer discrepancy vs. missing text-encoder weights; only superficial transformer/model similarity." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Both involve generation/caching, but one is chunked generation under compiled forward and the other is recurrent-state reset in modular Qwen3.5; different code paths and symptoms." + }, + { + "left": "issue:46832", + "right": "issue:46932", + "accept": false, + "reason": "Both concern RT-DETR/Gemma4 dtype or model wiring issues, but they are different models and different failure modes." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression across Qwen VL variants vs. num_labels propagation in Qwen3.5; not the same bug." + }, + { + "left": "issue:43761", + "right": "issue:46829", + "accept": false, + "reason": "Different outputs are wrong in different classes: CLIPVisionModel hidden_states regression vs. VideoPrism output field mismatch." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "One is a torch.compile/SDPA failure in Bamba, the other is a tokenizer AttributeError in MLukeTokenizer; unrelated." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "GGUF loading/tokenizer path breakage vs. Blip2 dtype loading bug; different subsystems and fixes." + }, + { + "left": "issue:42898", + "right": "issue:46491", + "accept": false, + "reason": "Both mention tokenization spacing, but 42898 is a v5 cleanup-behavior change while 46491 is a CodeLlama round-trip regression; not clearly the same concrete bug." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Multi-GPU training for Qwen2-VL vs. Trainer custom compute_loss signature mismatch; unrelated." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Qwen3 auto device mapping cuda assert vs. compiled-forward generation inconsistency; different failure mechanisms." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error vs. Qwen2-VL multi-GPU training; both distributed-training themed but not the same bug." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Serialization of lm_head weights vs. sinq quantized save/load corruption; both persistence-related but distinct problems." + }, + { + "left": "issue:44977", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug vs. FA2 long-context performance regression; same model family but different behavior and symptom." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Legacy-field model loading bug vs. torch.compile SDPA failure; different model-loading code paths." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency is not the same concrete issue as flash-attention generation failure." + }, + { + "left": "issue:45405", + "right": "issue:47254", + "accept": false, + "reason": "PEFT version pin complaint vs. checkpoint error with PEFT+DeepSpeed; related dependency, but not the same underlying bug." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Both concern tensor dtypes or normalization, but one is BF16/FP32 eval drift and the other is auxiliary-loss normalization; unrelated." + }, + { + "left": "issue:43782", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3VL weight_only load error vs. long-context FA2 performance regression; different models and failure modes." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "AutoConfig missing fields with trust_remote_code vs. trust_remote_code breakage from removed is_torch_fx_available; different root causes." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Aux-loss normalization bug vs. DeepseekV3MoE divergence from remote implementation; both model-accuracy issues but not the same code-path problem." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Legacy field loading for BEiT vs. BitNet packed-weights unpacking during accelerate loading; unrelated loading bugs." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet accelerate loading issue vs. tokenizer task AttributeError; entirely different components." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU offload device mismatch in bark-small vs. DiaConfig losing custom token IDs on save/load; both config/load-related but not the same bug." + }, + { + "left": "issue:44734", + "right": "issue:46612", + "accept": false, + "reason": "KV-cache continuation crash in serve vs. beam-search cache reorder skipped for several models; both cache-related but different execution paths and symptoms." + }, + { + "left": "issue:44625", + "right": "issue:47276", + "accept": false, + "reason": "Qwen3.5 num_labels propagation bug vs. misleading synced_gpus warning; unrelated." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Legacy preset loading bug vs. processor image-size mismatch; different model families and failures." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "Packed-weight loading bug vs. MLukeTokenizer task AttributeError; unrelated." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "Default max_shard_size question vs. AutoProcessor not passing kwargs to cached_file; not the same bug or fix." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config vs. missing pad_token_idx on StableLmConfig; similar naming, but different configs and likely separate omissions." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "Blip2 dtype remains float32 vs. MPS attention correctness issue; unrelated." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer adds an extra space after an added token vs. MacOS bus error loading CLIP; completely unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 29, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:39692", + "issue:42796", + "issue:43257", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43572", + "issue:43606", + "issue:43611", + "issue:43646", + "issue:43701", + "issue:43824", + "issue:43825", + "issue:43844", + "issue:43873", + "issue:43927", + "issue:43994", + "issue:44291", + "issue:44361", + "issue:44466", + "issue:44488", + "issue:44514", + "issue:44554", + "issue:44568", + "issue:44589", + "issue:44843", + "issue:44964", + "issue:44977", + "issue:45290", + "issue:45405", + "issue:45491", + "issue:45520", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46002", + "issue:46050", + "issue:46192", + "issue:46209", + "issue:46365", + "issue:46489", + "issue:46710", + "issue:46846", + "issue:47332", + "issue:47404" + ], + "soft_pairs": [ + "issue:44514|issue:45290", + "issue:44964|issue:45405", + "issue:43611|issue:46846", + "issue:45823|issue:46050", + "issue:44977|issue:46209", + "issue:39692|issue:43295", + "issue:43493|issue:43825", + "issue:43334|issue:45759", + "issue:44488|issue:44843", + "issue:43334|issue:43572", + "issue:43994|issue:46192", + "issue:42796|issue:44977", + "issue:45405|issue:47332", + "issue:43257|issue:47404", + "issue:44554|issue:46365", + "issue:45491|issue:46002", + "issue:43646|issue:44466", + "issue:44568|issue:46489", + "issue:43278|issue:43701", + "issue:43572|issue:44291", + "issue:44568|issue:45812", + "issue:43927|issue:44361", + "issue:43493|issue:43844", + "issue:46209|issue:46710", + "issue:43278|issue:43844", + "issue:43873|issue:47404", + "issue:43504|issue:43606", + "issue:44589|issue:45520", + "issue:43824|issue:44589" + ], + "prepared_review_unit_hash": "31adae51167a9d999f69e43e3e042a23d49200340b917a839042daa7da8bf9c8", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36175, + "estimated_input_tokens": 9044, + "estimated_eval_tokens": 18344 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is theme-based rather than duplicate-based: most items are distinct Transformers regressions across tokenizers, multimodal processors, loading, quantization, offloading, and v5 compatibility. None of the soft pairs look like the same underlying bug/change with enough specificity to merge.", + "confidence": 0.22, + "canonical_issue_reason": "If one issue must represent the cluster, issue 43646 is the broadest high-level regression report (custom model initialization breaks in Transformers 5.0.0) and is closest to the cluster\u2019s general v5-breakage theme.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43646 is the best single representative because it is the most generic and broadly applicable of the v5 regression reports in this set, though the overall cluster is still heterogeneous.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both involve apply_chat_template/processor code, but they are different failure modes: batched Qwen2.5-VL padding vs tool-call assistant messages." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "A model-loading failure for Phi-4 multimodal is unrelated to a PEFT version bump issue." + }, + { + "left": "issue:43611", + "right": "issue:46846", + "accept": false, + "reason": "Different bugs: base_model_prefix loading regression vs missing linear_attn TP plan entries causing OOM/shape errors." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Gemma4 device-mismatch with auto device_map is not the same as sinq quantization save/load breakage." + }, + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 flash-attn generation issues and FP8/B200 inference failures are separate code paths and hardware-specific problems." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Both mention processor/tokenizer usage, but one is a SigLIP2 docs example and the other is a regression in processor.tokenizer access with image passing." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation divergence and a pipeline error message for translation tasks are unrelated." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Both are multimodal Qwen config issues, but one is a missing pad_token_id attribute and the other is text_config unwrapping in from_config." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Loading a specific model failing is unrelated to AutoTokenizer's offline-breakage in _patch_mistral_regex." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Different missing config fields for different models: pad_token_id in Qwen3-VL vs pad_token_idx in StableLmConfig." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 wrong outputs and return_tensors being ignored are different API/behavior bugs." + }, + { + "left": "issue:42796", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer BPE marker handling for Ministral is unrelated to Qwen3.5 flash-attention generation." + }, + { + "left": "issue:45405", + "right": "issue:47332", + "accept": false, + "reason": "A dependency version bump and a disk-offloaded model save failure are unrelated." + }, + { + "left": "issue:43257", + "right": "issue:47404", + "accept": false, + "reason": "Qwen3 MOE weight conversion with accelerate/deepspeed is a different loading path from device_map OOMs during transient conversion ops." + }, + { + "left": "issue:44554", + "right": "issue:46365", + "accept": false, + "reason": "MPS attention correctness and Trainer import conflicts with datasets are unrelated." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Both mention NaNs, but one is Gemma3 sliding-window attention on GPU and the other is classification head initialization/checkpoint behavior." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage is not the same as inconsistent lm_head.weight serialization tied to device." + }, + { + "left": "issue:44568", + "right": "issue:46489", + "accept": false, + "reason": "These are separate tokenizer regressions affecting different model families and likely different causes." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype drift during evaluation is unrelated to resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Missing pad_token_idx in a config and init_empty_weights receiving an unexpected argument are different v5 compatibility issues." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and symptoms; not enough evidence of one underlying bug." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig save/load token-ID loss and MLukeTokenizer task AttributeError are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 JAX/HF discrepancy and a ZeRO-3 gradient increase issue do not share a concrete code path." + }, + { + "left": "issue:46209", + "right": "issue:46710", + "accept": false, + "reason": "FP8 inference on B200 and a post-upgrade accuracy regression are too different to be the same bug." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype changes in evaluation are unrelated to ZeRO-3 gradient anomalies." + }, + { + "left": "issue:43873", + "right": "issue:47404", + "accept": false, + "reason": "Quantization offloading behavior and device_map OOM from conversion transients are related only at a broad subsystem level, not the same bug." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "A legacy-field model loading failure and a CPU-offload device mismatch are separate issues." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure and a flash_attn import KeyError on Python 3.13 are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "Qwen2.5-VL import error and Float8 storage resolution failure are different problems in different areas." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 30, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:32090", + "issue:34567", + "issue:37428", + "issue:38175", + "issue:41720", + "issue:42490", + "issue:42947", + "issue:43381", + "issue:43493", + "issue:43504", + "issue:43606", + "issue:43653", + "issue:43696", + "issue:43701", + "issue:43844", + "issue:43873", + "issue:43957", + "issue:43994", + "issue:44112", + "issue:44423", + "issue:44464", + "issue:44493", + "issue:44734", + "issue:44849", + "issue:44964", + "issue:45072", + "issue:45084", + "issue:45200", + "issue:45229", + "issue:45245", + "issue:45357", + "issue:45520", + "issue:45584", + "issue:45588", + "issue:45823", + "issue:46050", + "issue:46209", + "issue:46285", + "issue:46710", + "issue:46821", + "issue:46832", + "issue:47013", + "issue:47145", + "issue:47254" + ], + "soft_pairs": [ + "issue:46821|issue:46832", + "issue:43504|issue:43653", + "issue:44493|issue:46285", + "issue:43994|issue:44493", + "issue:43606|issue:44112", + "issue:15354|issue:38175", + "issue:45245|issue:46710", + "issue:42490|issue:43696", + "issue:45072|issue:46832", + "issue:43873|issue:46050", + "issue:44464|issue:45584", + "issue:44849|issue:46209", + "issue:44849|issue:45357", + "issue:28218|issue:34567", + "issue:42947|issue:43381", + "issue:29159|issue:32090", + "issue:43844|issue:44964", + "issue:45084|issue:45520", + "issue:47013|issue:47254", + "issue:45229|issue:45823", + "issue:44423|issue:44734", + "issue:45200|issue:45229", + "issue:45588|issue:46050", + "issue:43493|issue:43701", + "issue:44849|issue:47145", + "issue:37428|issue:43957", + "issue:41720|issue:46710" + ], + "prepared_review_unit_hash": "15d64157a5b026637c75d6d3745c97fd8c2e57d4c2ea54f1aa8834775de8248a", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35398, + "estimated_input_tokens": 8850, + "estimated_eval_tokens": 17956 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is heterogeneous; the soft pairs mostly share broad subsystems or model families, but each describes a distinct bug or failure mode. I would reject all proposed duplicate links and avoid choosing a single canonical issue for the whole batch.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue is a clear canonical representative because the items span unrelated bugs across tokenizers, loading, training, quantization, serving, and model-specific regressions.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue is a strong global canonical choice; the list does not form one coherent duplicate cluster, so any pick would be arbitrary.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46821", + "right": "issue:46832", + "accept": false, + "reason": "Different models and failure modes: a TimesFM window_size AttributeError vs an RT-DETR forward shape/feature-level mismatch." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Unrelated components: Beit semantic segmentation loading a legacy field vs BigBirdTokenizer special-token decoding." + }, + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Both involve model structure/key shape concerns, but one is a broad unexpected-key regression and the other is a CLIP text-model attribute inconsistency." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "Different symptoms and models: SigLIP2 nonsensical outputs vs a generic unexpected-key issue in many models." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/device related, but one is a CPU offload mismatch in bark-small and the other is a stale GraniteSpeech test." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "TorchScript GeneratorExp export failure is unrelated to SigLIP2 zero-probability output behavior." + }, + { + "left": "issue:45245", + "right": "issue:46710", + "accept": false, + "reason": "A category-cardinality runtime error is unrelated to a downstream accuracy regression on a specific Llama model." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Local-path import warning and CUDA OOM on GPT-oss are unrelated failure classes." + }, + { + "left": "issue:45072", + "right": "issue:46832", + "accept": false, + "reason": "bfloat16 dtype mismatches in inference are not the same bug as RT-DETR having too many feature levels." + }, + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Quantization/offloading issues are adjacent, but one is runtime offloading behavior and the other is save/load corruption for sinq checkpoints." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency is unrelated to Whisper empty-transcription generation after token alignment." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 hidden-state handling and FP8 inference failure on B200 are different model/path problems." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Hidden-state output bug is unrelated to saving incorrect visual encoder keys." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing after added tokens does not match a TrainerState counter not updating." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is ineffective LoRA behavior while the other is an eval-mode restriction." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer inconsistency and a Trainer broadcast NoneType error are unrelated." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient inflation is not the same as a model-loading regression on transformers 5.x." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "A template-node compilation TypeError is unrelated to a Python 3.13 flash_attn import KeyError." + }, + { + "left": "issue:47013", + "right": "issue:47254", + "accept": false, + "reason": "xLSTM return_last_states unpacking and PEFT/DeepSpeed checkpoint errors are different training-path failures." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4 issues, but one is multi-GPU CUDA OOM and the other is a device-mismatch forward bug." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both touch transformers serve, but continuous batching crashes and KV-cache continuation tensor indexing are distinct defects." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "mm_token_type_ids defaulting and multi-GPU OOM are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Flash-attention sink handling and sinq checkpoint save/load corruption are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and resume_from_checkpoint key mismatch do not share the same underlying bug." + }, + { + "left": "issue:44849", + "right": "issue:47145", + "accept": false, + "reason": "Qwen3.5 hidden-state output bug and FP8 weight_scale omission are unrelated." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "A flash attention import error on Python 3.13 is unrelated to meta-device model loading breakage." + }, + { + "left": "issue:41720", + "right": "issue:46710", + "accept": false, + "reason": "A device-map cudaErrorAssert on A800 is unrelated to an accuracy regression after upgrading Transformers." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 31, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29870", + "issue:29942", + "issue:30333", + "issue:37428", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:43065", + "issue:43295", + "issue:43334", + "issue:43404", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43937", + "issue:43957", + "issue:44186", + "issue:44554", + "issue:44610", + "issue:44869", + "issue:44977", + "issue:45072", + "issue:45198", + "issue:45200", + "issue:45406", + "issue:45446", + "issue:45520", + "issue:45588", + "issue:45759", + "issue:45823", + "issue:46121", + "issue:46178", + "issue:46192", + "issue:46369", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821", + "issue:47332", + "issue:47404" + ], + "soft_pairs": [ + "issue:43937|issue:46701", + "issue:43295|issue:43450", + "issue:43065|issue:46369", + "issue:44554|issue:45446", + "issue:47332|issue:47404", + "issue:44977|issue:46701", + "issue:44869|issue:46752", + "issue:43404|issue:46710", + "issue:45072|issue:45198", + "issue:29942|issue:30333", + "issue:45588|issue:45759", + "issue:43295|issue:43873", + "issue:43572|issue:46821", + "issue:43540|issue:45200", + "issue:45406|issue:46121", + "issue:43572|issue:43957", + "issue:43295|issue:44610", + "issue:43450|issue:43575", + "issue:44186|issue:46192", + "issue:45823|issue:46192", + "issue:37428|issue:43824", + "issue:43550|issue:43844", + "issue:45520|issue:46178", + "issue:45200|issue:45823", + "issue:29870|issue:29942", + "issue:42490|issue:43493", + "issue:43334|issue:43525", + "issue:43295|issue:43493", + "issue:39692|issue:46192", + "issue:42915|issue:43824" + ], + "prepared_review_unit_hash": "2ebec9aa252096a70a321c3e30f78ab9b8e17575d2fe26732e5615a3c5cc8f66", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36298, + "estimated_input_tokens": 9075, + "estimated_eval_tokens": 18406 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is a loose collection of unrelated Transformers bugs. Several are multimodal/processor or loading regressions, but none of the soft pairs look like the same concrete underlying fix, so all soft edges should be rejected.", + "confidence": 0.37, + "canonical_issue_reason": "Issue 43295 is the broadest representative of the loose processor/tokenizer regression theme, and it is the closest hub among the candidates even though it is not a true duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43295 is the best single issue to anchor the cluster because it is the most general and centrally connected to the processor/tokenizer breakage pattern seen in a few neighboring reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43937", + "right": "issue:46701", + "accept": false, + "reason": "Different bugs: invalid GenerationConfig for GLM-5 vs a pipeline failure in DiffusionGemma output handling." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Both involve processors, but one is a regression around processor.tokenizer access while the other is batched video shape handling; not the same fix." + }, + { + "left": "issue:43065", + "right": "issue:46369", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to sam3-litetext memory overflow." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "MPS attention correctness with mismatched head dims is unrelated to a PyTorch version gate for AuxRequest import." + }, + { + "left": "issue:47332", + "right": "issue:47404", + "accept": false, + "reason": "Both mention loading/offloading memory, but one is save-time disk offload for Qwen3-VL and the other is device_map conversion OOM during weight conversion; not the same bug." + }, + { + "left": "issue:44977", + "right": "issue:46701", + "accept": false, + "reason": "Flash-attention generation failure in Qwen3.5 is unrelated to DiffusionGemma pipeline output issues." + }, + { + "left": "issue:44869", + "right": "issue:46752", + "accept": false, + "reason": "Whisper timestamp decoding on a trailing replacement character is different from apply_chat_template crashing on an empty list." + }, + { + "left": "issue:43404", + "right": "issue:46710", + "accept": false, + "reason": "Missing lm_head tying in Mistral3ForConditionalGeneration is a model-weight bug, not an accuracy regression after a Transformers upgrade." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "CI dtype mismatch for SwitchTransformers/TimmWrapperModel is unrelated to Wav2Vec2 save_pretrained/tokenization failure." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 test failures and MLflow reporting on failed training jobs are completely different areas." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Sink-less flash_attention crash on s_aux=None is unrelated to AutoModelForCausalLM not unwrapping text_config for multimodal configs." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "processor.tokenizer regression and quantized offloading behavior are different problems despite both touching multimodal loading paths." + }, + { + "left": "issue:43572", + "right": "issue:46821", + "accept": false, + "reason": "StableLmConfig missing pad_token_idx is unrelated to TimesFM window_size AttributeError." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to Gemma 4 mm_token_type_ids defaulting behavior." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer and rope dict conversion on JSON-loaded lists are unrelated failures." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is not the same as meta-device loading breaking some models." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Both are processor/model mismatches, but one is a processor.tokenizer regression and the other is a fixed input-size mismatch for OmDet-Turbo; different code paths." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Batched video processor shape issues are unrelated to Qwen2-57B-A14B-Instruct OOM on tensor-parallel load." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "LayoutLMv2Tokenizer crashes on NER padding/truncation are different from return_tensors being ignored when text_kwargs is passed." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Gemma4 PLE device_map auto mismatch is a device-placement bug, not the return_tensors/text_kwargs API issue." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Importing _flash_supports_window_size and importing Qwen2_5_VLForConditionalGeneration are unrelated import failures." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "torch.compile with SDPA in Bamba-9B-v2 is unrelated to gradient explosion under HfDeepSpeedConfig + ZeRO-3." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.13 flash_attn import KeyError is unrelated to an invalid BACKENDS_MAPPING backend value." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "mm_token_type_ids defaulting for fine-tuning is unrelated to Gemma4 PLE device mismatch at forward time." + }, + { + "left": "issue:29870", + "right": "issue:29942", + "accept": false, + "reason": "BART generate decoder_attention_mask UnboundLocalError is unrelated to Flash Attention 2 test failures." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "Trailing-slash local model import warning is unrelated to SigLIP2 implementation discrepancy vs JAX." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Qwen3-VL missing pad_token_id and Llama4 missing pad_token_id are similar in shape but refer to different model families and likely different config fixes." + }, + { + "left": "issue:43295", + "right": "issue:43493", + "accept": false, + "reason": "processor.tokenizer regression and SigLIP2 implementation discrepancy are different bugs in different subsystems." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 documentation/example errors are unrelated to the return_tensors ignored API bug." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "Qwen3Moe failing with FineGrainedFP8Config is unrelated to the Qwen2_5_VL import error." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 32, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:37428", + "issue:41720", + "issue:43262", + "issue:43278", + "issue:43335", + "issue:43425", + "issue:43450", + "issue:43452", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43856", + "issue:43950", + "issue:43975", + "issue:44112", + "issue:44442", + "issue:44466", + "issue:44479", + "issue:44521", + "issue:44561", + "issue:44610", + "issue:44737", + "issue:44857", + "issue:44898", + "issue:45083", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45325", + "issue:45381", + "issue:45412", + "issue:45520", + "issue:46192", + "issue:46209", + "issue:46234", + "issue:46285", + "issue:46512", + "issue:46894", + "issue:46899", + "issue:46932", + "issue:47405" + ], + "soft_pairs": [ + "issue:44521|issue:45290", + "issue:43824|issue:43975", + "issue:43278|issue:43856", + "issue:28218|issue:35141", + "issue:43262|issue:43450", + "issue:45381|issue:46209", + "issue:43975|issue:45520", + "issue:44737|issue:46894", + "issue:43950|issue:46285", + "issue:44466|issue:45127", + "issue:43653|issue:44112", + "issue:43452|issue:46512", + "issue:44442|issue:45198", + "issue:44561|issue:46234", + "issue:43611|issue:47405", + "issue:44479|issue:46192", + "issue:45200|issue:45325", + "issue:43335|issue:43550", + "issue:43819|issue:44112", + "issue:45200|issue:46899", + "issue:44898|issue:45412", + "issue:41720|issue:46209", + "issue:44610|issue:44857", + "issue:45200|issue:46932", + "issue:37428|issue:43425", + "issue:43819|issue:45083", + "issue:43335|issue:43653", + "issue:43606|issue:45083" + ], + "prepared_review_unit_hash": "61ab64d3629e490a05cb636fdd66f5ae76d50b8260770129a2f77bd433a2b96c", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35989, + "estimated_input_tokens": 8998, + "estimated_eval_tokens": 18252 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are mostly unrelated transformer bugs/regressions across tokenization, multimodal chat templates, model loading, dtype/device handling, and architecture-specific failures. No soft pair looks like the same underlying fix, so all soft edges are rejected. There are no PRs in the set.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44521 is the strongest anchor: it is open, highly discussed, and has the most inbound references, making it the best representative user-facing bug report in this mixed set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the best global issue candidate because it has the highest visible activity and references, and it describes a concrete, reproducible regression rather than an abstract compatibility note.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both involve chat templating, but one is about multimodal assistant masks and the other about tool-call messages with no content; different failure modes and fixes." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "Import error for a missing VL class is unrelated to a detokenization regression in DeepSeek Coder." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "BF16->FP32 dtype drift in eval is unrelated to Qwen3 MoE memory usage during training." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing after added tokens is a different bug from output embedding reinitialization when resizing embeddings." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and batched video output shape are distinct processor bugs." + }, + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Qwen2.5-VL vision position IDs regression is unrelated to FP8 inference failure on B200." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "A decoding regression and a Python 3.13 flash_attn import error are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:46894", + "accept": false, + "reason": "CPU device placement in XLNet positional encoding is not the same as the CohereASR double-shift training-loss bug." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Non-persistent buffer corruption on load is unrelated to CLIP text-model structure inconsistency." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied weights, but one is serialization inconsistency and the other is LoRA merge collapse with extended vocabulary; not the same code path." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBird tokenizer special-token registration is unrelated to a GraniteSpeech CI device-override test issue." + }, + { + "left": "issue:43452", + "right": "issue:46512", + "accept": false, + "reason": "GGUF loading breakage is unrelated to from_config(dtype=...) leaving config.dtype stale." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "FastSpeech2ConformerTokenizer loading failure is unrelated to Wav2Vec2 save_pretrained/tokenization failures." + }, + { + "left": "issue:44561", + "right": "issue:46234", + "accept": false, + "reason": "Both touch import_utils, but one is removal of an API breaking trust_remote_code models and the other is a missing availability helper; different fixes." + }, + { + "left": "issue:43611", + "right": "issue:47405", + "accept": false, + "reason": "Base-model-prefix loading breakage is different from silent conversion skipping on wrong Auto class/model_type mismatch." + }, + { + "left": "issue:44479", + "right": "issue:46192", + "accept": false, + "reason": "Qwen video-input regression is unrelated to return_tensors being ignored when text_kwargs is passed." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Gemma4 text-only mm_token_type_ids defaults and Qwen2.5-VL still-image position scaling are separate multimodal bugs." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "A config bug creating a sparse layer at zero layers is unrelated to a torch.compile/SDPA failure in Bamba-9B-v2." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC forward/latents mismatch is unrelated to a stale CI device override test." + }, + { + "left": "issue:45200", + "right": "issue:46899", + "accept": false, + "reason": "Defaulting mm_token_type_ids to zeros is not the same as unsafe dtype casting in quantized multimodal embedders." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver interpolation failure and RT-DETR memory reclamation are unrelated model behaviors." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3 auto device mapping cuda assert and FP8 inference on B200 are different hardware/runtime failures." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output resolution mismatch is unrelated to an AMP/CUDA loss crash." + }, + { + "left": "issue:45200", + "right": "issue:46932", + "accept": false, + "reason": "Both involve Gemma4 multimodal dtype behavior, but one is token-type defaults and the other is quantized embedder casting; not the same bug." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "A missing flash-attention helper import is not the same as generic Torch 2.10 incompatibility." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC latent reconstruction mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and BigBird tokenizer special-token registration are different subsystems and bugs." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU offload device mismatch in Bark is unrelated to qwen3_omni_moe feature-length calculation." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 33, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28282", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:37428", + "issue:38175", + "issue:42490", + "issue:43335", + "issue:43388", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43819", + "issue:43827", + "issue:43854", + "issue:43901", + "issue:44112", + "issue:44206", + "issue:44351", + "issue:44493", + "issue:44617", + "issue:44792", + "issue:44855", + "issue:44857", + "issue:45083", + "issue:45290", + "issue:45412", + "issue:45520", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45759", + "issue:45859", + "issue:45923", + "issue:46170", + "issue:46285", + "issue:46326", + "issue:46424", + "issue:46431", + "issue:46512", + "issue:46566", + "issue:46697", + "issue:46821", + "issue:46858", + "issue:47405" + ], + "soft_pairs": [ + "issue:44792|issue:45563", + "issue:44857|issue:45593", + "issue:28282|issue:33357", + "issue:46512|issue:46821", + "issue:38175|issue:45412", + "issue:43827|issue:43901", + "issue:45759|issue:46285", + "issue:43335|issue:43504", + "issue:29942|issue:32090", + "issue:44206|issue:45584", + "issue:43550|issue:44112", + "issue:45923|issue:47405", + "issue:44855|issue:45083", + "issue:46424|issue:46858", + "issue:43388|issue:43526", + "issue:46431|issue:46566", + "issue:43335|issue:44112", + "issue:43335|issue:43606", + "issue:43335|issue:45083", + "issue:37428|issue:44351", + "issue:45290|issue:46326", + "issue:42490|issue:43854", + "issue:44617|issue:46697", + "issue:45520|issue:46170", + "issue:43819|issue:44493", + "issue:45588|issue:45859" + ], + "prepared_review_unit_hash": "7594aa04d362b8c01525d9e32c4fc5d8cf6ba37e3fde19985e4a0c4c1586f186", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35338, + "estimated_input_tokens": 8835, + "estimated_eval_tokens": 17926 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose similarity bucket, not a duplicate set. The soft pairs span unrelated bugs across generation, model loading, config handling, cache behavior, docs, and CI, so none should be merged as the same underlying issue.", + "confidence": 0.19, + "canonical_issue_reason": "If one issue must anchor the cluster, issue 43335 is the best representative only because it is a concrete code bug with the broadest set of weak similarity links here; however, it is not actually a duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43335 is the least-bad representative for triage, but the set is too heterogeneous for a true canonical issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "Different problems: janus image generation test failure vs a stale paged-generate warning about num_return_sequences." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "Different code paths: AMP/CUDA loss crash vs D-FINE not using auxiliary losses when denoising is off." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Both are runtime failures, but one is a PyTorch import error and the other is a MacOS bus error for a CLIP model." + }, + { + "left": "issue:46512", + "right": "issue:46821", + "accept": false, + "reason": "Unrelated bugs: stale config dtype after from_config vs window_size raising AttributeError in TimesFM 2.5." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "Different behavior and subsystem: unexpected zero probabilities in a model vs RT-DETR memory not being released on deletion." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs-related, but they cover different APIs and different stale references; not the same bug." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Different model/config issues: Qwen multimodal from_config unwrap problem vs CLIP class structure inconsistency." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Both involve Switch/Beit config-loading bugs, but one creates sparse layers incorrectly and the other fails to load a pretrained preset due to a legacy field." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention test failures vs Trainer _gpu_broadcast_one TypeError are distinct failures in different paths." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Different regressions: LaserFeatureExtractor unsupported argument crash vs Whisper empty-transcription failure after token alignment." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/runtime failures, but one is a torch.compile SDPA issue and the other is a stale GraniteSpeech device override test." + }, + { + "left": "issue:45923", + "right": "issue:47405", + "accept": false, + "reason": "Different classes of bug: missing gradient_checkpointing flag vs silent model_type mismatch during auto-class loading." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Different parser/processor bugs: Python 3.13 torch.jit import failure vs helper length-calculation behavior in qwen3_omni_moe." + }, + { + "left": "issue:46424", + "right": "issue:46858", + "accept": false, + "reason": "Cache settings bug vs static-cache seq_length returning a device tensor; related theme only, not same issue." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both affect label handling, but one drops tuple elements in gather_for_metrics while the other returns only one label in BeitImageProcessorFast." + }, + { + "left": "issue:46431", + "right": "issue:46566", + "accept": false, + "reason": "Different loading failures: missing chat template on a processor vs DiffusionGemma weights left on meta with device_map=auto." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction bug is unrelated to a GraniteSpeech CI device-override test." + }, + { + "left": "issue:43335", + "right": "issue:43606", + "accept": false, + "reason": "SwitchTransformers config bug vs bark-small CPU offload device mismatch; same broad area, different root cause." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Different components: sparse layer creation in SwitchTransformers vs helper output-length behavior in qwen3_omni_moe." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Different import errors: missing flash_attention_utils symbol vs cannot import HybridCache." + }, + { + "left": "issue:45290", + "right": "issue:46326", + "accept": false, + "reason": "Different chat-template failures: tool-call assistant message crash vs continue_final_message bug for mistral_common tokenizers." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Both affect model loading, but one is a local-path trailing-slash warning and the other is a unit-test model-loading failure for GLM-4.7-Flash." + }, + { + "left": "issue:44617", + "right": "issue:46697", + "accept": false, + "reason": "Completely different image/video processing bugs: CUDA OOM in Sam3Video vs over-padding in Swin2SR image processor." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "Both are import-time issues on unusual platforms, but one is flash_attn on Python 3.13 and the other is Windows ROCm torch.distributed dependency handling." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "Different numeric/model bugs: DAC latent reconstruction mismatch vs unexpected position-id keys across many models." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Not the same bug: flash_attention integration crashes on s_aux=None, while Qwen3_5MoeTextRotaryEmbedding is incompatible with CPU offload." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 34, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:26413", + "issue:27957", + "issue:28218", + "issue:29127", + "issue:29870", + "issue:29942", + "issue:31501", + "issue:32090", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:39692", + "issue:42175", + "issue:42490", + "issue:43278", + "issue:43329", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43701", + "issue:43720", + "issue:43825", + "issue:44112", + "issue:44206", + "issue:44448", + "issue:44493", + "issue:44514", + "issue:44521", + "issue:44743", + "issue:44779", + "issue:44857", + "issue:45072", + "issue:45083", + "issue:45127", + "issue:45245", + "issue:45310", + "issue:45459", + "issue:45478", + "issue:45593", + "issue:45823", + "issue:46234", + "issue:47000" + ], + "soft_pairs": [ + "issue:44448|issue:44743", + "issue:34689|issue:35141", + "issue:39692|issue:43526", + "issue:44206|issue:45823", + "issue:39692|issue:44493", + "issue:43606|issue:44857", + "issue:10000|issue:15354", + "issue:44514|issue:44521", + "issue:29127|issue:32090", + "issue:28218|issue:32090", + "issue:43701|issue:45083", + "issue:43720|issue:44112", + "issue:44448|issue:44779", + "issue:44857|issue:45072", + "issue:38175|issue:44493", + "issue:32090|issue:33357", + "issue:42490|issue:43575", + "issue:43278|issue:43825", + "issue:43550|issue:45083", + "issue:26413|issue:31501", + "issue:45310|issue:45478", + "issue:37428|issue:42175", + "issue:45127|issue:45245", + "issue:45459|issue:46234", + "issue:27957|issue:34689", + "issue:29942|issue:35141", + "issue:45593|issue:47000", + "issue:29870|issue:32090", + "issue:43278|issue:43329" + ], + "prepared_review_unit_hash": "0c9dbcd5419aa58e2e3efc4b8bd673f8c1362466fb3d1bef1d6f1faca913a15f", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35968, + "estimated_input_tokens": 8992, + "estimated_eval_tokens": 18240 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated Transformers issues grouped by weak text similarity. I only found one clear duplicate pair: the two identical Qwen3.5 MoE `from_pretrained` errors. The rest share only broad subsystems (tokenizers, training, vision, dtype, etc.) but describe different bugs or code paths.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45310 is the earliest, clearest report of the duplicated Qwen3.5 MoE `from_pretrained` failure and can serve as the primary tracker for that bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the best issue to keep as the canonical representative because it is the first of the identical duplicate pair and has the same concrete failure mode as issue:45478.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Different models and failures: Pegasus output drift vs Qwen3 recurrent state reset." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "One is a Llama 3.2 model-loading regression; the other is embedding reinitialization after resizing. Not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Documentation/example errors vs BeitImageProcessorFast label reduction bug." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Different subsystems and symptoms: feature extractor arg crash vs PLE device mismatch." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 doc/example issues are unrelated to position-id key warnings." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "Both are runtime failures, but in different models and code paths (CPU offload vs AMP loss)." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBert issue vs TorchScript generator expression export error; unrelated." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Same broad processor family, but one is a batched padding crash and the other is incorrect assistant masks." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMv3 error messaging issue vs Trainer GPU broadcast NoneType error." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer added-space bug is unrelated to Trainer broadcast failure." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint key mismatch vs qwen3_omni_moe feature-extraction helper behavior; different bugs." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "Accelerate loading/unpacked weights bug vs GraniteSpeech CI test flake; not the same issue." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Pegasus output mismatch vs DeepSeek tokenizer regression; different components and failures." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype, but the affected models and failure mechanisms differ." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 zero probabilities vs unexpected position-id key warnings; unrelated." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer broadcast NoneType error vs MacOS bus error on CLIP; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Import warning on local path vs tensor-parallel OOM during model load; unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "BF16/FP32 dtype regression vs pipeline translation error message; not the same underlying bug." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "Torch.compile/SDPA failure in Bamba vs qwen3_omni_moe feature-length helper issue." + }, + { + "left": "issue:26413", + "right": "issue:31501", + "accept": false, + "reason": "Both affect training startup/resume, but one is checkpoint resume logic and the other is FineWeb training not starting." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same title, same version range, and same concrete `Qwen3.5 Moe from_pretrained` error." + }, + { + "left": "issue:37428", + "right": "issue:42175", + "accept": false, + "reason": "Missing import in flash attention utils vs installation backend dependency issue; unrelated." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/vocab merge collapse vs category cardinality limit error; different bugs." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf error masking vs missing utility function export; unrelated." + }, + { + "left": "issue:27957", + "right": "issue:34689", + "accept": false, + "reason": "Flash Attention 2 support issue vs model loading regression for a different model family." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention 2 test failures vs embedding reinitialization after resizing; different problems." + }, + { + "left": "issue:45593", + "right": "issue:47000", + "accept": false, + "reason": "Different object-detection loss failures: missing auxiliary losses vs infeasible cost matrix." + }, + { + "left": "issue:29870", + "right": "issue:32090", + "accept": false, + "reason": "BART generate decoder mask bug vs Trainer broadcast NoneType error; unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype regression vs multimodal token counting video-branch undefined names; unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative on the soft-edge pairs. The canonical duplicate call on issue:45310 / issue:45478 is well supported by the identical title, version range, and failure mode. The rest of the verdicts are phrased cautiously and do not overclaim duplication or shared root cause." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 35, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29870", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:39692", + "issue:40444", + "issue:42490", + "issue:43257", + "issue:43278", + "issue:43299", + "issue:43381", + "issue:43526", + "issue:43575", + "issue:43701", + "issue:43873", + "issue:43927", + "issue:44186", + "issue:44360", + "issue:44361", + "issue:44448", + "issue:44464", + "issue:44466", + "issue:44530", + "issue:44561", + "issue:44779", + "issue:44855", + "issue:44871", + "issue:44977", + "issue:45198", + "issue:45229", + "issue:45290", + "issue:45381", + "issue:45588", + "issue:45698", + "issue:45823", + "issue:45920", + "issue:46192", + "issue:46399", + "issue:46421", + "issue:46424", + "issue:46655", + "issue:46858", + "issue:47276" + ], + "soft_pairs": [ + "issue:44360|issue:44561", + "issue:45290|issue:45381", + "issue:25251|issue:32090", + "issue:29870|issue:33290", + "issue:44464|issue:46424", + "issue:44448|issue:44977", + "issue:32090|issue:33290", + "issue:45290|issue:46655", + "issue:40444|issue:43526", + "issue:43278|issue:43381", + "issue:44855|issue:46399", + "issue:44530|issue:47276", + "issue:43257|issue:43299", + "issue:43927|issue:45698", + "issue:44977|issue:45229", + "issue:42490|issue:43873", + "issue:45920|issue:46192", + "issue:44561|issue:44779", + "issue:29127|issue:34567", + "issue:33357|issue:39692", + "issue:44871|issue:44977", + "issue:43873|issue:45229", + "issue:29942|issue:30064", + "issue:44186|issue:45198", + "issue:43381|issue:43701", + "issue:44448|issue:44466", + "issue:44361|issue:45198", + "issue:45588|issue:45823", + "issue:43278|issue:43575", + "issue:46421|issue:46858" + ], + "prepared_review_unit_hash": "69a7297450500ab21512dbaccc2ec6658df7866b1bfdac1f2e01a521de082b33", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36835, + "estimated_input_tokens": 9209, + "estimated_eval_tokens": 18674 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issues sharing only broad subsystem similarity (tokenization, generation, cache, vision, trainer, etc.). I would reject all soft edges; none look like the same underlying bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "Different problems: DSA indexer/ReLU discussion vs trust_remote_code breakage from removing is_torch_fx_available." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Different code paths: chat templating with tool-call messages vs Qwen2.5-VL video vision_position_ids." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting bug is unrelated to Trainer _gpu_broadcast_one NoneType error." + }, + { + "left": "issue:29870", + "right": "issue:33290", + "accept": false, + "reason": "BART generate decoder_attention_mask assignment bug vs DeepSpeed AdaFactor OOM." + }, + { + "left": "issue:44464", + "right": "issue:46424", + "accept": false, + "reason": "Both cache/generation related, but one is compiled chunked generation inconsistency and the other is max_cache_length honoring for quantized cache." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Different models and failures: Pegasus v4/v5 output mismatch vs Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer broadcast NoneType vs optimizer OOM; no shared bug." + }, + { + "left": "issue:45290", + "right": "issue:46655", + "accept": false, + "reason": "Chat template tool-call crash is unrelated to SmolVLM video padding calculation." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "Different processors and symptoms: Qwen2.5-VL iterable dataset multi-image failure vs BeitImageProcessorFast reduce_labels bug." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Training/eval dtype mismatch is unrelated to gradient checkpointing being disallowed in eval mode." + }, + { + "left": "issue:44855", + "right": "issue:46399", + "accept": false, + "reason": "Python parser/torch.jit import issue vs Qwen3.5 kernel registration/decorator error." + }, + { + "left": "issue:44530", + "right": "issue:47276", + "accept": false, + "reason": "PagedAttentionCache crash on Qwen3.5 is unrelated to a warning message in continuous batching." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MOE, but one is accelerate+deepspeed weight conversion and the other is v5 model loading regression; not the same concrete bug." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both are save/load related, but one loses custom token IDs in config and the other loads the wrong custom module after save_pretrained." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug is unrelated to Gemma4 multi-GPU CUDA OOM." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Local-path import warning and quantized offloading failure are different issues." + }, + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "AutoTokenizer wrong IDs regression is unrelated to return_tensors being ignored with explicit text_kwargs." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Trust_remote_code breakage from API removal vs DeepSeek tokenizer regression; different failure modes." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 error messaging and TrainerState token counter updating are unrelated." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "Different vision-related issues: MacOS bus error on CLIP model vs SigLIP2 documentation/example errors." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma eos_token_id config inconsistency is unrelated to Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Quantization offloading not working is not the same as Gemma4 multi-GPU OOM." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures are unrelated to image processor void segmentation map processing." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2 tokenizer NER/padding crash and Wav2Vec2 save_pretrained/tokenization failure are different tokenizer bugs." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLukeTokenizer AttributeError and Wav2Vec2 save_pretrained/tokenization failure are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Flash-attention integration s_aux=None crash is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Eval-mode gradient checkpointing restriction vs resume_from_checkpoint key mismatch are different trainer issues." + }, + { + "left": "issue:46421", + "right": "issue:46858", + "accept": false, + "reason": "Chunked-prefill recompilation from lazy cache init differs from non-CUDA static-cache seq_length slice-arithmetic bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 36, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29127", + "issue:29870", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43334", + "issue:43381", + "issue:43412", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43976", + "issue:43994", + "issue:44077", + "issue:44206", + "issue:44265", + "issue:44387", + "issue:44442", + "issue:44479", + "issue:44568", + "issue:44743", + "issue:44977", + "issue:45072", + "issue:45198", + "issue:45229", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45803", + "issue:46121", + "issue:46693", + "issue:46701", + "issue:47217", + "issue:47276" + ], + "soft_pairs": [ + "issue:41720|issue:45229", + "issue:43976|issue:45440", + "issue:44206|issue:44479", + "issue:45198|issue:45803", + "issue:29870|issue:34689", + "issue:43611|issue:46121", + "issue:43653|issue:44442", + "issue:44265|issue:45198", + "issue:44077|issue:47276", + "issue:44977|issue:45563", + "issue:43412|issue:45588", + "issue:44568|issue:45491", + "issue:43653|issue:44743", + "issue:43756|issue:45440", + "issue:42915|issue:42994", + "issue:43065|issue:43994", + "issue:43575|issue:43701", + "issue:29127|issue:33357", + "issue:42490|issue:43381", + "issue:43334|issue:43531", + "issue:43575|issue:44387", + "issue:43065|issue:45412", + "issue:45491|issue:46693", + "issue:43994|issue:45412", + "issue:29127|issue:35141", + "issue:45406|issue:46701", + "issue:45372|issue:45588", + "issue:43526|issue:47217", + "issue:44442|issue:45072", + "issue:45440|issue:45593" + ], + "prepared_review_unit_hash": "7538b3f87669dc3b474b3c4e3a6d4ecade91aad3671ac1d24a496ebd4e8a3e81", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36828, + "estimated_input_tokens": 9207, + "estimated_eval_tokens": 18670 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issues that only share broad Transformers vocabulary (generation, loading, quantization, VL, OOM, etc.). I do not see any soft pair that looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue cleanly represents a duplicate cluster here; the items span many unrelated subsystems and failure modes.", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must anchor triage, issue 45229 is the strongest standalone report: it is concrete, high-activity, and describes a specific reproducible multi-GPU OOM. It is not a duplicate hub, just the most substantial issue in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Both are OOM-related, but one is a Qwen3 auto device-mapping cudaErrorAssert and the other is Gemma4 multi-GPU CUDA OOM; different models, triggers, and code paths." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python-version compatibility breakage vs a model divergence bug; no shared failure mode beyond Transformers version context." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Feature extractor argument regression vs video-input regression across several Qwen VL models; unrelated components and symptoms." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer save/tokenization failure vs a bare-except swallowing errors in a batch feature converter; different classes of bugs." + }, + { + "left": "issue:29870", + "right": "issue:34689", + "accept": false, + "reason": "BART generate() UnboundLocalError vs Llama 3.2 Vision model loading breakage; unrelated models and code paths." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "base_model_prefix loading regression vs rope validation helper TypeError; no concrete overlap in behavior or fix." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token decode bug vs AutoTokenizer failing on FastSpeech2ConformerTokenizer; both tokenizer-related but distinct failures." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export/exportability issue with torch_compilable_check vs Wav2Vec2 save/tokenization failure; separate subsystems." + }, + { + "left": "issue:44077", + "right": "issue:47276", + "accept": false, + "reason": "patchtsmixer post_init policy issue vs a misleading synced_gpus warning message; one is a model class constraint, the other is logging text." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Qwen3.5 flash-attention generation failure vs a stale warning in paged generate(); different bug surface and no shared fix." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export failure on gemma3n vs flash_attention crash on s_aux=None; both export/integration-adjacent but clearly different code paths." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer special-token behavior vs Gemma3 NaN embeddings under sliding-window attention; unrelated layers of the stack." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "Tokenizer decode/special-token issue vs Qwen3.5 recurrent state reset in cached generation; no substantive overlap." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Smollm3 RoPE-layer count mismatch vs DeepSeekV3MoE implementation divergence; both model-architecture issues, but different models and fixes." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Qwen3Moe FP8 config failure vs quantized model saving failure; both quantization-adjacent but not the same bug." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Sam3PixelDecoder dummy Conv2d concern vs SigLIP2 AutoModel/pipeline incorrect outputs; unrelated model families and symptoms." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Tensor-parallel OOM when loading Qwen2-57B vs resume_from_checkpoint key mismatch; different stages and failure modes." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 box-information error message vs CLIP community-model bus error on macOS; only superficial model-loading/build similarity." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Trailing-slash local model import warning vs gradient checkpointing disallowed in eval mode; unrelated behaviors." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Qwen3-VL missing pad_token_id in config vs sliding_window issue for Qwen3-MoE; different model types and root causes." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Loading a large Qwen2 model with tp causing OOM vs increased CUDA reserved memory under int4 quantization; both OOMs, but not the same underlying regression." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Dummy Conv2d concern in Sam3PixelDecoder vs RT-DETR memory not released on deletion; unrelated model architecture vs lifecycle/memory leak bug." + }, + { + "left": "issue:45491", + "right": "issue:46693", + "accept": false, + "reason": "Gemma3 NaNs from all-padding sliding windows vs FA2 long-context performance regression; one is correctness/NaN, the other is performance." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "Wrong outputs from SigLIP2 AutoModel/pipeline vs RT-DETR memory leak after deletion; entirely different symptoms." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "Error-message clarity for LayoutLMv3 boxes vs token embedding resize reinitialization with untied embeddings; unrelated functionality." + }, + { + "left": "issue:45406", + "right": "issue:46701", + "accept": false, + "reason": "Gemma4 processor attribute error in serve vs DiffusionGemmaGenerationOutput pipeline failure; same broad model family only, but not the same bug." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 processor loading breakage from mistral_common import vs flash_attention crash on sink-less models; different failure points and remedies." + }, + { + "left": "issue:43526", + "right": "issue:47217", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels bug vs VL image-placeholder-in-user-text failure; both involve vision inputs, but not the same defect." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure vs dtype mismatch in bfloat16 inference; unrelated tokenizer vs inference dtype bug." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepseekV3MoE implementation divergence vs D-FINE auxiliary-loss omission when denoising is off; different models and behaviors." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 37, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:27957", + "issue:29127", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:42915", + "issue:43066", + "issue:43262", + "issue:43316", + "issue:43381", + "issue:43421", + "issue:43526", + "issue:43824", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44206", + "issue:44279", + "issue:44442", + "issue:44462", + "issue:44514", + "issue:44625", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44898", + "issue:45245", + "issue:45250", + "issue:45325", + "issue:45372", + "issue:45381", + "issue:45406", + "issue:45412", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45712", + "issue:45834", + "issue:46121", + "issue:46369", + "issue:46399", + "issue:46640", + "issue:46693", + "issue:46697", + "issue:46821", + "issue:47312" + ], + "soft_pairs": [ + "issue:44625|issue:45584", + "issue:43421|issue:45491", + "issue:45459|issue:45712", + "issue:45412|issue:46369", + "issue:43066|issue:44462", + "issue:45250|issue:46640", + "issue:43526|issue:45834", + "issue:45372|issue:45406", + "issue:45245|issue:45563", + "issue:27957|issue:33290", + "issue:45372|issue:45459", + "issue:33290|issue:34689", + "issue:43824|issue:44279", + "issue:43526|issue:46693", + "issue:33290|issue:35141", + "issue:27957|issue:29127", + "issue:43994|issue:46369", + "issue:43381|issue:43873", + "issue:44857|issue:45588", + "issue:42915|issue:43931", + "issue:45588|issue:46821", + "issue:44514|issue:45381", + "issue:44442|issue:44898", + "issue:43316|issue:47312", + "issue:45325|issue:46697", + "issue:46121|issue:46399", + "issue:44743|issue:44811", + "issue:43262|issue:44206" + ], + "prepared_review_unit_hash": "c95c4429e965b2a40020438db7ec170e00aedcd9f70729efb08a2a3678f7802d", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36041, + "estimated_input_tokens": 9011, + "estimated_eval_tokens": 18278 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the soft-similarity links mostly connect superficially related transformer/model issues, but they do not describe the same underlying bug or change. I would reject all soft edges and treat the set as unrelated reports.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue fits the cluster; the items span unrelated bugs across tokenization, vision, audio, quantization, and Flash Attention paths with only broad keyword overlap.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "Issue 44514 is the most concrete and narrowly scoped bug report in the set, with a clear failing path (`Qwen2_5_VLProcessor.apply_chat_template` on batched input with `padding=False`).", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Different components and failures: Qwen3.5 config propagation vs Whisper empty-transcription generation after token alignment." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizers backend post-processor sync is unrelated to Gemma3 NaN embeddings from sliding-window attention." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer error masking from protobuf handling is unrelated to leftover dummy class cleanup." + }, + { + "left": "issue:45412", + "right": "issue:46369", + "accept": false, + "reason": "Both mention memory, but one is RT-DETR cleanup/GC and the other is SAM model memory overflow; different bug paths." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Both are tokenizer-loading themed, but one is decoder-type mismatch in v5 and the other is ignoring repository `tokenizer.json`." + }, + { + "left": "issue:45250", + "right": "issue:46640", + "accept": false, + "reason": "Same broad Flash Attention label, but one is a general issue and the other is a specific regression/performance complaint." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to Kosmos2.5 OCR index errors." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Both involve Gemma 4 loading, but one is a missing `ReasoningEffort` import and the other is a missing `_tokenizer` attribute in serve." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-limit runtime error is unrelated to a stale warning in paged generation." + }, + { + "left": "issue:27957", + "right": "issue:33290", + "accept": false, + "reason": "XLMRoberta Flash Attention setup and Adafactor/DeepSpeed OOM are different subsystems and different failures." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma 4 processor import failure is unrelated to protobuf-backed tokenizer error masking." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Optimizer/deepspeed OOM is unrelated to model-loading breakage for Llama 3.2 Vision." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Importing a specific Qwen2.5-VL class is not the same as a generic dependency issue with transformers." + }, + { + "left": "issue:43526", + "right": "issue:46693", + "accept": false, + "reason": "Label reduction in BEiT image processing is unrelated to Flash Attention long-context performance regression." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "DeepSpeed OOM and token-embedding reinitialization are unrelated bugs." + }, + { + "left": "issue:27957", + "right": "issue:29127", + "accept": false, + "reason": "Flash Attention support for XLMRoberta is unrelated to LayoutLMv3 error-message clarity." + }, + { + "left": "issue:43994", + "right": "issue:46369", + "accept": false, + "reason": "SigLIP2 auto/pipeline misbehavior is unrelated to SAM memory overflow." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient-checkpointing eval-mode restriction is unrelated to quantization offloading behavior." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "AMP/CUDA loss crash in LwDetr is not the same as Flash Attention sink-less model failure on `s_aux=None`." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are different model/config issues." + }, + { + "left": "issue:45588", + "right": "issue:46821", + "accept": false, + "reason": "Flash Attention sink-less crash and TimesFM `window_size` attribute error do not share the same code path." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL, but one is batched chat template padding behavior and the other is video vision position IDs." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading for FastSpeech2Conformer is unrelated to Perceiver interpolation behavior." + }, + { + "left": "issue:43316", + "right": "issue:47312", + "accept": false, + "reason": "Gemma3TextConfig API mismatch is unrelated to a kernel-compatibility report." + }, + { + "left": "issue:45325", + "right": "issue:46697", + "accept": false, + "reason": "Qwen2.5-VL rope/position-id scaling bug is unrelated to Swin2SR image padding logic." + }, + { + "left": "issue:46121", + "right": "issue:46399", + "accept": false, + "reason": "Rope-validation list typing error is unrelated to a missing decorator causing Qwen3.5 kernel registration failure." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Recurrent state reset in modular Qwen3.5 is unrelated to Whisper `batch_decode` skipping special tokens." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Default audio sampling-rate handling and unsupported `center` argument in a feature extractor are different bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 38, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:28282", + "issue:29870", + "issue:32090", + "issue:42175", + "issue:42490", + "issue:42617", + "issue:43316", + "issue:43493", + "issue:43525", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43643", + "issue:43646", + "issue:43701", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44743", + "issue:44855", + "issue:44871", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:44987", + "issue:45137", + "issue:45216", + "issue:45245", + "issue:45440", + "issue:45759", + "issue:45859", + "issue:45907", + "issue:46178", + "issue:46234", + "issue:46506" + ], + "soft_pairs": [ + "issue:43646|issue:44077", + "issue:44977|issue:45245", + "issue:43761|issue:45859", + "issue:43720|issue:44442", + "issue:42175|issue:46234", + "issue:25251|issue:28282", + "issue:43873|issue:44871", + "issue:27957|issue:29870", + "issue:44077|issue:44479", + "issue:42490|issue:44743", + "issue:43643|issue:44913", + "issue:43316|issue:45759", + "issue:43493|issue:43827", + "issue:44062|issue:45907", + "issue:44987|issue:46178", + "issue:43316|issue:43525", + "issue:43749|issue:43931", + "issue:44964|issue:45440", + "issue:42617|issue:43575", + "issue:43550|issue:43701", + "issue:43575|issue:43742", + "issue:44855|issue:45137", + "issue:43606|issue:43701", + "issue:43701|issue:46506", + "issue:43761|issue:44079", + "issue:44190|issue:44496", + "issue:43931|issue:45216", + "issue:27957|issue:32090" + ], + "prepared_review_unit_hash": "3c82ebf1ef61628ac00076f2d7c51479bb82371f7be10aef6d9be27a9958559f", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36100, + "estimated_input_tokens": 9025, + "estimated_eval_tokens": 18306 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a true duplicate set: the soft pairs mostly share generic terms like loading, config, offload, or regression, but they describe different code paths and failure modes. I would reject every soft edge and treat the cluster as heterogeneous.", + "confidence": 0.95, + "canonical_issue_reason": "No single issue is a strong canonical duplicate target here. If forced to pick a loose representative, issue 43316 is the broadest of the config/API-discrepancy reports, but the cluster does not cohere around one underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43316 is the closest thing to a representative because it frames a generic API discrepancy, but it is still not a suitable true canonical since the surrounding issues span unrelated subsystems and symptoms.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Both concern model init/post-init behavior, but one is about Transformers 5 breaking custom initialization and the other about a specific model class allowing optional post_init; not the same bug." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Different problems: flash-attention generation failure for Qwen3.5 versus a categorical cardinality runtime limit." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "Both involve model behavior regressions, but one is CLIPVision hidden_states not returned and the other is Qwen3.5 MoE rotary embedding CPU offload compatibility." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "Packed-weight loading in accelerate versus AutoTokenizer failing for a specific tokenizer class are unrelated failures." + }, + { + "left": "issue:42175", + "right": "issue:46234", + "accept": false, + "reason": "Both mention backend/import packaging, but one is TensorFlow being absent from an install extra and the other is a missing availability helper symbol." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline output nesting and an AutoModel ImportError are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization offload behavior and Gemma eos token configuration mismatch do not describe the same underlying bug." + }, + { + "left": "issue:27957", + "right": "issue:29870", + "accept": false, + "reason": "Both are generation-related model issues, but XLM-RoBERTa Flash Attention 2 support and BART decoder_attention_mask assignment are different code paths." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "Optional post_init gating is unrelated to the Qwen video-input regression." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "A trailing-slash local model path warning and Qwen recurrent-state reset are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Both involve config persistence, but one is missing fields from trust_remote_code loading and the other is rotary_pct reverting on reload; different root causes." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both are about composite config API shape, but one is a Gemma3TextConfig discrepancy and the other is AutoModelForCausalLM not unwrapping text_config for Qwen multimodal configs." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 HF-vs-JAX divergence and docs still referencing pipeline() are entirely different issues." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are unrelated API bugs." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "Loading a specific model on transformers>=5.1.0 and a backend mapping error for tensorflow_text are not the same problem." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both are config attribute inconsistencies, but Gemma3TextConfig API mismatch and Llama4Config missing pad_token_id are separate model-specific issues." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and Qwen3-VL weight-shape mismatch are different loading failures." + }, + { + "left": "issue:44964", + "right": "issue:45440", + "accept": false, + "reason": "Phi-4 multimodal loading and DeepseekV3MoE divergence are unrelated model families and code paths." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "A script startup failure and tp-related OOM on Qwen2-57B are not duplicates." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "torch.compile SDPA failure for Bamba is unrelated to resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Tensor-parallel OOM on Qwen2-57B and key errors loading MobileLLM-125M are different issues." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 indentation/@torch.jit.script parsing issue and DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch and checkpoint-resume key mismatch do not share a code path." + }, + { + "left": "issue:43701", + "right": "issue:46506", + "accept": false, + "reason": "Checkpoint key mismatch and differing CPU/CUDA outputs in PPDocLayoutV3 are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "Hidden_states being None in CLIPVisionModel and ModelOutput key assignment with prior None are different bugs." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Local dataset loading in an image-classification script and an unrecognized-model config.json error are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Weight-shape mismatch on model load and a broken save_pretrained regression for Qwen3.5 are different failure modes." + }, + { + "left": "issue:27957", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 support for XLM-RoBERTa and a Trainer _gpu_broadcast_one NoneType error are not the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 39, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:25251", + "issue:27957", + "issue:42490", + "issue:42907", + "issue:43065", + "issue:43232", + "issue:43295", + "issue:43316", + "issue:43352", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43452", + "issue:43540", + "issue:43572", + "issue:43646", + "issue:43716", + "issue:43761", + "issue:43825", + "issue:44077", + "issue:44162", + "issue:44188", + "issue:44442", + "issue:44464", + "issue:44496", + "issue:44514", + "issue:44521", + "issue:44534", + "issue:44617", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:45584", + "issue:45698", + "issue:45803", + "issue:46050", + "issue:46178", + "issue:46234", + "issue:46291", + "issue:46365", + "issue:46421", + "issue:46640", + "issue:46829", + "issue:46932", + "issue:47217", + "issue:47276" + ], + "soft_pairs": [ + "issue:44442|issue:45803", + "issue:42490|issue:47276", + "issue:46829|issue:46932", + "issue:43295|issue:44871", + "issue:43316|issue:43716", + "issue:44871|issue:45584", + "issue:43295|issue:43825", + "issue:43232|issue:43388", + "issue:43761|issue:44188", + "issue:16998|issue:22355", + "issue:44521|issue:47217", + "issue:46234|issue:46291", + "issue:43646|issue:44913", + "issue:43452|issue:44496", + "issue:44077|issue:44534", + "issue:42907|issue:45698", + "issue:44871|issue:46050", + "issue:44464|issue:46421", + "issue:43404|issue:43441", + "issue:44162|issue:44792", + "issue:43065|issue:44617", + "issue:43352|issue:46640", + "issue:45803|issue:46365", + "issue:25251|issue:27957", + "issue:43540|issue:44514", + "issue:43572|issue:46178" + ], + "prepared_review_unit_hash": "57a8e92d915f9b5093ba09f01870880c15e831f1fb7e7fd4bdc600ffb43d23c6", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35341, + "estimated_input_tokens": 8836, + "estimated_eval_tokens": 17928 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: most items are unrelated model/config/regression bugs rather than duplicates. I would not merge any of the soft pairs. If a single issue must represent the set, #44521 is the best candidate because it is open, has the most inbound references/discussion, and describes a concrete multimodal bug.", + "confidence": 0.94, + "canonical_issue_reason": "#44521 is the strongest representative: it is open, has the highest external interest in this set, and centers on a clearly reproducible multimodal bug. There is no convincing duplicate group strong enough to make another issue a better canonical target.", + "canonical_pr_reason": null, + "best_issue_reason": "#44521 is the best standalone issue in the set because it has the broadest impact signals and a precise failure mode that is likely actionable for maintainers.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Different failure classes: tokenizer load/import error vs exception-handling bug in tensor conversion." + }, + { + "left": "issue:42490", + "right": "issue:47276", + "accept": false, + "reason": "Both mention warnings, but they concern unrelated warning messages and code paths." + }, + { + "left": "issue:46829", + "right": "issue:46932", + "accept": false, + "reason": "Different model families and bugs: output field wiring vs quantized dtype casting." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "One is a processor/tokenizer regression; the other is an eos_token_id config mismatch." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "Gemma3 config API inconsistency is unrelated to Mistral-3 image dtype mismatch." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Both are generation/config issues, but the concrete bugs are different models and different root causes." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Regression in custom processor usage is unrelated to a pipeline error-message wording issue." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs update after sync_gpus is unrelated to metrics gathering on the last batch." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "VisionModel hidden_states regression and attention-kernel divergence are different subsystems and symptoms." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Model max length question and missing transformers.onnx module are unrelated." + }, + { + "left": "issue:44521", + "right": "issue:47217", + "accept": false, + "reason": "Both touch multimodal chat templates, but one is an assistant_mask bug and the other is a failure when the image placeholder appears in user text." + }, + { + "left": "issue:46234", + "right": "issue:46291", + "accept": false, + "reason": "Missing import_utils helper and pipeline import failure from kernels versioning are distinct issues." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Custom model initialization breakage is not the same as GPTNeoX rotary_pct not persisting on reload." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "gguf_file loading breaks for tokenizer/model loading, while the other is an unrecognized-model/config.json issue." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "Optional post_init deprecation is unrelated to junk-filled non-persistent buffers." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Dequantized save failure is a quantization bug, not a custom-module resolution bug after save/load." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Eos-token mismatch and sinq quantization round-trip failure are different problems." + }, + { + "left": "issue:44464", + "right": "issue:46421", + "accept": false, + "reason": "Compiled-forward generation inconsistency is not the same as chunked prefill recompilation/cache initialization." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Weight tying in Mistral3 is unrelated to FlashAttention support gating for Ministral-3." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 broken workflow and janus image-generation test failure are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Dummy Conv2d in a decoder is not the same as CUDA OOM in Sam3Video." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "Flash Attention support error for NemotronH is not clearly the same as a generic Flash Attention 2 issue." + }, + { + "left": "issue:45803", + "right": "issue:46365", + "accept": false, + "reason": "Bare except swallowing errors is unrelated to Trainer import failure caused by datasets." + }, + { + "left": "issue:25251", + "right": "issue:27957", + "accept": false, + "reason": "Pipeline top_k nesting change and XLMRoberta Flash Attention support are different behaviors." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Video input ValueError in Qwen3OmniMoe is not the same as batched apply_chat_template padding=False crash." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "Missing config field after v5 update and backend mapping error for tensorflow_text are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 40, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29870", + "issue:35141", + "issue:41950", + "issue:42490", + "issue:43388", + "issue:43475", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43650", + "issue:43673", + "issue:43688", + "issue:43827", + "issue:43931", + "issue:43994", + "issue:44060", + "issue:44190", + "issue:44242", + "issue:44479", + "issue:44534", + "issue:44610", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45381", + "issue:45405", + "issue:45542", + "issue:45588", + "issue:45600", + "issue:45632", + "issue:45698", + "issue:45803", + "issue:46082", + "issue:46170", + "issue:46202", + "issue:46234", + "issue:46369", + "issue:46424", + "issue:46693", + "issue:46697", + "issue:46894", + "issue:47217" + ], + "soft_pairs": [ + "issue:44610|issue:46369", + "issue:45381|issue:47217", + "issue:44792|issue:44977", + "issue:44479|issue:47217", + "issue:45542|issue:46234", + "issue:44534|issue:44913", + "issue:43525|issue:43531", + "issue:43504|issue:45698", + "issue:44792|issue:45600", + "issue:45632|issue:46424", + "issue:45803|issue:46170", + "issue:35141|issue:42490", + "issue:46693|issue:47217", + "issue:45127|issue:45200", + "issue:44242|issue:46894", + "issue:41950|issue:43475", + "issue:43650|issue:46202", + "issue:43673|issue:44743", + "issue:43931|issue:44060", + "issue:44855|issue:45588", + "issue:44792|issue:45405", + "issue:44190|issue:44964", + "issue:43388|issue:43688", + "issue:29870|issue:35141", + "issue:46082|issue:46697", + "issue:43827|issue:43994" + ], + "prepared_review_unit_hash": "2dca31212ce3fc3ca6cf8585c9052108d28bd79da74c8656183a759f613f2659", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35141, + "estimated_input_tokens": 8786, + "estimated_eval_tokens": 17828 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a noisy mix of unrelated issue reports across many models, pipelines, and config/load paths. None of the soft-paired items look like true duplicates, so there is no credible single canonical issue or PR here.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44610", + "right": "issue:46369", + "accept": false, + "reason": "Processor image-size mismatch vs memory overflow; different model and failure mode." + }, + { + "left": "issue:45381", + "right": "issue:47217", + "accept": false, + "reason": "Video vision_position_ids regression vs image-placeholder-in-text failure; different code paths." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Janus image-generation test failure vs Qwen3.5 flash-attention generation bug." + }, + { + "left": "issue:44479", + "right": "issue:47217", + "accept": false, + "reason": "v5.3.0 video-input regression across Qwen VL models vs text-placeholder handling." + }, + { + "left": "issue:45542", + "right": "issue:46234", + "accept": false, + "reason": "TensorBoard/TF backend packaging issue vs missing import_utils symbol." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer initialization issue vs GPTNeoX config reload bug." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config vs Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Legacy field breaks Beit preset load vs wrong custom module loaded after save_pretrained." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "Janus test failure vs stale auto_mappings entry for removed Sam3LiteText configs." + }, + { + "left": "issue:45632", + "right": "issue:46424", + "accept": false, + "reason": "Leaf-dir cache collision vs max_cache_length cache policy bug." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Bare except in tensor conversion vs Windows ROCm import dependency issue." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Embedding resize/reinit bug vs trailing-slash local model import warning." + }, + { + "left": "issue:46693", + "right": "issue:47217", + "accept": false, + "reason": "FA2 long-context performance regression vs VL image-placeholder bug." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA/vocab collapse with tied embeddings vs mm_token_type_ids defaulting." + }, + { + "left": "issue:44242", + "right": "issue:46894", + "accept": false, + "reason": "Missing load-balancing loss vs ASR training double-shift bug." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Video-classification processor mismatch vs missing vision encoder attribute." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "Unrelated placeholder issue vs Gemma pipeline roundtrip bug." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Chunked-prefill cache missing vs recurrent state reset; both cache-related but different bugs." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch vs Qwen3-Next tied-weights warning." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 JIT parse error vs flash_attention s_aux=None crash." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Janus test failure vs unreleased PEFT version constraint." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset script loading issue vs Phi-4 multimodal model load failure." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Metric gather truncation vs auxiliary-loss normalization." + }, + { + "left": "issue:29870", + "right": "issue:35141", + "accept": false, + "reason": "BART generate mask error vs embedding resize reinit." + }, + { + "left": "issue:46082", + "right": "issue:46697", + "accept": false, + "reason": "Llama head_dim validation vs Swin2SR over-padding." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 41, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:25251", + "issue:29870", + "issue:39692", + "issue:41720", + "issue:43065", + "issue:43295", + "issue:43335", + "issue:43526", + "issue:43540", + "issue:43550", + "issue:43644", + "issue:43827", + "issue:43854", + "issue:44190", + "issue:44242", + "issue:44479", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44779", + "issue:44792", + "issue:44855", + "issue:44869", + "issue:44913", + "issue:44933", + "issue:45072", + "issue:45245", + "issue:45276", + "issue:45362", + "issue:45381", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45632", + "issue:46369", + "issue:46399", + "issue:46489", + "issue:46617", + "issue:46655", + "issue:46697", + "issue:46832", + "issue:46858", + "issue:46897", + "issue:46932", + "issue:47013", + "issue:47312", + "issue:47321" + ], + "soft_pairs": [ + "issue:44617|issue:45584", + "issue:45245|issue:46858", + "issue:45072|issue:46932", + "issue:43644|issue:44913", + "issue:45588|issue:47013", + "issue:43335|issue:44855", + "issue:44734|issue:45406", + "issue:25251|issue:29870", + "issue:46399|issue:47312", + "issue:44869|issue:46832", + "issue:45381|issue:46369", + "issue:41720|issue:45245", + "issue:45632|issue:46617", + "issue:43827|issue:47321", + "issue:43550|issue:44190", + "issue:43854|issue:44792", + "issue:45563|issue:45584", + "issue:39692|issue:47321", + "issue:45381|issue:46655", + "issue:44617|issue:46369", + "issue:43065|issue:44610", + "issue:44479|issue:44734", + "issue:43335|issue:44190", + "issue:44933|issue:45362", + "issue:39692|issue:44190", + "issue:44779|issue:46489", + "issue:43295|issue:45276", + "issue:44242|issue:46897", + "issue:43526|issue:46697", + "issue:43540|issue:44734" + ], + "prepared_review_unit_hash": "dbe82f58e1a7f1f37d96b9b1906780283a4d7e66ae8ce95932148ecca5110917", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36838, + "estimated_input_tokens": 9210, + "estimated_eval_tokens": 18676 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs are mostly same-subsystem or same-model-family near-misses, but not the same concrete bug/change. I would not merge any of the candidate pairs.", + "confidence": 0.88, + "canonical_issue_reason": "No canonical issue: the items span unrelated models, regressions, docs, and runtime failures, so there isn\u2019t a single duplicate target for the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44734 is the strongest standalone bug report here: it describes a specific crash in `transformers serve` with a clear failing code path and a likely actionable fix.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Different bugs: Sam3Video OOM vs Whisper empty-transcription crash." + }, + { + "left": "issue:45245", + "right": "issue:46858", + "accept": false, + "reason": "Both involve runtime errors, but one is a category-limit error and the other is a cache-length/device-tensor arithmetic bug." + }, + { + "left": "issue:45072", + "right": "issue:46932", + "accept": false, + "reason": "Both mention dtype issues, but they affect different models and different code paths." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Unrelated failures: non-persistent buffer initialization vs GPTNeoX config reload behavior." + }, + { + "left": "issue:45588", + "right": "issue:47013", + "accept": false, + "reason": "Both are model/runtime crashes, but the underlying faults are different (`s_aux` handling vs xLSTM return shape unpacking)." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "Completely different problems: SwitchTransformers config bug vs Python 3.13 import/`torch.jit` parsing error." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "Serve crash with tensor indexing is not the same as a missing `_tokenizer` attribute on Gemma4Processor." + }, + { + "left": "issue:25251", + "right": "issue:29870", + "accept": false, + "reason": "Pipeline output nesting and Bart `generate()` mask initialization are unrelated bugs." + }, + { + "left": "issue:46399", + "right": "issue:47312", + "accept": false, + "reason": "One is a missing decorator registration crash; the other is a broad kernels version incompatibility." + }, + { + "left": "issue:44869", + "right": "issue:46832", + "accept": false, + "reason": "Whisper trailing-character decoding crash is unrelated to RT-DETR feature-level handling." + }, + { + "left": "issue:45381", + "right": "issue:46369", + "accept": false, + "reason": "Same broad family of model issues, but one is video position IDs and the other is memory overflow." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Different failures in different subsystems: Qwen3 device-mapping assert vs category-count limit." + }, + { + "left": "issue:45632", + "right": "issue:46617", + "accept": false, + "reason": "Both touch trust_remote_code cache loading, but one is a leaf-directory collision and the other is a FileNotFoundError; not clearly the same bug." + }, + { + "left": "issue:43827", + "right": "issue:47321", + "accept": false, + "reason": "Docs still referencing pipeline() has nothing to do with AMD Quark class updates." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "torch.compile/SDPA failure is unrelated to loading a local dataset in the image-classification example." + }, + { + "left": "issue:43854", + "right": "issue:44792", + "accept": false, + "reason": "Two separate model/test failures with no shared code-path evidence." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "A stale warning in paged generate is not the same as Whisper empty-transcription generation failure." + }, + { + "left": "issue:39692", + "right": "issue:47321", + "accept": false, + "reason": "SigLIP2 doc example errors and AMD Quark class updates are unrelated." + }, + { + "left": "issue:45381", + "right": "issue:46655", + "accept": false, + "reason": "Both are video-processing issues, but they concern different models and different defects (position IDs vs padding calculation)." + }, + { + "left": "issue:44617", + "right": "issue:46369", + "accept": false, + "reason": "Same SAM3 area, but one is OOM in video and the other is memory overflow in litetext; not the same bug." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to OmDet-Turbo input-size mismatch." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Both are regressions, but one is Qwen video-input behavior and the other is a serve/KV-cache indexing crash." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "Config-layer creation bug is unrelated to local dataset loading in the training script." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Missing import from image_utils is unrelated to Qwen3.5 chat crashes." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "SigLIP2 docs errors and dataset loading failure do not share the same underlying bug." + }, + { + "left": "issue:44779", + "right": "issue:46489", + "accept": false, + "reason": "Both concern DeepSeek tokenization, but one is a general v5 regression and the other is a specific gap in PR #44801; not the same concrete defect." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Processor/tokenizer regression and Gemma4 resize-token-embeddings propagation bug are different issues." + }, + { + "left": "issue:44242", + "right": "issue:46897", + "accept": false, + "reason": "Load-balancing-loss omission and Florence2 double-shift are different training-loss bugs." + }, + { + "left": "issue:43526", + "right": "issue:46697", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to Swin2SR over-padding." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Qwen3OmniMoe video-input ValueError is not the same as the serve KV-cache indexing crash." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 42, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:27957", + "issue:32090", + "issue:35141", + "issue:36246", + "issue:38617", + "issue:41628", + "issue:42175", + "issue:42796", + "issue:43317", + "issue:43335", + "issue:43404", + "issue:43531", + "issue:43611", + "issue:43643", + "issue:43673", + "issue:43854", + "issue:43994", + "issue:44112", + "issue:44246", + "issue:44279", + "issue:44462", + "issue:44496", + "issue:44521", + "issue:44737", + "issue:44938", + "issue:44964", + "issue:45003", + "issue:45127", + "issue:45216", + "issue:45276", + "issue:45290", + "issue:45357", + "issue:45381", + "issue:45542", + "issue:45584", + "issue:45593", + "issue:45600", + "issue:45632", + "issue:45803", + "issue:45907", + "issue:46234", + "issue:46421", + "issue:46752", + "issue:46858", + "issue:47145", + "issue:47217", + "issue:47312", + "issue:47321" + ], + "soft_pairs": [ + "issue:43673|issue:46421", + "issue:45127|issue:45357", + "issue:43854|issue:44964", + "issue:45276|issue:45584", + "issue:47145|issue:47321", + "issue:44462|issue:45907", + "issue:43643|issue:44496", + "issue:44521|issue:45381", + "issue:43994|issue:47321", + "issue:44112|issue:45600", + "issue:43317|issue:47312", + "issue:43611|issue:47321", + "issue:27957|issue:35141", + "issue:42175|issue:44279", + "issue:44246|issue:45003", + "issue:45290|issue:46752", + "issue:44112|issue:44964", + "issue:43673|issue:46858", + "issue:42796|issue:44521", + "issue:32090|issue:35141", + "issue:44964|issue:45600", + "issue:44737|issue:45803", + "issue:43404|issue:45216", + "issue:44496|issue:47217", + "issue:43531|issue:45600", + "issue:44938|issue:45542", + "issue:43335|issue:45593", + "issue:45632|issue:46234", + "issue:36246|issue:41628", + "issue:36246|issue:38617" + ], + "prepared_review_unit_hash": "678e35c832599605a19a66725bdd6e35730aaaa52105bee0225b65999801fc9d", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36905, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated Transformers issues. None of the soft pairs look like the same concrete bug/change, so I reject all soft edges. If a single issue must represent the cluster, issue:44521 is the least-bad anchor because it has the most engagement, but it is not a true duplicate hub.", + "confidence": 0.77, + "canonical_issue_reason": "issue:44521 is the most central/referenced item here, but the cluster is not a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 has the strongest engagement and broadest relevance among these mostly unrelated bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43673", + "right": "issue:46421", + "accept": false, + "reason": "Both mention chunked prefill/cache, but they describe different failures: missing cache in v5.0.0 vs lazy-init recompilation." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "Different models and different problems: LoRA/extended-vocab collapse vs incorrect visual encoder keys in save_pretrained." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are model-loading failures, but for different models and likely different root causes." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "One is about resize_token_embeddings not updating outputs; the other is Whisper empty-transcription handling." + }, + { + "left": "issue:47145", + "right": "issue:47321", + "accept": false, + "reason": "FP8 frozen-checkpoint scaling and AMD quark class updates are unrelated." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer.json selection and list_repo_files revision forwarding are different repository-access bugs." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Both involve config loading, but one is trust_remote_code field loss and the other is missing model_type in config.json." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both are multimodal/VL-related, but the actual failures differ: assistant masks vs vision_position_ids." + }, + { + "left": "issue:43994", + "right": "issue:47321", + "accept": false, + "reason": "Nonsensical SigLIP outputs and AMD quark class updates are unrelated." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Both touch CI, but they are separate failures in different subsystems." + }, + { + "left": "issue:43317", + "right": "issue:47312", + "accept": false, + "reason": "Dequantized model offload and kernels 0.16 compatibility are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:47321", + "accept": false, + "reason": "base_model_prefix loading regression and AMD quark class update are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention support for XLMRoberta and post_init reinitializing output embeddings are different bugs." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "One is a missing TensorFlow backend in extras; the other is a generic dependency issue." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Import performance and unsafe sys.modules access are not the same bug." + }, + { + "left": "issue:45290", + "right": "issue:46752", + "accept": false, + "reason": "Both are apply_chat_template crashes, but the triggering inputs and failure modes differ." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "Different model-specific CI/load failures with no shared concrete code path." + }, + { + "left": "issue:43673", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache/prefill, but one is lazy initialization and the other is device-tensor arithmetic in static cache." + }, + { + "left": "issue:42796", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer BPE-marker handling and assistant-mask computation are separate tokenizer/chat-template issues." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error is unrelated to embedding resize/reinitialization." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Model loading failure and auto_mappings CI breakage are different problems." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "XLNet device placement bug and bare-except error masking are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Mistral3 weight tying and Qwen3.5 save_pretrained regression are different model-specific bugs." + }, + { + "left": "issue:44496", + "right": "issue:47217", + "accept": false, + "reason": "Unrecognized model config and image-placeholder handling in VL models are not the same issue." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and removed-config CI breakage are unrelated." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 import failure and TensorBoard-without-TF backend error are different dependency/runtime issues." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "Sparse-layer creation and auxiliary-loss disabling are unrelated model-architecture bugs." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "trust_remote_code cache-path collisions and missing import_utils helper are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different export paths." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Qwen2_5_VLImageProcessor import failure and layer_type_validation import failure are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 43, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:39692", + "issue:41950", + "issue:42796", + "issue:42907", + "issue:42994", + "issue:43295", + "issue:43381", + "issue:43441", + "issue:43526", + "issue:43650", + "issue:43673", + "issue:43824", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44220", + "issue:44279", + "issue:44295", + "issue:44327", + "issue:44336", + "issue:44496", + "issue:44683", + "issue:44779", + "issue:44792", + "issue:44933", + "issue:44987", + "issue:45020", + "issue:45042", + "issue:45083", + "issue:45335", + "issue:45405", + "issue:45446", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45759", + "issue:46082", + "issue:46121", + "issue:46424", + "issue:46506", + "issue:46515", + "issue:46752", + "issue:46932", + "issue:47217", + "issue:47332", + "issue:47333" + ], + "soft_pairs": [ + "issue:44496|issue:45042", + "issue:45083|issue:46506", + "issue:43295|issue:45335", + "issue:43976|issue:44987", + "issue:44295|issue:46506", + "issue:41950|issue:44016", + "issue:39692|issue:46932", + "issue:44683|issue:45446", + "issue:44496|issue:45759", + "issue:43673|issue:46424", + "issue:42994|issue:47333", + "issue:43526|issue:43901", + "issue:43673|issue:44336", + "issue:43381|issue:45593", + "issue:44016|issue:44220", + "issue:44279|issue:46515", + "issue:46082|issue:46121", + "issue:43650|issue:44220", + "issue:44016|issue:44062", + "issue:44792|issue:46424", + "issue:42994|issue:47332", + "issue:42907|issue:43441", + "issue:43441|issue:45588", + "issue:44327|issue:46752", + "issue:43976|issue:44279", + "issue:43824|issue:44933", + "issue:45335|issue:45584", + "issue:41950|issue:45020", + "issue:43295|issue:44779", + "issue:43901|issue:44496", + "issue:44279|issue:45405", + "issue:43901|issue:47217", + "issue:42796|issue:43650", + "issue:39692|issue:43901" + ], + "prepared_review_unit_hash": "48a62fd27b2ce58754e68ac3d2ddcb3bac55dcae2d098e01ac67a2c9186b5552", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 34, + "serialized_chars": 37038, + "estimated_input_tokens": 9260, + "estimated_eval_tokens": 18776 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose thematic grouping of Transformers bugs across different models, pipelines, and version regressions. I do not see any soft pair that is clearly the same underlying issue, so I would not merge any of these as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:42994 is the cleanest, most general bug report in the set: a concrete quantized-model saving failure with likely broader applicability than the more model-specific offloading variants.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Both are transformer/model loading or vision-related failures, but one is an unrecognized model/config issue and the other is PIL image-processor torchvision gating; different code paths and symptoms." + }, + { + "left": "issue:45083", + "right": "issue:46506", + "accept": false, + "reason": "These are both numerical/output anomalies, but they affect different models and different helpers; no sign they are the same bug or could be fixed by one change." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "One is a regression in processor/tokenizer access for custom multimodal code; the other is a T5Gemma embedding resize inconsistency. Different components and failure modes." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Both mention version compatibility, but one is a Python-version support break and the other is a model-specific loading failure in transformers>=5.1.0; not the same underlying bug." + }, + { + "left": "issue:44295", + "right": "issue:46506", + "accept": false, + "reason": "One is about reading a registered buffer for position_ids; the other is CPU/CUDA divergence in a document layout model. Different mechanisms and fixes." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "A pipeline processor lookup bug and a notebook syntax error are unrelated; they do not share a concrete code path." + }, + { + "left": "issue:39692", + "right": "issue:46932", + "accept": false, + "reason": "SigLIP2 docs/quantization example problems and Gemma4 quantized embedder dtype casting are different models and different failure modes." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Both involve version/platform checks around attention code, but one is a compiled flex_attention failure on torch>=2.9 and the other is an incorrect PyTorch version guard for AuxRequest import." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Both concern model/config loading, but one is an unrecognized model error and the other is composite Qwen text_config unwrapping; not the same bug." + }, + { + "left": "issue:43673", + "right": "issue:46424", + "accept": false, + "reason": "One is a missing GenerationMixin cache in chunked_prefill; the other is cache max length handling for quantized cache only. Related area, but distinct defects." + }, + { + "left": "issue:42994", + "right": "issue:47333", + "accept": false, + "reason": "Both are saving/offloading-related, but one is a generic quantized model save failure and the other is a specific disk-offloaded Qwen3-30B-A3B save failure; too model-specific to treat as one bug." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "A BeitImageProcessorFast label reduction bug is unrelated to text-classification docs mentioning return_all_scores." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Generation cache behavior and ANSI escape output in loading_report are unrelated subsystems." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are different training-time concerns with no shared code path." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "A notebook syntax error and a _torch_extract_fbank_features issue are unrelated." + }, + { + "left": "issue:44279", + "right": "issue:46515", + "accept": false, + "reason": "A dependency issue with transformers and an incompatibility with torch 2.4.1 are both environment-related, but not the same concrete failure." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "One is a LlamaConfig validation rule for head_dim; the other is rope-parameter dict conversion with JSON-loaded lists. Different validation logic." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "A vague 'ADD THE DATA' issue and an audio feature extraction bug are not the same problem." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax errors and AddedToken keyword-argument duplication are unrelated." + }, + { + "left": "issue:44792", + "right": "issue:46424", + "accept": false, + "reason": "A janus image-generation test failure and cache length handling for quantized cache affect different model paths." + }, + { + "left": "issue:42994", + "right": "issue:47332", + "accept": false, + "reason": "Both involve saving offloaded/quantized models, but the reproductions target different architectures and storage/offload setups; not safely mergeable as one PR." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized Ministrals/Devstrals and FlashAttention failures in Ministral-3 are separate bugs in different parts of the model stack." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention flash attention, but one is a Ministral FlashAttention failure and the other is an AttributeError on s_aux=None for sink-less models; different concrete failures." + }, + { + "left": "issue:44327", + "right": "issue:46752", + "accept": false, + "reason": "A QA pipeline top-k bounds crash and apply_chat_template on empty list are different index/error conditions in different APIs." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python-version compatibility and a generic dependency issue are too broad and not the same underlying defect." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "ImportError for Qwen2_5_VLForConditionalGeneration and a nonexistent import from image_utils are both import problems, but for different symbols and modules." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "T5Gemma embedding resize behavior and Whisper empty-transcription generation after align_special_tokens are unrelated model behaviors." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "A video-classification processor lookup bug and remote_code breakages are not the same code path." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "One is a processor/tokenizer regression with images; the other is DeepSeek tokenizer output changes in v5. Different tokenizer bugs." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Docs behavior mismatch for return_all_scores and an unrecognized model/config error do not describe the same issue." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A dependency issue and a bumped MIN_PEFT_VERSION not yet on PyPI are both packaging-related, but not the same bug." + }, + { + "left": "issue:43901", + "right": "issue:47217", + "accept": false, + "reason": "Text-classification docs and VL image-placeholder handling are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "A tokenizer BPE-marker bug and a placeholder request issue are unrelated; no shared code path." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "SigLIP2 documentation/quantization errors and text-classification docs mentioning return_all_scores are separate documentation or example issues, not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 44, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:41950", + "issue:43299", + "issue:43381", + "issue:43479", + "issue:43577", + "issue:43644", + "issue:43650", + "issue:43653", + "issue:43854", + "issue:43873", + "issue:43931", + "issue:43976", + "issue:44016", + "issue:44075", + "issue:44242", + "issue:44291", + "issue:44351", + "issue:44368", + "issue:44462", + "issue:44466", + "issue:44683", + "issue:44779", + "issue:45081", + "issue:45200", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45357", + "issue:45375", + "issue:45405", + "issue:45542", + "issue:45588", + "issue:45593", + "issue:45812", + "issue:45823", + "issue:45907", + "issue:46082", + "issue:46123", + "issue:46209", + "issue:46369", + "issue:46399", + "issue:46512", + "issue:46614", + "issue:47000", + "issue:47085", + "issue:47145", + "issue:47405" + ], + "soft_pairs": [ + "issue:43299|issue:45216", + "issue:44779|issue:45812", + "issue:43479|issue:45375", + "issue:46614|issue:47085", + "issue:41950|issue:45823", + "issue:46123|issue:46369", + "issue:43577|issue:45216", + "issue:44466|issue:45276", + "issue:43976|issue:45405", + "issue:44075|issue:44368", + "issue:43653|issue:45593", + "issue:44462|issue:45081", + "issue:43381|issue:47000", + "issue:47145|issue:47405", + "issue:43299|issue:45357", + "issue:43650|issue:46209", + "issue:43644|issue:47145", + "issue:44351|issue:45278", + "issue:43299|issue:43854", + "issue:44683|issue:45542", + "issue:46082|issue:46399", + "issue:44016|issue:44291", + "issue:43931|issue:47145", + "issue:46082|issue:46512", + "issue:45081|issue:45907", + "issue:44242|issue:45200", + "issue:45588|issue:47000", + "issue:43873|issue:47145" + ], + "prepared_review_unit_hash": "71978428e799810ce435bb3fd8bd104b1694703f057f06e639733ba8c28cf08e", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35807, + "estimated_input_tokens": 8952, + "estimated_eval_tokens": 18160 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: the candidate pairs mostly share only broad themes like tokenizers, loading, or regressions, but not the same concrete bug. I found no pair with enough evidence to merge as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45216 is the clearest concrete regression report with a specific version boundary and a well-scoped model-saving failure, making it the best representative issue in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45216 is the most actionable and specific of the reports; it has a precise regression claim and is easier to anchor than the broader or more ambiguous items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Both involve Qwen model regressions, but one is a loading failure for Qwen3VL MoE and the other is an incorrect save_pretrained output for Qwen3.5; different code paths." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes (incorrect tokenization vs wrong IDs for Granite)." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Both concern config handling, but they are different models and different missing/defaulted fields; no same underlying bug is evident." + }, + { + "left": "issue:46614", + "right": "issue:47085", + "accept": false, + "reason": "One is a stale language backend in a specific tokenizer class, the other is a broad thread-safety regression in fast tokenizers; not the same issue." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "These are unrelated subsystems: a pipeline processor lookup bug versus a device-mapping mismatch in Gemma4 forward passes." + }, + { + "left": "issue:46123", + "right": "issue:46369", + "accept": false, + "reason": "Different symptoms and code paths: partial-batch handling in a pipeline versus a memory overflow in a model invocation." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Both touch loading/saving dtype or regression behavior, but one is a BLIP2 dtype propagation issue and the other is a Qwen3.5 save_pretrained regression." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Both mention tied weights or embeddings, but one is serialization inconsistency and the other is resize_token_embeddings not updating Gemma4 internals." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "A Python version compatibility break and a dependency version bump are unrelated." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Different bugs: optimizer argument handling versus a model config warning about tied embeddings." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer special-token registration is unrelated to D-FINE auxiliary-loss handling." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-loading related, but one ignores tokenizer.json while the other crashes in a Mistral regex patch; different failures." + }, + { + "left": "issue:43381", + "right": "issue:47000", + "accept": false, + "reason": "Gradient checkpointing in eval mode and RTDetr loss infeasibility are separate model/training bugs." + }, + { + "left": "issue:47145", + "right": "issue:47405", + "accept": false, + "reason": "Both are silent corruption/loading issues, but one is FP8 weight_scale application and the other is wrong Auto class/model_type conversion skipping." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Same broad Qwen3.5 family, but one is a load-time model initialization failure and the other is incorrect save_pretrained key serialization." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "No overlap beyond being user-reported failures; one is placeholder text, the other is an FP8 inference issue on B200." + }, + { + "left": "issue:43644", + "right": "issue:47145", + "accept": false, + "reason": "Different loading/saving corruption mechanisms; non-persistent buffers filled with junk is not the same as missing FP8 weight scaling." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "A specific missing export cannot be merged with a broad report of many import errors without evidence they share the same root cause." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Different models and different load failures; there is no indication they stem from the same bug." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch 2.9 is unrelated to tensorboard/TensorFlow backend installation issues." + }, + { + "left": "issue:46082", + "right": "issue:46399", + "accept": false, + "reason": "Different components and failure modes: LlamaConfig validation versus a Qwen3.5 kernel registration error." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "A notebook syntax error and an init_empty_weights TypeError are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:47145", + "accept": false, + "reason": "A Qwen3-VL weight-shape mismatch is not the same as FP8 frozen checkpoints missing weight_scale application." + }, + { + "left": "issue:46082", + "right": "issue:46512", + "accept": false, + "reason": "Both concern config/dtype semantics, but one is head_dim validation and the other is stale model.config.dtype after from_config." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer regex patching and list_repo_files revision forwarding are unrelated API bugs." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Both involve missing defaults in model behavior, but they affect different models and different inputs/loss logic." + }, + { + "left": "issue:45588", + "right": "issue:47000", + "accept": false, + "reason": "An attention integration AttributeError and a cost-matrix infeasibility in RTDetrLoss are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:47145", + "accept": false, + "reason": "Quantized offloading behavior is not the same bug as FP8 compressed-tensors loading without applying weight_scale." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 45, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:33290", + "issue:42175", + "issue:42796", + "issue:42994", + "issue:43317", + "issue:43441", + "issue:43479", + "issue:43525", + "issue:43575", + "issue:43643", + "issue:43673", + "issue:43761", + "issue:43784", + "issue:43824", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44186", + "issue:44188", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44336", + "issue:44589", + "issue:44623", + "issue:44671", + "issue:44749", + "issue:44779", + "issue:45103", + "issue:45325", + "issue:45362", + "issue:45372", + "issue:45397", + "issue:45542", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45800", + "issue:45823", + "issue:45834", + "issue:45859", + "issue:46399", + "issue:46846", + "issue:46851", + "issue:47000", + "issue:47145", + "issue:47217" + ], + "soft_pairs": [ + "issue:45103|issue:45372", + "issue:45859|issue:46851", + "issue:44336|issue:46851", + "issue:42994|issue:43317", + "issue:43441|issue:45834", + "issue:33290|issue:45656", + "issue:45542|issue:45800", + "issue:44623|issue:44749", + "issue:44671|issue:44779", + "issue:44246|issue:47217", + "issue:42175|issue:43976", + "issue:44016|issue:45823", + "issue:44623|issue:45397", + "issue:43525|issue:43881", + "issue:44186|issue:47000", + "issue:43479|issue:43525", + "issue:43673|issue:43761", + "issue:43784|issue:45542", + "issue:44261|issue:44263", + "issue:43525|issue:43643", + "issue:43901|issue:44246", + "issue:45325|issue:46851", + "issue:43575|issue:47145", + "issue:45600|issue:45712", + "issue:42796|issue:43441", + "issue:44188|issue:44336", + "issue:44589|issue:45362", + "issue:46399|issue:46846", + "issue:43824|issue:43976" + ], + "prepared_review_unit_hash": "835948a0d236e3c72f98b780e1bc98bc7c973129c9bf498884ac887ec66c88f4", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36300, + "estimated_input_tokens": 9075, + "estimated_eval_tokens": 18406 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the issues span unrelated models, config bugs, tokenizer errors, loading/offload problems, docs, and CI cleanup. None of the soft pairs look like the same underlying bug or change, so all are rejected.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "Different failure modes and code paths: auto-docstring annotation handling vs a mistral_common import error in Gemma 4 processor loading." + }, + { + "left": "issue:45859", + "right": "issue:46851", + "accept": false, + "reason": "Same model family but different bugs: CPU-offload rotary embedding compatibility vs variable-length position_ids handling." + }, + { + "left": "issue:44336", + "right": "issue:46851", + "accept": false, + "reason": "Unrelated: ANSI output handling in loading_report vs Qwen3.5 variable-length input behavior." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Both quantization-related, but one is checkpoint saving and the other is dequantized model loading with device_map auto/offload." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Completely different models and failures: FlashAttention import/runtime issue vs long OCR input index error." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Both mention deepspeed, but one is OOM with Adafactor and the other is optimizer.step being called twice; not the same bug." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "Different problems: missing TensorFlow backend dependency vs general torch/transformers incompatibility." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "No shared underlying issue: missing files on processor.save_pretrained vs a slowdown report after upgrading Transformers." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Both are incorrect model outputs, but for different models and causes; not the same code-path bug." + }, + { + "left": "issue:44246", + "right": "issue:47217", + "accept": false, + "reason": "Import performance issue vs VL image-placeholder handling bug; unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Different scope: packaging backend selection vs Python version compatibility regression." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error and Gemma4 device mismatch are unrelated." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Processor save_pretrained file emission bug vs Gemma-4 zero3 from_pretrained issue; different code paths." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Both are model-loading/config-related, but one is a missing pad_token_id attribute and the other is a GLM-4V loading failure; not enough to treat as one bug." + }, + { + "left": "issue:44186", + "right": "issue:47000", + "accept": false, + "reason": "Tokenizer padding/NER crash vs RTDetrLoss infeasible cost matrix; unrelated." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Both involve config defaults, but for different models and different missing-field behavior." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Different regressions in different APIs: GenerationMixin chunked_prefill cache vs CLIPVisionModel hidden_states output." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Import-time NameError in sentence-transformers integration vs TensorFlow backend installation problem; unrelated." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Both are model-specific issues, but one is an rms_norm_eps precision/config problem and the other is torch.split return-value handling; not the same bug." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Both mention missing config fields, but one is a model config attribute omission and the other is trust_remote_code AutoConfig returning incomplete objects; too different to merge." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs mismatch for return_all_scores vs import performance issue; unrelated." + }, + { + "left": "issue:45325", + "right": "issue:46851", + "accept": false, + "reason": "Both touch position_ids, but they affect different Qwen models and distinct logic paths." + }, + { + "left": "issue:43575", + "right": "issue:47145", + "accept": false, + "reason": "OOM during tensor-parallel loading vs silent FP8 checkpoint scaling corruption; unrelated." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both are cleanup/maintenance-type reports, but they concern different files and different repository checks." + }, + { + "left": "issue:42796", + "right": "issue:43441", + "accept": false, + "reason": "Tokenizer BPE-marker handling and FlashAttention failure are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile vs ANSI code emission in loading_report; unrelated." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Storage-object lookup failure vs Qwen3.5 chat crash; unrelated." + }, + { + "left": "issue:46399", + "right": "issue:46846", + "accept": false, + "reason": "Same model family, but one is a kernel registration/decorator crash and the other is a tensor-parallel plan omission causing OOM/shape errors; different fixes." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Import error for a model class vs Python version compatibility; unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 46, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36246", + "issue:42491", + "issue:42617", + "issue:42915", + "issue:43299", + "issue:43526", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43723", + "issue:43976", + "issue:44162", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44373", + "issue:44496", + "issue:44871", + "issue:45003", + "issue:45072", + "issue:45083", + "issue:45103", + "issue:45362", + "issue:45397", + "issue:45561", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:46291", + "issue:46399", + "issue:46424", + "issue:46506", + "issue:46614", + "issue:46655", + "issue:46665", + "issue:46726", + "issue:46752", + "issue:46821", + "issue:46829", + "issue:46851", + "issue:47176", + "issue:47177", + "issue:47217", + "issue:47312", + "issue:47333" + ], + "soft_pairs": [ + "issue:45362|issue:47333", + "issue:44336|issue:44373", + "issue:44246|issue:44496", + "issue:43723|issue:44162", + "issue:42617|issue:43976", + "issue:44297|issue:45397", + "issue:47176|issue:47177", + "issue:46291|issue:46399", + "issue:45083|issue:46851", + "issue:46655|issue:46726", + "issue:46399|issue:46851", + "issue:42491|issue:43299", + "issue:45003|issue:47217", + "issue:45072|issue:46829", + "issue:43526|issue:45003", + "issue:43540|issue:46752", + "issue:44295|issue:45397", + "issue:45103|issue:46399", + "issue:44246|issue:44871", + "issue:45588|issue:46851", + "issue:45003|issue:46821", + "issue:45561|issue:45563", + "issue:44336|issue:46424", + "issue:43550|issue:46506", + "issue:46614|issue:46665", + "issue:36246|issue:43596", + "issue:43540|issue:46655", + "issue:42915|issue:47312", + "issue:44242|issue:45593" + ], + "prepared_review_unit_hash": "7ee6d2124596d96fe4f0e2b8016b64a68c040d20bfbdb34ae94ea17e0934851a", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35808, + "estimated_input_tokens": 8952, + "estimated_eval_tokens": 18160 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a mixed bag of mostly unrelated Transformers issues spanning Qwen, tokenizers, model loading, kernels, security, CI, and docs. None of the soft pairs look like the same underlying bug or change, so there is no credible duplicate canonical here.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45362", + "right": "issue:47333", + "accept": false, + "reason": "Different Qwen runtime problems: chat/template crash vs offloading save failure; no shared code path." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "Unrelated doc/report formatting vs a docstring typo; not the same bug." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import slowness and unrecognized-model config errors are different failure modes." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading regression vs ESM2 model breakage; different subsystems." + }, + { + "left": "issue:42617", + "right": "issue:43976", + "accept": false, + "reason": "A script/runtime failure and a Python-version compatibility issue are not the same underlying problem." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save metadata mismatch vs Gemma-4 ZeRO/from_pretrained failure; different paths." + }, + { + "left": "issue:47176", + "right": "issue:47177", + "accept": false, + "reason": "Distinct security issues: shell-substitution allowlist bypass vs archive path traversal/RCE." + }, + { + "left": "issue:46291", + "right": "issue:46399", + "accept": false, + "reason": "Both involve kernels, but one is an import-time dependency/version constraint and the other is a missing registration decorator in Qwen3.5." + }, + { + "left": "issue:45083", + "right": "issue:46851", + "accept": false, + "reason": "Both mention Qwen and positions, but one is a feature-extract length helper and the other is variable-length position_ids handling." + }, + { + "left": "issue:46655", + "right": "issue:46726", + "accept": false, + "reason": "Video padding bug vs missing-vision-backend processor error; not the same bug." + }, + { + "left": "issue:46399", + "right": "issue:46851", + "accept": false, + "reason": "Different Qwen3.5 failures: kernel registration vs variable-length input handling." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both are Qwen MoE regressions, but one is LoRA compatibility across hf4.x/hf5.x and the other is model loading on transformers v5; too different to merge." + }, + { + "left": "issue:45003", + "right": "issue:47217", + "accept": false, + "reason": "Unsafe sys.modules access is unrelated to VL image-placeholder parsing." + }, + { + "left": "issue:45072", + "right": "issue:46829", + "accept": false, + "reason": "Different inference bugs: dtype mismatch vs incorrect output field assignment." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BEIT label reduction logic and sys.modules access are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:46752", + "accept": false, + "reason": "Qwen3OmniMoe video-input error vs empty-list chat template crash; different code paths." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Buffer access for position_ids vs Gemma-4 ZeRO loading; not the same underlying issue." + }, + { + "left": "issue:45103", + "right": "issue:46399", + "accept": false, + "reason": "Auto-docstring future-annotations crash is unrelated to Qwen kernel registration." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import-time performance regression and eos_token_id config inconsistency are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:46851", + "accept": false, + "reason": "Flash-attention sink handling vs variable-length position_ids; different failures." + }, + { + "left": "issue:45003", + "right": "issue:46821", + "accept": false, + "reason": "sys.modules access and TimesFM window_size errors are unrelated." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "A test-file race and a stale warning in generate() are different issues." + }, + { + "left": "issue:44336", + "right": "issue:46424", + "accept": false, + "reason": "ANSI output formatting bug vs cache-length logic bug; no shared code path." + }, + { + "left": "issue:43550", + "right": "issue:46506", + "accept": false, + "reason": "torch.compile/SDPA failure and CPU-vs-CUDA divergence are distinct model/runtime problems." + }, + { + "left": "issue:46614", + "right": "issue:46665", + "accept": false, + "reason": "Phonemizer backend caching and from_pretrained-in-class behavior are different issues." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Qwen2_5_VLImageProcessor import error is unrelated to BertModel ZeRO init indexing." + }, + { + "left": "issue:42915", + "right": "issue:47312", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and kernels 0.16 incompatibility are different regressions." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Both concern missing losses, but they affect different models and mechanisms, so they are not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 47, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36246", + "issue:41950", + "issue:42757", + "issue:42886", + "issue:43257", + "issue:43295", + "issue:43352", + "issue:43475", + "issue:43901", + "issue:44038", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44220", + "issue:44295", + "issue:44351", + "issue:44451", + "issue:44466", + "issue:44479", + "issue:44623", + "issue:44749", + "issue:44779", + "issue:44877", + "issue:44987", + "issue:44995", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45250", + "issue:45292", + "issue:45335", + "issue:45375", + "issue:45459", + "issue:45561", + "issue:45588", + "issue:45600", + "issue:45701", + "issue:45759", + "issue:46132", + "issue:46170", + "issue:46369", + "issue:46506", + "issue:46655", + "issue:46726", + "issue:46752", + "issue:47217", + "issue:47321" + ], + "soft_pairs": [ + "issue:44779|issue:45292", + "issue:41950|issue:44220", + "issue:44479|issue:46655", + "issue:46726|issue:47217", + "issue:43901|issue:45003", + "issue:43257|issue:45137", + "issue:45375|issue:45759", + "issue:36246|issue:44351", + "issue:36246|issue:42757", + "issue:42886|issue:46132", + "issue:45292|issue:45701", + "issue:43295|issue:45292", + "issue:44295|issue:44749", + "issue:45137|issue:45600", + "issue:43352|issue:45250", + "issue:44466|issue:45335", + "issue:44162|issue:45200", + "issue:44451|issue:46752", + "issue:46170|issue:47321", + "issue:44995|issue:45200", + "issue:44060|issue:44077", + "issue:44877|issue:45375", + "issue:43475|issue:46369", + "issue:45003|issue:45561", + "issue:44987|issue:45459", + "issue:44479|issue:46170", + "issue:44995|issue:45588", + "issue:44038|issue:46506", + "issue:44623|issue:46726" + ], + "prepared_review_unit_hash": "876797a17597f25408310bcbf1dd637103861bcedc21e2d8fc352ff035f39cfd", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35692, + "estimated_input_tokens": 8923, + "estimated_eval_tokens": 18102 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "A heterogeneous cluster of Transformers bug reports: mostly model-specific regressions around v5/v5.3, multimodal processors/configs, DeepSpeed, imports, and tokenizer behavior. The items share broad theme overlap but do not look like true duplicates of one underlying bug.", + "confidence": 0.22, + "canonical_issue_reason": "Issue 44479 is the broadest, most representative regression report in the set: it describes a v5.3.0 video-input regression affecting multiple related Qwen VL model variants, which is closer to the center of the cluster than the narrower single-model or unrelated infrastructure bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44479 is the best overall representative because it is a concrete, user-facing regression with multiple affected models and a clear version boundary, making it a stronger triage anchor than the many narrower or unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer output changes and embedding-resize wiring are different code paths and different failure modes." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Both mention pipelines/features, but one is about processor selection and the other is a feature-extraction bug." + }, + { + "left": "issue:44479", + "right": "issue:46655", + "accept": false, + "reason": "Both are video-related regressions, but they affect different model families and the reported defects are different." + }, + { + "left": "issue:46726", + "right": "issue:47217", + "accept": false, + "reason": "AutoProcessor backend detection and VL image-placeholder handling are distinct problems." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch for pipeline arguments is unrelated to unsafe sys.modules access." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Both involve DeepSpeed, but one is Qwen3 MoE weight conversion and the other is a ZeRO3 deque crash." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Both touch composite Qwen configs, but one is strict-schema field loss and the other is from_config unwrapping behavior." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Different missing exports: Qwen2.5 VL image processor vs HybridCache." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Different packages and causes: Transformers export vs huggingface_hub import." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline cache loading and AttentionInterface registration are unrelated." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Embedding resize behavior is not the same bug as tokenization changes across versions." + }, + { + "left": "issue:43295", + "right": "issue:45292", + "accept": false, + "reason": "Processor/tokenizer regression is unrelated to token embedding resizing." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer access and a filtering slowdown are unrelated." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed ZeRO3 runtime failure and removed auto-mapping config references are different issues." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "A model explicitly not supporting Flash Attention 2.0 is not the same as the generic Flash Attention issue." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Serialization of tied weights and t5gemma embedding resize behavior are different bugs." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma 4 multimodal token-type defaults are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:46752", + "accept": false, + "reason": "Model loading failure and empty-list chat-template IndexError are unrelated." + }, + { + "left": "issue:46170", + "right": "issue:47321", + "accept": false, + "reason": "Windows ROCm import/dependency failure and AMD quark-class update are different concerns." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "Moe cache crash on second forward pass is unrelated to Gemma 4 token-type IDs." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Incorrect tied-weights warnings and patchtsmixer post_init policy are different code paths." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Both involve config strictness, but they are distinct models and opposite schema problems." + }, + { + "left": "issue:45459", + "right": "issue:46132", + "accept": false, + "reason": "Tokenizer error masking from protobuf handling is unrelated to AttentionInterface registration." + }, + { + "left": "issue:45561", + "right": "issue:45003", + "accept": false, + "reason": "A pytest-xdist temp-file race is unrelated to sys.modules access in modeling_utils." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Loading a specific model failing and protobuf-related tokenizer error masking are different root causes." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "Video-input regression and Windows ROCm import failure are unrelated." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Second-forward cache crash and flash_attention s_aux handling are different runtime bugs." + }, + { + "left": "issue:44038", + "right": "issue:46506", + "accept": false, + "reason": "Qwen3-VL-Moe regression and CPU/CUDA output divergence are not the same issue." + }, + { + "left": "issue:44623", + "right": "issue:46726", + "accept": false, + "reason": "Missing save_pretrained files and backend-missing image processor warnings affect different phases of processor handling." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 48, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:42199", + "issue:42886", + "issue:42994", + "issue:43065", + "issue:43408", + "issue:43475", + "issue:43540", + "issue:43653", + "issue:43723", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44162", + "issue:44188", + "issue:44242", + "issue:44361", + "issue:44368", + "issue:44483", + "issue:44749", + "issue:44857", + "issue:44877", + "issue:44977", + "issue:45103", + "issue:45229", + "issue:45397", + "issue:45412", + "issue:45563", + "issue:45593", + "issue:45656", + "issue:45685", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46132", + "issue:46170", + "issue:46202", + "issue:46431", + "issue:46710", + "issue:46726", + "issue:46752", + "issue:46851", + "issue:46858", + "issue:47020", + "issue:47068", + "issue:47145", + "issue:47312" + ], + "soft_pairs": [ + "issue:45685|issue:46431", + "issue:29870|issue:30333", + "issue:44877|issue:46170", + "issue:44242|issue:44857", + "issue:45397|issue:45656", + "issue:43475|issue:43540", + "issue:44368|issue:44977", + "issue:43653|issue:44162", + "issue:42994|issue:47312", + "issue:43723|issue:46710", + "issue:42199|issue:46851", + "issue:43065|issue:43408", + "issue:45412|issue:46123", + "issue:42886|issue:45229", + "issue:43931|issue:44368", + "issue:43906|issue:46202", + "issue:44188|issue:46202", + "issue:44361|issue:44749", + "issue:45987|issue:46082", + "issue:44368|issue:45563", + "issue:46132|issue:47145", + "issue:46752|issue:47020", + "issue:45103|issue:46858", + "issue:44483|issue:46202", + "issue:43994|issue:46726", + "issue:45593|issue:47068" + ], + "prepared_review_unit_hash": "17af3efcc080e5389ead5bd18ae9fdc3db5d6c4cecfad2ac3e34a0f2774a24d5", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35614, + "estimated_input_tokens": 8904, + "estimated_eval_tokens": 18064 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are a loose collection of unrelated bug reports across tokenizers, generation, config loading, vision models, and distributed training. None of the soft pairs look like the same underlying bug or a mergeable PR-level duplicate.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43906 is the most actionable standalone report because it is explicitly framed as an isolated reproduction, but it is not a canonical duplicate target for the rest of this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45685", + "right": "issue:46431", + "accept": false, + "reason": "MPS histogram/int error in MoE is unrelated to a chat-template processor missing a chat template." + }, + { + "left": "issue:29870", + "right": "issue:30333", + "accept": false, + "reason": "Bart generation UnboundLocalError and MLflow job-state reporting are different subsystems and failure modes." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Strict config loading for granite_speech is unrelated to Windows/ROCm torch.distributed import issues." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Load-balancing loss omission and LwDetrImageLoss AMP crash are different code paths and bugs." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma-4 Zero3 from_pretrained failure is unrelated to Deepspeed optimizer stepping twice." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "Missing fpn_position_embeddings in SAM 3 video output is not the same as Qwen3OmniMoe video-input validation." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "A tied-embeddings warning during LoRA fine-tuning is not the same bug as flash-attention generation failure." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are different tokenizer/model issues." + }, + { + "left": "issue:42994", + "right": "issue:47312", + "accept": false, + "reason": "Quantized model save failure and kernel version incompatibility are not the same underlying problem." + }, + { + "left": "issue:43723", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer loading in v5 and DeepSeek accuracy regression after upgrade are distinct regressions." + }, + { + "left": "issue:42199", + "right": "issue:46851", + "accept": false, + "reason": "DETR-derived cardinality logic and Qwen3.5 variable-length position_ids are unrelated model-specific bugs." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Dummy Conv2d in SAM3 pixel decoder is unrelated to sam3_video/sam3_tracker config-type mismatch." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "RT-DETR memory release and MaskGenerationPipeline partial-batch dropping are different issues." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU OOM are unrelated failures." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch is not the same as a Qwen3.5 tied-embeddings warning." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "An isolated reproduction of another issue is too generic to match a Gemma pipeline roundtrip failure on \u2581; no concrete shared code path is evident." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "Torch.compile attention-kernel divergence is unrelated to Gemma tokenizer/pipeline roundtrip behavior." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "MLLukeTokenizer AttributeError and Transformers slowdown after upgrade are different bugs." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "StaticCache seq-length return type and LlamaConfig head_dim validation are separate config/cache issues." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Two warning reports with different triggers and semantics; not the same bug." + }, + { + "left": "issue:46132", + "right": "issue:47145", + "accept": false, + "reason": "AttentionInterface.register behavior change and FP8 weight_scale loading corruption are unrelated." + }, + { + "left": "issue:46752", + "right": "issue:47020", + "accept": false, + "reason": "Empty-list chat-template IndexError and XLMRobertaTokenizer Unigram init type mismatch are different tokenizer bugs." + }, + { + "left": "issue:45103", + "right": "issue:46858", + "accept": false, + "reason": "Docstring annotation parsing crash and static-cache slice arithmetic on non-CUDA backends are unrelated." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "Chat/completions request rejection is unrelated to Gemma pipeline roundtrip failures." + }, + { + "left": "issue:43994", + "right": "issue:46726", + "accept": false, + "reason": "SigLIP2 nonsense outputs and missing vision-backend processor errors are different auto-model/processor issues." + }, + { + "left": "issue:45593", + "right": "issue:47068", + "accept": false, + "reason": "D-FINE auxiliary-loss omission and Deepspeed loss host sync are different training bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 49, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:36010", + "issue:39692", + "issue:41628", + "issue:43065", + "issue:43232", + "issue:43475", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43701", + "issue:43723", + "issue:43784", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44373", + "issue:44623", + "issue:44661", + "issue:44749", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44964", + "issue:45137", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45397", + "issue:45478", + "issue:45561", + "issue:45656", + "issue:46123", + "issue:46132", + "issue:46326", + "issue:46399", + "issue:46421", + "issue:46463", + "issue:46726", + "issue:46821", + "issue:46851", + "issue:46877", + "issue:47000", + "issue:47059", + "issue:47217" + ], + "soft_pairs": [ + "issue:44188|issue:46421", + "issue:43701|issue:44749", + "issue:44749|issue:45656", + "issue:44297|issue:44623", + "issue:43065|issue:46123", + "issue:43646|issue:44162", + "issue:43475|issue:46821", + "issue:43475|issue:43643", + "issue:44623|issue:45656", + "issue:44964|issue:45137", + "issue:39692|issue:46726", + "issue:43784|issue:46399", + "issue:44373|issue:46851", + "issue:46463|issue:46877", + "issue:28282|issue:30990", + "issue:44871|issue:46132", + "issue:43723|issue:46326", + "issue:45561|issue:47217", + "issue:45230|issue:45397", + "issue:44188|issue:46851", + "issue:44869|issue:45356", + "issue:46877|issue:47000", + "issue:44038|issue:44623", + "issue:43232|issue:43673", + "issue:45375|issue:45478", + "issue:41628|issue:44933", + "issue:45310|issue:45375", + "issue:36010|issue:41628", + "issue:44661|issue:47059" + ], + "prepared_review_unit_hash": "f5f18bc565a52db4079178c1005c133bbac08f0ef0cf12d10bc2b691b2cd48bf", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36174, + "estimated_input_tokens": 9044, + "estimated_eval_tokens": 18344 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports and regressions. The clearest duplicate-looking item is the Qwen3.5 MoE `from_pretrained` error pair, with 45310 as the better canonical issue and 45478 as a later near-duplicate.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45310 is the earliest and cleanest representative of the near-identical Qwen3.5 MoE `from_pretrained` failure; issue:45478 repeats the same title/symptom later.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the strongest global representative here because it has a clear concrete regression, a close duplicate, and is more coherent than the many unrelated issues in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44188", + "right": "issue:46421", + "accept": false, + "reason": "Different bugs: one is attention-kernel divergence under `torch.compile`, the other is cache lazy-init causing recompilation in chunked prefill." + }, + { + "left": "issue:43701", + "right": "issue:44749", + "accept": false, + "reason": "Unrelated symptoms: checkpoint key mismatch vs a preprocessing slowdown after upgrading Transformers." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "One is a performance regression; the other is a deepspeed optimizer step-counting bug." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both are save-related, but they affect different APIs and failures: tokenizer config mismatch vs processor files missing." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "Different models and failure modes: SAM3 decoder placeholder Conv2d vs MaskGenerationPipeline dropping a final partial batch." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Broad v5 custom init breakage is not the same as the ESM2-specific breakage described in 44162." + }, + { + "left": "issue:43475", + "right": "issue:46821", + "accept": false, + "reason": "Different code paths and attributes: SAM3 vision encoder output missing a field vs TimesFM `window_size` attribute access." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM3 missing `fpn_position_embeddings` is unrelated to `trust_remote_code=True` dropping fields in `AutoConfig`." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Different APIs and root causes: processor serialization vs deepspeed optimizer behavior." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Different subsystems and bugs: Phi-4 multimodal loading vs DeepSpeed ZeRO3 deque underflow." + }, + { + "left": "issue:39692", + "right": "issue:46726", + "accept": false, + "reason": "Docs-example errors are not the same as missing-vision-backend handling in AutoProcessor." + }, + { + "left": "issue:43784", + "right": "issue:46399", + "accept": false, + "reason": "Import-time `nn` NameError in sentence-transformers is unrelated to Qwen3.5 kernel registration failure." + }, + { + "left": "issue:44373", + "right": "issue:46851", + "accept": false, + "reason": "A wrong docstring for `position_ids` is not the same as a runtime variable-length input bug." + }, + { + "left": "issue:46463", + "right": "issue:46877", + "accept": false, + "reason": "Incorrect validation error text in T5 is unrelated to docstring lint output being printed at import time." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "Missing PyTorch import error and a loading hang are different failure modes." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Gemma EOS config mismatch does not match the behavior change in `AttentionInterface.register`." + }, + { + "left": "issue:43723", + "right": "issue:46326", + "accept": false, + "reason": "Tokenizer loading in v5 and mistral_common final-message handling are separate code paths." + }, + { + "left": "issue:45561", + "right": "issue:47217", + "accept": false, + "reason": "A pytest-xdist file race is unrelated to VL models failing on image placeholders in user text." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "`Bug report` is too generic to be a duplicate of a specific Zero3 `from_pretrained` issue." + }, + { + "left": "issue:44188", + "right": "issue:46851", + "accept": false, + "reason": "Both mention generation-related behavior, but one is about compiled attention divergence and the other about Qwen3.5 variable-length `position_ids`." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper timestamp decoding crash and Kimi tokenizer regression are different tokenization/decoding bugs." + }, + { + "left": "issue:46877", + "right": "issue:47000", + "accept": false, + "reason": "Doc-lint stdout noise is unrelated to RTDetrLoss cost-matrix infeasibility." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe loading problems are not the same as processor save-pretrained file omissions." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Both concern generation internals, but one is a sync-gpus kwargs update bug and the other is a missing cache in chunked prefill." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "These are related in model family, but one is a missing config field silently dropped and the other is a broader `from_pretrained` error report." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Different import failures: missing `AutoImageProcessor` export vs a non-existent import from `image_utils`." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "A Qwen3.5 MoE loading error is not the same as a missing `deepstack_visual_indexes` config field." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different symbols and modules (`GenerationMixin` vs `AutoImageProcessor`)." + }, + { + "left": "issue:44661", + "right": "issue:47059", + "accept": false, + "reason": "Model-mapping code failure and CLI completion breakage in huggingface_hub are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 50, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:33290", + "issue:35141", + "issue:41628", + "issue:42175", + "issue:42371", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:43232", + "issue:43278", + "issue:43704", + "issue:43784", + "issue:43824", + "issue:43873", + "issue:43976", + "issue:44038", + "issue:44186", + "issue:44222", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44743", + "issue:44792", + "issue:44843", + "issue:44857", + "issue:44933", + "issue:44945", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45137", + "issue:45310", + "issue:45405", + "issue:45406", + "issue:45561", + "issue:45593", + "issue:45712", + "issue:45753", + "issue:45812", + "issue:46202", + "issue:46701", + "issue:46877", + "issue:47059", + "issue:47068", + "issue:47085" + ], + "soft_pairs": [ + "issue:43873|issue:44945", + "issue:42371|issue:45092", + "issue:45081|issue:45103", + "issue:43278|issue:45137", + "issue:45406|issue:46202", + "issue:44857|issue:46877", + "issue:46202|issue:46701", + "issue:45593|issue:46877", + "issue:43232|issue:44792", + "issue:44038|issue:45310", + "issue:41628|issue:44351", + "issue:42175|issue:44933", + "issue:44222|issue:45310", + "issue:44945|issue:45753", + "issue:45561|issue:47059", + "issue:30333|issue:33290", + "issue:45405|issue:46877", + "issue:44743|issue:47068", + "issue:42898|issue:44568", + "issue:42831|issue:44222", + "issue:43784|issue:43824", + "issue:42175|issue:45712", + "issue:43976|issue:44933", + "issue:44496|issue:45561", + "issue:30990|issue:35141", + "issue:44568|issue:47085", + "issue:44843|issue:47085", + "issue:42898|issue:45812", + "issue:42673|issue:43704", + "issue:44186|issue:47085" + ], + "prepared_review_unit_hash": "40f2a97d79900cf27b0aec6a94a951f894a5c359474de963ade6b2d2db0fbd65", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36097, + "estimated_input_tokens": 9025, + "estimated_eval_tokens": 18306 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose set of unrelated issues. Only one soft pair looks like the same underlying bug: the duplicated Qwen3ForCausalLM VRAM leak report. The rest are too semantically different to merge as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "issue:42673 is the earlier, more detailed report of the same Qwen3ForCausalLM VRAM leak seen again in issue:43704, so it is the better canonical issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the best global issue candidate because it is the clearest true duplicate pair in the cluster and represents a concrete, repeated bug report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Both involve model/runtime behavior, but one is quantization offloading and the other is pipeline-parallel output correctness; different bugs and code paths." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 settings and old InternVL2 meta-init incompatibility are unrelated failures." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Mistral regex loading crash vs auto_docstring future-annotations crash; different subsystems and root causes." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift in evaluate and a DeepSpeed ZeRO3 deque IndexError are not the same issue." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Both mention Gemma, but one is a missing processor attribute and the other is a tokenizer roundtrip failure; not the same bug." + }, + { + "left": "issue:44857", + "right": "issue:46877", + "accept": false, + "reason": "A CUDA float16 loss crash is unrelated to auto_docstring printing diagnostics at import time." + }, + { + "left": "issue:46202", + "right": "issue:46701", + "accept": false, + "reason": "Gemma tokenizer roundtrip failure and DiffusionGemma image-text pipeline failure are different concrete problems." + }, + { + "left": "issue:45593", + "right": "issue:46877", + "accept": false, + "reason": "D-FINE auxiliary-loss behavior and auto_docstring stdout logging are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus is unrelated to a janus test failure." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both are Qwen/MoE-related, but they describe different model variants and different from_pretrained failures." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Two separate import errors for different symbols; no shared underlying bug." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Missing TensorFlow in an extra dependency install and a nonexistent image_utils import are different packaging/API issues." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained on MoE and Qwen3.5 from_pretrained failure are not the same code-path problem." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output correctness and Qwen3_5 image-specific infinite loop are unrelated." + }, + { + "left": "issue:45561", + "right": "issue:47059", + "accept": false, + "reason": "A pytest-xdist file race and a huggingface_hub CLI compatibility error are unrelated." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLFlow reporting on failed training and a DeepSpeed Adafactor OOM are different failures." + }, + { + "left": "issue:45405", + "right": "issue:46877", + "accept": false, + "reason": "A version bump complaint for PEFT and auto_docstring stdout diagnostics are unrelated." + }, + { + "left": "issue:44743", + "right": "issue:47068", + "accept": false, + "reason": "Qwen3.5 recurrent-state reset logic and a DeepSpeed host-sync warning are different bugs." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer regressions, but one is clean_up_tokenization_spaces semantics and the other is add_special_tokens BOS/EOS behavior." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Both mention FP8, but one is accuracy and the other is save_pretrained serialization; not mergeable as one fix." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Separate import errors for sentence-transformers and Qwen2.5-VL symbols." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Dependency backend packaging and leftover dummy class cleanup are different maintenance tasks." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version compatibility and an image_utils import regression are unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model/config issue and xdist race condition are not the same bug." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "SentenceTransformers loading hang and embedding reinitialization during resize are different model-init problems." + }, + { + "left": "issue:44568", + "right": "issue:47085", + "accept": false, + "reason": "Tokenizer add_special_tokens behavior and tokenizer thread-safety are distinct issues." + }, + { + "left": "issue:44843", + "right": "issue:47085", + "accept": false, + "reason": "Offline-mode mistral regex patching and tokenizer thread-safety are different root causes." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one concerns cleanup space behavior and the other wrong token IDs for Granite models." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same title, same Qwen3ForCausalLM VRAM leak symptom, and clearly the same underlying bug reported twice." + }, + { + "left": "issue:44186", + "right": "issue:47085", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes on NER padding/truncation is unrelated to tokenizers thread-safety." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: issue:42673 and issue:43704 are the only clearly duplicated reports, and the rest of the soft pairs are rejected conservatively with reasons that do not overstate the evidence. No pull-request pairing concerns apply here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 51, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28282", + "issue:30333", + "issue:34689", + "issue:36331", + "issue:39692", + "issue:42222", + "issue:42673", + "issue:42796", + "issue:42898", + "issue:43116", + "issue:43295", + "issue:43479", + "issue:43723", + "issue:43874", + "issue:44016", + "issue:44060", + "issue:44075", + "issue:44079", + "issue:44190", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44448", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44869", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45200", + "issue:45310", + "issue:45405", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:46077", + "issue:46123", + "issue:46231", + "issue:46693", + "issue:46829", + "issue:46909", + "issue:47059", + "issue:47085" + ], + "soft_pairs": [ + "issue:43479|issue:46829", + "issue:44623|issue:45310", + "issue:42673|issue:46693", + "issue:39692|issue:46909", + "issue:43723|issue:44568", + "issue:44060|issue:45125", + "issue:45600|issue:47059", + "issue:30333|issue:34689", + "issue:27957|issue:30333", + "issue:44016|issue:46231", + "issue:43479|issue:44079", + "issue:42898|issue:44779", + "issue:42673|issue:47085", + "issue:42898|issue:43295", + "issue:30333|issue:36331", + "issue:43479|issue:44869", + "issue:42898|issue:44448", + "issue:46077|issue:46123", + "issue:22355|issue:28282", + "issue:44295|issue:45656", + "issue:42222|issue:45200", + "issue:43874|issue:45092", + "issue:44734|issue:45103", + "issue:42796|issue:43723", + "issue:44075|issue:46231", + "issue:43116|issue:44190", + "issue:44222|issue:44297", + "issue:45405|issue:45712" + ], + "prepared_review_unit_hash": "f64e5b0d62efa82dccf73c2d5d98e3fcc0520f30a65b0e113e4d9e55ae1cdd25", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35685, + "estimated_input_tokens": 8922, + "estimated_eval_tokens": 18100 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is broad and mostly heterogeneous, with only loose surface similarity around Transformers v5 regressions, tokenizers, and model loading. None of the soft-edge pairs look like the same concrete bug/change, so I would not merge any of them.", + "confidence": 0.76, + "canonical_issue_reason": "issue:43723 is the broadest and most actionable representative of the tokenizer/loading regression subset in this cluster, especially the v5-related failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43723 is the best single issue anchor for triage because it targets a general AutoTokenizer/v5 loading problem that is closer to several nearby regressions than the more model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43479", + "right": "issue:46829", + "accept": false, + "reason": "Different model families and different failure modes: config initialization vs output field mix-up." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "One is processor save_pretrained file omission; the other is a Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:42673", + "right": "issue:46693", + "accept": false, + "reason": "VRAM leak in multi-threaded dataloaders is unrelated to FA2 long-context performance regression." + }, + { + "left": "issue:39692", + "right": "issue:46909", + "accept": false, + "reason": "Both are documentation examples, but they are different examples with different concrete errors." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Both involve tokenizers in v5, but one is loading failure and the other is add_special_tokens behavior for a specific model." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Different Qwen3 variants and different bugs: tied-weights warning vs missing tensor-parallel plan." + }, + { + "left": "issue:45600", + "right": "issue:47059", + "accept": false, + "reason": "auto_mappings config cleanup is unrelated to a CLI completion API break in huggingface_hub." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow reporting on failed training jobs is unrelated to a Llama model-loading regression." + }, + { + "left": "issue:27957", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention model support and MLflow failure handling are different subsystems and bugs." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "Notebook syntax error and docs parameter-name mismatch are not the same underlying issue." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config defaulting bug and ModelOutput key assignment bug are distinct code paths." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they describe different model behavior changes and likely different fixes." + }, + { + "left": "issue:42673", + "right": "issue:47085", + "accept": false, + "reason": "VRAM leak in dataloader threads and tokenizer thread-safety regression are different concurrency problems." + }, + { + "left": "issue:42898", + "right": "issue:43295", + "accept": false, + "reason": "General clean_up_tokenization_spaces change and processor.tokenizer/images regression are not the same bug." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "Failed-job MLflow reporting and CustomTrainer compute_loss signature mismatch are unrelated training issues." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config defaults are unrelated to Whisper timestamp decoding on trailing replacement characters." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Different output changes for different models; not a single concrete code-path problem." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Bidirectional attention mask construction and partial-batch pipeline bookkeeping are unrelated." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Importing transformers.onnx and missing PyTorch for AutoModel are separate dependency/import failures." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "position_ids buffer read error and deepspeed double optimizer step are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken vitpose model files and Gemma 4 mm_token_type_ids defaults are different model-specific issues." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Missing image-patch API and remote-code meta-init incompatibility are both multimodal, but not the same bug." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "Serve KV-cache indexing crash and auto-docstring future-annotations crash are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43723", + "accept": false, + "reason": "Tokenizer BPE marker handling and tokenizer loading in v5 are different failures." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "Optimizer SGD args bug is not a documentation parameter-name typo." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Multi-label classification output bug and local dataset loading failure are separate example-script issues." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "FP8 MoE save_pretrained and tokenizer_class metadata mismatch are different save_pretrained paths." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "PEFT version pinning and leftover dummy class cleanup are unrelated repository maintenance issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 52, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:30990", + "issue:38175", + "issue:41628", + "issue:42757", + "issue:42831", + "issue:42907", + "issue:43122", + "issue:43232", + "issue:43475", + "issue:43479", + "issue:43504", + "issue:43644", + "issue:43723", + "issue:44190", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44327", + "issue:44360", + "issue:44373", + "issue:44448", + "issue:44484", + "issue:44485", + "issue:44623", + "issue:44933", + "issue:44977", + "issue:45071", + "issue:45092", + "issue:45245", + "issue:45310", + "issue:45656", + "issue:45712", + "issue:45834", + "issue:46050", + "issue:46082", + "issue:46132", + "issue:46209", + "issue:46305", + "issue:46369", + "issue:46424", + "issue:46431", + "issue:46614", + "issue:46762", + "issue:46909", + "issue:46932", + "issue:47059", + "issue:47276" + ], + "soft_pairs": [ + "issue:46209|issue:46431", + "issue:44448|issue:46132", + "issue:44297|issue:45656", + "issue:42831|issue:44623", + "issue:43479|issue:43644", + "issue:44222|issue:44623", + "issue:45245|issue:46132", + "issue:44933|issue:45712", + "issue:44373|issue:44485", + "issue:44297|issue:45310", + "issue:43122|issue:46209", + "issue:43723|issue:44977", + "issue:41628|issue:42757", + "issue:43232|issue:47276", + "issue:43504|issue:47059", + "issue:46614|issue:46932", + "issue:45071|issue:46082", + "issue:43232|issue:45834", + "issue:43475|issue:46305", + "issue:44222|issue:44484", + "issue:30990|issue:38175", + "issue:44295|issue:44623", + "issue:44360|issue:46762", + "issue:44263|issue:45092", + "issue:44327|issue:47059", + "issue:46369|issue:46909", + "issue:43232|issue:46424", + "issue:44263|issue:44373", + "issue:44190|issue:46082", + "issue:42907|issue:46050" + ], + "prepared_review_unit_hash": "253db94a4441074f64da4fa1138731e4e33a7d4d6781ddd03610623bc50f554f", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36293, + "estimated_input_tokens": 9074, + "estimated_eval_tokens": 18404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is noisy and heterogeneous: the issues share superficial keywords like loading/import/save_pretrained/tokenization, but they point to distinct bugs, models, and code paths. I do not see a safe canonical issue or PR in this set.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46209", + "right": "issue:46431", + "accept": false, + "reason": "FP8 inference on B200 vs missing chat template; unrelated failure modes and code paths." + }, + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "Different problems: Pegasus output regression vs AttentionInterface.register behavior change." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to deepspeed optimizer stepping twice." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue is not the same as processor.save_pretrained missing files." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Phi4MultimodalConfig init bug and non-persistent buffer junk are different bugs in different components." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Both mention saving, but one is FP8 moe serialization and the other is processor file output." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category limit runtime error is unrelated to AttentionInterface.register semantics." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Both are import-related, but they concern different modules and different root causes." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Wrong docstring for position_ids is not the GLM-5 RoPE implementation bug." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save_pretrained mismatch and Qwen3.5 MoE from_pretrained error are distinct issues." + }, + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenization differences across versions are unrelated to FP8 inference on B200." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading in v5 is not the same as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols and dependency paths." + }, + { + "left": "issue:43232", + "right": "issue:47276", + "accept": false, + "reason": "Sync-gpus generation bug and misleading warning text are related only broadly, not the same concrete bug." + }, + { + "left": "issue:43504", + "right": "issue:47059", + "accept": false, + "reason": "Legacy field loading in Beit is unrelated to the hub CLI completion attribute regression." + }, + { + "left": "issue:46614", + "right": "issue:46932", + "accept": false, + "reason": "Stale phonemizer backend state is unrelated to Gemma4 dtype casting in quantized embedders." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Type-checking regression and LlamaConfig head_dim validation are different problems." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs handling and Kosmos2.5 OCR index error are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM3 missing attribute and MoonshineStreaming off-by-one are different model/runtime bugs." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 save_pretrained moe bug is unrelated to the 50GB max_shard_size question." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Sentence Transformers loading stall and SigLIP zero probabilities are unrelated model-specific issues." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading position_ids after buffer registration is not the same as missing processor save files." + }, + { + "left": "issue:44360", + "right": "issue:46762", + "accept": false, + "reason": "DSA indexer lacking ReLU and Minimax MSA numerical differences are different implementation issues." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return values in GlmMoeDsaIndexer is unrelated to InternVL2 meta-init incompatibility." + }, + { + "left": "issue:44327", + "right": "issue:47059", + "accept": false, + "reason": "QA decode_spans top_k bounds error and hub CLI completion error are unrelated." + }, + { + "left": "issue:46369", + "right": "issue:46909", + "accept": false, + "reason": "SAM3-LiteText memory overflow is not the broken DIA documentation example." + }, + { + "left": "issue:43232", + "right": "issue:46424", + "accept": false, + "reason": "Generation sync_gpus handling and cache max_cache_length behavior are different bugs." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "GlmMoeDsaIndexer torch.split behavior and a docstring fix for position_ids do not match." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Local dataset loading script issue is unrelated to LlamaConfig head_dim validation." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both involve save/load around quantization, but they are different models and failure modes, so not a duplicate pair." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 53, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28218", + "issue:39401", + "issue:41628", + "issue:42796", + "issue:42831", + "issue:42886", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43408", + "issue:43540", + "issue:43576", + "issue:43644", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:43986", + "issue:44038", + "issue:44060", + "issue:44188", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44360", + "issue:44373", + "issue:44483", + "issue:44568", + "issue:44829", + "issue:44964", + "issue:45092", + "issue:45310", + "issue:45357", + "issue:45381", + "issue:45406", + "issue:45600", + "issue:45753", + "issue:46002", + "issue:46305", + "issue:46328", + "issue:46369", + "issue:46402", + "issue:46655", + "issue:46726", + "issue:46909", + "issue:47059", + "issue:47217" + ], + "soft_pairs": [ + "issue:45092|issue:45357", + "issue:44279|issue:46328", + "issue:43986|issue:46726", + "issue:22355|issue:27957", + "issue:43408|issue:46402", + "issue:46305|issue:46655", + "issue:45381|issue:46305", + "issue:42831|issue:45310", + "issue:41628|issue:43824", + "issue:44263|issue:44360", + "issue:44483|issue:45406", + "issue:22355|issue:28218", + "issue:44964|issue:47059", + "issue:42831|issue:44038", + "issue:44261|issue:45092", + "issue:44829|issue:46002", + "issue:44188|issue:44373", + "issue:43784|issue:45600", + "issue:42796|issue:43122", + "issue:39401|issue:44568", + "issue:43576|issue:43784", + "issue:43986|issue:47217", + "issue:43644|issue:46305", + "issue:42831|issue:45753", + "issue:43756|issue:44060", + "issue:43408|issue:46369", + "issue:42886|issue:43066", + "issue:43540|issue:46909", + "issue:45381|issue:46909", + "issue:42886|issue:42994" + ], + "prepared_review_unit_hash": "d5be645108d01bdab29aa085ce3cdc089e40b1b7eb6be1616bd7ddc663e86032", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36442, + "estimated_input_tokens": 9111, + "estimated_eval_tokens": 18478 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No clear duplicate cluster emerges here: the soft pairs mostly share only broad subsystem labels (tokenizers, multimodal loading, sam3, DSA, etc.) but describe different concrete failures. I would reject all soft edges and avoid picking a single canonical issue/PR for the set.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Both involve model-loading behavior, but one is remote-code/meta-init incompatibility and the other is incorrect visual encoder keys on save_pretrained; different bugs and code paths." + }, + { + "left": "issue:44279", + "right": "issue:46328", + "accept": false, + "reason": "Generic dependency complaint vs a SequenceFeatureExtractor.pad performance issue; no shared failure mode." + }, + { + "left": "issue:43986", + "right": "issue:46726", + "accept": false, + "reason": "Both mention missing vision backends in AutoProcessor, but one is a crash for video models without torchvision and the other is an unrecognized-image-processor error; not the same concrete bug." + }, + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "Importing transformers.onnx is unrelated to XLMRoberta Flash Attention 2 behavior." + }, + { + "left": "issue:43408", + "right": "issue:46402", + "accept": false, + "reason": "Same sam3 family, but one is a model-type warning and the other is missing text encoder weights; different problems." + }, + { + "left": "issue:46305", + "right": "issue:46655", + "accept": false, + "reason": "Both are off-by-one style issues in different video/streaming components and different models; insufficient evidence of one underlying bug." + }, + { + "left": "issue:45381", + "right": "issue:46305", + "accept": false, + "reason": "Different models and different position/padding logic; only a broad video-processing similarity." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy regression and Qwen3.5 MoE from_pretrained error are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different symbols/models with no shared root cause shown." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Both reference the DSA indexer, but one is about torch.split outputs and the other about a missing ReLU; not the same bug." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat/completions request handling and Gemma4Processor _tokenizer attribute crash are different code paths." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "ONNX import failure and tokenizer extra-space behavior are unrelated." + }, + { + "left": "issue:44964", + "right": "issue:47059", + "accept": false, + "reason": "Model-loading failure and huggingface_hub CLI completion regression are unrelated." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue does not match Qwen3-VL-Moe loading bug." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "A layernorm epsilon precision mismatch is not the same as remote-code checkpoint meta-initialization incompatibility." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Training degeneration with flash_attention_3 and NaNs after checkout/init are different failure modes." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "torch.compile kernel branching bug and docstring typo are unrelated." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import-time NameError in sentence-transformers differs from stale auto_mappings references and CI breakage." + }, + { + "left": "issue:42796", + "right": "issue:43122", + "accept": false, + "reason": "Both are tokenizer regressions, but one is a specific BPE-marker handling bug and the other is a broad version-to-version tokenization change; not enough to treat as the same issue." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Wrong offset_mapping and missing BOS/EOS insertion are different tokenizer behaviors." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Broken transformers env CLI and import-time NameError are unrelated." + }, + { + "left": "issue:43986", + "right": "issue:47217", + "accept": false, + "reason": "Both involve multimodal model input handling, but the former is missing torchvision during AutoProcessor loading and the latter is image-placeholder handling in user text; different bugs." + }, + { + "left": "issue:43644", + "right": "issue:46305", + "accept": false, + "reason": "Non-persistent buffer junk and MoonshineStreaming mask off-by-one are unrelated." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FP8 accuracy regression and Qwen3_5 infinite-loop-on-image bug do not share a concrete code path." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "RoPE-layer count bug and tied-weights warning bug are separate model-definition issues." + }, + { + "left": "issue:43408", + "right": "issue:46369", + "accept": false, + "reason": "Both are sam3-related, but one is a warning mismatch and the other is memory overflow; different failures." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline cache loading failure and wrong tokenizer decoder type are distinct tokenizer problems." + }, + { + "left": "issue:43540", + "right": "issue:46909", + "accept": false, + "reason": "Qwen3OmniMoe video processing ValueError and a broken DIA documentation example are unrelated." + }, + { + "left": "issue:45381", + "right": "issue:46909", + "accept": false, + "reason": "Qwen2.5-VL video position IDs bug and a documentation example issue are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Tokenizer cache/offline loading and quantized model saving are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 54, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:29159", + "issue:30990", + "issue:33357", + "issue:39379", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42907", + "issue:43116", + "issue:43122", + "issue:43329", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43646", + "issue:43824", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44117", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44488", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45478", + "issue:46002", + "issue:46172", + "issue:46326", + "issue:46762", + "issue:47059" + ], + "soft_pairs": [ + "issue:39379|issue:46172", + "issue:42757|issue:45278", + "issue:27957|issue:30990", + "issue:43646|issue:46002", + "issue:30990|issue:33357", + "issue:43122|issue:46326", + "issue:44295|issue:44297", + "issue:43329|issue:47059", + "issue:39401|issue:44488", + "issue:44625|issue:44829", + "issue:42907|issue:44964", + "issue:39401|issue:44521", + "issue:43976|issue:44945", + "issue:41628|issue:45278", + "issue:44117|issue:44704", + "issue:42907|issue:44945", + "issue:43122|issue:44521", + "issue:22355|issue:29159", + "issue:44704|issue:45290", + "issue:42371|issue:44263", + "issue:43408|issue:43475", + "issue:42907|issue:43937", + "issue:42907|issue:43452", + "issue:44038|issue:44297", + "issue:43824|issue:44351", + "issue:44038|issue:44295", + "issue:44360|issue:44485", + "issue:45216|issue:45478", + "issue:43116|issue:45200", + "issue:43122|issue:44779", + "issue:44038|issue:44222", + "issue:46172|issue:46762" + ], + "prepared_review_unit_hash": "0df8eae8f07b6b699b9b30d79f29b2767e42b4af05c56bd94a94b0619f2535b7", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36859, + "estimated_input_tokens": 9215, + "estimated_eval_tokens": 18686 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous, with a few loose theme overlaps around import and tokenizer regressions, but no clear duplicate pairs. The best representative issue is the broad import-regression report.", + "confidence": 0.64, + "canonical_issue_reason": "issue:45278 is the broadest and most central issue in the set, capturing the shared import-regression theme more clearly than the narrower model- or tokenizer-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best overall representative because it is recent, broad, and plausibly umbrellas several neighboring import-failure reports without overcommitting to a specific model or API path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39379", + "right": "issue:46172", + "accept": false, + "reason": "Different multimodal code paths: mllama mask handling vs DeepSeek-V4 tensor transpose logic." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-related, but one is a specific missing symbol and the other is a broad upgrade-related import breakage; not the same bug." + }, + { + "left": "issue:27957", + "right": "issue:30990", + "accept": false, + "reason": "XLM-RoBERTa Flash Attention 2 support and Sentence Transformers loading hang are unrelated failures." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage and NaNs in classification heads are different failure modes." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "A loading hang and a MacOS bus error are different symptoms with no clear shared code-path." + }, + { + "left": "issue:43122", + "right": "issue:46326", + "accept": false, + "reason": "Both concern tokenization, but one is a version-to-version tokenization change and the other is a continue_final_message bug for mistral_common tokenizers." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Reading a registered buffer and tokenizer_class mismatch in saved config are separate save/load issues." + }, + { + "left": "issue:43329", + "right": "issue:47059", + "accept": false, + "reason": "Undefined variables in multimodal token counting and a CLI completion attribute error are unrelated." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Qwen3 offset mapping is a tokenizer alignment bug; loading cjvt/sleng-bert is a model-loading issue." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "num_labels propagation and flash_attention_3 training degeneration are unrelated model/config bugs." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Saving dequantized Ministral/Devstral models and loading Phi-4 multimodal are different model-specific failures." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Wrong offset mappings and zero assistant masks are different tokenizer/template defects." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility and pipeline-parallel output corruption are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Both are import problems, but one is a missing AutoImageProcessor symbol and the other is a broad post-upgrade import regression; not the same concrete bug." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES handling and AutoProcessor kwargs forwarding are distinct code paths." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Quantized model save failure and incorrect pipeline-parallel output are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "General tokenizer output differences and multimodal assistant-mask generation are not the same bug." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "onnx import failure and slow/fast tokenizer inconsistency are unrelated." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "cached_file argument forwarding and chat-template crashes with tool calls are different API paths." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 API guidance and a torch.split return-value bug in a DSA indexer are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Both are SAM3 video-related, but one is a model-type warning and the other is a missing attribute on an encoder output; not clearly the same bug." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Model save/load failure and GLM-5 GenerationConfig validation are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Dequantized model saving and gguf_file loading failures affect different code paths." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe loading and tokenizer_class mismatch are different problems." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Missing Qwen2_5_VL export and missing HybridCache export are separate import regressions." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe model loading and position_ids buffer access are unrelated." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "A missing ReLU in the DSA indexer and GLM-5 RoPE implementation are different algorithmic issues." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 and 5.4.0+, but one is save_pretrained checkpoint corruption and the other is a from_pretrained error; too little evidence they are the same bug." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Example-script multi-label classification output and Gemma 4 mm_token_type_ids defaults are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they concern different models and different failure descriptions; not enough to treat as one duplicate." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Qwen3-VL-Moe loading and FP8 MoE save_pretrained are different features and failure modes." + }, + { + "left": "issue:46172", + "right": "issue:46762", + "accept": false, + "reason": "DeepSeek-V4 tensor-transpose logic and Minimax attention numerical mismatch are different numerical bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 55, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:22355", + "issue:28348", + "issue:29127", + "issue:36296", + "issue:38617", + "issue:39379", + "issue:39401", + "issue:42175", + "issue:42222", + "issue:42673", + "issue:42757", + "issue:42898", + "issue:43066", + "issue:43116", + "issue:43232", + "issue:43278", + "issue:43493", + "issue:43611", + "issue:43976", + "issue:44075", + "issue:44188", + "issue:44351", + "issue:44483", + "issue:44610", + "issue:44625", + "issue:44779", + "issue:44936", + "issue:44938", + "issue:44945", + "issue:44995", + "issue:45072", + "issue:45083", + "issue:45278", + "issue:45357", + "issue:45446", + "issue:45478", + "issue:45753", + "issue:46002", + "issue:46231", + "issue:46328", + "issue:46614", + "issue:46640", + "issue:46697", + "issue:46877", + "issue:46909", + "issue:47217", + "issue:47312" + ], + "soft_pairs": [ + "issue:22355|issue:29127", + "issue:42175|issue:44938", + "issue:43278|issue:44945", + "issue:39401|issue:46614", + "issue:39379|issue:43232", + "issue:42222|issue:47312", + "issue:42757|issue:44351", + "issue:36296|issue:43116", + "issue:43493|issue:45753", + "issue:44995|issue:46697", + "issue:44188|issue:46328", + "issue:28348|issue:46640", + "issue:46231|issue:46877", + "issue:43976|issue:45446", + "issue:44483|issue:47217", + "issue:42673|issue:45072", + "issue:42898|issue:46614", + "issue:45357|issue:45478", + "issue:43611|issue:46002", + "issue:44625|issue:46002", + "issue:38617|issue:45278", + "issue:44075|issue:44936", + "issue:43066|issue:44779", + "issue:45083|issue:45446", + "issue:44610|issue:46909" + ], + "prepared_review_unit_hash": "f47be8e0d08d5e6c12d17001569c2077fea3066366996660801c474fcaaf48a8", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 25, + "serialized_chars": 34081, + "estimated_input_tokens": 8521, + "estimated_eval_tokens": 17298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is largely heterogeneous: most soft-similarity pairs are not true duplicates and only share broad subsystem or regression themes. The closest-to-related items are the import/version-regression reports, but even those point to different missing symbols or different failure modes.", + "confidence": 0.2, + "canonical_issue_reason": "issue:45278 is the broadest representative of the import/regression subset in this set, but there is no strong single canonical duplicate because the cluster spans many unrelated bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most general and representative issue here; it best matches the recurring pattern of post-upgrade import breakages, while the rest are narrower or unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Different problems: missing transformers.onnx module vs LayoutLMv3 box-content validation/error messaging." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Both mention environment/loading issues, but one is about missing TensorFlow in a torch install and the other is Python 3.14 loading failure; not the same bug." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "BF16/FP32 dtype drift in evaluation is unrelated to incorrect output under pipeline parallelism." + }, + { + "left": "issue:39401", + "right": "issue:46614", + "accept": false, + "reason": "Qwen3 tokenizer offset mapping and stale phonemizer language backend are different code paths and failure modes." + }, + { + "left": "issue:39379", + "right": "issue:43232", + "accept": false, + "reason": "mllama mask handling vs generation kwargs update after sync_gpus are unrelated defects." + }, + { + "left": "issue:42222", + "right": "issue:47312", + "accept": false, + "reason": "VitPose model breakage and kernels 0.16 incompatibility are not the same underlying issue." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different packages; too broad to treat as one bug." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor parallel training bug is unrelated to an example script returning empty multi-label results." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "SigLIP2 JAX parity discrepancy and Qwen3.5 infinite-loop-on-image are different model-specific bugs." + }, + { + "left": "issue:44995", + "right": "issue:46697", + "accept": false, + "reason": "Stale indexer cache crash and image over-padding are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:46328", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to a padding performance issue." + }, + { + "left": "issue:28348", + "right": "issue:46640", + "accept": false, + "reason": "FlashAttention 2 support request for GPT2LMHeadModel is not the same as a separate Flash Attention 2 issue." + }, + { + "left": "issue:46231", + "right": "issue:46877", + "accept": false, + "reason": "Documentation parameter-name mismatch and import-time doc-lint stdout spam are different issues." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Both involve version checks, but one is a Python-version compatibility break and the other is a PyTorch version guard for AuxRequest import." + }, + { + "left": "issue:44483", + "right": "issue:47217", + "accept": false, + "reason": "Chat/completions request rejection and image-placeholder handling in VL models are different API/model behaviors." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in dataloader threads and bfloat16 dtype mismatches are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:46614", + "accept": false, + "reason": "Tokenizer cleanup behavior change and stale phonemizer backend state are different tokenizer-related bugs." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both concern Qwen3.5, but one is save_pretrained key serialization and the other is from_pretrained loading failure; not the same concrete defect." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "base_model_prefix loading breakage and NaNs in classification heads are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and NaNs after checkout/init are different bugs." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "The latter is a broad umbrella for many import errors, but this specific missing-symbol ImportError is not the same concrete issue." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Unused SGD args and trainer.evaluate() failing after trainer.train() are unrelated." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Tokenizer decoder-type regression and Deepseek tokenizer incorrect results are too broad and not clearly the same bug." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Qwen3_omni_moe feature-length helper behavior and flex_attention version guard are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:46909", + "accept": false, + "reason": "OmDet-Turbo processor sizing bug and a broken documentation example are not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 56, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:33453", + "issue:39379", + "issue:39401", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:42796", + "issue:43064", + "issue:43066", + "issue:43116", + "issue:43493", + "issue:43618", + "issue:43673", + "issue:43688", + "issue:43824", + "issue:43825", + "issue:43827", + "issue:43874", + "issue:43937", + "issue:43992", + "issue:44038", + "issue:44117", + "issue:44291", + "issue:44355", + "issue:44380", + "issue:44466", + "issue:44485", + "issue:44704", + "issue:45071", + "issue:45081", + "issue:45092", + "issue:45405", + "issue:45446", + "issue:45478", + "issue:45563", + "issue:45588", + "issue:46002", + "issue:46077", + "issue:46231", + "issue:46328", + "issue:46710", + "issue:46726", + "issue:46762", + "issue:46829", + "issue:46897", + "issue:46981", + "issue:47000", + "issue:47116" + ], + "soft_pairs": [ + "issue:43937|issue:44485", + "issue:43116|issue:43493", + "issue:44038|issue:45478", + "issue:43066|issue:46710", + "issue:44355|issue:46231", + "issue:42673|issue:42796", + "issue:46726|issue:46981", + "issue:42371|issue:43874", + "issue:43116|issue:43825", + "issue:43618|issue:46829", + "issue:43827|issue:45446", + "issue:43992|issue:44704", + "issue:45405|issue:45446", + "issue:46328|issue:47116", + "issue:33453|issue:39401", + "issue:43824|issue:46981", + "issue:39379|issue:43688", + "issue:44291|issue:45071", + "issue:44117|issue:45588", + "issue:43673|issue:46328", + "issue:43064|issue:43673", + "issue:45092|issue:46762", + "issue:44117|issue:45081", + "issue:43937|issue:45563", + "issue:44466|issue:46002", + "issue:44380|issue:46077", + "issue:42222|issue:47000", + "issue:39379|issue:46897" + ], + "prepared_review_unit_hash": "47b05f058c35f42463c721ff612f65161b37e2c6d99dd69dd643afd422a5149f", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36133, + "estimated_input_tokens": 9034, + "estimated_eval_tokens": 18324 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose grab bag of mostly Transformers v5 regressions and model-specific bugs (tokenizers, loading, generation, attention, docs, and a few performance issues). None of the soft pairs look like the same underlying bug or a mergeable PR-level duplicate.", + "confidence": 0.89, + "canonical_issue_reason": "issue:43066 is the broadest core regression here: a v5 tokenizer/decoder loading problem in the main library path, which best matches the cluster\u2019s recurring tokenizer/loading breakages.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43066 is the clearest representative issue for the cluster\u2019s main theme (v5-era tokenizer/loading regressions) without being overly model-specific.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Both mention GLM-5, but one is a GenerationConfig validity error and the other is a RoPE implementation discussion; different code paths." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Multi-label classification script output and SigLIP2 numerical discrepancy are unrelated bugs in different areas." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3-family v5 issues, but one is Qwen3-VL-Moe loading and the other is Qwen3.5-Moe from_pretrained; not the same defect." + }, + { + "left": "issue:43066", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer decoder type regression and an accuracy regression after upgrade are distinct symptoms with no clear shared code-path." + }, + { + "left": "issue:44355", + "right": "issue:46231", + "accept": false, + "reason": "Compiled Python file errors and a docs parameter-name mismatch are unrelated." + }, + { + "left": "issue:42673", + "right": "issue:42796", + "accept": false, + "reason": "VRAM leakage in multithreaded Qwen3ForCausalLM and tokenizer BPE marker handling are different bugs." + }, + { + "left": "issue:46726", + "right": "issue:46981", + "accept": false, + "reason": "AutoProcessor missing-vision-backend messaging and repeated HTTP 429 retries are separate processor/network behaviors." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 settings guidance and a GLM46V image-patch API crash are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Classification example emptiness and a pipeline translation error message are different example/UI issues." + }, + { + "left": "issue:43618", + "right": "issue:46829", + "accept": false, + "reason": "Both are output-field assignment bugs, but they affect different model classes and different outputs; not the same concrete fix." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() and a flex_attention import version check are unrelated." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Missing embed_tokens.weight in UMT5Encoder and cached_file kwargs passthrough are separate loading bugs." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "PEFT minimum-version bump and PyTorch version-check logic are unrelated maintenance issues." + }, + { + "left": "issue:46328", + "right": "issue:47116", + "accept": false, + "reason": "SequenceFeatureExtractor.pad performance waste and streaming pipeline regression do not share a concrete failure mode." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Both are tokenizer-related, but one is loading regression and the other is wrong offset_mapping for Qwen3; different defects." + }, + { + "left": "issue:43824", + "right": "issue:46981", + "accept": false, + "reason": "ImportError for Qwen2.5-VL and repeated 429 retries during config fetch are unrelated." + }, + { + "left": "issue:39379", + "right": "issue:43688", + "accept": false, + "reason": "Incorrect mllama mask handling and auxiliary-loss normalization in OLMoE/GPT Oss are different model-training bugs." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Unexpected _is_hf_initialized arg during init_empty_weights and PretrainedConfig type-checking breakage are separate v5 compatibility issues." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer mapping None handling and flash_attention sink-less-model crash involve different code paths and failure causes." + }, + { + "left": "issue:43673", + "right": "issue:46328", + "accept": false, + "reason": "Generation cache missing during chunked_prefill and a pad() performance issue are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Wrong optimizer states on nonzero ranks in Trainer.train and missing generation cache are distinct distributed-training vs generation bugs." + }, + { + "left": "issue:45092", + "right": "issue:46762", + "accept": false, + "reason": "Remote-code meta-initialization incompatibility and Minimax attention numerical differences are not the same bug." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer mapping assumptions and Mistral regex patch crashes are separate tokenizer-loading problems." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "Invalid GenerationConfig and a stale num_return_sequences warning are different generation-config issues." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "Serialization inconsistency for lm_head.weight and NaNs in classification heads are unrelated." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT2 attention scaling ignored in SDPA/FlashAttention and Gemma4 causal-mask construction are different attention bugs." + }, + { + "left": "issue:42222", + "right": "issue:47000", + "accept": false, + "reason": "VitPose loading breakage and RTDetrLoss infeasible cost matrix are unrelated model failures." + }, + { + "left": "issue:39379", + "right": "issue:46897", + "accept": false, + "reason": "mllama mask handling and Florence2 double-shift training loss are different bugs, only superficially similar as training regressions." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 57, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42617", + "issue:42831", + "issue:43122", + "issue:43208", + "issue:43408", + "issue:43421", + "issue:43441", + "issue:43450", + "issue:43742", + "issue:43937", + "issue:43986", + "issue:44075", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44410", + "issue:44483", + "issue:44496", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44877", + "issue:44964", + "issue:44977", + "issue:45071", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45464", + "issue:45561", + "issue:45632", + "issue:45656", + "issue:46231", + "issue:46298", + "issue:46328", + "issue:46512", + "issue:46752", + "issue:46762", + "issue:46894", + "issue:46981", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:42617|issue:45341", + "issue:43937|issue:44977", + "issue:43408|issue:43742", + "issue:43450|issue:43986", + "issue:42222|issue:43441", + "issue:44263|issue:46762", + "issue:42222|issue:46752", + "issue:45071|issue:46512", + "issue:43208|issue:46894", + "issue:39401|issue:44779", + "issue:44829|issue:45216", + "issue:44877|issue:46231", + "issue:45071|issue:47059", + "issue:44410|issue:45310", + "issue:45632|issue:46981", + "issue:45341|issue:45561", + "issue:45341|issue:46328", + "issue:44704|issue:47110", + "issue:39401|issue:43421", + "issue:38617|issue:41628", + "issue:43122|issue:46298", + "issue:43937|issue:44496", + "issue:38617|issue:44220", + "issue:44483|issue:45464", + "issue:44964|issue:45278", + "issue:42371|issue:44261", + "issue:44075|issue:45656", + "issue:44261|issue:46762", + "issue:42371|issue:42831" + ], + "prepared_review_unit_hash": "d47b1e91de0fb01f5d78cd1abcad573ec92ddf1697e01217012ed5e9ec936c44", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36052, + "estimated_input_tokens": 9013, + "estimated_eval_tokens": 18282 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: most items are distinct model-, tokenizer-, processor-, training-, or testing-specific bugs, and none of the soft pairs look like true duplicates.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45278 is the broadest umbrella regression report in the set (many import errors after upgrading), and it best serves as a representative anchor for this otherwise mixed cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most general, user-facing issue here and the closest thing to a cluster-level representative; the rest are mostly narrower, unrelated bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "Script startup failure vs a small testing_utils bug; different code paths and symptoms." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "Both mention generation, but they are different models and different failures (invalid config vs flash-attention generation bug)." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "Model-type warning/mismatch vs key error while loading another model; not the same underlying bug." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Video processor batch-shape bug vs AutoProcessor failure without torchvision; separate processor/loading issues." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "Vitpose loading break and Ministral FlashAttention crash are unrelated model-specific problems." + }, + { + "left": "issue:44263", + "right": "issue:46762", + "accept": false, + "reason": "Both involve tensor math, but one is a torch.split indexing issue and the other is attention numerical mismatch." + }, + { + "left": "issue:42222", + "right": "issue:46752", + "accept": false, + "reason": "Model loading failure vs empty chat-template IndexError; no shared concrete code path." + }, + { + "left": "issue:45071", + "right": "issue:46512", + "accept": false, + "reason": "Type-checking regression vs stale dtype metadata after from_config; distinct config/model state bugs." + }, + { + "left": "issue:43208", + "right": "issue:46894", + "accept": false, + "reason": "xLSTM training blockers vs CohereASR loss double-shift bug; same broad domain, different defect." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Tokenizer offset_mapping bug vs Deepseek tokenization regression; related area, but not the same issue." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "FlashAttention training degeneration and save_pretrained checkpoint regression are different failures." + }, + { + "left": "issue:44877", + "right": "issue:46231", + "accept": false, + "reason": "Strict config loading vs documentation typo in set_training; unrelated." + }, + { + "left": "issue:45071", + "right": "issue:47059", + "accept": false, + "reason": "PretrainedConfig typing regression vs CLI completion attribute error; no common bug." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing qwen3next projections vs Qwen3.5 Moe from_pretrained error; different model internals and failure modes." + }, + { + "left": "issue:45632", + "right": "issue:46981", + "accept": false, + "reason": "Local cache-path collision vs repeated 429 retry behavior; separate AutoProcessor problems." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "Both touch testing utilities, but one is a small bug and the other is a pytest-xdist race condition." + }, + { + "left": "issue:45341", + "right": "issue:46328", + "accept": false, + "reason": "Testing_utils bug vs SequenceFeatureExtractor.pad performance issue; not the same defect." + }, + { + "left": "issue:44704", + "right": "issue:47110", + "accept": false, + "reason": "Kwargs not forwarded in AutoProcessor.from_pretrained vs special_tokens list/dict error; different API paths." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Tokenizer offset mapping bug vs runtime post-processor update issue; different functionality." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and likely separate regressions." + }, + { + "left": "issue:43122", + "right": "issue:46298", + "accept": false, + "reason": "Tokenizer behavior regression vs noisy logging message; unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GenerationConfig validation for GLM vs unrecognized model/config loading failure; different root causes." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "ImportError vs audio feature extraction bug; no common code path." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Both hit chat/completions, but one is request acceptance and the other is streaming inference failure; distinct bugs." + }, + { + "left": "issue:44964", + "right": "issue:45278", + "accept": false, + "reason": "Both are load/import regressions, but one is a specific multimodal model load failure and the other is a broad upgrade-related import issue." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API warning and MLA q_a_layernorm epsilon precision bug are unrelated." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args ignored vs optimizer.step being called twice under deepspeed; different optimizer bugs." + }, + { + "left": "issue:44261", + "right": "issue:46762", + "accept": false, + "reason": "Different numerical issues in different components: missing rms_norm_eps vs MSA attention mismatch." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 settings warning vs FineGrainedFP8 accuracy regression; not the same problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 58, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:27957", + "issue:41628", + "issue:43012", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43208", + "issue:43352", + "issue:43441", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43582", + "issue:43673", + "issue:43874", + "issue:43931", + "issue:43957", + "issue:43986", + "issue:44038", + "issue:44077", + "issue:44117", + "issue:44230", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44351", + "issue:44360", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44829", + "issue:44987", + "issue:45071", + "issue:45310", + "issue:45341", + "issue:45372", + "issue:45440", + "issue:45478", + "issue:45588", + "issue:45753", + "issue:46050", + "issue:46172", + "issue:46369", + "issue:46752", + "issue:46897", + "issue:46981", + "issue:47116" + ], + "soft_pairs": [ + "issue:43097|issue:44493", + "issue:44484|issue:45310", + "issue:43066|issue:43122", + "issue:44077|issue:46981", + "issue:43931|issue:44410", + "issue:44410|issue:45478", + "issue:44263|issue:44485", + "issue:43673|issue:45341", + "issue:41628|issue:45071", + "issue:43208|issue:46897", + "issue:43097|issue:43452", + "issue:43931|issue:44038", + "issue:43441|issue:44829", + "issue:43012|issue:43577", + "issue:43874|issue:44261", + "issue:46752|issue:47116", + "issue:27957|issue:43352", + "issue:44484|issue:45478", + "issue:44117|issue:44987", + "issue:44263|issue:46172", + "issue:44360|issue:46172", + "issue:43957|issue:44351", + "issue:43531|issue:45341", + "issue:43097|issue:45588", + "issue:43582|issue:45341", + "issue:44261|issue:44485", + "issue:44038|issue:44230", + "issue:43986|issue:45341", + "issue:43097|issue:46050", + "issue:43986|issue:45372", + "issue:45753|issue:46369", + "issue:44246|issue:46981", + "issue:45440|issue:46172" + ], + "prepared_review_unit_hash": "208c57134b883f97679d9efc04172dc99a0e320b19e13605bdf0ed7b929f07e1", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37244, + "estimated_input_tokens": 9311, + "estimated_eval_tokens": 18878 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly false-positive similarity around unrelated Transformers issues. The only clear duplicate-looking item is the Qwen3.5 MoE from_pretrained regression (45310/45478); I\u2019d use 45310 as canonical because it is the earlier, clearer representative. No PRs are present.", + "confidence": 0.63, + "canonical_issue_reason": "Issue 45310 is the best canonical issue because it is the earlier of an exact duplicate-title pair with 45478 and clearly describes a concrete, reproducible Qwen3.5 MoE from_pretrained regression.", + "canonical_pr_reason": null, + "best_issue_reason": "45310 is the strongest overall issue representative in this set: it is specific, bug-oriented, and has an exact duplicate elsewhere in the items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Both are v5 regressions, but one is about a removed config flag and the other about unexpected position-id keys; different bugs." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "Max shard size question vs. Qwen3.5 MoE from_pretrained error; unrelated symptoms and code paths." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Both concern tokenizer behavior changes, but one is a decoder-type issue and the other is a broader tokenization regression." + }, + { + "left": "issue:44077", + "right": "issue:46981", + "accept": false, + "reason": "patchtsmixer post_init validation and HTTP 429 retry behavior are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Different model families and failure modes: VL weight-shape mismatch vs. missing Qwen3next projections." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Different model families; missing projections in qwen3next is not the same as Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "DSA indexer split semantics and GLM-5 RoPE implementation are separate model internals issues." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Generation cache/chunked_prefill regression vs. a testing_utils bug; no shared concrete code path." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "ImportError for AutoImageProcessor is unrelated to PretrainedConfig type checking." + }, + { + "left": "issue:43208", + "right": "issue:46897", + "accept": false, + "reason": "xLSTM training bugs and Florence2 double-shift loss bug are distinct model-specific failures." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Removed tie_embeddings behavior and gguf_file/from_pretrained loading breakage are different regressions." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Same broad Qwen3-VL area, but the concrete failures differ: one is a load shape mismatch, the other a Qwen3-VL-Moe bug." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "FlashAttention 2 support crash and FlashAttention 3 degenerate training are different problems." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "One is a PyTorch warning during compile, the other is incorrect dtype propagation when loading BLIP2." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Missing image-processor method vs. missing rms_norm_eps in MLA q_a_layernorm; unrelated." + }, + { + "left": "issue:46752", + "right": "issue:47116", + "accept": false, + "reason": "Empty-chat-template IndexError is not the same as a streaming pipeline regression." + }, + { + "left": "issue:27957", + "right": "issue:43352", + "accept": false, + "reason": "Both mention Flash Attention 2, but on different models and with different support/status issues." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "A save_pretrained shard-size question is unrelated to the Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer_MAPPING_NAMES returning None is a different failure from a specific model-loading regression." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "DSA torch.split behavior and deepseek-v4 comb transpose are unrelated code paths." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "Missing ReLU in the DSA indexer is unrelated to the deepseek-v4 dtype/transpose bug." + }, + { + "left": "issue:43957", + "right": "issue:44351", + "accept": false, + "reason": "Meta-device model loading issues and missing HybridCache import are not the same bug." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Qwen3-MoE sliding_window handling and a testing_utils bug are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Removed tie_embeddings behavior is unrelated to flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:43582", + "right": "issue:45341", + "accept": false, + "reason": "Apple Silicon caching_allocator_warmup TypeError is unrelated to testing_utils." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "LayerNorm epsilon precision issue vs. RoPE implementation bug; different model components." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "A Qwen3-VL-Moe runtime bug and an fp8 support request are not the same underlying issue." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "AutoProcessor without torchvision crash and ReasoningEffort import error are separate processor-loading problems." + }, + { + "left": "issue:44246", + "right": "issue:46981", + "accept": false, + "reason": "Slow import and repeated 429 retries are both performance-related, but they are different mechanisms and code paths." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "DeepseekV3MoE divergence from remote implementation is not the same as the deepseek-v4 comb transpose bug." + }, + { + "left": "issue:45753", + "right": "issue:46369", + "accept": false, + "reason": "Qwen3_5 infinite loop on an image and SAM3 LiteText memory overflow are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "Removed tie_embeddings behavior and quantized save/load corruption are different regression classes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 59, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:38617", + "issue:42757", + "issue:42890", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43502", + "issue:43550", + "issue:43673", + "issue:43873", + "issue:43874", + "issue:43906", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44062", + "issue:44117", + "issue:44263", + "issue:44315", + "issue:44351", + "issue:44380", + "issue:44410", + "issue:44451", + "issue:44484", + "issue:44655", + "issue:44704", + "issue:44936", + "issue:44938", + "issue:45103", + "issue:45375", + "issue:45712", + "issue:45753", + "issue:46123", + "issue:46172", + "issue:46178", + "issue:46298", + "issue:46365", + "issue:46421", + "issue:46655", + "issue:46664", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821", + "issue:47116" + ], + "soft_pairs": [ + "issue:43097|issue:43873", + "issue:44410|issue:45375", + "issue:44062|issue:44117", + "issue:44117|issue:46701", + "issue:43673|issue:46123", + "issue:43012|issue:46821", + "issue:43874|issue:44263", + "issue:44038|issue:44410", + "issue:38617|issue:42757", + "issue:46655|issue:46664", + "issue:43208|issue:43550", + "issue:44936|issue:46365", + "issue:46298|issue:46752", + "issue:43064|issue:44315", + "issue:46172|issue:46710", + "issue:43986|issue:44484", + "issue:42890|issue:46298", + "issue:44704|issue:45103", + "issue:38617|issue:44351", + "issue:43502|issue:43992", + "issue:44451|issue:47116", + "issue:44380|issue:46421", + "issue:44938|issue:46178", + "issue:43064|issue:47116", + "issue:43012|issue:43550", + "issue:42907|issue:45753", + "issue:45712|issue:46298", + "issue:43066|issue:43906", + "issue:42907|issue:44655" + ], + "prepared_review_unit_hash": "e42ef0affa779d49b0bfa5171ddd12a3949b4ab12ac121a0e2c096335f1defea", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36089, + "estimated_input_tokens": 9023, + "estimated_eval_tokens": 18302 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated transformer bugs; the soft links are mostly superficial similarity around imports, model families, or training/inference features. I would not merge any of the provided soft pairs.", + "confidence": 0.86, + "canonical_issue_reason": "issue:43873 is the strongest standalone issue in the set: it has high discussion/inbound activity, is still open, and describes a concrete, impactful bug with clear reproduction around quantization/offloading.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best representative issue because it appears well-developed, broadly relevant, and actionable compared with the rest of the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Different problems: one is a v5 API removal/compatibility complaint, the other is a quantization/offloading failure. No shared code path or fix." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen-related, but one is missing projections in a layer and the other is a config field being dropped. Different bugs and likely different files." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Unrelated tokenizer constructor error vs tokenizer-mapping lookup behavior. Same area of code, but not the same concrete defect." + }, + { + "left": "issue:44117", + "right": "issue:46701", + "accept": false, + "reason": "Tokenizer mapping fallback bug and a DiffusionGemma pipeline failure are separate issues with no clear common root cause." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Both mention generation/pipeline behavior, but one is cache availability in chunked prefill and the other is final partial-batch handling in mask generation. Different code paths." + }, + { + "left": "issue:43012", + "right": "issue:46821", + "accept": false, + "reason": "A PyTorch precision warning and a TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Both are GLM-family issues, but one is a missing image-processor method and the other is a torch.split return-value problem in a different component." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Different Qwen variants and different failure modes: one is a Transformers 5.0/Qwen3-VL-Moe bug, the other is missing layer projections in qwen3next." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different symbols, modules, and dependencies. Not the same underlying bug." + }, + { + "left": "issue:46655", + "right": "issue:46664", + "accept": false, + "reason": "Same model family, but one is padding math and the other is num_frames semantics in the video processor. Related area, different defects." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs and a Bamba torch.compile/SDPA failure are separate model-specific issues." + }, + { + "left": "issue:44936", + "right": "issue:46365", + "accept": false, + "reason": "Trainer.evaluate after train and inability to import Trainer due to datasets are different failures with different causes." + }, + { + "left": "issue:46298", + "right": "issue:46752", + "accept": false, + "reason": "A noisy logging message and an empty-list apply_chat_template crash are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Different training-stack problems: FSDP2/PEFT/optimizer-state corruption versus Liger Kernel not being applied with model_init." + }, + { + "left": "issue:46172", + "right": "issue:46710", + "accept": false, + "reason": "A deepseek-v4 tensor operation issue and a broad accuracy regression after upgrade are not the same concrete bug." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "AutoProcessor/torchvision crash and save_pretrained shard size behavior are unrelated." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "Missing set_seed in tests and a noisy logging message have nothing in common beyond being maintenance issues." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "AutoProcessor kwargs forwarding and auto_docstring annotation handling are different subsystems and bugs." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import-related, but they expose different missing exports and likely different release-compatibility problems. Not mergeable as one fix." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "local_files_only still making API requests and a missing embed_tokens weight in from_pretrained are unrelated loading issues." + }, + { + "left": "issue:44451", + "right": "issue:47116", + "accept": false, + "reason": "A model load regression for ScandiBERT and a streaming pipeline regression are distinct failures." + }, + { + "left": "issue:44380", + "right": "issue:46421", + "accept": false, + "reason": "Attention scaling ignored in GPT2 backends and chunked-prefill cache lazy init causing recompilation are separate runtime issues." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 import failure and a backend mapping error for tensorflow_text are unrelated compatibility problems." + }, + { + "left": "issue:43064", + "right": "issue:47116", + "accept": false, + "reason": "Wrong optimizer state on distributed ranks and a streaming pipeline regression do not share the same code path or fix." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "Precision warning versus compile-time SDPA failure; different symptoms and likely different fixes." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Saving dequantized Ministrals/Devstrals and Qwen3_5 infinite loop on one image are distinct model-specific bugs." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Left is a repo hygiene/dummy-object cleanup issue; right is logging noise. No overlap in underlying bug." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Wrong tokenizer decoder type and a reproduction of a different issue are not the same defect." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Dequantized model saving and pipeline save_pretrained behavior are both save-related, but they concern different object types and failure modes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 60, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:38617", + "issue:42175", + "issue:42757", + "issue:42831", + "issue:42890", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43502", + "issue:43526", + "issue:43582", + "issue:43618", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43825", + "issue:43827", + "issue:43976", + "issue:43992", + "issue:44077", + "issue:44162", + "issue:44222", + "issue:44261", + "issue:44368", + "issue:44380", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44625", + "issue:44655", + "issue:44734", + "issue:45276", + "issue:45478", + "issue:45479", + "issue:45584", + "issue:46077", + "issue:46132", + "issue:46172", + "issue:46298", + "issue:46365", + "issue:47059", + "issue:47116", + "issue:47276" + ], + "soft_pairs": [ + "issue:43208|issue:43526", + "issue:45584|issue:46365", + "issue:43825|issue:44655", + "issue:43012|issue:44368", + "issue:43992|issue:44484", + "issue:43352|issue:43976", + "issue:44734|issue:47116", + "issue:43827|issue:47116", + "issue:44077|issue:46365", + "issue:44162|issue:46172", + "issue:44380|issue:46132", + "issue:36683|issue:43582", + "issue:43352|issue:44380", + "issue:42757|issue:43824", + "issue:44380|issue:45479", + "issue:43097|issue:46298", + "issue:42175|issue:46365", + "issue:43012|issue:47276", + "issue:43756|issue:44485", + "issue:36010|issue:42757", + "issue:44222|issue:45478", + "issue:43064|issue:44380", + "issue:42890|issue:47059", + "issue:42831|issue:44261", + "issue:44483|issue:44514", + "issue:43618|issue:43761", + "issue:38617|issue:43502", + "issue:44625|issue:46365", + "issue:45276|issue:46077" + ], + "prepared_review_unit_hash": "836fae356273cdc1733c4f2735d264ab7a89e30ce9f1a64a3de42ed73f427531", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36354, + "estimated_input_tokens": 9089, + "estimated_eval_tokens": 18434 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues (imports, model/config bugs, backend regressions, warnings, docs, and packaging). None of the soft pairs look like the same underlying bug or change, so all are rejected.", + "confidence": 0.94, + "canonical_issue_reason": "No single issue is a true duplicate anchor for the whole set; issue:43097 is the broadest v5-breaking-change report and the best representative issue if one must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue:43097 is the most generally framed, high-impact regression in the set; the others are narrower model-, backend-, or packaging-specific bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training bugs vs BeitImageProcessorFast label reduction; different models and failure modes." + }, + { + "left": "issue:45584", + "right": "issue:46365", + "accept": false, + "reason": "Whisper empty-transcription generation bug vs Trainer import conflict with datasets; unrelated subsystems." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "pipeline() error-message regression vs Pipeline save_pretrained serialization; different problems." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "bfloat16 compile warning vs tie_word_embeddings warning during Qwen fine-tuning; unrelated warnings." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing embed_tokens.weight on load vs max_shard_size default on save; load-path vs save-path issue." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Flash Attention support for Nemotron vs Python version compatibility; not the same bug." + }, + { + "left": "issue:44734", + "right": "issue:47116", + "accept": false, + "reason": "KV-cache continuation indexing in /v1/responses vs streaming pipeline regression; different serving paths." + }, + { + "left": "issue:43827", + "right": "issue:47116", + "accept": false, + "reason": "Docs still referencing pipeline() vs runtime streaming regression; docs issue is separate from code bug." + }, + { + "left": "issue:44077", + "right": "issue:46365", + "accept": false, + "reason": "patchtsmixer post_init validation vs Trainer import blocked by datasets; unrelated." + }, + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "ESM2 broken vs deepseek-v4 comb transpose bug; different models and code paths." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "GPT2 attention scaling ignored in SDPA/FlashAttention vs AttentionInterface.register behavior change; distinct attention issues." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size vs Apple Silicon caching_allocator_warmup TypeError; config bug vs platform/runtime bug." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Nemotron FlashAttention 2 unsupported vs GPT2 attention scaling ignored; both backend-related but not the same failure." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Different missing imports from different packages/symbols; same symptom class, but not the same underlying issue." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention scaling backend bug vs zero-loss classification bug; unrelated." + }, + { + "left": "issue:43097", + "right": "issue:46298", + "accept": false, + "reason": "Removed API without indication vs noisy logging message; unrelated severity and root cause." + }, + { + "left": "issue:42175", + "right": "issue:46365", + "accept": false, + "reason": "Missing TensorFlow backend in torch install vs Trainer import conflict with datasets; different packaging problems." + }, + { + "left": "issue:43012", + "right": "issue:47276", + "accept": false, + "reason": "bfloat16 compile warning vs misleading synced_gpus warning in batching; unrelated warnings." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE layer drop vs GLM-5 RoPE implementation; both RoPE-related but clearly different bugs." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Different import errors for different symbols/modules; not the same missing-export regression." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained MoE bug vs Qwen3.5 MoE from_pretrained error; different stages and symptoms." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Wrong optimizer states with FSDP2+PEFT vs GPT2 attention scaling bug; unrelated code paths." + }, + { + "left": "issue:42890", + "right": "issue:47059", + "accept": false, + "reason": "Flaky tests due to missing set_seed vs CLI failure from huggingface_hub API change; different causes." + }, + { + "left": "issue:42831", + "right": "issue:44261", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue vs missing rms_norm_eps precision drift; different precision bugs." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Chat completions API rejection vs batched processor crash; different endpoints and models." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "CLIP attentions missing vs CLIPVision hidden_states missing; same family, but different outputs and regressions." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "ImportError for layer_type_validation vs local_files_only still making API requests; unrelated import vs networking bug." + }, + { + "left": "issue:44625", + "right": "issue:46365", + "accept": false, + "reason": "Qwen3.5 num_labels propagation vs Trainer import conflict with datasets; unrelated." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Gemma4 embedding resize bug vs Gemma4 causal mask construction bug; same model family but different concrete failures." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 61, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29870", + "issue:33453", + "issue:36296", + "issue:36683", + "issue:42994", + "issue:43010", + "issue:43064", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43575", + "issue:43618", + "issue:43867", + "issue:43976", + "issue:43992", + "issue:44075", + "issue:44077", + "issue:44222", + "issue:44263", + "issue:44355", + "issue:44380", + "issue:44393", + "issue:44492", + "issue:44625", + "issue:44655", + "issue:44855", + "issue:44908", + "issue:44938", + "issue:45125", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45440", + "issue:45593", + "issue:46077", + "issue:46202", + "issue:46365", + "issue:46512", + "issue:46851", + "issue:46877", + "issue:47000", + "issue:47059", + "issue:47110", + "issue:47317", + "issue:47321", + "issue:47332", + "issue:47337" + ], + "soft_pairs": [ + "issue:44655|issue:47321", + "issue:43299|issue:43867", + "issue:44655|issue:47332", + "issue:43618|issue:44380", + "issue:42994|issue:44655", + "issue:44625|issue:46077", + "issue:36683|issue:43475", + "issue:43352|issue:45593", + "issue:36683|issue:44077", + "issue:43976|issue:44938", + "issue:44355|issue:46877", + "issue:46077|issue:46512", + "issue:43867|issue:45216", + "issue:44263|issue:45440", + "issue:44938|issue:47059", + "issue:43618|issue:45216", + "issue:36683|issue:47110", + "issue:36683|issue:44625", + "issue:29870|issue:36296", + "issue:44355|issue:44855", + "issue:44908|issue:46851", + "issue:43010|issue:45278", + "issue:43867|issue:46202", + "issue:43992|issue:47317", + "issue:44075|issue:46365", + "issue:43992|issue:45310", + "issue:29870|issue:33453", + "issue:43064|issue:44492", + "issue:44393|issue:47000", + "issue:44393|issue:47337", + "issue:44222|issue:45125", + "issue:36296|issue:43575", + "issue:43618|issue:44077" + ], + "prepared_review_unit_hash": "6867ec84c3726f0bb4b5f9dcc053640abc1fc88104a82589eabfb9ddbf4c6932", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 36444, + "estimated_input_tokens": 9111, + "estimated_eval_tokens": 18478 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These artifacts are mostly unrelated issue reports; the soft-similarity pairs do not look like true duplicates or the same fix. No PRs are present. The most representative issue is the active, high-activity Gemma3Config bug.", + "confidence": 0.91, + "canonical_issue_reason": "issue:36683 is the most central standalone report here: open, very high discussion activity, and a concrete, narrowly described runtime error (`Gemma3Config` missing `vocab_size`).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:36683 is the best issue candidate because it is specific, actionable, and active; the rest are either narrower, closed, or clearly unrelated to one another.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44655", + "right": "issue:47321", + "accept": false, + "reason": "Both mention framework/model issues, but one is about `save_pretrained` on Pipeline objects and the other is an AMD Quark class update bug; different code paths." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Both are loading failures, but they concern different models and different failure modes; not the same underlying bug." + }, + { + "left": "issue:44655", + "right": "issue:47332", + "accept": false, + "reason": "Both involve saving, but one is generic Pipeline saving and the other is disk-offloaded Qwen3-VL saving; too different to merge." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "Different models and different regressions: CLIP attentions assignment vs GPT2 attention scaling under SDPA/FlashAttention." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure is a different bug from Pipeline `save_pretrained` failure." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Different models and different config propagation issues (`num_labels` vs bidirectional attention masking)." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Both are missing-attribute errors, but on unrelated model classes and different missing fields." + }, + { + "left": "issue:43352", + "right": "issue:45593", + "accept": false, + "reason": "Nemotron Flash Attention 2 support and D-FINE auxiliary-loss behavior are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "`vocab_size` missing on Gemma3Config is unrelated to patchtsmixer `post_init` handling." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility complaints, but they concern different versions and different breakages; not the same bug." + }, + { + "left": "issue:44355", + "right": "issue:46877", + "accept": false, + "reason": "Compiled-file import errors and auto-docstring stdout diagnostics are unrelated." + }, + { + "left": "issue:46077", + "right": "issue:46512", + "accept": false, + "reason": "Different config/state bugs: attention-mask construction vs stale `config.dtype` after `from_config`." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Both are model loading/saving regressions, but the titles point to different mechanisms and scopes; not enough to treat as the same issue." + }, + { + "left": "issue:44263", + "right": "issue:45440", + "accept": false, + "reason": "`torch.split()` behavior in one MoE indexer and DeepSeekV3 implementation divergence are distinct problems." + }, + { + "left": "issue:44938", + "right": "issue:47059", + "accept": false, + "reason": "Python 3.14 import failure and a CLI completion regression from `huggingface_hub` are unrelated." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIP attentions regression and Qwen3.5 checkpoint-saving regression are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:47110", + "accept": false, + "reason": "A missing config attribute and a `special_tokens` type error are different bugs." + }, + { + "left": "issue:36683", + "right": "issue:44625", + "accept": false, + "reason": "Both are config-related, but one is a missing `vocab_size` attribute and the other is `num_labels` propagation for Qwen3.5; different fixes." + }, + { + "left": "issue:29870", + "right": "issue:36296", + "accept": false, + "reason": "Bart generation `decoder_attention_mask` bug and tensor-parallel training bug are unrelated." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Compiled Python file errors and a Python 3.13 `@torch.jit.script` parsing issue are different import-time failures." + }, + { + "left": "issue:44908", + "right": "issue:46851", + "accept": false, + "reason": "Scheduler kwargs handling and Qwen3.5 variable-length position IDs are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "`@torch.no_grad` decoration for cache updates and broad import errors after upgrading Transformers are not the same bug." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "A generic state-dict-sorting load error and Gemma pipeline roundtrip failure are different code paths." + }, + { + "left": "issue:43992", + "right": "issue:47317", + "accept": false, + "reason": "UMT5 encoder weight loading and PPFormulaNet training-loss bug are unrelated." + }, + { + "left": "issue:44075", + "right": "issue:46365", + "accept": false, + "reason": "SGD args being ignored is not the same as an import failure caused by `datasets`." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "Both mention loading, but they are different Qwen/UMT5 issues with different missing pieces." + }, + { + "left": "issue:29870", + "right": "issue:33453", + "accept": false, + "reason": "Bart generation error and tokenizer loading regression are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "FSDP/PEFT optimizer-state corruption and a cache-strategy typo are not the same bug." + }, + { + "left": "issue:44393", + "right": "issue:47000", + "accept": false, + "reason": "Qwen3-VL bbox hallucination/error and RTDetrLoss infeasible cost matrix are both vision-related, but they are different failures in different models." + }, + { + "left": "issue:44393", + "right": "issue:47337", + "accept": false, + "reason": "A Qwen3-VL bounding-box issue and a flaky bbox-range test are not the same underlying defect." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 MoE save_pretrained failure and missing `_tp_plan` for tensor parallelism are different issues." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug and tensor-parallel OOM on Qwen2-57B are different problems; OOM alone does not imply the same defect." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIP attentions regression and patchtsmixer `post_init` policy are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 62, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:41628", + "issue:41669", + "issue:42548", + "issue:42890", + "issue:42898", + "issue:43010", + "issue:43065", + "issue:43232", + "issue:43502", + "issue:43525", + "issue:43576", + "issue:43618", + "issue:43723", + "issue:43824", + "issue:43867", + "issue:43874", + "issue:43976", + "issue:44062", + "issue:44112", + "issue:44246", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44734", + "issue:44749", + "issue:44964", + "issue:44977", + "issue:45071", + "issue:45125", + "issue:45357", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45561", + "issue:45850", + "issue:46132", + "issue:46617", + "issue:46664", + "issue:47110", + "issue:47317" + ], + "soft_pairs": [ + "issue:41669|issue:44246", + "issue:43010|issue:43976", + "issue:43867|issue:45375", + "issue:42890|issue:45561", + "issue:43867|issue:45357", + "issue:28282|issue:33453", + "issue:43010|issue:45850", + "issue:42548|issue:44351", + "issue:43618|issue:44977", + "issue:45125|issue:45478", + "issue:45125|issue:45375", + "issue:44360|issue:45440", + "issue:43502|issue:44062", + "issue:43525|issue:47317", + "issue:42890|issue:44112", + "issue:42890|issue:44964", + "issue:44734|issue:46617", + "issue:42898|issue:46664", + "issue:43723|issue:44556", + "issue:44387|issue:44556", + "issue:43010|issue:43065", + "issue:43824|issue:46617", + "issue:45071|issue:47317", + "issue:43576|issue:46132", + "issue:43502|issue:44485", + "issue:44355|issue:44749", + "issue:45850|issue:47110", + "issue:43874|issue:46617", + "issue:30333|issue:33453", + "issue:30990|issue:31795", + "issue:41628|issue:42548", + "issue:43010|issue:43232" + ], + "prepared_review_unit_hash": "ed495b25fc46412d2dab85413feef3413d34029380871716db6c2498706d5e64", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36824, + "estimated_input_tokens": 9206, + "estimated_eval_tokens": 18668 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated import/load regressions, tokenizer/model compatibility bugs, and test flakes. Only one pair looks plausibly like the same underlying issue: import-time slowness around `transformers` import.", + "confidence": 0.56, + "canonical_issue_reason": "issue:44246 is the broadest user-facing report in the only obvious duplicate pair, and it best represents the import-performance problem that issue:41669 appears to explain more concretely.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44246 is the best overall representative because it states the externally visible symptom in a general way and can absorb the more specific import-cost complaint in issue:41669.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41669", + "right": "issue:44246", + "accept": true, + "reason": "Both concern `transformers` import taking too long; 41669 gives a likely cause (`import *` in models) for the same performance problem described in 44246." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Different problems: one is a missing `@torch.no_grad` decoration on cache/layer updates, the other is Python version compatibility for Transformers 5.1.0." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "Different failure modes: sorted `state_dict` loading vs a missing Qwen3.5 vision config field silently dropped by strict validation." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "One is a flaky model integration test needing `set_seed()`, the other is an xdist race on `captured_info.txt`; both are test issues but not the same bug." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Both involve loading/saving models, but the concrete problems differ: `state_dict` sorting vs incorrect Qwen3.5 visual encoder keys on save." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "Different import/load failures: missing PyTorch for AutoModel vs a tokenizer-loading regression." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "`@torch.no_grad` on cache updates is a specific fix; the other is a broad complaint that 5.0 breaks too many models." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols (`PreTrainedModel` vs `HybridCache`) and likely different code paths." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "CLIP attentions not being assigned is unrelated to Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3.5 MoE, but one is about tensor-parallel plan metadata and the other is a `from_pretrained` error; not the same concrete bug." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Different Qwen3.5 MoE issues: missing tensor-parallel plan vs missing vision config field." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "The DSA indexer/ReLU discussion is unrelated to divergence between native and remote DeepSeekV3 implementations." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "`local_files_only=True` still making requests is unrelated to `AddedToken()` receiving duplicate `special` arguments." + }, + { + "left": "issue:43525", + "right": "issue:47317", + "accept": false, + "reason": "A missing `pad_token_id` on `Llama4Config` is unrelated to a PPFormulaNet training-loss bug." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "One is a seed-related flaky test, the other is a CI device-override test failure; different artifacts and failure mechanisms." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Seeded test flakiness is unrelated to failure loading `microsoft/Phi-4-multimodal-instruct`." + }, + { + "left": "issue:44734", + "right": "issue:46617", + "accept": false, + "reason": "KV-cache continuation tensor indexing in serve is unrelated to loading a trust-remote-code model from cache." + }, + { + "left": "issue:42898", + "right": "issue:46664", + "accept": false, + "reason": "Tokenization-space cleanup behavior in v5 is unrelated to the SmolVLM `num_frames` meaning inconsistency." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both are loading-related, but one is tokenizer loading via `AutoTokenizer`, while the other is checkpoint reload compatibility after upgrading from v4.57 to v5.2/5.3." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Int4 quantization OOM from extra CUDA reserved memory is a different issue from checkpoint reload incompatibility after upgrade." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "`no_grad` decoration on cache updates is unrelated to a dummy `nn.Conv2d` in `Sam3PixelDecoder`." + }, + { + "left": "issue:43824", + "right": "issue:46617", + "accept": false, + "reason": "Missing `Qwen2_5_VLForConditionalGeneration` import and trust-remote-code cache loading are different import/load failures." + }, + { + "left": "issue:45071", + "right": "issue:47317", + "accept": false, + "reason": "PretrainedConfig type-checking regression is unrelated to PPFormulaNet training loss." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "A broken `transformers env` command is not the same as `AttentionInterface.register` changing registered-function behavior." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "Unexpected API requests with `local_files_only=True` is unrelated to GLM-5 RoPE implementation concerns." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled Python file errors are unrelated to the slowdown when filtering data after upgrading Transformers." + }, + { + "left": "issue:45850", + "right": "issue:47110", + "accept": false, + "reason": "A broad complaint that 5.0 breaks many models is too general to equate with the specific `special_tokens` list-vs-dict attribute error." + }, + { + "left": "issue:43874", + "right": "issue:46617", + "accept": false, + "reason": "Missing `get_number_of_image_patches` in GLM46V is unrelated to trust-remote-code cache loading." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow reporting on failed training jobs is unrelated to tokenizer loading regression." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers getting stuck loading is unrelated to confusing `input_ids`/`past_key_values` documentation." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different parts of the package." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Cache/layer `no_grad` decoration and generation after `sync_gpus` are different generation-path issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted soft-edge pair (issue:41669 / issue:44246) is plausibly the same import-time slowness problem, while the other pairs are rejected with appropriately cautious reasoning. No overstatement stands out." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 63, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:31795", + "issue:33357", + "issue:33453", + "issue:36296", + "issue:38617", + "issue:42757", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43525", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43827", + "issue:43901", + "issue:44038", + "issue:44077", + "issue:44222", + "issue:44230", + "issue:44273", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44861", + "issue:44961", + "issue:45071", + "issue:45125", + "issue:45362", + "issue:45375", + "issue:45468", + "issue:45542", + "issue:45850", + "issue:46095", + "issue:46143", + "issue:46144", + "issue:46209", + "issue:46617", + "issue:46664", + "issue:46726", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:44222|issue:44230", + "issue:46143|issue:46144", + "issue:46664|issue:46726", + "issue:43502|issue:44222", + "issue:43299|issue:44556", + "issue:30333|issue:36296", + "issue:43827|issue:44961", + "issue:29870|issue:31795", + "issue:33357|issue:33453", + "issue:44038|issue:45125", + "issue:44355|issue:45071", + "issue:43502|issue:44961", + "issue:43576|issue:43824", + "issue:43576|issue:45071", + "issue:44230|issue:46095", + "issue:43576|issue:47059", + "issue:42757|issue:43576", + "issue:31795|issue:38617", + "issue:36296|issue:43901", + "issue:43475|issue:46617", + "issue:44351|issue:46617", + "issue:44315|issue:44360", + "issue:45362|issue:45850", + "issue:45375|issue:47110", + "issue:43525|issue:47110", + "issue:44485|issue:44961", + "issue:43673|issue:44492", + "issue:43352|issue:44961", + "issue:44315|issue:46209", + "issue:44961|issue:45468", + "issue:44273|issue:45542", + "issue:44077|issue:44861" + ], + "prepared_review_unit_hash": "adb099f2c93389d8eb40b4ed249a7c07fb6ef94717109f4a696c75c12ad18d52", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36727, + "estimated_input_tokens": 9182, + "estimated_eval_tokens": 18620 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the issues span unrelated regressions, model-specific bugs, docs, import errors, and CLI failures. No soft pair looks like the same underlying defect with enough specificity to merge.", + "confidence": 0.93, + "canonical_issue_reason": "There is no true duplicate anchor here. If one issue must represent the cluster, 43576 is the broadest and most discussed v5 regression report, with inbound references and overlap in the general upgrade-breakage theme, but it should not be treated as a duplicate hub.", + "canonical_pr_reason": null, + "best_issue_reason": "43576 is the best representative issue only because it is broad, active, and centrally situated among several v5-related regressions. It is still not a good canonical duplicate target because the cluster does not describe one single bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "Both mention FP8/moe, but one is a save_pretrained bug and the other is inference support; different code paths and fixes." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "One is about forwarding kwargs in RoFormer model methods; the other is about the attention implementation not using the attention interface. Related area, but not the same defect." + }, + { + "left": "issue:46664", + "right": "issue:46726", + "accept": false, + "reason": "Both concern vision/video processing, but one is an inconsistent num_frames semantic and the other is a missing vision backend causing an unrecognized image processor error." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "local_files_only API requests and FP8 moe save_pretrained are unrelated bugs." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both are version-v5 compatibility complaints, but one is Qwen3VL MoE loading and the other is checkpoint reload compatibility after upgrading; too broad to be the same bug." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLflow reporting on failed jobs and tensor parallel training are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs still referencing pipeline() and a random 'racoon' issue are clearly unrelated." + }, + { + "left": "issue:29870", + "right": "issue:31795", + "accept": false, + "reason": "Bart generate() UnboundLocalError and documentation confusion for input_ids/past_key_values are different problems." + }, + { + "left": "issue:33357", + "right": "issue:33453", + "accept": false, + "reason": "MacOS CLIP bus error and tokenizer loading regression are separate failures." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL-Moe loading breakage and missing _tp_plan for tensor parallelism are not the same concrete bug." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled Python file errors and PretrainedConfig type checking regressions differ in symptom, cause, and fix." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "local_files_only network leakage and a meaningless 'racoon' issue are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Broken transformers env command and missing Qwen2_5_VL import are different CLI/import issues." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "env command breakage and PretrainedConfig type-checking regression do not share a concrete code path." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "FP8 inference support for Qwen models and a deepseekv4 weight-conversion question are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:47059", + "accept": false, + "reason": "Broken env command and CLI test failure with huggingface_hub are distinct CLI incompatibilities." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing is_offline_mode import and the transformers env command failure are unrelated." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Forward-doc confusion and missing layer_type_validation import are different issue types and code paths." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor parallel training bug and TextClassificationPipeline docs mismatch are not the same underlying defect." + }, + { + "left": "issue:43475", + "right": "issue:46617", + "accept": false, + "reason": "Missing fpn_position_embeddings in SAM 3 Video and trust_remote_code cache FileNotFoundError are unrelated." + }, + { + "left": "issue:44351", + "right": "issue:46617", + "accept": false, + "reason": "Missing HybridCache import and cache-loading FileNotFoundError are different import/loading failures." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel not applied in model_init and a DSA indexer ReLU discussion are unrelated." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Qwen3.5-35B chat crash and the vague '5.0 breaks too many models' complaint are not the same concrete bug." + }, + { + "left": "issue:45375", + "right": "issue:47110", + "accept": false, + "reason": "Missing deepstack_visual_indexes in a specific config and special_tokens list/keys error are different model/config bugs." + }, + { + "left": "issue:43525", + "right": "issue:47110", + "accept": false, + "reason": "Llama4Config missing pad_token_id and special_tokens list/keys error are separate attribute/type problems." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation and the unrelated 'racoon' issue are not duplicates." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Missing GenerationMixin cache during chunked_prefill and a cache-strategy typo are not the same issue." + }, + { + "left": "issue:43352", + "right": "issue:44961", + "accept": false, + "reason": "Nemotron Flash Attention 2.0 support and the unrelated 'racoon' issue are different problems." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "Liger Kernel not applied in model_init and FP8 inference on B200 are distinct performance/runtime issues." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "The 'racoon' issue has no meaningful technical overlap with Gemma4AudioRelPositionalEncoding." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading behavior and undefined tf backend due to tensorboard-only install are unrelated." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "patchtsmixer post_init policy and _get_tied_weight_keys list-handling crash are separate code changes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 64, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:43298", + "issue:43408", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43867", + "issue:43976", + "issue:44246", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44393", + "issue:44466", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44655", + "issue:44704", + "issue:44861", + "issue:44908", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45230", + "issue:45341", + "issue:45381", + "issue:45753", + "issue:45850", + "issue:46018", + "issue:46095", + "issue:46172", + "issue:46218", + "issue:46328", + "issue:46655", + "issue:46664", + "issue:47059", + "issue:47110", + "issue:47176" + ], + "soft_pairs": [ + "issue:43867|issue:47110", + "issue:44623|issue:45230", + "issue:45381|issue:46664", + "issue:44273|issue:45003", + "issue:44485|issue:46664", + "issue:41669|issue:43673", + "issue:44466|issue:44556", + "issue:43525|issue:45850", + "issue:43572|issue:45850", + "issue:44393|issue:45753", + "issue:44393|issue:45071", + "issue:44556|issue:45092", + "issue:42371|issue:44492", + "issue:45850|issue:47059", + "issue:44246|issue:44273", + "issue:44655|issue:44861", + "issue:43408|issue:44315", + "issue:44492|issue:45850", + "issue:41669|issue:46218", + "issue:44295|issue:45230", + "issue:44492|issue:46328", + "issue:43976|issue:44393", + "issue:44297|issue:45230", + "issue:42890|issue:44908", + "issue:44704|issue:44908", + "issue:44908|issue:45071", + "issue:41669|issue:45341", + "issue:43576|issue:44908", + "issue:43576|issue:47176", + "issue:41669|issue:44351", + "issue:46018|issue:46172", + "issue:43976|issue:46218", + "issue:43298|issue:46095", + "issue:43976|issue:47176", + "issue:31795|issue:33453", + "issue:43519|issue:46655" + ], + "prepared_review_unit_hash": "bfcc5a85855ff685deb7531150187f48d25da199869870c9fd23b9188fb6436e", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 36, + "serialized_chars": 37880, + "estimated_input_tokens": 9470, + "estimated_eval_tokens": 19196 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated issues with a few loose v5 migration/loading/import-time similarities; none of the soft pairs looks like the same underlying bug or change, so all soft edges should be rejected. The best overall issue anchor is the checkpoint reload regression after upgrading to v5, since it is concrete and broadly representative of the migration breakages in this set.", + "confidence": 0.7, + "canonical_issue_reason": "issue:44556 is the clearest, most concrete v5 regression in the set: it describes a real checkpoint reload failure after upgrading from v4.57 to v5.2/v5.3, which makes it a useful duplicate-triage anchor for neighboring migration/loading reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44556 is the strongest global representative among these issues because it is specific, actionable, and centrally about v5 compatibility/loading failures rather than a vague or model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43867", + "right": "issue:47110", + "accept": false, + "reason": "Different bugs: state_dict loading error vs special_tokens/list-vs-dict handling." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "processor.save_pretrained missing files is specific; 45230 is just a generic bug report." + }, + { + "left": "issue:45381", + "right": "issue:46664", + "accept": false, + "reason": "Same model family, but distinct video-processing bugs: wrong vision_position_ids vs num_frames semantics." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Both touch loading/import behavior, but one is lazy loading not working and the other is unsafe sys.modules access." + }, + { + "left": "issue:44485", + "right": "issue:46664", + "accept": false, + "reason": "GLM-5 RoPE implementation is unrelated to SmolVLM video padding/frames issues." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-* slowdown is a performance issue; 43673 is a generation cache regression during chunked prefill." + }, + { + "left": "issue:44466", + "right": "issue:44556", + "accept": false, + "reason": "Tied-weight serialization inconsistency is a different code path from checkpoint reload compatibility after v5 upgrade." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Llama4 pad_token_id AttributeError is model-specific; 45850 is only a broad 'many models break' complaint." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLm pad_token_idx missing is a specific config bug, not the same as the broad v5 breakage report." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Different Qwen3_5 issues: 2D bbox hallucination/error vs infinite loop on a specific image." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Qwen3-VL bounding-box output bug is unrelated to PretrainedConfig type-checking failures." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both are v5 compatibility issues, but one is generic checkpoint reload after upgrade and the other is InternVL2 remote-code/meta-init incompatibility." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 behavior API change is unrelated to a cache-strategy typo." + }, + { + "left": "issue:45850", + "right": "issue:47059", + "accept": false, + "reason": "Broad v5 breakage claim is not the same as the CLI test failure with huggingface_hub 1.22.0." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Both involve import/lazy-loading, but one is intermittent slow import and the other is lazy loading not functioning correctly." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Saving Pipeline objects is a different problem from _get_tied_weight_keys crashing on a list." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "SAM3 tracker/model-type warning is unrelated to Liger Kernel not being applied via model_init." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A typo in cache strategies is not the same bug as the broad 'many models break in v5' report." + }, + { + "left": "issue:41669", + "right": "issue:46218", + "accept": false, + "reason": "Import-* slowdown is a performance regression; 46218 is a missing Qwen3_5 import/export path." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "position_ids buffer access error is specific; 45230 is too generic to indicate the same issue." + }, + { + "left": "issue:44492", + "right": "issue:46328", + "accept": false, + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad inefficiency are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version compatibility problem is not the same as a Qwen3-VL output error." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "tokenizer_class mismatch in save_pretrained is specific; 45230 is just a generic bug report." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Missing set_seed in tests and inverse_sqrt scheduler kwargs being ignored are different test vs scheduler bugs." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor kwargs passthrough to cached_file is unrelated to scheduler kwarg handling." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs bug and PretrainedConfig type-checking regression are unrelated subsystems." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Import-star performance regression is different from a testing_utils bug." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "Broken transformers env command is unrelated to inverse_sqrt scheduler kwargs." + }, + { + "left": "issue:43576", + "right": "issue:47176", + "accept": false, + "reason": "CLI env command breakage is unrelated to a shell-operator allowlist bypass security issue." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star slowdown is unrelated to missing HybridCache import." + }, + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "DeepSeek-V4 gating issue is not the same as the comb.to(...).transpose(...) bug." + }, + { + "left": "issue:43976", + "right": "issue:46218", + "accept": false, + "reason": "Python version incompatibility and Qwen3_5 import failure are different failure modes." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Continuous batching audio support is unrelated to DeepSeek-V4 weight conversion format/support." + }, + { + "left": "issue:43976", + "right": "issue:47176", + "accept": false, + "reason": "Python compatibility issue is unrelated to a shell command-substitution allowlist bypass." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Documentation confusion in model.forward is unrelated to tokenizer loading regression." + }, + { + "left": "issue:43519", + "right": "issue:46655", + "accept": false, + "reason": "Incorrect timestamp calculation in Qwen3VL is unrelated to SmolVLM video padding math." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 65, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:28348", + "issue:29870", + "issue:30333", + "issue:31515", + "issue:31795", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:43066", + "issue:43298", + "issue:43352", + "issue:43502", + "issue:43519", + "issue:43576", + "issue:43606", + "issue:43618", + "issue:43704", + "issue:43976", + "issue:44230", + "issue:44263", + "issue:44273", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:44998", + "issue:45230", + "issue:45250", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45656", + "issue:45850", + "issue:46218", + "issue:46894", + "issue:47059", + "issue:47110", + "issue:47176", + "issue:47177" + ], + "soft_pairs": [ + "issue:44315|issue:44829", + "issue:44485|issue:47176", + "issue:45230|issue:45656", + "issue:45468|issue:46894", + "issue:44360|issue:45468", + "issue:43976|issue:47177", + "issue:43576|issue:44273", + "issue:43519|issue:44485", + "issue:44485|issue:47177", + "issue:44273|issue:44908", + "issue:31795|issue:34689", + "issue:44861|issue:47110", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:43576|issue:46218", + "issue:42913|issue:43066", + "issue:44485|issue:45468", + "issue:43704|issue:44485", + "issue:45850|issue:47177", + "issue:45850|issue:47176", + "issue:42371|issue:43704", + "issue:43352|issue:47177", + "issue:43606|issue:45468", + "issue:43502|issue:43519", + "issue:44861|issue:47059", + "issue:43519|issue:44623", + "issue:45850|issue:46218", + "issue:44263|issue:45468", + "issue:43704|issue:45341", + "issue:28348|issue:45250", + "issue:43352|issue:46218", + "issue:44351|issue:46218", + "issue:44861|issue:45290", + "issue:44230|issue:44623", + "issue:44908|issue:44998", + "issue:44998|issue:47177", + "issue:44998|issue:47176", + "issue:28348|issue:43352", + "issue:44492|issue:44998", + "issue:43618|issue:44998", + "issue:44230|issue:45310", + "issue:31515|issue:38617", + "issue:29870|issue:31515", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:43298|issue:44230", + "issue:31515|issue:31795" + ], + "prepared_review_unit_hash": "b5f65af377cb6913a8f7a957e8b763ebe0100aeaa7f9f5b81f457320bcde661f", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 47, + "serialized_chars": 40919, + "estimated_input_tokens": 10230, + "estimated_eval_tokens": 20716 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All listed soft pairs appear to be unrelated issues across different models, features, or failure modes, so I reject every soft edge. There are no PRs in the set.", + "confidence": 0.93, + "canonical_issue_reason": "If one issue must be treated as the representative item, issue 44829 is the clearest, most concrete bug report: it names a specific model, a specific setting (`flash_attention_3`), and a concrete training failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44829 is the strongest standalone issue for triage quality because it is specific, reproducible, and describes an exact symptom rather than a broad or vague request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Different code paths: model_init/Liger Kernel application vs Flash Attention 3 training degeneration." + }, + { + "left": "issue:44485", + "right": "issue:47176", + "accept": false, + "reason": "Unrelated topics: GLM-5 RoPE implementation vs shell-command allowlist bypass." + }, + { + "left": "issue:45230", + "right": "issue:45656", + "accept": false, + "reason": "Generic bug report vs a specific DeepSpeed double optimizer-step bug; no shared underlying change." + }, + { + "left": "issue:45468", + "right": "issue:46894", + "accept": false, + "reason": "Different models and symptoms: Gemma-4 audio positional encoding vs CohereASR double-shift training loss bug." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "DSA/ReLU discussion is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:43976", + "right": "issue:47177", + "accept": false, + "reason": "Python-version compatibility issue vs tar/zip-slip RCE security bug." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Both mention v5, but one is `transformers env` and the other is lazy loading; different failures." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Timestamp calculation bug in Qwen3VL processor is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:44485", + "right": "issue:47177", + "accept": false, + "reason": "RoPE implementation bug vs command-substitution allowlist bypass." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading regression and inverse_sqrt scheduler kwargs are different subsystems and symptoms." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion about forward args vs a model-loading regression in Llama 3.2 vision." + }, + { + "left": "issue:44861", + "right": "issue:47110", + "accept": false, + "reason": "Different crashes in different helpers: tied weights keys vs special_tokens dict/list handling." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Processor timestamp math is unrelated to scheduler keyword propagation." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-star cleanup/performance issue vs VRAM leak in Qwen3ForCausalLM." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "`transformers env` command failure is unrelated to Qwen3_5 import errors." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both concern tokenizer v5 behavior, but the reported bugs are different enough to keep separate." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "GLM-5 RoPE implementation and Gemma-4 audio positional encoding are distinct model-specific bugs." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak in Qwen3ForCausalLM is unrelated to RoPE implementation." + }, + { + "left": "issue:45850", + "right": "issue:47177", + "accept": false, + "reason": "Broad regressions after v5 vs a tar/zip-slip arbitrary file-write RCE." + }, + { + "left": "issue:45850", + "right": "issue:47176", + "accept": false, + "reason": "General breakage report vs shell-operator allowlist bypass; different vulnerabilities." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 configuration guidance is unrelated to a VRAM leak bug." + }, + { + "left": "issue:43352", + "right": "issue:47177", + "accept": false, + "reason": "Flash Attention 2.0 support gap for Nemotron is unrelated to file-write RCE." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "Local-files-only network request bug vs timestamp calculation bug." + }, + { + "left": "issue:44861", + "right": "issue:47059", + "accept": false, + "reason": "Different failures: tied-weight key handling vs CLI completion attribute error." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Timestamp bug in Qwen3VL processor vs missing files in processor.save_pretrained." + }, + { + "left": "issue:45850", + "right": "issue:46218", + "accept": false, + "reason": "General v5 regressions vs Qwen3_5 import failure; not the same bug." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "torch.split return-value bug is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:43704", + "right": "issue:45341", + "accept": false, + "reason": "VRAM leak in model usage vs a testing_utils typo." + }, + { + "left": "issue:28348", + "right": "issue:45250", + "accept": false, + "reason": "GPT2-specific Flash Attention 2.0 support request vs a generic Flash Attention 2.0 issue title." + }, + { + "left": "issue:43352", + "right": "issue:46218", + "accept": false, + "reason": "Nemotron Flash Attention 2.0 support vs Qwen3_5 import failure." + }, + { + "left": "issue:44351", + "right": "issue:46218", + "accept": false, + "reason": "HybridCache import failure and Qwen3_5 import failure are different modules/problems." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weight-key crash vs chat-template crash with tool calls; different code paths." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request vs missing files on processor.save_pretrained." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs bug is unrelated to the placeholder/irrelevant issue." + }, + { + "left": "issue:44998", + "right": "issue:47177", + "accept": false, + "reason": "Irrelevant placeholder issue vs security RCE." + }, + { + "left": "issue:44998", + "right": "issue:47176", + "accept": false, + "reason": "Irrelevant placeholder issue vs shell-operator bypass." + }, + { + "left": "issue:28348", + "right": "issue:43352", + "accept": false, + "reason": "Same broad Flash Attention theme, but different models and concrete requests." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "Cache-strategy typo vs irrelevant placeholder issue." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression vs irrelevant placeholder issue." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 support request for qwen3-vl/qwen3.5 moe vs from_pretrained error in a different version/model." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow checkpoint loading vs ImportError for `layer_type_validation`; unrelated symptoms." + }, + { + "left": "issue:29870", + "right": "issue:31515", + "accept": false, + "reason": "BART generate decoder_attention_mask bug vs slow from_pretrained checkpoint loading." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow job-finalization bug vs checkpoint loading performance issue." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "from_pretrained slowdown and tensor-parallel training bug are different problems." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching support for audio models is unrelated to FP8 qwen3-vl/qwen3.5 support." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 27, + "review_unit_index": 66, + "review_unit_count": 66, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-533", + "nodes": [ + "issue:37934", + "issue:42936", + "issue:43240", + "issue:43668", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:46018", + "issue:46144" + ], + "soft_pairs": [ + "issue:43717|issue:46144", + "issue:42936|issue:46018", + "issue:43668|issue:44031", + "issue:42936|issue:43240", + "issue:43240|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:46144", + "issue:42936|issue:43813", + "issue:43668|issue:43813", + "issue:43240|issue:43668", + "issue:37934|issue:43668", + "issue:42936|issue:43668" + ], + "prepared_review_unit_hash": "d060d6efa04dc84d3ebf55f6253a97de7c60dade99b33fe9b3ee0279a19b4b25", + "component_budget": { + "node_count": 533, + "item_count": 533, + "soft_pair_count": 1964, + "serialized_chars": 1018550, + "estimated_input_tokens": 254638, + "estimated_eval_tokens": 509532 + }, + "budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 12, + "serialized_chars": 8611, + "estimated_input_tokens": 2153, + "estimated_eval_tokens": 4562 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are not duplicates: they span unrelated models, configs, warnings, and typo/behavior bugs across different code paths. No pair looks like the same underlying defect.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out; the set is heterogeneous and does not cluster around one bug or feature change.", + "canonical_pr_reason": "No pull requests are present.", + "best_issue_reason": "Issue 42936 is the strongest standalone report because it names a specific model, a concrete missing-use bug, and has the clearest actionable symptom.", + "best_pr_reason": "No pull requests are present.", + "soft_edge_verdicts": [ + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Different model families and different bugs: Mamba init-weights behavior vs RoFormer attention interface usage." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Mask2Former ignore_value handling is unrelated to DeepSeek-V4 shared expert gating." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "A type-hint issue in ModernBERTConfig is not the same as a tokenizer regex warning regression." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "One is about a model-specific ignore_value parameter; the other is about kwargs not being forwarded to a loss function." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss-function kwargs forwarding and RoFormer attention-interface usage are separate code paths and bugs." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2Former ignore_value handling does not match a tokenizer warning regression." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "These affect different architectures and distinct implementation layers." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "A Mask2Former behavior bug is not the same as a quantization-attribute typo." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Incorrect type annotation in config is unrelated to a typo in a quantization field name." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Passing kwargs to loss calculation and fixing a config type hint are different defects." + }, + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "RMS-norm implementation question for Llama4TextL2Norm is unrelated to a ModernBERT norm_eps type hint." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2Former ignore_value usage and ModernBERTConfig norm_eps typing do not describe the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10519-3", + "nodes": [ + "issue:10519", + "issue:11005", + "issue:11307" + ], + "soft_pairs": [ + "issue:10519|issue:11307", + "issue:10519|issue:11005", + "issue:11005|issue:11307" + ], + "prepared_review_unit_hash": "cac86e5bce768c272656a354c1b0666c7281c2bad37cb1cb977ef06d66c2dadf", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These three issues are unrelated feature requests: token truncation behavior, ReduceLROnPlateau-like scheduler functionality, and Wav2Vec2 word time offsets. The cluster looks spurious; no duplicates to merge.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Different areas and requests: tokenizer truncation direction vs Wav2Vec2 word boundary timestamps." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "Unrelated functionality: truncation strategy option vs learning-rate scheduler behavior." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "Different code paths and goals: scheduler adaptation vs extracting token/word offsets from audio model outputs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 3, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-13244-3", + "nodes": [ + "issue:13244", + "issue:15980", + "issue:16618" + ], + "soft_pairs": [ + "issue:13244|issue:16618", + "issue:15980|issue:16618" + ], + "prepared_review_unit_hash": "123b400b83b96fcddc280b4b2633058af1b4c23ae9f3712be93748ff166c9663", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2477, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2477, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is not a true duplicate set: one issue is about Tapas tokenization, and the other two are about model loading/download failures, but they point to different failure modes. Issue 16618 is the broadest representative of the loading-error side, while 15980 is specifically about private-model auth messaging.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 16618 is the most general representative of the cluster\u2019s model-loading problems; the other two are either unrelated (13244) or narrower/auth-specific (15980).", + "canonical_pr_reason": null, + "best_issue_reason": "16618 is the best issue to anchor the cluster because it captures the broader user-facing load failure theme, whereas 15980 is a specific login/error-message case and 13244 is a different tokenization problem.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tapas tokenization mismatch is a different underlying bug from failing to load bert-base-uncased; they affect different code paths and symptoms." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "15980 is about an unauthenticated private-model download and bad error messaging, while 16618 is a public-model load failure; not the same concrete bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 4, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-19631-3", + "nodes": [ + "issue:19631", + "issue:20362", + "issue:22372" + ], + "soft_pairs": [ + "issue:19631|issue:22372", + "issue:19631|issue:20362" + ], + "prepared_review_unit_hash": "d37841b74d599ded462731174b3434f6e1b58fa16d0709548690e4b1ea7a321d", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2351, + "estimated_input_tokens": 588, + "estimated_eval_tokens": 1432 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2351, + "estimated_input_tokens": 588, + "estimated_eval_tokens": 1432 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate feature requests for different model additions (EDSR/MDSR, FlexiBERT, Restormer). They share only a generic \"Add \" shape, so there is no duplicate relationship here.", + "confidence": 0.97, + "canonical_issue_reason": "issue:19631 is the oldest and most discussed item, so it is the best cluster representative, but it is still a distinct request rather than a duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:19631 is the strongest representative by age and discussion activity; however, all three issues should remain separate because they request different models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:19631", + "right": "issue:20362", + "accept": false, + "reason": "Different model targets: EDSR/MDSR vs FlexiBERT. The similarity is only the generic request pattern, not the same underlying issue." + }, + { + "left": "issue:19631", + "right": "issue:22372", + "accept": false, + "reason": "Different model targets: EDSR/MDSR vs Restormer. No shared code-path or concrete duplicate problem is evident." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 5, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-26350-15", + "nodes": [ + "issue:26350", + "issue:27088", + "issue:28936", + "issue:30191", + "issue:30240", + "issue:33260", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370", + "issue:46222", + "issue:46226", + "issue:46931", + "issue:47299", + "issue:47402" + ], + "soft_pairs": [ + "issue:26350|issue:33260", + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:47402", + "issue:27088|issue:43834", + "issue:27088|issue:30191", + "issue:28936|issue:43834", + "issue:27088|issue:28936", + "issue:28936|issue:44365", + "issue:28936|issue:30240", + "issue:28936|issue:38435", + "issue:30191|issue:30240", + "issue:28936|issue:30191", + "issue:30240|issue:44370", + "issue:30240|issue:44365", + "issue:30240|issue:47402", + "issue:30240|issue:43834", + "issue:30240|issue:47299", + "issue:38435|issue:44370", + "issue:38435|issue:44365", + "issue:38435|issue:47402", + "issue:38435|issue:43834", + "issue:38435|issue:47299", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:43834|issue:47402", + "issue:43834|issue:47299", + "issue:43834|issue:46226", + "issue:44365|issue:44370", + "issue:44365|issue:47402", + "issue:44365|issue:47299", + "issue:44365|issue:46226", + "issue:44370|issue:47402", + "issue:44370|issue:47299", + "issue:44370|issue:46226", + "issue:46222|issue:47299", + "issue:44370|issue:46222", + "issue:44365|issue:46222", + "issue:46222|issue:47402", + "issue:43834|issue:46222", + "issue:46226|issue:47402", + "issue:46226|issue:46931", + "issue:44370|issue:46931", + "issue:44365|issue:46931", + "issue:43834|issue:46931", + "issue:46931|issue:47402", + "issue:47299|issue:47402", + "issue:46226|issue:47299" + ], + "prepared_review_unit_hash": "bcc7d539f32b36453c8b5b015d51b06f1183f2fbe4908db89edce127392aa048", + "component_budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 48, + "serialized_chars": 26216, + "estimated_input_tokens": 6554, + "estimated_eval_tokens": 13364 + }, + "budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 48, + "serialized_chars": 26216, + "estimated_input_tokens": 6554, + "estimated_eval_tokens": 13364 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster mixes one unrelated feature request pair (Flash Attention vs GGUF) with many separate docs-translation issues for different languages. The i18n items are superficially similar, but they are not the same underlying change, so none of the soft edges should be merged.", + "confidence": 0.59, + "canonical_issue_reason": "issue:43834 is the cleanest generic i18n template and the most central representative of the translation-request subgroup, but the overall set is not a true duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43834 is the closest thing to a hub among the translation issues, but it still represents a distinct language-specific request rather than a mergeable duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Flash Attention support and GGUF support are different feature requests in different code paths." + }, + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Both are docs translation requests, but for separate language targets; not the same issue." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Both are docs translation requests, but for separate language targets; not the same issue." + }, + { + "left": "issue:27088", + "right": "issue:47402", + "accept": false, + "reason": "Both are translation-template issues, but they appear to be distinct language-specific requests." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Same translation template, but the issues are language-specific and not the same underlying task." + }, + { + "left": "issue:27088", + "right": "issue:30191", + "accept": false, + "reason": "Turkish docs translation and Persian docs translation are separate language requests." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish translation and the generic translation template are not the same concrete request." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Turkish and Spanish docs translations are separate language issues." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Different target languages; these are separate translation requests, not duplicates." + }, + { + "left": "issue:28936", + "right": "issue:30240", + "accept": false, + "reason": "Spanish and Polish docs translations are distinct language-specific issues." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Spanish and Romanian docs translations are distinct language-specific issues." + }, + { + "left": "issue:30191", + "right": "issue:30240", + "accept": false, + "reason": "Persian and Polish translation issues are separate language requests." + }, + { + "left": "issue:28936", + "right": "issue:30191", + "accept": false, + "reason": "Spanish and Persian translation issues are separate language requests." + }, + { + "left": "issue:30240", + "right": "issue:44370", + "accept": false, + "reason": "Polish translation and the generic template are not the same concrete translation task." + }, + { + "left": "issue:30240", + "right": "issue:44365", + "accept": false, + "reason": "Polish translation and the generic template are not the same concrete translation task." + }, + { + "left": "issue:30240", + "right": "issue:47402", + "accept": false, + "reason": "Polish translation and the generic template are not the same concrete translation task." + }, + { + "left": "issue:30240", + "right": "issue:43834", + "accept": false, + "reason": "Polish translation and the generic template are not the same concrete translation task." + }, + { + "left": "issue:30240", + "right": "issue:47299", + "accept": false, + "reason": "Polish docs translation and the Thai/branded translation-like issue are not the same request." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian translation and the generic template are not the same concrete request." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Romanian translation and the generic template are not the same concrete request." + }, + { + "left": "issue:38435", + "right": "issue:47402", + "accept": false, + "reason": "Romanian translation and the generic template are not the same concrete request." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Romanian translation and the generic template are not the same concrete request." + }, + { + "left": "issue:38435", + "right": "issue:47299", + "accept": false, + "reason": "Romanian translation and the Thai/branded translation-like issue are not the same request." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "Both are template-like translation issues, but they do not identify the same language or change." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "Both are template-like translation issues, but they do not identify the same language or change." + }, + { + "left": "issue:43834", + "right": "issue:47402", + "accept": false, + "reason": "Both are template-like translation issues, but they do not identify the same language or change." + }, + { + "left": "issue:43834", + "right": "issue:47299", + "accept": false, + "reason": "Template-like translation issue versus a different translation-like request; not the same underlying task." + }, + { + "left": "issue:43834", + "right": "issue:46226", + "accept": false, + "reason": "These look like separate translation/request artifacts, not one mergeable issue." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "Both are translation-template issues, but they are not the same language-specific request." + }, + { + "left": "issue:44365", + "right": "issue:47402", + "accept": false, + "reason": "Both are translation-template issues, but they are not the same language-specific request." + }, + { + "left": "issue:44365", + "right": "issue:47299", + "accept": false, + "reason": "Template-like translation issue versus a different translation-like request; not the same underlying task." + }, + { + "left": "issue:44365", + "right": "issue:46226", + "accept": false, + "reason": "These appear to be separate translation/request artifacts, not duplicates." + }, + { + "left": "issue:44370", + "right": "issue:47402", + "accept": false, + "reason": "Both are translation-template issues, but they are not the same language-specific request." + }, + { + "left": "issue:44370", + "right": "issue:47299", + "accept": false, + "reason": "Template-like translation issue versus a different translation-like request; not the same underlying task." + }, + { + "left": "issue:44370", + "right": "issue:46226", + "accept": false, + "reason": "These appear to be separate translation/request artifacts, not duplicates." + }, + { + "left": "issue:46222", + "right": "issue:47299", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:44370", + "right": "issue:46222", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:44365", + "right": "issue:46222", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:46222", + "right": "issue:47402", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:43834", + "right": "issue:46222", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:46226", + "right": "issue:47402", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:46226", + "right": "issue:46931", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:44370", + "right": "issue:46931", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:44365", + "right": "issue:46931", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:43834", + "right": "issue:46931", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:46931", + "right": "issue:47402", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + }, + { + "left": "issue:47299", + "right": "issue:47402", + "accept": false, + "reason": "Both are translation-like issues, but they do not point to the same concrete language request." + }, + { + "left": "issue:46226", + "right": "issue:47299", + "accept": false, + "reason": "These are different translation-like requests with different text/targets, so not the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-36718-22", + "nodes": [ + "issue:36718", + "issue:36817", + "issue:36879", + "issue:37015", + "issue:37334", + "issue:41115", + "issue:41211", + "issue:42503", + "issue:42738", + "issue:42954", + "issue:43671", + "issue:43837", + "issue:43909", + "issue:44121", + "issue:44205", + "issue:45027", + "issue:45177", + "issue:45322", + "issue:46133", + "issue:46645", + "issue:46653", + "issue:47004" + ], + "soft_pairs": [ + "issue:36718|issue:47004", + "issue:36718|issue:41115", + "issue:36817|issue:41211", + "issue:36817|issue:43671", + "issue:36817|issue:36879", + "issue:36817|issue:42954", + "issue:36879|issue:41211", + "issue:36879|issue:43909", + "issue:36879|issue:44121", + "issue:36879|issue:43837", + "issue:36879|issue:43671", + "issue:37015|issue:43837", + "issue:37015|issue:44121", + "issue:37015|issue:43909", + "issue:37015|issue:37334", + "issue:36879|issue:37015", + "issue:37334|issue:41115", + "issue:37334|issue:43837", + "issue:37334|issue:43671", + "issue:37334|issue:42503", + "issue:41115|issue:43837", + "issue:41115|issue:44121", + "issue:41115|issue:42503", + "issue:41211|issue:43671", + "issue:37015|issue:42503", + "issue:42503|issue:43837", + "issue:42503|issue:43671", + "issue:42738|issue:45027", + "issue:42503|issue:42738", + "issue:42738|issue:42954", + "issue:42738|issue:46653", + "issue:42954|issue:44205", + "issue:37334|issue:42954", + "issue:42954|issue:43671", + "issue:42503|issue:42954", + "issue:42954|issue:45177", + "issue:43671|issue:43837", + "issue:43671|issue:45027", + "issue:43671|issue:45177", + "issue:43671|issue:44121", + "issue:43837|issue:43909", + "issue:43837|issue:45322", + "issue:43837|issue:44121", + "issue:43909|issue:44205", + "issue:43909|issue:46645", + "issue:44121|issue:45322", + "issue:41115|issue:45027", + "issue:43837|issue:45027", + "issue:37015|issue:45027", + "issue:37334|issue:45027", + "issue:43837|issue:45177", + "issue:45322|issue:46133", + "issue:42954|issue:46653", + "issue:42503|issue:46653", + "issue:45027|issue:46653" + ], + "prepared_review_unit_hash": "3cc60960208b1d71e2fbeaa898c87d00a5e84e3ad84c76ef09c7a875ffae7506", + "component_budget": { + "node_count": 22, + "item_count": 22, + "soft_pair_count": 55, + "serialized_chars": 31690, + "estimated_input_tokens": 7923, + "estimated_eval_tokens": 16102 + }, + "budget": { + "node_count": 22, + "item_count": 22, + "soft_pair_count": 55, + "serialized_chars": 31690, + "estimated_input_tokens": 7923, + "estimated_eval_tokens": 16102 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a loose set of model-request/support issues across unrelated models and modalities. The titles are superficially similar, but they do not describe the same underlying change, so there is no good canonical duplicate target here.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36718", + "right": "issue:47004", + "accept": false, + "reason": "Both are keypoint matcher requests, but RoMa and LoMa are different models/changes." + }, + { + "left": "issue:36718", + "right": "issue:41115", + "accept": false, + "reason": "Different domains: keypoint matching vs a MiniCPM3 model architecture request." + }, + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "EuroBert config support and DEIMv2 are unrelated model additions." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert config and Qwen3-TTS are different model families and tasks." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "A config update for EuroBert is not the same as adding RF-DETR." + }, + { + "left": "issue:36817", + "right": "issue:42954", + "accept": false, + "reason": "EuroBert config work is unrelated to Xiaomi MiMo support." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are distinct detection model requests." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "Different model requests: RF-DETR vs LFM2.5 Audio 1.5B." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR and a weight-sparse transformer request are not the same change." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "Object detection model support and Qwen3-ASR support are unrelated." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR and Qwen3-TTS are different models and tasks." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are separate model support requests." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT support is unrelated to the OpenAI weight-sparse transformer request." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT and LFM2.5 Audio are different architectures and tasks." + }, + { + "left": "issue:37015", + "right": "issue:37334", + "accept": false, + "reason": "NeoBERT and LLMDet are distinct model additions, not one bug/change." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are unrelated model requests." + }, + { + "left": "issue:37334", + "right": "issue:41115", + "accept": false, + "reason": "LLMDet support and MiniCPM3 architecture work are different model integrations." + }, + { + "left": "issue:37334", + "right": "issue:43837", + "accept": false, + "reason": "LLMDet and Qwen3-ASR are unrelated model support requests." + }, + { + "left": "issue:37334", + "right": "issue:43671", + "accept": false, + "reason": "LLMDet and Qwen3-TTS are different tasks and code paths." + }, + { + "left": "issue:37334", + "right": "issue:42503", + "accept": false, + "reason": "LLMDet support and ModernVBERT models are not the same underlying change." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture and Qwen3-ASR support are different model requests." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 and the OpenAI sparse transformer request do not describe the same change." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT are unrelated model additions." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 and Qwen3-TTS are distinct model support requests." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are different model families." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT model support is unrelated to Qwen3-ASR support." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are different model additions." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like RoPE support and Voxtral TTS support are not the same feature." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "ModernVBERT models and generic BERT-like RoPE support are related only broadly, not the same change." + }, + { + "left": "issue:42738", + "right": "issue:42954", + "accept": false, + "reason": "BERT-like RoPE support is unrelated to Xiaomi MiMo." + }, + { + "left": "issue:42738", + "right": "issue:46653", + "accept": false, + "reason": "BERT-like RoPE support and a personal-name issue are unrelated." + }, + { + "left": "issue:42954", + "right": "issue:44205", + "accept": false, + "reason": "MiMo and SAM3-LiteText are different model support requests." + }, + { + "left": "issue:37334", + "right": "issue:42954", + "accept": false, + "reason": "LLMDet support and Xiaomi MiMo are unrelated model requests." + }, + { + "left": "issue:42954", + "right": "issue:43671", + "accept": false, + "reason": "Xiaomi MiMo and Qwen3-TTS are different model families." + }, + { + "left": "issue:42503", + "right": "issue:42954", + "accept": false, + "reason": "ModernVBERT and Xiaomi MiMo are not the same underlying request." + }, + { + "left": "issue:42954", + "right": "issue:45177", + "accept": false, + "reason": "MiMo and DeepSeek-OCR2 are unrelated model additions." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Qwen3-TTS and Qwen3-ASR are different modalities and separate support requests." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-TTS and Voxtral TTS are different model support tasks." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS support and DeepSeek-OCR2 support are unrelated." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-TTS and the sparse transformer request are not the same code-path problem." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are different audio models." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR and EUPE are unrelated model support requests." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "ASR model support and weight-sparse transformer support are different changes." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "LFM2.5 Audio and SAM3-LiteText are unrelated model requests." + }, + { + "left": "issue:43909", + "right": "issue:46645", + "accept": false, + "reason": "LFM2.5 Audio support and Granite SWA+sink options are different changes." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Weight-sparse transformer support and EUPE are distinct model additions." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 architecture work and Voxtral TTS support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral-4B-TTS are different speech-model requests." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral TTS have no shared underlying bug/change." + }, + { + "left": "issue:37334", + "right": "issue:45027", + "accept": false, + "reason": "LLMDet support and Voxtral TTS support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are separate model support requests." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "EUPE and TIPSv2 are different Meta/DeepMind model requests." + }, + { + "left": "issue:42954", + "right": "issue:46653", + "accept": false, + "reason": "MiMo and a name-only issue are clearly not the same change." + }, + { + "left": "issue:42503", + "right": "issue:46653", + "accept": false, + "reason": "ModernVBERT and a name-only issue are unrelated." + }, + { + "left": "issue:45027", + "right": "issue:46653", + "accept": false, + "reason": "Voxtral TTS support and a name-only issue are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 7, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-38740-2", + "nodes": [ + "issue:38740", + "issue:46443" + ], + "soft_pairs": [ + "issue:38740|issue:46443" + ], + "prepared_review_unit_hash": "5e041c3e70b2184d9600b1e23c70090aaa65e0f482c96f82f2b0bedb26217907", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1505, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1505, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are unrelated: one is a docs request to add Pruna as an optimization framework, and the other is a Vision RoPE refactor. No duplicate or same-underlying-change relationship is evident.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38740", + "right": "issue:46443", + "accept": false, + "reason": "Different topics and scopes: documentation/framework listing vs a Vision RoPE code refactor. They do not describe the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 8, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-41084-2", + "nodes": [ + "issue:41084", + "issue:43037" + ], + "soft_pairs": [ + "issue:41084|issue:43037" + ], + "prepared_review_unit_hash": "b3ab2f23bfbf83089200bdd35408b71073ef6a3eb73fbe7fdb47e02fb045dac1", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1479, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1479, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one requests Set Block Decoding, the other asks for DeepSeek v3.2 support. They describe different feature areas and different underlying work.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "No single canonical issue fits this cluster because the issues are about different requests and do not share the same underlying bug or change.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Reject: the titles and implied scopes are unrelated; block decoding behavior is a codec/runtime feature request, while DeepSeek v3.2 support is model compatibility work." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 9, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-41306-2", + "nodes": [ + "issue:41306", + "issue:41826" + ], + "soft_pairs": [ + "issue:41306|issue:41826" + ], + "prepared_review_unit_hash": "5f3b891ba5f8b465396304c3e6ee07c0dfc862e72ef572ef41a97e1912661d71", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two distinct feature requests in the same general 'Tiled' area, but they target different components and changes: one is about a from_pretrained() flag for TiledFusedLogitsLoss/FLCE, the other is about integrating TiledMLP for lower memory use.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 41306 is the more concrete, narrowly scoped API request and is the better anchor for this cluster; 41826 is a separate TiledMLP feature proposal.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 41306 is the best representative issue because it describes a specific user-facing change with a clear code path, whereas 41826 is a different feature area.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41306", + "right": "issue:41826", + "accept": false, + "reason": "Different underlying requests: one asks for a from_pretrained() flag for TiledFusedLogitsLoss/FLCE, the other proposes TiledMLP integration for memory reduction. They are related only at a broad subsystem level, not the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 10, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43598-2", + "nodes": [ + "issue:43598", + "issue:43600" + ], + "soft_pairs": [ + "issue:43598|issue:43600" + ], + "prepared_review_unit_hash": "7b7bd12133201876507696bf8cc5698b84410e251b1d434fe7e3f7f85f14c61e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1582, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1582, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are related to Trainer customization, but they describe different problems: one is about `remove_unused_column` behavior and the other about FLOPS estimation for custom trainers. They should not be deduplicated.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 43598 is the more concrete Trainer-behavior bug report, centered on a specific setting (`remove_unused_column`) and its customizability impact.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43598 is the better anchor for this cluster because it is specific, actionable, and directly tied to a single Trainer code path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "They share a broad theme of custom Trainer support, but one concerns column removal/customizability while the other concerns FLOPS counting. Different underlying bugs and likely different fixes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 11, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43684-2", + "nodes": [ + "issue:43684", + "issue:44405" + ], + "soft_pairs": [ + "issue:43684|issue:44405" + ], + "prepared_review_unit_hash": "a2d7233a7f05c12dcec207254326c83cf7110dd953df637d747e4ebe9ce74263", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related only at a broad \u201cadd model support to AutoModel\u201d level, but they target different Qwen model families and different auto classes, so they should not be deduplicated.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 43684 is the broader and more central representative: it requests registration for Qwen3-Omni in AutoModel/AutoModelForConditionalGeneration, whereas the other issue is a narrower sequence-classification request for Qwen3.5.", + "canonical_pr_reason": null, + "best_issue_reason": "43684 is the better cluster representative because it is the more general model-registration request and closer to the likely architectural change, while 44405 is a separate, narrower capability gap.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying changes: Qwen3-Omni registration for AutoModel/AutoModelForConditionalGeneration versus Qwen3.5 sequence-classification support for Qwen3_5Config. Same broad subsystem, but not the same bug or fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 12, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43835-4", + "nodes": [ + "issue:43835", + "issue:43845", + "issue:43846", + "issue:43859" + ], + "soft_pairs": [ + "issue:43835|issue:43845", + "issue:43835|issue:43846", + "issue:43835|issue:43859", + "issue:43845|issue:43846", + "issue:43845|issue:43859", + "issue:43846|issue:43859" + ], + "prepared_review_unit_hash": "625c8bbbf53745591b6c9087b551d4ac0c02b0dc10e102c75cb869f682556466", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4260, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4260, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Three issues are near-identical placeholders titled \"huggingface\" and likely duplicate each other; \"Fett\" appears unrelated. No PRs are present.", + "confidence": 0.62, + "canonical_issue_reason": "Issue 43845 is the earliest of the repeated \"huggingface\" reports and is the clearest representative of that duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43845 is the best single issue to keep because it matches the repeated topic and predates the other near-duplicate \"huggingface\" reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "\"Fett\" does not match the repeated \"huggingface\" topic, and there is no concrete evidence they describe the same problem." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "\"Fett\" appears unrelated to the \"huggingface\" reports; title similarity is too weak to treat them as the same issue." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "No meaningful overlap beyond being closed issues; the topics do not line up." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": true, + "reason": "They have the same title, were opened minutes apart, and nothing in the available metadata distinguishes them as different bugs." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": true, + "reason": "This appears to be the same repeated \"huggingface\" issue report, with no evidence of a distinct underlying problem." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": true, + "reason": "Same repeated \"huggingface\" title and no differentiating content; likely duplicate reports of the same issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet: the three \"huggingface\" issues share the same title and are close enough in timing to plausibly form a duplicate cluster, while \"Fett\" is not well supported as related. The choice of 43845 as the canonical/earliest representative is reasonable, and the soft-edge verdicts stay mostly conservative." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44371-6", + "nodes": [ + "issue:44371", + "issue:44418", + "issue:44572", + "issue:44573", + "issue:44937", + "issue:47036" + ], + "soft_pairs": [ + "issue:44371|issue:44572", + "issue:44371|issue:44573", + "issue:44371|issue:44418", + "issue:44371|issue:47036", + "issue:44371|issue:44937", + "issue:44418|issue:44572", + "issue:44418|issue:44573", + "issue:44572|issue:44573", + "issue:44572|issue:47036", + "issue:44572|issue:44937", + "issue:44573|issue:47036", + "issue:44573|issue:44937" + ], + "prepared_review_unit_hash": "0037db997ec137ca3beba49762766d561fbfee8119d46ad07a8f83aa320db6c8", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 12, + "serialized_chars": 7346, + "estimated_input_tokens": 1837, + "estimated_eval_tokens": 3930 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 12, + "serialized_chars": 7346, + "estimated_input_tokens": 1837, + "estimated_eval_tokens": 3930 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues look like unrelated low-signal/spam reports rather than duplicates of one concrete bug or change. No soft edge has enough evidence to merge.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44418 is the only non-spam title and is the cleanest representative if one must be chosen, but it still lacks enough detail to serve as a true canonical duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are titled \"\", but there is no issue text or concrete shared defect/change to show they are the same underlying report." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Same spam-like title, but no substantive content links them to the same bug or change." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "One is spam and the other is a vague documentation suggestion; they do not describe the same issue." + }, + { + "left": "issue:44371", + "right": "issue:47036", + "accept": false, + "reason": "Spam placeholder versus a community benchmark tip; no shared concrete problem." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus a distinct titled issue about \"Google LLC\"; no evidence of duplication." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "A vague documentation suggestion is not the same underlying report as a spam placeholder." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "No concrete overlap beyond low-quality metadata; not enough to treat as duplicates." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Both are spam placeholders, but identical template text alone is insufficient to prove the same underlying bug or change." + }, + { + "left": "issue:44572", + "right": "issue:47036", + "accept": false, + "reason": "Spam placeholder and benchmark tip are unrelated in title and implied intent." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "No shared concrete defect or code-path issue is visible." + }, + { + "left": "issue:44573", + "right": "issue:47036", + "accept": false, + "reason": "No evidence these refer to the same underlying problem; titles suggest different subjects." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Distinct issue titles with no supporting body or targets; not a safe duplicate merge." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 14, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45145-2", + "nodes": [ + "issue:45145", + "issue:45151" + ], + "soft_pairs": [ + "issue:45145|issue:45151" + ], + "prepared_review_unit_hash": "02c2ed00cf0a0e06ef6048df6cddac76e154669b0dd78a241d0a9169d6042cc1", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1645, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1645, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two closed issues appear to be duplicates of the same report: identical titles, highly similar content size, and no distinguishing links or targets. The earlier issue is the better canonical choice.", + "confidence": 0.98, + "canonical_issue_reason": "Issue #45145 is earlier and has an inbound reference, so it is the better canonical representative of this duplicate pair.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #45145 is the strongest global issue candidate because it is the earliest instance of the duplicated report and already has some reference activity.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": true, + "reason": "Same title and evidently the same underlying energy-optimization report; no evidence of different bugs or scope." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet: both issues share the exact title, have very similar body lengths, and there are no explicit issue links or targets to distinguish them. Choosing the earlier issue with the inbound reference as canonical is conservative and reasonable. The duplicate/soft-edge verdict is also sufficiently cautious given the available evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 15, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45295-2", + "nodes": [ + "issue:45295", + "issue:45373" + ], + "soft_pairs": [], + "prepared_review_unit_hash": "f5fd95fada2f530ab0999e343c17c16266cde0c7182bb22509972ca44827cd49", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 0, + "serialized_chars": 1286, + "estimated_input_tokens": 322, + "estimated_eval_tokens": 900 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 0, + "serialized_chars": 1286, + "estimated_input_tokens": 322, + "estimated_eval_tokens": 900 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues describe the same missing Gemma 4 sequence-classification support; the later, more specific open issue is the better cluster representative.", + "confidence": 0.94, + "canonical_issue_reason": "issue #45373 is the clearest canonical tracker: it is open, explicitly names the missing class `Gemma4ForSequenceClassification`, and states the module-level gap relative to Gemma 2/3.", + "canonical_pr_reason": null, + "best_issue_reason": "issue #45373 best captures the concrete request and likely remaining work, while #45295 is a shorter earlier duplicate with less detail and is already closed.", + "best_pr_reason": null, + "soft_edge_verdicts": [] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 16, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45762-2", + "nodes": [ + "issue:45762", + "issue:45763" + ], + "soft_pairs": [ + "issue:45762|issue:45763" + ], + "prepared_review_unit_hash": "9013b8829ce7cd36308f9c021a22ad23ee36fe83098e0b63403438fbb7a16047", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two closed issues do not appear to be duplicates; they are separate name/date entries with no shared targets, files, or discussion context.", + "confidence": 0.96, + "canonical_issue_reason": "No canonical issue: the titles refer to different people and different dates, with no evidence they describe the same underlying problem.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good global representative because they appear to be distinct records rather than duplicate reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "Only superficial title similarity; the issues name different people and contain no shared evidence of the same underlying bug or request." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 17, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45764-2", + "nodes": [ + "issue:45764", + "issue:45765" + ], + "soft_pairs": [ + "issue:45764|issue:45765" + ], + "prepared_review_unit_hash": "a2c7e544d9389d583affd9db25301989eaebd605bd7e34f7d776c19b5e16aded", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster contains two unrelated closed issues with only superficial similarity in title format (personal names and dates). They do not describe the same bug, feature, or failure mode, so this is not a duplicate cluster.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue: both entries are isolated, content-free, and unrelated beyond a similar name/date pattern.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a strong canonical choice because neither contains problem details; if forced, no issue is meaningfully better than the other.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "Reject: the titles refer to different people and different dates, with no shared bug report, code path, or explicit linkage." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 18, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45810-2", + "nodes": [ + "issue:45810", + "issue:46295" + ], + "soft_pairs": [ + "issue:45810|issue:46295" + ], + "prepared_review_unit_hash": "0f85cf7bfea147db706c444a40e2cd69934285d49e73204308cbdb6ca1c04477", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two open issues are not duplicates: one requests a new Qwen3_5 token-classification/value-model class, while the other is about FP8 training support. They concern different code paths and different feature areas.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "No canonical issue emerges here because the cluster does not represent the same bug or request. If one must be chosen as the broader umbrella, issue #46295 is more general, but it is still unrelated to #45810.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45810", + "right": "issue:46295", + "accept": false, + "reason": "The titles and descriptions point to unrelated features: adding a Qwen3_5 token-classification model versus enabling FP8 training support. They do not describe the same underlying problem or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 19, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45869-5", + "nodes": [ + "issue:45869", + "issue:45870", + "issue:45871", + "issue:45872", + "issue:45873" + ], + "soft_pairs": [ + "issue:45869|issue:45870", + "issue:45869|issue:45871", + "issue:45870|issue:45871", + "issue:45870|issue:45872", + "issue:45870|issue:45873", + "issue:45871|issue:45872", + "issue:45871|issue:45873", + "issue:45872|issue:45873" + ], + "prepared_review_unit_hash": "2bbb35c717746627a9072f3c52742765a965e529473deb9ad8717a6438b5e8e7", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5727, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5727, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is actually two separate exact-duplicate issue pairs: 45870/45871 and 45872/45873. Issue 45869 is unrelated marketing/dataset content and should not be merged with the others.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue for the whole cluster: it mixes two distinct duplicate topics. For the first pair, 45870 is the cleanest representative because it is the earliest exact duplicate of 45871 and has the same title/body.", + "canonical_pr_reason": null, + "best_issue_reason": "45870 is the best single issue to keep if one representative is needed, but only for the 'Example: Financial time series prediction with DataForge' pair; the cluster should otherwise be split because 45872/45873 are a different duplicate pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Dataset bundle promotion vs. an example/tutorial issue; same financial theme but not the same underlying bug or change." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Dataset bundle promotion vs. an example/tutorial issue; not the same underlying request." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Identical title and matching body size/content pattern; these are the same issue." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Both mention financial time series, but one is an example prediction workflow and the other is a scraping-service feature request; different underlying issues." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Same as above: related domain, but different feature/problem statement." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Different requests: example prediction workflow vs. financial time series plus scraping service." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Different requests: example prediction workflow vs. financial time series plus scraping service." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Identical title and matching body size/content pattern; these are the same issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet: 45870/45871 and 45872/45873 have identical titles and matching body-length/Jaccard signals, while 45869 is a different dataset-promo issue. The cluster-splitting conclusion and the conservative cross-pair non-duplicates are reasonable." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 20, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46439-2", + "nodes": [ + "issue:46439", + "issue:46441" + ], + "soft_pairs": [ + "issue:46439|issue:46441" + ], + "prepared_review_unit_hash": "029eb312bcb93d95de774f0ddb0667905ebd6bcc9d050b33a6b17153b18f5b31", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1641, + "estimated_input_tokens": 411, + "estimated_eval_tokens": 1078 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1641, + "estimated_input_tokens": 411, + "estimated_eval_tokens": 1078 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are related in theme (`static` cache + linear-attention), but the evidence here is only titles/metadata, so they should not be treated as the same bug. The safer triage is to keep them separate and use the broader issue as the representative one.", + "confidence": 0.68, + "canonical_issue_reason": "Issue 46439 is the broader, more general title: it describes a `static` cache problem affecting linear-attention layers, while 46441 is a narrower Qwen3.5-specific crash report with a different surfaced exception.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46439 is the best standalone issue because its title captures the underlying subsystem-level failure more generally than the model-specific crash in 46441.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46439", + "right": "issue:46441", + "accept": false, + "reason": "The titles suggest a shared area (`static` cache / linear-attention), but there is not enough evidence here to conclude they are the same underlying bug; the surfaced failures are different and the packet lacks bodies or diffs to connect them." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The result overstates what is supported by the packet. We only have titles and metadata here, not the issue bodies, so claims like `early_initialization`, shape corruption details, and a specific root-cause cache initialization bug are not grounded. The pair may be related by theme (`static` cache + linear-attention), but the soft-edge acceptance should stay conservative and not assert they are the same bug on this evidence alone." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 21, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46551-2", + "nodes": [ + "issue:46551", + "issue:46552" + ], + "soft_pairs": [ + "issue:46551|issue:46552" + ], + "prepared_review_unit_hash": "412cce0fb379a567bef368b4b66d73a0121b71d305ac32237936286ce1cf7082", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1707, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1707, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are effectively identical reports of the same Wav2Vec2CTCTokenizer v5 bug; the newer open issue is the better canonical thread.", + "confidence": 0.99, + "canonical_issue_reason": "Issue 46552 is the newer, still-open report with inbound references, so it is the best place to track and consolidate discussion for this bug.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46552 best represents the cluster because it is active and duplicates the exact same title/problem statement as 46551.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46551", + "right": "issue:46552", + "accept": true, + "reason": "Exact same title and same underlying tokenizer bug report; these are duplicates of the same issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both issues have the same title, the same body length, and the soft-similarity signal is maximal. Choosing the newer open issue (46552) as the canonical thread is a conservative inference supported by its open state and inbound references. The soft-edge duplicate verdict is also appropriately cautious." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 22, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46657-2", + "nodes": [ + "issue:46657", + "issue:46692" + ], + "soft_pairs": [ + "issue:46657|issue:46692" + ], + "prepared_review_unit_hash": "dcc35bcf6b4b00ae895e7f33006ac3b2b442c9860e4a77c72983b5658bd4e884", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1569, + "estimated_input_tokens": 393, + "estimated_eval_tokens": 1042 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1569, + "estimated_input_tokens": 393, + "estimated_eval_tokens": 1042 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one reports a tensor shape mismatch during object-detection training, while the other reports an out-of-memory failure after several epochs in RT-DETR v2. They describe different failure modes with no clear shared code-path evidence.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #46657 is the better canonical representative because it describes a concrete, specific training error rather than a broad resource issue.", + "canonical_pr_reason": null, + "best_issue_reason": "#46657 is the most representative issue in the cluster due to its clearer error signature and tighter problem statement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46657", + "right": "issue:46692", + "accept": false, + "reason": "Different underlying bugs: a tensor-size mismatch versus delayed OOM during training. The titles and problem descriptions point to separate failure modes, not the same code-path defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 23, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46822-2", + "nodes": [ + "issue:46822", + "issue:46823" + ], + "soft_pairs": [ + "issue:46822|issue:46823" + ], + "prepared_review_unit_hash": "c3e3cdcdadc5bdb1047b71953eb38c2df0a8ef7f5015fd798f744f34b10170c4", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1673, + "estimated_input_tokens": 419, + "estimated_eval_tokens": 1094 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1673, + "estimated_input_tokens": 419, + "estimated_eval_tokens": 1094 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are about different bugs in different loading paths: ZeRO-3/zero.Init partitioning for a multimodal submodule vs. 4-bit device_map/max_memory dispatch behavior. They should not be deduplicated.", + "confidence": 0.98, + "canonical_issue_reason": "No true canonical duplicate exists here; the issues describe unrelated failure modes and code paths.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46822 is the more concrete and narrowly scoped report, but it is not a duplicate of 46823.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46822", + "right": "issue:46823", + "accept": false, + "reason": "Different underlying problems: one is ZeRO-3 zero.Init failing to partition a composite submodule causing OOM, the other is 4-bit loading with device_map='auto' and max_memory incorrectly dispatching to CPU. Same general ML loading area, but not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 24, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46906-21", + "nodes": [ + "issue:46906", + "issue:46940", + "issue:46948", + "issue:46969", + "issue:46985", + "issue:46998", + "issue:47037", + "issue:47049", + "issue:47053", + "issue:47091", + "issue:47102", + "issue:47120", + "issue:47182", + "issue:47227", + "issue:47256", + "issue:47265", + "issue:47279", + "issue:47309", + "issue:47377", + "issue:47390", + "issue:47406" + ], + "soft_pairs": [ + "issue:46906|issue:46940", + "issue:46906|issue:46948", + "issue:46906|issue:46969", + "issue:46906|issue:46985", + "issue:46906|issue:47037", + "issue:46940|issue:46948", + "issue:46940|issue:46969", + "issue:46940|issue:46985", + "issue:46940|issue:46998", + "issue:46948|issue:46969", + "issue:46948|issue:46985", + "issue:46948|issue:46998", + "issue:46969|issue:46985", + "issue:46969|issue:47037", + "issue:46969|issue:47091", + "issue:46985|issue:47037", + "issue:46985|issue:47049", + "issue:46998|issue:47053", + "issue:46998|issue:47037", + "issue:46998|issue:47049", + "issue:46985|issue:46998", + "issue:46998|issue:47091", + "issue:47037|issue:47049", + "issue:47037|issue:47091", + "issue:47037|issue:47053", + "issue:47049|issue:47053", + "issue:47049|issue:47091", + "issue:47053|issue:47091", + "issue:47053|issue:47102", + "issue:47091|issue:47102", + "issue:46985|issue:47091", + "issue:47037|issue:47102", + "issue:46985|issue:47102", + "issue:47049|issue:47102", + "issue:47120|issue:47182", + "issue:47120|issue:47256", + "issue:47120|issue:47279", + "issue:47120|issue:47265", + "issue:47120|issue:47377", + "issue:47182|issue:47377", + "issue:47182|issue:47256", + "issue:47182|issue:47265", + "issue:47182|issue:47279", + "issue:47227|issue:47377", + "issue:47227|issue:47256", + "issue:47227|issue:47279", + "issue:47182|issue:47227", + "issue:47227|issue:47265", + "issue:47256|issue:47265", + "issue:47256|issue:47279", + "issue:47102|issue:47256", + "issue:47265|issue:47279", + "issue:47265|issue:47309", + "issue:47265|issue:47377", + "issue:47279|issue:47309", + "issue:47279|issue:47377", + "issue:47309|issue:47377", + "issue:47182|issue:47309", + "issue:47227|issue:47309", + "issue:47390|issue:47406", + "issue:47377|issue:47390", + "issue:47279|issue:47390", + "issue:47227|issue:47390", + "issue:47256|issue:47390", + "issue:47377|issue:47406", + "issue:47279|issue:47406", + "issue:47120|issue:47406", + "issue:47256|issue:47406" + ], + "prepared_review_unit_hash": "1dbf57ba8b456c7a8de6e9596403b7c5c962de70a06a8711c7238eba1439d363", + "component_budget": { + "node_count": 21, + "item_count": 21, + "soft_pair_count": 68, + "serialized_chars": 36197, + "estimated_input_tokens": 9050, + "estimated_eval_tokens": 18356 + }, + "budget": { + "node_count": 21, + "item_count": 21, + "soft_pair_count": 68, + "serialized_chars": 36197, + "estimated_input_tokens": 9050, + "estimated_eval_tokens": 18356 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a repeated series of daily '[serge] integration failure triage' issues, but they appear to be separate day-by-day reports rather than duplicates of one underlying bug. If one issue has to represent the set, issue 47279 is the best hub because it has the strongest inbound-reference/activity signal.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 47279 is the most central report in the series by inbound references and discussion activity, so it is the best canonical anchor if one must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 47279 has the strongest linkage/attention in the cluster, but the rest look like distinct daily triage reports rather than exact duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46906", + "right": "issue:46940", + "accept": false, + "reason": "Same triage template, but different daily reports; no evidence they describe the same underlying failure." + }, + { + "left": "issue:46906", + "right": "issue:46948", + "accept": false, + "reason": "Different dates in the triage series; similar wording alone is not enough to merge as duplicates." + }, + { + "left": "issue:46906", + "right": "issue:46969", + "accept": false, + "reason": "These look like separate daily failure reports, not the same bug." + }, + { + "left": "issue:46906", + "right": "issue:46985", + "accept": false, + "reason": "Same broad series, but distinct day-specific triage issues." + }, + { + "left": "issue:46906", + "right": "issue:47037", + "accept": false, + "reason": "Different daily triage reports; no concrete shared bug/change evidence." + }, + { + "left": "issue:46940", + "right": "issue:46948", + "accept": false, + "reason": "Adjacent triage days, but still separate reports rather than one duplicate issue." + }, + { + "left": "issue:46940", + "right": "issue:46969", + "accept": false, + "reason": "No sign of the same underlying failure; only the same report template." + }, + { + "left": "issue:46940", + "right": "issue:46985", + "accept": false, + "reason": "Different daily triage entries, not the same issue." + }, + { + "left": "issue:46940", + "right": "issue:46998", + "accept": false, + "reason": "Similar title series only; these are distinct day-by-day reports." + }, + { + "left": "issue:46948", + "right": "issue:46969", + "accept": false, + "reason": "Separate daily triage issues with the same pattern, but not the same bug." + }, + { + "left": "issue:46948", + "right": "issue:46985", + "accept": false, + "reason": "No concrete evidence of one underlying problem shared by both reports." + }, + { + "left": "issue:46948", + "right": "issue:46998", + "accept": false, + "reason": "Different triage dates; similar wording is not enough to merge." + }, + { + "left": "issue:46969", + "right": "issue:46985", + "accept": false, + "reason": "Distinct day-specific integration-failure triage reports." + }, + { + "left": "issue:46969", + "right": "issue:47037", + "accept": false, + "reason": "Separate daily reports; no shared concrete failure description." + }, + { + "left": "issue:46969", + "right": "issue:47091", + "accept": false, + "reason": "Same series, but not the same underlying bug/change." + }, + { + "left": "issue:46985", + "right": "issue:46998", + "accept": false, + "reason": "These are consecutive triage reports, not duplicates." + }, + { + "left": "issue:46985", + "right": "issue:47037", + "accept": false, + "reason": "Different dated triage entries; no merge-worthy same bug evidence." + }, + { + "left": "issue:46985", + "right": "issue:47049", + "accept": false, + "reason": "Same template, different day and likely different failure instance." + }, + { + "left": "issue:46985", + "right": "issue:47091", + "accept": false, + "reason": "No indication these are the same concrete problem." + }, + { + "left": "issue:46985", + "right": "issue:47102", + "accept": false, + "reason": "Separate daily integration-failure reports." + }, + { + "left": "issue:46998", + "right": "issue:47037", + "accept": false, + "reason": "Different daily triage reports; similar title is not enough." + }, + { + "left": "issue:46998", + "right": "issue:47049", + "accept": false, + "reason": "They belong to the same triage series but are distinct reports." + }, + { + "left": "issue:46998", + "right": "issue:47053", + "accept": false, + "reason": "No shared concrete bug/change, just the same triage template." + }, + { + "left": "issue:46998", + "right": "issue:47091", + "accept": false, + "reason": "Separate day-by-day reports, not duplicates." + }, + { + "left": "issue:47037", + "right": "issue:47049", + "accept": false, + "reason": "Same broad series, but different daily failure reports." + }, + { + "left": "issue:47037", + "right": "issue:47053", + "accept": false, + "reason": "Different dates and no concrete same-code-path evidence." + }, + { + "left": "issue:47037", + "right": "issue:47091", + "accept": false, + "reason": "These are separate triage issues, not one underlying bug." + }, + { + "left": "issue:47037", + "right": "issue:47102", + "accept": false, + "reason": "Different daily reports in the same integration-failure sequence." + }, + { + "left": "issue:47049", + "right": "issue:47053", + "accept": false, + "reason": "Similar title series only; not the same underlying failure." + }, + { + "left": "issue:47049", + "right": "issue:47091", + "accept": false, + "reason": "Distinct triage days, no duplicate evidence." + }, + { + "left": "issue:47049", + "right": "issue:47102", + "accept": false, + "reason": "Separate reports with no concrete shared bug." + }, + { + "left": "issue:47053", + "right": "issue:47091", + "accept": false, + "reason": "Same report pattern, but different issue instances." + }, + { + "left": "issue:47053", + "right": "issue:47102", + "accept": false, + "reason": "No sign they are the same underlying failure." + }, + { + "left": "issue:47091", + "right": "issue:47102", + "accept": false, + "reason": "Different daily triage issues, not duplicates." + }, + { + "left": "issue:47102", + "right": "issue:47256", + "accept": false, + "reason": "Same series, but these are separate day-specific reports." + }, + { + "left": "issue:47120", + "right": "issue:47182", + "accept": false, + "reason": "Both are triage reports with similar shape, but no evidence of the same bug." + }, + { + "left": "issue:47120", + "right": "issue:47256", + "accept": false, + "reason": "Different daily failures; same template only." + }, + { + "left": "issue:47120", + "right": "issue:47265", + "accept": false, + "reason": "Separate reports in the series, not one duplicate issue." + }, + { + "left": "issue:47120", + "right": "issue:47279", + "accept": false, + "reason": "No concrete indication these describe the same underlying problem." + }, + { + "left": "issue:47120", + "right": "issue:47377", + "accept": false, + "reason": "Distinct daily triage entries, not the same change/bug." + }, + { + "left": "issue:47120", + "right": "issue:47406", + "accept": false, + "reason": "Open/latest triage issue, but still a separate report from the earlier ones." + }, + { + "left": "issue:47182", + "right": "issue:47227", + "accept": false, + "reason": "Adjacent triage dates, but not the same underlying failure." + }, + { + "left": "issue:47182", + "right": "issue:47256", + "accept": false, + "reason": "Similar report series only; distinct reports." + }, + { + "left": "issue:47182", + "right": "issue:47265", + "accept": false, + "reason": "No concrete shared bug/change evidence." + }, + { + "left": "issue:47182", + "right": "issue:47279", + "accept": false, + "reason": "Different day in the same triage sequence; not a duplicate." + }, + { + "left": "issue:47182", + "right": "issue:47309", + "accept": false, + "reason": "Separate daily integration-failure reports." + }, + { + "left": "issue:47182", + "right": "issue:47377", + "accept": false, + "reason": "Same broad triage topic, but not the same issue." + }, + { + "left": "issue:47227", + "right": "issue:47256", + "accept": false, + "reason": "Distinct daily reports; the titles are similar because they are part of the same series." + }, + { + "left": "issue:47227", + "right": "issue:47265", + "accept": false, + "reason": "No evidence of one shared underlying bug." + }, + { + "left": "issue:47227", + "right": "issue:47279", + "accept": false, + "reason": "Separate triage issues, not a duplicate pair." + }, + { + "left": "issue:47227", + "right": "issue:47309", + "accept": false, + "reason": "Different daily failure reports with the same triage template." + }, + { + "left": "issue:47227", + "right": "issue:47377", + "accept": false, + "reason": "Same series, but not the same concrete code-path problem." + }, + { + "left": "issue:47227", + "right": "issue:47390", + "accept": false, + "reason": "Distinct reports; no sign they should be merged." + }, + { + "left": "issue:47256", + "right": "issue:47265", + "accept": false, + "reason": "Adjacent daily triage reports, but still separate issues." + }, + { + "left": "issue:47256", + "right": "issue:47279", + "accept": false, + "reason": "No concrete evidence they are duplicates." + }, + { + "left": "issue:47256", + "right": "issue:47390", + "accept": false, + "reason": "Different day and likely different failure instance." + }, + { + "left": "issue:47256", + "right": "issue:47406", + "accept": false, + "reason": "Same triage series only; this is not the same underlying bug." + }, + { + "left": "issue:47265", + "right": "issue:47279", + "accept": false, + "reason": "Separate daily triage reports, not one duplicate issue." + }, + { + "left": "issue:47265", + "right": "issue:47309", + "accept": false, + "reason": "No shared concrete failure beyond the template." + }, + { + "left": "issue:47265", + "right": "issue:47377", + "accept": false, + "reason": "Different dated reports in the same series." + }, + { + "left": "issue:47279", + "right": "issue:47309", + "accept": false, + "reason": "Same generic triage pattern, but distinct reports." + }, + { + "left": "issue:47279", + "right": "issue:47377", + "accept": false, + "reason": "No evidence of the same underlying bug/change." + }, + { + "left": "issue:47279", + "right": "issue:47390", + "accept": false, + "reason": "Separate triage issues; titles are similar because the series is repetitive." + }, + { + "left": "issue:47279", + "right": "issue:47406", + "accept": false, + "reason": "Latest issue in the series, but still a distinct daily report rather than a duplicate." + }, + { + "left": "issue:47309", + "right": "issue:47377", + "accept": false, + "reason": "Different day-specific integration-failure reports." + }, + { + "left": "issue:47377", + "right": "issue:47390", + "accept": false, + "reason": "Same series, but not the same concrete problem." + }, + { + "left": "issue:47377", + "right": "issue:47406", + "accept": false, + "reason": "Separate triage reports; no duplicate evidence." + }, + { + "left": "issue:47390", + "right": "issue:47406", + "accept": false, + "reason": "Different dated reports, not one underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 25, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46983-2", + "nodes": [ + "issue:46983", + "issue:46984" + ], + "soft_pairs": [ + "issue:46983|issue:46984" + ], + "prepared_review_unit_hash": "b60aea8c07af99bce316c3f5bd6ebaa5d88eefa8162e2ba7039863c3c755f34b", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1492, + "estimated_input_tokens": 373, + "estimated_eval_tokens": 1002 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1492, + "estimated_input_tokens": 373, + "estimated_eval_tokens": 1002 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues share the exact same title, and the packet provides a strong soft-similarity signal between them. With no body text or explicit linkage exposed, they look closely related and possibly duplicate, but the evidence does not prove that conclusively.", + "confidence": 0.81, + "canonical_issue_reason": "Issue #46983 is a reasonable canonical choice because it has the exact same title as #46984 and there is no differentiating metadata in the packet; the duplicate call is plausible, but limited by the lack of body text.", + "canonical_pr_reason": null, + "best_issue_reason": "#46983 is a suitable representative for the issue cluster because it is an exact title match and there are no visible signals that distinguish it from #46984.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46983", + "right": "issue:46984", + "accept": true, + "reason": "Accept: the issues have the exact same title and a strong soft-similarity signal, which makes them look like the same feature request, though the missing body text means the match cannot be proven definitively." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The analysis overstates the evidence. The packet supports an exact title match and a soft similarity signal between the two issues, but it does not expose the body text, so claims like \u201chighly similar body content,\u201d \u201csame problem statement,\u201d or a definitive duplicate conclusion are too strong. A more conservative summary would note that the two closed issues share the same title and show a strong soft-similarity signal, suggesting they may be related or duplicate, but not prove it." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 26, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-47007-2", + "nodes": [ + "issue:47007", + "issue:47015" + ], + "soft_pairs": [ + "issue:47007|issue:47015" + ], + "prepared_review_unit_hash": "9576b9d27f30e50f96dcc6e52edcadc60c39742547f5a3a1b8788c347b549cbd", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1626, + "estimated_input_tokens": 407, + "estimated_eval_tokens": 1070 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1626, + "estimated_input_tokens": 407, + "estimated_eval_tokens": 1070 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both items are closed issue-style posts about AI-assisted vulnerability discovery, but they read like related discussion/proposal titles rather than the same concrete bug or change. The overlap is thematic, not a clear duplicate report.", + "confidence": 0.64, + "canonical_issue_reason": "Issue 47015 is the broader, more general framing of the topic, so it is the better representative if this cluster must have a canonical issue.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 47015 is the best single issue representative because its title is the most umbrella-like and least specific to one angle of the topic.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:47007", + "right": "issue:47015", + "accept": false, + "reason": "The titles overlap on AI-based vulnerability discovery, but they describe different angles ('lessons from using advanced LLMs' vs 'novel application of transformers') and do not identify the same underlying issue or concrete change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 27, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-47138-2", + "nodes": [ + "issue:47138", + "issue:47139" + ], + "soft_pairs": [ + "issue:47138|issue:47139" + ], + "prepared_review_unit_hash": "018885568b514bff565dc29daa5dc1a441a1779cdebe828334c412c260e1c6bd", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two open issues are effectively duplicates: they have identical titles, similar metadata, and the same reported routing bug involving chat management endpoints under /v1 when base_url includes /v1.", + "confidence": 0.99, + "canonical_issue_reason": "Issue #47138 is as good a canonical choice as #47139; with identical titles and no distinguishing content, either could serve as the primary tracker, but #47138 can be used as the representative by convention.", + "canonical_pr_reason": null, + "best_issue_reason": "#47138 is the best issue candidate because it is indistinguishable from #47139 on the available evidence and can serve as the single tracker for the duplicated report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:47138", + "right": "issue:47139", + "accept": true, + "reason": "Same title, same described bug, and no evidence of different scope; these are duplicate reports of the same underlying routing issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The conclusion is grounded in the packet: the two issues have identical titles, identical body length, no conflicting metadata, and the only pair evidence is soft_similarity with Jaccard 1.0. The duplicate/one-canonical-tracker reading is conservative enough here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 4, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40546-83", + "nodes": [ + "pull_request:43251", + "pull_request:43254", + "pull_request:44775", + "pull_request:45157", + "pull_request:46196", + "pull_request:46266", + "pull_request:46395", + "pull_request:46836", + "pull_request:46903", + "pull_request:46912", + "pull_request:46954", + "pull_request:46957", + "pull_request:46958", + "pull_request:46974", + "pull_request:46995", + "pull_request:47003", + "pull_request:47077", + "pull_request:47078", + "pull_request:47082", + "pull_request:47090", + "pull_request:47093", + "pull_request:47105", + "pull_request:47108", + "pull_request:47129", + "pull_request:47163", + "pull_request:47166", + "pull_request:47173", + "pull_request:47179", + "pull_request:47184", + "pull_request:47202", + "pull_request:47204", + "pull_request:47210", + "pull_request:47212", + "pull_request:47216", + "pull_request:47243", + "pull_request:47268", + "pull_request:47275", + "pull_request:47284", + "pull_request:47291", + "pull_request:47292", + "pull_request:47301", + "pull_request:47360", + "pull_request:47374", + "pull_request:47383", + "pull_request:47384", + "pull_request:47395", + "pull_request:47398", + "pull_request:47409" + ], + "soft_pairs": [ + "pull_request:47398|pull_request:47409", + "pull_request:43251|pull_request:43254", + "pull_request:46903|pull_request:47090", + "pull_request:47090|pull_request:47395", + "pull_request:46903|pull_request:46958", + "pull_request:46903|pull_request:47395", + "pull_request:47268|pull_request:47284", + "pull_request:46958|pull_request:46974", + "pull_request:46954|pull_request:46958", + "pull_request:47173|pull_request:47284", + "pull_request:47202|pull_request:47204", + "pull_request:47204|pull_request:47374", + "pull_request:47284|pull_request:47292", + "pull_request:47108|pull_request:47202", + "pull_request:47077|pull_request:47078", + "pull_request:47204|pull_request:47243", + "pull_request:47202|pull_request:47360", + "pull_request:47108|pull_request:47360", + "pull_request:47105|pull_request:47129", + "pull_request:47077|pull_request:47212", + "pull_request:46266|pull_request:46957", + "pull_request:47301|pull_request:47360", + "pull_request:47360|pull_request:47383", + "pull_request:47173|pull_request:47184", + "pull_request:47275|pull_request:47360", + "pull_request:47291|pull_request:47384", + "pull_request:47166|pull_request:47173", + "pull_request:47077|pull_request:47275", + "pull_request:44775|pull_request:47275", + "pull_request:47275|pull_request:47301", + "pull_request:47216|pull_request:47275", + "pull_request:46912|pull_request:47179", + "pull_request:46196|pull_request:47202", + "pull_request:47163|pull_request:47284", + "pull_request:46912|pull_request:47284", + "pull_request:46836|pull_request:47202", + "pull_request:46196|pull_request:47210", + "pull_request:47003|pull_request:47204", + "pull_request:47093|pull_request:47275", + "pull_request:45157|pull_request:47093", + "pull_request:46995|pull_request:47163", + "pull_request:45157|pull_request:46395", + "pull_request:47163|pull_request:47173", + "pull_request:46836|pull_request:47275", + "pull_request:46912|pull_request:47082" + ], + "prepared_review_unit_hash": "76844adcde1a0b8496e9fcd30bac122a6969f581cf488cc05ef3794bb6a2f3a4", + "component_budget": { + "node_count": 83, + "item_count": 83, + "soft_pair_count": 227, + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 45, + "serialized_chars": 124004, + "estimated_input_tokens": 31001, + "estimated_eval_tokens": 62258 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated PRs, but there are a few clear duplicate-like subgroups around the same underlying fixes: `fixed_cross_entropy` kwargs forwarding, and several training-loss double-shift fixes. Most soft edges should be rejected because they only share a broad subsystem or test/doc surface area.", + "confidence": 0.78, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single PR covers the whole cluster; among the duplicate-like candidates, 47090 is the strongest representative because it is issue-linked and fixes the same training-loss double-shift root cause across multiple affected models, not just one narrow case.", + "best_issue_reason": null, + "best_pr_reason": "47090 is the best global representative of the meaningful duplicate cluster: it addresses the shared loss-shift bug with the broadest concrete code coverage and is more complete than the narrower CohereASR-only variants.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47398", + "right": "pull_request:47409", + "accept": true, + "reason": "Same `num_experts` back-compat fallback bug in the GLM MoE config path; 47409 is a broader superset and could plausibly absorb 47398." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Same `fixed_cross_entropy` signature/forwarding change to pass supported kwargs into `nn.functional.cross_entropy`; these are effectively the same fix." + }, + { + "left": "pull_request:46903", + "right": "pull_request:47090", + "accept": true, + "reason": "Same training-loss double-shift bug class in the same model family; 47090 broadens the same fix across additional affected models and is compatible as one combined PR." + }, + { + "left": "pull_request:47090", + "right": "pull_request:47395", + "accept": false, + "reason": "Both involve loss shifting, but they are different concrete code paths and models (`cohere_asr`/`moonshine` vs `git`), so this is too broad to merge as duplicates." + }, + { + "left": "pull_request:46903", + "right": "pull_request:46958", + "accept": true, + "reason": "Same CohereASR double-shift training-loss bug; the two PRs use different implementation styles, but they target the same concrete failure mode." + }, + { + "left": "pull_request:46903", + "right": "pull_request:47395", + "accept": false, + "reason": "Shared theme only; `GitForCausalLM` has a separate model-specific loss path, so this is not the same underlying change." + }, + { + "left": "pull_request:47268", + "right": "pull_request:47284", + "accept": false, + "reason": "Unrelated test fixes in different model suites; no common bug or change." + }, + { + "left": "pull_request:46958", + "right": "pull_request:46974", + "accept": false, + "reason": "Different bugs in different files: CohereASR loss shifting vs Musicgen cache-length typo." + }, + { + "left": "pull_request:46954", + "right": "pull_request:46958", + "accept": false, + "reason": "One is a Pytest fp32-strictness cleanup, the other is a training-loss fix; they are not the same change." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47284", + "accept": false, + "reason": "Separate test-suite expectation updates for different models; not the same underlying issue." + }, + { + "left": "pull_request:47202", + "right": "pull_request:47204", + "accept": false, + "reason": "One removes a Docker workflow, the other adds exporter test dependencies to a Dockerfile; related to CI, but not the same fix." + }, + { + "left": "pull_request:47204", + "right": "pull_request:47374", + "accept": false, + "reason": "Dockerfile/runtime-test setup vs exporter documentation rewrite; these are unrelated changes." + }, + { + "left": "pull_request:47284", + "right": "pull_request:47292", + "accept": false, + "reason": "Different model test failures and different model codepaths; no duplicate bug here." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47202", + "accept": false, + "reason": "Workflow comment/pin cleanup is unrelated to the Docker image builder removal." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47078", + "accept": false, + "reason": "Both touch Japanese docs, but one is a translation PR and the other is wording cleanup; not the same artifact change." + }, + { + "left": "pull_request:47204", + "right": "pull_request:47243", + "accept": false, + "reason": "Exporter test-dependency setup is different from executorch exporter code fixes; same area, different bug/change." + }, + { + "left": "pull_request:47202", + "right": "pull_request:47360", + "accept": false, + "reason": "Docker image builder removal and OTEL CI pilot workflow are distinct CI changes." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47360", + "accept": false, + "reason": "Dependabot-compatible comment cleanup is unrelated to the OTEL workflow pilot." + }, + { + "left": "pull_request:47105", + "right": "pull_request:47129", + "accept": false, + "reason": "VITS test re-enable and RT-DETR converter key fixes are unrelated." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47212", + "accept": false, + "reason": "Japanese docs edits vs KV cache documentation updates; different docs issues." + }, + { + "left": "pull_request:46266", + "right": "pull_request:46957", + "accept": false, + "reason": "These are broad subsystem refactors, but they fix different concrete problems in different areas." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47360", + "accept": false, + "reason": "Docs for MTP support and an OTEL CI workflow change are unrelated." + }, + { + "left": "pull_request:47360", + "right": "pull_request:47383", + "accept": false, + "reason": "CI workflow pilot vs Kimi integration tests; no shared underlying bug or change." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47360", + "accept": false, + "reason": "Model addition PR vs CI workflow PR; unrelated." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47184", + "accept": false, + "reason": "Different test fixes for different models/devices; not duplicates." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47173", + "accept": false, + "reason": "Quantization test expansion vs XPU expectation updates; different changes." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47275", + "accept": false, + "reason": "Japanese doc translation vs new model code/tests; unrelated." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47275", + "accept": false, + "reason": "Docs parallelism navigation update and EfficientViT-SAM model addition are unrelated." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47301", + "accept": false, + "reason": "New model addition vs assisted-decoding documentation; no duplicate change." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47275", + "accept": false, + "reason": "Compressed-tensors quantization support vs a new model architecture; different concrete work." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47179", + "accept": false, + "reason": "RoPE test/model refactors and Granite-SWA model support are different changes." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47202", + "accept": false, + "reason": "SDPA dispatch support and Docker image builder removal are unrelated." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47284", + "accept": false, + "reason": "Paged-attention XPU test coverage vs Mimo V2 Flash fixes; different issues." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47284", + "accept": false, + "reason": "RoPE-related test/model work is unrelated to Mimo V2 Flash test fixes." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47202", + "accept": false, + "reason": "New Unlimited OCR model support is unrelated to Docker workflow cleanup." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47210", + "accept": false, + "reason": "Both touch attention internals, but one is model dispatch support and the other is packed SDPA varlen routing; different concrete fixes." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47204", + "accept": false, + "reason": "OpenVINO exporter addition is not the same change as adding ONNX/ExecuTorch deps to torch-light." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47275", + "accept": false, + "reason": "OpenPangu-V2 model support and EfficientViT-SAM model support are different model integrations." + }, + { + "left": "pull_request:45157", + "right": "pull_request:47093", + "accept": false, + "reason": "PrismML GGUF dequantization support and OpenPangu-V2 model addition are unrelated." + }, + { + "left": "pull_request:45157", + "right": "pull_request:46395", + "accept": false, + "reason": "GGUF dequantization support and Param2MoE model support are unrelated." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47173", + "accept": false, + "reason": "XPU paged-attention tests and XPU expectation updates in model tests are separate test changes." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47275", + "accept": false, + "reason": "Unlimited OCR model addition and EfficientViT-SAM model addition are unrelated model integrations." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47082", + "accept": false, + "reason": "RoPE test/model work and RoMa keypoint matcher model addition are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 31, + "review_unit_index": 2, + "review_unit_count": 4, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40546-83", + "nodes": [ + "pull_request:44517", + "pull_request:44775", + "pull_request:45157", + "pull_request:46196", + "pull_request:46266", + "pull_request:46267", + "pull_request:46509", + "pull_request:46604", + "pull_request:46658", + "pull_request:46836", + "pull_request:46895", + "pull_request:46912", + "pull_request:46957", + "pull_request:46995", + "pull_request:47003", + "pull_request:47077", + "pull_request:47078", + "pull_request:47082", + "pull_request:47093", + "pull_request:47105", + "pull_request:47108", + "pull_request:47113", + "pull_request:47143", + "pull_request:47163", + "pull_request:47166", + "pull_request:47175", + "pull_request:47179", + "pull_request:47184", + "pull_request:47204", + "pull_request:47210", + "pull_request:47216", + "pull_request:47243", + "pull_request:47275", + "pull_request:47292", + "pull_request:47294", + "pull_request:47295", + "pull_request:47298", + "pull_request:47301", + "pull_request:47304", + "pull_request:47306", + "pull_request:47320", + "pull_request:47322", + "pull_request:47336", + "pull_request:47352", + "pull_request:47360", + "pull_request:47383", + "pull_request:47384", + "pull_request:47398" + ], + "soft_pairs": [ + "pull_request:47204|pull_request:47295", + "pull_request:47163|pull_request:47384", + "pull_request:44775|pull_request:47301", + "pull_request:46196|pull_request:46995", + "pull_request:47301|pull_request:47384", + "pull_request:47216|pull_request:47384", + "pull_request:47077|pull_request:47301", + "pull_request:47336|pull_request:47384", + "pull_request:44775|pull_request:47077", + "pull_request:47294|pull_request:47320", + "pull_request:47166|pull_request:47292", + "pull_request:46267|pull_request:47383", + "pull_request:46509|pull_request:46995", + "pull_request:47093|pull_request:47143", + "pull_request:46995|pull_request:47143", + "pull_request:46196|pull_request:47179", + "pull_request:46266|pull_request:47301", + "pull_request:46266|pull_request:47383", + "pull_request:44775|pull_request:47304", + "pull_request:47082|pull_request:47210", + "pull_request:47093|pull_request:47163", + "pull_request:47179|pull_request:47210", + "pull_request:46604|pull_request:47003", + "pull_request:47082|pull_request:47108", + "pull_request:47078|pull_request:47275", + "pull_request:46266|pull_request:47216", + "pull_request:47108|pull_request:47301", + "pull_request:47082|pull_request:47275", + "pull_request:46995|pull_request:47301", + "pull_request:46266|pull_request:47336", + "pull_request:47175|pull_request:47179", + "pull_request:44775|pull_request:46836", + "pull_request:46604|pull_request:46658", + "pull_request:47163|pull_request:47179", + "pull_request:44517|pull_request:45157", + "pull_request:47082|pull_request:47216", + "pull_request:47275|pull_request:47336", + "pull_request:47210|pull_request:47304", + "pull_request:47108|pull_request:47383", + "pull_request:46957|pull_request:47216", + "pull_request:47216|pull_request:47336", + "pull_request:46912|pull_request:47078", + "pull_request:47179|pull_request:47301", + "pull_request:47166|pull_request:47352", + "pull_request:46912|pull_request:47294", + "pull_request:47184|pull_request:47352", + "pull_request:47143|pull_request:47320", + "pull_request:47175|pull_request:47294", + "pull_request:46895|pull_request:47306", + "pull_request:46196|pull_request:47301", + "pull_request:47105|pull_request:47298", + "pull_request:47275|pull_request:47304", + "pull_request:45157|pull_request:46658", + "pull_request:47078|pull_request:47322", + "pull_request:47078|pull_request:47243", + "pull_request:44517|pull_request:47301", + "pull_request:46895|pull_request:47398", + "pull_request:47294|pull_request:47360", + "pull_request:47113|pull_request:47304" + ], + "prepared_review_unit_hash": "942a9a9b224ac2407dea998023d0f0bfd27946cc1858392b892f17614977abd0", + "component_budget": { + "node_count": 83, + "item_count": 83, + "soft_pair_count": 227, + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 59, + "serialized_chars": 134391, + "estimated_input_tokens": 33598, + "estimated_eval_tokens": 67452 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a broad set of unrelated PRs that share only superficial keyword similarity (docs/model additions/tests/infrastructure), so none of the soft edges look like true duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No clear canonical PR: the cluster is heterogeneous and does not center on one underlying bug or change.", + "best_issue_reason": null, + "best_pr_reason": "PR 47352 is the strongest representative only because it is a substantive core-runtime change with an explicit issue target, but it is still not a duplicate hub for the cluster.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47204", + "right": "pull_request:47295", + "accept": false, + "reason": "Unrelated: exporter test-environment setup vs a DiffusionGemma mask bugfix." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47384", + "accept": false, + "reason": "Unrelated tests: XPU paged-attention coverage vs Gemma 4 heterogeneous-attention config changes." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47301", + "accept": false, + "reason": "Different docs topics: distributed-training navigation vs assisted-decoding/MTP docs." + }, + { + "left": "pull_request:46196", + "right": "pull_request:46995", + "accept": false, + "reason": "Different code paths: SDPA dispatch for wav2vec2 models vs flash-linear-attention availability check." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47384", + "accept": false, + "reason": "Docs vs model config: MTP documentation change vs Gemma 4 per-layer config validation." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47384", + "accept": false, + "reason": "Quantization docs/code vs model architecture config; no shared underlying bug." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47301", + "accept": false, + "reason": "Translation/docs update for DistilBERT vs docs for multi-token prediction." + }, + { + "left": "pull_request:47336", + "right": "pull_request:47384", + "accept": false, + "reason": "Offloaded-save bugfix vs Gemma 4 config refactor; unrelated problems." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47077", + "accept": false, + "reason": "Both are documentation edits, but for different content and languages; not the same change." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47320", + "accept": false, + "reason": "Kernel integration work vs chat response-schema deprecation; unrelated subsystems." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47292", + "accept": false, + "reason": "Different test suites for different quantization/model areas; not a shared bug." + }, + { + "left": "pull_request:46267", + "right": "pull_request:47383", + "accept": false, + "reason": "Serve/chat parsing refactor vs Kimi integration tests; unrelated." + }, + { + "left": "pull_request:46509", + "right": "pull_request:46995", + "accept": false, + "reason": "New Nemotron Omni model support vs flash-linear-attention availability; unrelated." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47143", + "accept": false, + "reason": "OpenPangu-V2 model addition vs NeuCodec model addition; different models and code paths." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47143", + "accept": false, + "reason": "Utility/backend availability change vs new model support; no overlap in bug or fix." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47179", + "accept": false, + "reason": "Audio SDPA dispatch change vs Granite-SWA model support; unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:47301", + "accept": false, + "reason": "DINO/DINOv3 conversion/model modernization vs MTP docs; unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:47383", + "accept": false, + "reason": "Model conversion/architecture work vs Kimi integration tests; not the same underlying issue." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47304", + "accept": false, + "reason": "Different docs maintenance changes; one is distributed-training nav, the other is MPS graph-cache docs cleanup." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47210", + "accept": false, + "reason": "New RoMa model support vs SDPA packed-sequence optimization; unrelated." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47163", + "accept": false, + "reason": "OpenPangu-V2 model support vs XPU paged-attention tests; unrelated." + }, + { + "left": "pull_request:47179", + "right": "pull_request:47210", + "accept": false, + "reason": "Model support vs attention-kernel optimization; different concrete changes." + }, + { + "left": "pull_request:46604", + "right": "pull_request:47003", + "accept": false, + "reason": "Mistral tekken tokenizer utilities vs OpenVINO exporter implementation; different feature areas." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47108", + "accept": false, + "reason": "RoMa model support vs a docs-build comment tweak; unrelated." + }, + { + "left": "pull_request:47078", + "right": "pull_request:47275", + "accept": false, + "reason": "Japanese docs wording updates vs new EfficientViTSam model support; unrelated." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47301", + "accept": false, + "reason": "Keypoint matcher model support vs MTP documentation; unrelated." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47301", + "accept": false, + "reason": "MLU support for flash-linear-attention availability vs MTP docs; unrelated." + }, + { + "left": "pull_request:47179", + "right": "pull_request:47301", + "accept": false, + "reason": "Granite-SWA model support vs MTP docs; unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:47336", + "accept": false, + "reason": "DINO modernization/conversion changes vs fully-offloaded save bugfix; unrelated." + }, + { + "left": "pull_request:47175", + "right": "pull_request:47179", + "accept": false, + "reason": "Olmo3 CI expectation updates vs Granite-SWA model support; unrelated." + }, + { + "left": "pull_request:44775", + "right": "pull_request:46836", + "accept": false, + "reason": "Docs navigation edits vs Unlimited OCR model support; unrelated." + }, + { + "left": "pull_request:46604", + "right": "pull_request:46658", + "accept": false, + "reason": "Mistral tokenizer utilities vs Step-3.7 Flash model/scripts; unrelated." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47179", + "accept": false, + "reason": "Paged-attention XPU test coverage vs Granite-SWA model addition; unrelated." + }, + { + "left": "pull_request:44517", + "right": "pull_request:45157", + "accept": false, + "reason": "Qwen3 TTS support vs Prism GGUF dequantization/import fixes; unrelated." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47216", + "accept": false, + "reason": "RoMa model addition vs compressed-tensors fp8 quantization work; unrelated." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47336", + "accept": false, + "reason": "EfficientViTSam model support vs offloaded-save bugfix; unrelated." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47304", + "accept": false, + "reason": "Packed SDPA runtime optimization vs docs cleanup for MPS graph cache; unrelated." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47383", + "accept": false, + "reason": "Doc-builder pin comment tweak vs Kimi integration tests; unrelated." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47216", + "accept": false, + "reason": "Finegrained FP8 decode optimization vs compressed-tensors quantization/docs; not the same issue." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47336", + "accept": false, + "reason": "Quantization format support vs fully-offloaded save bugfix; unrelated." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47078", + "accept": false, + "reason": "RoPE test expansion vs Japanese docs wording; unrelated." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47352", + "accept": false, + "reason": "Quantization test updates vs distributed runtime/mixin refactor; unrelated." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47301", + "accept": false, + "reason": "SDPA backend dispatch change vs MTP docs; unrelated." + }, + { + "left": "pull_request:47105", + "right": "pull_request:47298", + "accept": false, + "reason": "VITS test re-enable vs MI450 runtime/model-failure fixes; different scopes." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47304", + "accept": false, + "reason": "Model support vs docs cleanup; unrelated." + }, + { + "left": "pull_request:45157", + "right": "pull_request:46658", + "accept": false, + "reason": "Prism GGUF dequantization/import fixes vs Step-3.7 Flash model build/conversion scripts; unrelated." + }, + { + "left": "pull_request:47078", + "right": "pull_request:47322", + "accept": false, + "reason": "Japanese docs wording vs AMD Quark quantization config changes; unrelated." + }, + { + "left": "pull_request:47078", + "right": "pull_request:47243", + "accept": false, + "reason": "Japanese docs wording vs Executorch exporter fixes; unrelated." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47301", + "accept": false, + "reason": "Qwen3 TTS model support vs assisted-decoding/MTP docs; unrelated." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47398", + "accept": false, + "reason": "CohereASR training-loss bugfix vs GLM MoE config fallback fix; unrelated." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47360", + "accept": false, + "reason": "Kernel function integration work vs CI workflow instrumentation change; unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47304", + "accept": false, + "reason": "Audio feature/proc requirement fixes vs docs cleanup; unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 31, + "review_unit_index": 3, + "review_unit_count": 4, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40546-83", + "nodes": [ + "pull_request:41855", + "pull_request:42668", + "pull_request:43251", + "pull_request:43424", + "pull_request:43451", + "pull_request:43823", + "pull_request:43995", + "pull_request:44517", + "pull_request:44827", + "pull_request:45170", + "pull_request:45213", + "pull_request:45655", + "pull_request:46048", + "pull_request:46381", + "pull_request:46509", + "pull_request:46604", + "pull_request:46658", + "pull_request:46836", + "pull_request:46895", + "pull_request:46912", + "pull_request:46957", + "pull_request:46958", + "pull_request:47003", + "pull_request:47077", + "pull_request:47090", + "pull_request:47096", + "pull_request:47113", + "pull_request:47143", + "pull_request:47163", + "pull_request:47166", + "pull_request:47173", + "pull_request:47175", + "pull_request:47184", + "pull_request:47210", + "pull_request:47243", + "pull_request:47275", + "pull_request:47291", + "pull_request:47292", + "pull_request:47294", + "pull_request:47301", + "pull_request:47304", + "pull_request:47306", + "pull_request:47322", + "pull_request:47352", + "pull_request:47360", + "pull_request:47367", + "pull_request:47383", + "pull_request:47399" + ], + "soft_pairs": [ + "pull_request:46912|pull_request:47291", + "pull_request:47163|pull_request:47352", + "pull_request:46895|pull_request:47399", + "pull_request:43451|pull_request:44517", + "pull_request:47352|pull_request:47383", + "pull_request:44517|pull_request:47383", + "pull_request:47243|pull_request:47294", + "pull_request:44517|pull_request:47143", + "pull_request:43451|pull_request:47352", + "pull_request:47210|pull_request:47291", + "pull_request:46957|pull_request:47383", + "pull_request:46604|pull_request:47163", + "pull_request:46604|pull_request:47383", + "pull_request:46957|pull_request:47360", + "pull_request:47113|pull_request:47175", + "pull_request:46381|pull_request:47322", + "pull_request:43451|pull_request:46604", + "pull_request:46957|pull_request:47301", + "pull_request:47113|pull_request:47275", + "pull_request:47113|pull_request:47243", + "pull_request:46912|pull_request:47113", + "pull_request:47077|pull_request:47367", + "pull_request:47294|pull_request:47306", + "pull_request:47275|pull_request:47367", + "pull_request:47210|pull_request:47367", + "pull_request:43451|pull_request:46381", + "pull_request:46381|pull_request:47096", + "pull_request:41855|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:42668|pull_request:43823", + "pull_request:43424|pull_request:47301", + "pull_request:43424|pull_request:47383", + "pull_request:43823|pull_request:45655", + "pull_request:44827|pull_request:47383", + "pull_request:44827|pull_request:45170", + "pull_request:43424|pull_request:44827", + "pull_request:43823|pull_request:43995", + "pull_request:43251|pull_request:43823", + "pull_request:44827|pull_request:46048", + "pull_request:43424|pull_request:43823", + "pull_request:43251|pull_request:44827", + "pull_request:44827|pull_request:45655", + "pull_request:43995|pull_request:45213", + "pull_request:44827|pull_request:45213", + "pull_request:43995|pull_request:44827", + "pull_request:43823|pull_request:45213", + "pull_request:46895|pull_request:47090", + "pull_request:46958|pull_request:47090", + "pull_request:46895|pull_request:46958", + "pull_request:47173|pull_request:47292", + "pull_request:47166|pull_request:47184", + "pull_request:46912|pull_request:47175", + "pull_request:47096|pull_request:47383", + "pull_request:47184|pull_request:47292", + "pull_request:47301|pull_request:47304", + "pull_request:47077|pull_request:47304", + "pull_request:46658|pull_request:46836", + "pull_request:47210|pull_request:47301", + "pull_request:47003|pull_request:47243", + "pull_request:46509|pull_request:47143", + "pull_request:47163|pull_request:47184", + "pull_request:47163|pull_request:47166", + "pull_request:46509|pull_request:47163" + ], + "prepared_review_unit_hash": "54a016e2814bf3b7d6b0458ffd8cc667e6d5c990d0e8932c582d27fdfa9deafb", + "component_budget": { + "node_count": 83, + "item_count": 83, + "soft_pair_count": 227, + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 63, + "serialized_chars": 132890, + "estimated_input_tokens": 33223, + "estimated_eval_tokens": 66702 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most links are false positives across unrelated model-add, docs, test, and exporter PRs. The only real duplicate cluster is CohereASR training-loss double-shift fixes: 46895, 46958, and 47090 all target the same bug, with 47090 being the best representative.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47090 is the best canonical choice: it fixes the same CohereASR double-shift loss bug as 46895/46958, has the explicit issue target, and uses the clearest final approach (`shift_labels` to preserve loss reduction behavior).", + "best_issue_reason": null, + "best_pr_reason": "PR 47090 is the strongest overall representative of the actual duplicate cluster because it is the latest, most complete fix for the same concrete training-loss bug and subsumes the earlier variants.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46912", + "right": "pull_request:47291", + "accept": false, + "reason": "RoPE test-value updates vs generation-config deprecation; different code paths and problems." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47352", + "accept": false, + "reason": "Paged-attention XPU test coverage vs distributed runtime utilities; unrelated changes." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47399", + "accept": false, + "reason": "CohereASR loss bug fix vs Cosmos 3 Edge patch packing order; different models and bugs." + }, + { + "left": "pull_request:43451", + "right": "pull_request:44517", + "accept": false, + "reason": "Separate new-model additions (Molmo2 vs Qwen3 TTS); not the same model or change." + }, + { + "left": "pull_request:47352", + "right": "pull_request:47383", + "accept": false, + "reason": "Distributed runtime utilities vs Kimi integration tests; only broad framework similarity." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47383", + "accept": false, + "reason": "Qwen3 TTS model add vs Kimi integration tests; unrelated artifacts." + }, + { + "left": "pull_request:47243", + "right": "pull_request:47294", + "accept": false, + "reason": "ExecuTorch exporter fixes vs kernel-native causal_conv1d support; different concrete functionality." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47143", + "accept": false, + "reason": "Different model additions (Qwen3 TTS vs NeuCodec), not the same underlying change." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47352", + "accept": false, + "reason": "Molmo2 model addition and distributed runtime utilities are unrelated." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47291", + "accept": false, + "reason": "Packed SDPA varlen-attention routing vs continuous-batching config deprecation; different bugs." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47383", + "accept": false, + "reason": "Finegrained FP8 decode optimization vs Kimi integration tests; not the same fix." + }, + { + "left": "pull_request:46604", + "right": "pull_request:47163", + "accept": false, + "reason": "Mistral tekken tokenizer utilities vs paged-attention XPU tests; unrelated." + }, + { + "left": "pull_request:46604", + "right": "pull_request:47383", + "accept": false, + "reason": "Mistral tekken utilities vs Kimi integration tests; different code and intent." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47360", + "accept": false, + "reason": "FP8 grouped/batched kernel optimization vs CI OTEL workflow change; unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47175", + "accept": false, + "reason": "Audio feature/proc requirements fix vs Olmo3 CI expectation refresh; different issues." + }, + { + "left": "pull_request:46381", + "right": "pull_request:47322", + "accept": false, + "reason": "Tokenizer version bump vs AMD Quark config-class rename; unrelated maintenance." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46604", + "accept": false, + "reason": "Molmo2 model add vs Mistral tekken tokenizer utilities; not the same artifact." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47301", + "accept": false, + "reason": "FP8 decode optimization vs MTP documentation; different subsystems." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47275", + "accept": false, + "reason": "Audio requirements fix vs EfficientViTSam model addition; unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47243", + "accept": false, + "reason": "Audio requirements fix vs ExecuTorch exporter fixes; different bugs." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47175", + "accept": false, + "reason": "RoPE-related tests vs Olmo3 CI expectation updates; different test concerns." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47367", + "accept": false, + "reason": "Japanese DistilBERT docs translation vs pipeline device-placement clarification; unrelated docs." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47306", + "accept": false, + "reason": "Kernel-native causal_conv1d functions vs Neuron experiment scripts; different features." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47367", + "accept": false, + "reason": "EfficientViTSam model add vs pipeline device-placement docs/code; unrelated." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47367", + "accept": false, + "reason": "Packed SDPA varlen attention vs pipeline device-placement clarification; different changes." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46381", + "accept": false, + "reason": "Molmo2 model add vs tokenizer dependency bump; not the same work." + }, + { + "left": "pull_request:46381", + "right": "pull_request:47096", + "accept": false, + "reason": "Tokenizer dependency changes vs ExecuTorch/Kimi exporter work; unrelated." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Mistral tokenizer missing methods vs ExecuTorch dynamic-shape tests; different artifacts." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Mistral tokenizer methods vs MobileLLM model addition; unrelated." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor auto-loading robustness vs MobileLLM model addition; different changes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:47301", + "accept": false, + "reason": "ExecuTorch dynamic-shape test vs MTP docs; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:47383", + "accept": false, + "reason": "Mistral4 test fixes vs Kimi integration tests; different models and bugs." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 tests vs CLIP-like `layrnorm` rename mapping; unrelated." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "ExecuTorch export tests vs Mistral4 tests; only broad testing similarity." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "MobileLLM model addition vs model-creation skill draft; not the same change." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs fix vs MobileLLM model addition; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 test fixes vs TDT loss kernel addition; different code paths." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon output-interface refactor vs model-creation skill draft; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 tests vs model-creation skill draft; not the same bug or feature." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-interface refactor vs Mistral4 tests; different models/problems." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "MobileLLM model add vs model-creation skill draft; unrelated artifacts." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs fix vs MobileLLM model add; different concern." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 tests vs model-creation skill draft; not a duplicate." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor vs model-creation skill draft; unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "MobileLLM model add vs model-creation skill draft; different scope." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47090", + "accept": true, + "reason": "Same CohereASR training-loss double-shift bug; same files and same issue target, with 47090 refining the fix." + }, + { + "left": "pull_request:46958", + "right": "pull_request:47090", + "accept": true, + "reason": "Same CohereASR loss bug in the same code path; 47090 is the more complete final variant." + }, + { + "left": "pull_request:46895", + "right": "pull_request:46958", + "accept": true, + "reason": "Both address the same CohereASR double-shift training-loss issue in the same files." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47292", + "accept": false, + "reason": "XPU expectation updates in different test suites for different models; not the same bug." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47184", + "accept": false, + "reason": "Quantization test tweaks vs Qwen3 Next accelerator test rename; unrelated." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47175", + "accept": false, + "reason": "RoPE tests vs Olmo3 CI numeric adjustments; different test goals." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47383", + "accept": false, + "reason": "ExecuTorch exporter fixes vs Kimi integration tests; same model family but different concrete changes." + }, + { + "left": "pull_request:47184", + "right": "pull_request:47292", + "accept": false, + "reason": "Qwen3 Next accelerator test rename vs TipsV2-DPT XPU expectation updates; unrelated." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47304", + "accept": false, + "reason": "MTP support docs vs MPS graph-cache docs; different documentation topics." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47304", + "accept": false, + "reason": "Japanese DistilBERT docs translation vs MPS graph-cache docs; unrelated." + }, + { + "left": "pull_request:46658", + "right": "pull_request:46836", + "accept": false, + "reason": "Step3.7 Flash model scaffolding vs Unlimited OCR model addition; different models." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47301", + "accept": false, + "reason": "Packed SDPA varlen-attention code vs MTP docs; unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47243", + "accept": false, + "reason": "OpenVINO exporter work vs ExecuTorch exporter fixes; different exporters and code paths." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47143", + "accept": false, + "reason": "Nemotron Omni model add vs NeuCodec model add; different models." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47184", + "accept": false, + "reason": "Paged-attention XPU tests vs Qwen3 Next accelerator tests; different test suites." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47166", + "accept": false, + "reason": "Paged-attention XPU tests vs quantization test updates; unrelated." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47163", + "accept": false, + "reason": "Nemotron Omni model add vs paged-attention XPU tests; unrelated." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47292", + "accept": false, + "reason": "Quantization test updates vs TipsV2-DPT XPU failures; different models and fixes." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47292", + "accept": false, + "reason": "Generic XPU expectation fixes vs TipsV2-DPT-specific test values; not the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 31, + "review_unit_index": 4, + "review_unit_count": 4, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40546-83", + "nodes": [ + "pull_request:40546", + "pull_request:43451", + "pull_request:45157", + "pull_request:46196", + "pull_request:46509", + "pull_request:46658", + "pull_request:46836", + "pull_request:46912", + "pull_request:47003", + "pull_request:47096", + "pull_request:47143", + "pull_request:47163", + "pull_request:47166", + "pull_request:47175", + "pull_request:47184", + "pull_request:47210", + "pull_request:47216", + "pull_request:47243", + "pull_request:47291", + "pull_request:47292", + "pull_request:47294", + "pull_request:47298", + "pull_request:47301", + "pull_request:47304", + "pull_request:47306", + "pull_request:47322", + "pull_request:47336", + "pull_request:47367", + "pull_request:47383", + "pull_request:47387", + "pull_request:47398", + "pull_request:47399" + ], + "soft_pairs": [ + "pull_request:45157|pull_request:46836", + "pull_request:46836|pull_request:47387", + "pull_request:47163|pull_request:47292", + "pull_request:47291|pull_request:47322", + "pull_request:45157|pull_request:46509", + "pull_request:47301|pull_request:47336", + "pull_request:47216|pull_request:47383", + "pull_request:47336|pull_request:47383", + "pull_request:46509|pull_request:47387", + "pull_request:47304|pull_request:47367", + "pull_request:46912|pull_request:47298", + "pull_request:47301|pull_request:47383", + "pull_request:47216|pull_request:47301", + "pull_request:46509|pull_request:47301", + "pull_request:47175|pull_request:47298", + "pull_request:47163|pull_request:47210", + "pull_request:47294|pull_request:47298", + "pull_request:45157|pull_request:47387", + "pull_request:46658|pull_request:47387", + "pull_request:46196|pull_request:47143", + "pull_request:47143|pull_request:47163", + "pull_request:46658|pull_request:47143", + "pull_request:46658|pull_request:47322", + "pull_request:47143|pull_request:47387", + "pull_request:47143|pull_request:47322", + "pull_request:45157|pull_request:47143", + "pull_request:47301|pull_request:47367", + "pull_request:43451|pull_request:47243", + "pull_request:47096|pull_request:47243", + "pull_request:43451|pull_request:47291", + "pull_request:47166|pull_request:47383", + "pull_request:47184|pull_request:47383", + "pull_request:43451|pull_request:47175", + "pull_request:40546|pull_request:43451", + "pull_request:47306|pull_request:47367", + "pull_request:47298|pull_request:47399", + "pull_request:47398|pull_request:47399", + "pull_request:43451|pull_request:47143", + "pull_request:43451|pull_request:47163", + "pull_request:43451|pull_request:47383", + "pull_request:43451|pull_request:46509", + "pull_request:47298|pull_request:47398", + "pull_request:46658|pull_request:47399", + "pull_request:43451|pull_request:47096", + "pull_request:43451|pull_request:47003", + "pull_request:47003|pull_request:47096", + "pull_request:43451|pull_request:47322", + "pull_request:47096|pull_request:47322", + "pull_request:43451|pull_request:47298", + "pull_request:43451|pull_request:47398", + "pull_request:43451|pull_request:47399", + "pull_request:47096|pull_request:47398", + "pull_request:47096|pull_request:47399", + "pull_request:43451|pull_request:47306", + "pull_request:47096|pull_request:47306", + "pull_request:47003|pull_request:47398", + "pull_request:47003|pull_request:47306", + "pull_request:47096|pull_request:47387", + "pull_request:47003|pull_request:47387", + "pull_request:47096|pull_request:47143" + ], + "prepared_review_unit_hash": "c5ba8e54f93ea214a77f939e52e2e6935f6b09324ddb1bdecdbb17dcba723aaa", + "component_budget": { + "node_count": 83, + "item_count": 83, + "soft_pair_count": 227, + "serialized_chars": 283383, + "estimated_input_tokens": 70846, + "estimated_eval_tokens": 141948 + }, + "budget": { + "node_count": 32, + "item_count": 32, + "soft_pair_count": 60, + "serialized_chars": 92588, + "estimated_input_tokens": 23147, + "estimated_eval_tokens": 46550 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These soft links are mostly false positives: the items span unrelated PRs for model additions, docs, tests, exporters, quantization, and bug fixes. I found no pair that clearly shares the same concrete underlying change or bug, so no duplicate cluster emerges.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:45157", + "right": "pull_request:46836", + "accept": false, + "reason": "Different model-support PRs for unrelated architectures (PrismML Bonsai quantization vs Unlimited OCR)." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47387", + "accept": false, + "reason": "Unlimited OCR model support is unrelated to one-to-many conversion support in core loading." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47292", + "accept": false, + "reason": "Both mention XPU, but one adds paged-attention test coverage and the other fixes a specific model test; different issues." + }, + { + "left": "pull_request:47291", + "right": "pull_request:47322", + "accept": false, + "reason": "Different config changes for different subsystems (gen deprecation vs AMD Quark config class rename)." + }, + { + "left": "pull_request:45157", + "right": "pull_request:46509", + "accept": false, + "reason": "Unrelated model-support additions for PrismML Bonsai and Nemotron Omni." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47336", + "accept": false, + "reason": "Docs for MTP support and offloaded-model saving bugfix are unrelated changes." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47383", + "accept": false, + "reason": "Compressed-tensors FP8 support and Kimi integration tests do not target the same problem." + }, + { + "left": "pull_request:47336", + "right": "pull_request:47383", + "accept": false, + "reason": "Offloaded-model saving bugfix and Kimi integration tests are unrelated." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47387", + "accept": false, + "reason": "Nemotron Omni support is unrelated to one-to-many conversion support." + }, + { + "left": "pull_request:47304", + "right": "pull_request:47367", + "accept": false, + "reason": "Both are docs-oriented, but they address different topics: MPS graph cache vs pipeline device placement." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47298", + "accept": false, + "reason": "RoPE test expansion and MI450 image/model failure fixes are different issues." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47383", + "accept": false, + "reason": "MTP docs and Kimi integration tests are unrelated." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47301", + "accept": false, + "reason": "Compressed-tensors FP8 support is unrelated to MTP docs." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47301", + "accept": false, + "reason": "Nemotron Omni support and MTP docs do not share the same underlying change." + }, + { + "left": "pull_request:47175", + "right": "pull_request:47298", + "accept": false, + "reason": "Olmo3 CI test updates and MI450 failure fixes are not the same bug." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47210", + "accept": false, + "reason": "Paged-attention XPU tests and packed-SDPA varlen routing are different test/code-path changes." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47298", + "accept": false, + "reason": "Kernel-native causal_conv1d support is unrelated to MI450 image/model fixes." + }, + { + "left": "pull_request:45157", + "right": "pull_request:47387", + "accept": false, + "reason": "PrismML Bonsai support and one-to-many conversion support are unrelated." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47387", + "accept": false, + "reason": "Step-3.7 Flash model work and one-to-many conversion support are unrelated." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47143", + "accept": false, + "reason": "SDPA dispatch support across audio models and NeuCodec model addition are different changes." + }, + { + "left": "pull_request:47143", + "right": "pull_request:47163", + "accept": false, + "reason": "NeuCodec model support and paged-attention XPU tests are unrelated." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47143", + "accept": false, + "reason": "Step-3.7 Flash model support and NeuCodec support are different model additions." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47322", + "accept": false, + "reason": "Step-3.7 Flash work is unrelated to AMD Quark config changes." + }, + { + "left": "pull_request:47143", + "right": "pull_request:47387", + "accept": false, + "reason": "NeuCodec support and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47143", + "right": "pull_request:47322", + "accept": false, + "reason": "NeuCodec support and Quark config changes are unrelated." + }, + { + "left": "pull_request:45157", + "right": "pull_request:47143", + "accept": false, + "reason": "PrismML Bonsai support and NeuCodec support are unrelated model additions." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47367", + "accept": false, + "reason": "MTP docs and pipeline device-placement docs are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47243", + "accept": false, + "reason": "Molmo2 model addition and Executorch exporter fixes are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47243", + "accept": false, + "reason": "Kimi exportability work and Executorch exporter fixes target different concrete code paths." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47291", + "accept": false, + "reason": "Molmo2 support and continuous-batching config deprecation are unrelated." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47383", + "accept": false, + "reason": "Quantization test updates and Kimi integration tests are unrelated." + }, + { + "left": "pull_request:47184", + "right": "pull_request:47383", + "accept": false, + "reason": "Qwen3 Next multi-accelerator test changes and Kimi integration tests are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47175", + "accept": false, + "reason": "Molmo2 model addition and Olmo3 CI test updates are unrelated." + }, + { + "left": "pull_request:40546", + "right": "pull_request:43451", + "accept": false, + "reason": "VibeVoice and Molmo2 are separate model-addition PRs with different code paths." + }, + { + "left": "pull_request:47306", + "right": "pull_request:47367", + "accept": false, + "reason": "Neuron training experiments and pipeline device-placement docs are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47298", + "accept": false, + "reason": "Molmo2 support and MI450 image/model fixes are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47398", + "accept": false, + "reason": "Molmo2 support and GLM MoE DSA config fallback are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47398", + "accept": false, + "reason": "Kimi exporter work and GLM MoE config fallback are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47399", + "accept": false, + "reason": "Kimi exporter work and Cosmos 3 Edge patch packing order fixes are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47306", + "accept": false, + "reason": "Molmo2 support and Neuron experiments are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47387", + "accept": false, + "reason": "Kimi exportability work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47387", + "accept": false, + "reason": "OpenVINO exporter work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47143", + "accept": false, + "reason": "Kimi exporter work and NeuCodec support are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47398", + "accept": false, + "reason": "OpenVINO exporter work and GLM MoE config fallback are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47306", + "accept": false, + "reason": "OpenVINO exporter work and Neuron experiments are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47387", + "accept": false, + "reason": "OpenVINO exporter work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47387", + "accept": false, + "reason": "Kimi exportability work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47387", + "accept": false, + "reason": "OpenVINO exporter work and one-to-many conversions are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47143", + "accept": false, + "reason": "Kimi exporter work and NeuCodec support are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47398", + "accept": false, + "reason": "OpenVINO exporter work and GLM MoE config fallback are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47306", + "accept": false, + "reason": "OpenVINO exporter work and Neuron experiments are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 2, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40755-2", + "nodes": [ + "pull_request:40755", + "pull_request:41224" + ], + "soft_pairs": [ + "pull_request:40755|pull_request:41224" + ], + "prepared_review_unit_hash": "0aaa3b2aa82331cfed11462fa4a70f7c07630bc65c4285cf1aa2c85354d4e381", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two pull requests are unrelated: one adds TimesFM covariate forecasting, the other adds DINOv3 image classification support. They touch different model families, different code paths, and do not represent the same underlying change.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different models and features: TimesFM forecasting with covariates versus DINOv3 image classification. No shared bug or mergeable change scope." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 3, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-41356-2", + "nodes": [ + "pull_request:41356", + "pull_request:44339" + ], + "soft_pairs": [ + "pull_request:41356|pull_request:44339" + ], + "prepared_review_unit_hash": "aead1308c167b0bae6ff11bf9358340baa5e76f816618383418b505a10a9cd01", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6050, + "estimated_input_tokens": 1513, + "estimated_eval_tokens": 3282 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6050, + "estimated_input_tokens": 1513, + "estimated_eval_tokens": 3282 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two DEIMv2 pull requests target the same model family, but they are not the same change: PR 41356 is an early/basic model + image processor add, while PR 44339 is a much broader follow-up that adds loss code, auto-mappings, conversion tooling, modular files, and more integration work.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44339 is the more complete and representative DEIMv2 implementation; PR 41356 looks like an earlier partial version of the same model bring-up.", + "best_issue_reason": null, + "best_pr_reason": "PR 44339 is the strongest single artifact for the cluster because it subsumes the basic model work and includes the broader integration needed for a full Transformers addition.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41356", + "right": "pull_request:44339", + "accept": false, + "reason": "Same model area, but not the same concrete change: 44339 expands the implementation substantially and adds new code paths beyond the basic PR 41356. They look like related iterations, not mergeable duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 4, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42975-2", + "nodes": [ + "pull_request:42975", + "pull_request:42976" + ], + "soft_pairs": [ + "pull_request:42975|pull_request:42976" + ], + "prepared_review_unit_hash": "27758471f7b85818ab4e27aa6a8c1c3e2f6737b666fa79c1e1bee65076332e04", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5933, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5933, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two unrelated workflow-maintenance pull requests: one upgrades several GitHub Actions for Node 24 compatibility, the other bumps Docker build actions to newer versions. They overlap in some workflow files but do not fix the same underlying problem or represent a mergeable single change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "pull_request:42975 is the more coherent representative: it has a single explicit goal (Node 24 compatibility) and a consistent set of action-version upgrades across workflows.", + "best_issue_reason": null, + "best_pr_reason": "pull_request:42975 is the best overall representative because its intent is clearer and broader, while 42976 is a narrower dependency bump focused on Docker build actions.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Different maintenance updates: 42975 upgrades checkout/cache/upload-artifact for Node 24 compatibility; 42976 upgrades docker/build-push-action v5->v6. Shared workflow files are incidental, not the same concrete change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 5, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42978-4", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:45490", + "pull_request:46562" + ], + "soft_pairs": [ + "pull_request:42978|pull_request:43448", + "pull_request:42978|pull_request:45490", + "pull_request:43448|pull_request:45490", + "pull_request:45490|pull_request:46562" + ], + "prepared_review_unit_hash": "3bf7fb48b380a2a02621681d15d297bc6f84abec588215ae7f3d407f6ddb5f91", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 13047, + "estimated_input_tokens": 3262, + "estimated_eval_tokens": 6780 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 13047, + "estimated_input_tokens": 3262, + "estimated_eval_tokens": 6780 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are four separate model-addition PRs that only overlap on shared infrastructure files like docs/auto mappings; they do not look like the same underlying change, so none should be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: each PR introduces a different model (ViT NEPA, Molmo, CTSM, Step3p7) with distinct code paths and tests, despite some shared registry/docs edits.", + "best_issue_reason": null, + "best_pr_reason": "No single PR is a good duplicate representative for this cluster because the overlapping files are generic framework plumbing rather than evidence of one shared fix/change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo); shared auto/docs files are generic registration work, not the same concrete change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "These add unrelated models (ViT NEPA vs CTSM) and touch different model implementations; overlap is only broad model registry/docs plumbing." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are distinct model integrations with different code paths; shared files do not imply a single duplicate PR." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46562", + "accept": false, + "reason": "CTSM and Step3p7 are different model additions; despite heavier overlap in auto-mapping files, they are not the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 6, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43291-8", + "nodes": [ + "pull_request:43291", + "pull_request:43665", + "pull_request:44815", + "pull_request:46380", + "pull_request:46539", + "pull_request:46886", + "pull_request:46922", + "pull_request:47393" + ], + "soft_pairs": [ + "pull_request:43291|pull_request:43665", + "pull_request:43291|pull_request:44815", + "pull_request:43291|pull_request:46380", + "pull_request:43291|pull_request:46922", + "pull_request:43665|pull_request:46539", + "pull_request:43665|pull_request:47393", + "pull_request:43665|pull_request:44815", + "pull_request:43665|pull_request:46380", + "pull_request:43665|pull_request:46922", + "pull_request:44815|pull_request:46539", + "pull_request:44815|pull_request:46380", + "pull_request:44815|pull_request:46922", + "pull_request:46380|pull_request:46539", + "pull_request:46380|pull_request:46922", + "pull_request:46539|pull_request:46922", + "pull_request:46886|pull_request:46922" + ], + "prepared_review_unit_hash": "602097d29f649eee4712ce34ce9ef23fc1bfe3bd34b778d44d308797e8529f27", + "component_budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 16, + "serialized_chars": 21115, + "estimated_input_tokens": 5279, + "estimated_eval_tokens": 10814 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 16, + "serialized_chars": 21115, + "estimated_input_tokens": 5279, + "estimated_eval_tokens": 10814 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Most items are unrelated PRs bundled by soft similarity. The only strong duplicate-like pair is the YOLOS device-mismatch fixes (46886 and 46922), which target the same issue and code path; the rest should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46922 is the best representative because it addresses the concrete YOLOS forward/device-mismatch fix directly in model code and also includes the split-module safeguard; it is the more complete version of the same change as 46886.", + "best_issue_reason": null, + "best_pr_reason": "PR 46922 is the strongest standalone PR in this set for triage purposes: it has an explicit issue target, touches the real production code path, and is narrowly about one bug fix rather than test churn or unrelated maintenance.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both are test-oriented PRs, but they affect different models and do not fix the same underlying bug or code path." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper tokenizer/test fix and FP8 dequant logic are unrelated subsystems with different failure modes." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46380", + "accept": false, + "reason": "Whisper decoding tests and generation-watermarking validation are separate issues." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46922", + "accept": false, + "reason": "Whisper tests and YOLOS device-mismatch fixes are unrelated." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46539", + "accept": false, + "reason": "These are both test-heavy PRs, but they target different models and different behaviors; no shared bug is evident." + }, + { + "left": "pull_request:43665", + "right": "pull_request:47393", + "accept": false, + "reason": "CLIP/ViT test edits and a README URL change are unrelated." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "Model test failures in CLIP/ViT are not the same as the FP8 dequant fix." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46380", + "accept": false, + "reason": "Different components and different bug classes; test changes do not indicate the same underlying fix." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46922", + "accept": false, + "reason": "CLIP/ViT test updates are unrelated to YOLOS runtime device placement." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46539", + "accept": false, + "reason": "FP8 quantization/dequantization logic and CI Phimoe integration-test infrastructure are different concerns." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46380", + "accept": false, + "reason": "The FP8 kernel/model-loading fix and watermarking/generation validation are unrelated code paths." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46922", + "accept": false, + "reason": "These PRs fix different model/runtime issues; no same bug or mergeable combined change is shown." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46539", + "accept": false, + "reason": "Generation watermarking validation and Phimoe CI/device-memory setup are unrelated." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46922", + "accept": false, + "reason": "A generation-config validation fix is not the same bug as YOLOS device mismatch." + }, + { + "left": "pull_request:46539", + "right": "pull_request:46922", + "accept": false, + "reason": "Phimoe CI memory/device-map handling and YOLOS model device placement are different issues." + }, + { + "left": "pull_request:46886", + "right": "pull_request:46922", + "accept": true, + "reason": "Both PRs address the same YOLOS device-mismatch problem under device_map=\"auto\" in the same file, and 46922 plausibly subsumes 46886's fix." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. The duplicate-like relationship between 46886 and 46922 is well supported: they share the same issue target and file, and 46922 clearly subsumes 46886 by adding the runtime device-placement fix plus the split-module safeguard. The other pairwise rejections are appropriately cautious given only soft similarity evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 7, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43488-3", + "nodes": [ + "pull_request:43488", + "pull_request:43532", + "pull_request:46747" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:46747", + "pull_request:43488|pull_request:43532", + "pull_request:43532|pull_request:46747" + ], + "prepared_review_unit_hash": "d0cd80cd8783ddd95b6a182b6a787f047a9458af93948071d32182c68cd30a42", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 6349, + "estimated_input_tokens": 1588, + "estimated_eval_tokens": 3432 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 6349, + "estimated_input_tokens": 1588, + "estimated_eval_tokens": 3432 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three items are synthetic '[don't merge]' pull requests, but they target different behaviors: version/setup bot checks, workflow diff/show tooling, and a security-gate workflow trigger change. The title similarity is superficial; there is no clear duplicate PR cluster here.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "If one PR must be treated as the cluster representative, 46747 is the most self-contained and concrete change: a single workflow file with a clear trigger shift.", + "best_issue_reason": null, + "best_pr_reason": "46747 is the best standalone PR candidate because it has the narrowest scope and the clearest code-path change, while the others are broader or mix multiple unrelated workflow/test changes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:46747", + "accept": false, + "reason": "Different purposes and files: 43488 is a setup/version-format and test bot probe, while 46747 changes a PR security-gate workflow trigger. Not the same underlying change." + }, + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "43488 checks repository formatting/versioning behavior; 43532 rewires several CI/notification workflows and testing helpers. Shared test-file overlap is not enough to call them the same bug/change." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46747", + "accept": false, + "reason": "Both touch workflows, but they alter different CI paths for different goals: 43532 is a show-diff/commenting workflow change, while 46747 changes the security-gate trigger semantics. Not mergeable as one fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43612-2", + "nodes": [ + "pull_request:43612", + "pull_request:43613" + ], + "soft_pairs": [ + "pull_request:43612|pull_request:43613" + ], + "prepared_review_unit_hash": "e3de25ce051f50682a9e3eedabb4cd41cec242fe47e84f4499f00d0864cb70f9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related segmentation-feature PRs, but they implement different pipelines with different model integrations and task docs, so they should not be merged as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: 43612 adds Promptable Concept Segmentation, while 43613 adds Promptable Visual Segmentation; they target different features and code paths.", + "best_issue_reason": null, + "best_pr_reason": "43613 is the better representative if one must be chosen because it is the broader and more fully integrated feature addition, but it is still distinct from 43612.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Reject: these are separate pipeline additions for different tasks, with different filenames, model hookups, and tests. They share scaffolding but not the same underlying change or fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 9, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43651-2", + "nodes": [ + "pull_request:43651", + "pull_request:43663" + ], + "soft_pairs": [ + "pull_request:43651|pull_request:43663" + ], + "prepared_review_unit_hash": "35fb69ad37cc0d5e4879915b38652f88706472648425f607270415895289615d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4293, + "estimated_input_tokens": 1074, + "estimated_eval_tokens": 2404 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4293, + "estimated_input_tokens": 1074, + "estimated_eval_tokens": 2404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate trainer customizability changes in the same file: one adds a hook for gradient-accumulation loss scaling, the other adds a hook for signature-column filtering. They are related at a high level but do not fix the same bug or change the same code path.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43651 is the better canonical representative because it introduces the more concrete behavioral change in the training step and clearly defines a new loss-scaling hook. PR #43663 is a separate refactor for input-column filtering.", + "best_issue_reason": null, + "best_pr_reason": "PR #43651 is the strongest standalone candidate: it targets a specific training-step bug/behavior around gradient accumulation and has a clear, self-contained code-path change. PR #43663 is also valid but addresses a different trainer extensibility point.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Both touch trainer.py and add customization hooks, but one is about loss scaling during gradient accumulation while the other is about signature-column filtering. They are different behaviors and would not plausibly be merged as one PR fixing the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 10, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43656-2", + "nodes": [ + "pull_request:43656", + "pull_request:43842" + ], + "soft_pairs": [ + "pull_request:43656|pull_request:43842" + ], + "prepared_review_unit_hash": "19e95ebbd9da51e24862e164239093779e7e185e4f493ca817b129b130ead3ef", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3663, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3663, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs both address a TypeAdapter-related NameError in `src/transformers/cli/serve.py`, with one broader typing cleanup and one narrower pydantic-absence fix; they look closely related enough to treat as the same bug cluster.", + "confidence": 0.74, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43842 is the more direct and narrowly scoped fix for the reported runtime NameError when pydantic is unavailable, with explicit `TypeAdapter` handling under `TYPE_CHECKING`.", + "best_issue_reason": null, + "best_pr_reason": "PR #43842 is the best representative because it most clearly targets the concrete failure mode and has the cleaner, more specific fix description.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": true, + "reason": "Both PRs target the same TypeAdapter NameError in the CLI serve code path and their changes overlap enough that they could plausibly be merged into one fix." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded in the packet: both PRs target a TypeAdapter-related NameError in `src/transformers/cli/serve.py`, and the cluster summary stays appropriately cautious about their relationship. The soft-edge verdict is conservative enough because it only claims the changes could plausibly be merged into one fix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 11, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43779-2", + "nodes": [ + "pull_request:43779", + "pull_request:43816" + ], + "soft_pairs": [ + "pull_request:43779|pull_request:43816" + ], + "prepared_review_unit_hash": "e7c674adcc031aa8883109a5dfc678b4407fb20b10764302092e384d031ba83a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4362, + "estimated_input_tokens": 1091, + "estimated_eval_tokens": 2438 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4362, + "estimated_input_tokens": 1091, + "estimated_eval_tokens": 2438 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests appear to implement the same SwanLab integration change: adding run id/resume support in `integration_utils.py` for issue #43698, with one variant also expanding the docs. They are strong duplicate PR candidates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43816 is the cleaner representative: it carries the same SwanLab id/resume support plus the accompanying documentation updates, making it the more complete version of the shared change.", + "best_issue_reason": null, + "best_pr_reason": "PR #43816 is the best global representative because it appears to include the full feature change and docs for SwanLab run id/resume support, while #43779 is the narrower implementation-focused variant.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43779", + "right": "pull_request:43816", + "accept": true, + "reason": "Same underlying SwanLab callback enhancement for id/resume support in the same file and same issue target; the doc differences do not change the core code-path change, so these are duplicate PRs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both PRs target issue #43698, touch the same file, and the previews indicate the same SwanLab id/resume support with #43816 adding docs. The duplicate/mergeable verdict is conservative enough given the shared target and overlapping code change." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43996", + "pull_request:43997", + "pull_request:44007", + "pull_request:44010", + "pull_request:44013", + "pull_request:44017", + "pull_request:44018", + "pull_request:44019", + "pull_request:44024", + "pull_request:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44066|pull_request:44071", + "pull_request:44018|pull_request:44068", + "pull_request:44066|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44018|pull_request:44116", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44085", + "pull_request:44018|pull_request:44086", + "pull_request:44068|pull_request:44086", + "pull_request:44071|pull_request:44085", + "pull_request:44085|pull_request:44086", + "pull_request:44059|pull_request:44086", + "pull_request:44098|pull_request:44101", + "pull_request:44059|pull_request:44722", + "pull_request:44018|pull_request:44071", + "pull_request:44017|pull_request:44161", + "pull_request:44017|pull_request:44073", + "pull_request:44017|pull_request:44722", + "pull_request:44068|pull_request:44071", + "pull_request:43997|pull_request:44073", + "pull_request:44017|pull_request:44071", + "pull_request:44071|pull_request:44086", + "pull_request:44073|pull_request:44101", + "pull_request:44073|pull_request:44161", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44018|pull_request:44059", + "pull_request:44059|pull_request:44066", + "pull_request:43997|pull_request:44017", + "pull_request:44101|pull_request:44161", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request:44056|pull_request:44161", + "pull_request:43997|pull_request:44018", + "pull_request:43997|pull_request:44161", + "pull_request:44074|pull_request:44161", + "pull_request:44018|pull_request:44161", + "pull_request:44010|pull_request:44071", + "pull_request:44018|pull_request:44129", + "pull_request:44010|pull_request:44073", + "pull_request:43997|pull_request:44074", + "pull_request:44010|pull_request:44018", + "pull_request:43997|pull_request:44129", + "pull_request:44071|pull_request:44129", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154", + "pull_request:44024|pull_request:44076" + ], + "prepared_review_unit_hash": "8db6007d91526401d4a8fdac61f61195bfd7c00ca0be8203f4170ec5c36a0927", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 + }, + "budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 72, + "serialized_chars": 76941, + "estimated_input_tokens": 19236, + "estimated_eval_tokens": 38728 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a batch of model-specific output-tracing refactors, not one duplicate set. Only a few same-family GPT-Neo/GPT-J PRs look like true rerolls of the same change; the rest touch different model code paths and should stay separate.", + "confidence": 0.61, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44722 is the strongest representative of the true duplicate subgroup: it is the latest and most complete GPT-J/CodeGen output-tracing refactor, with the same concrete file-level change as the earlier GPT-J PRs.", + "best_issue_reason": null, + "best_pr_reason": "PR 44722 is the best global PR candidate because it most clearly captures the shared GPT-J/CodeGen output-tracing refactor and is broader/more complete than the earlier GPT-J variants.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families and code paths; same tracing theme/issue tag is not enough." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "CVT/FNet vs DeBERTa v2 are unrelated implementations; not the same underlying change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor family and overlapping file-level change; this looks like a reroll/duplicate." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "ResNet/RT-DETR ResNet vs EfficientNet are different model paths; not mergeable as one fix." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen output tracing vs EfficientNet output tracing are different model-specific changes." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa v2 are unrelated; same broad initiative does not make them duplicates." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen and MGP-STR are different models and concrete fixes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen vs MPT are separate model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both are GPT-Neo output-tracing refactors on the same file family; one is effectively a rerun of the same change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen vs GPT-Neo are different model-specific refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo and GPT-J are related but distinct code paths; not the same concrete fix." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet output tracing and GPT-J/CodeGen output tracing are unrelated patches." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are different model implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet vs GPT-J are different fixes, even if both mention output tracing." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are different model families and code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet and MPT are unrelated model-specific changes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are different models with different output-tracing code." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are not the same underlying patch." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated model refactors." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are related but distinct model-specific refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are different concrete changes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated model implementations." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44085", + "accept": false, + "reason": "MPT and GPT-J are separate model code paths." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are different architectures and fixes." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 and MGP-STR are unrelated concrete changes." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and XLM are different model families; not a duplicate fix." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are separate output-tracing refactors." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are unrelated code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "SegFormer and VisualBERT are different model-specific changes." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer and GPT-J/CodeGen are unrelated fixes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44073", + "accept": false, + "reason": "RegNet and VisualBERT are unrelated implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44071", + "accept": false, + "reason": "SegFormer and MPT are different concrete changes." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR are different model code paths." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44101", + "accept": false, + "reason": "VisualBERT and XLM are distinct model families." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and XLM are different concrete output-tracing refactors." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are unrelated model changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT-2 are different model families and code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are not the same underlying fix." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are different model-specific refactors." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "XLM and LongT5 are unrelated code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are different model implementations." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44129", + "accept": false, + "reason": "MPT and SpeechT5 are unrelated model-specific changes." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet and XLM are different output-tracing refactors." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 are unrelated concrete fixes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet and GPT-Neo are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 are unrelated patches." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet and LongT5 are different model-specific changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44161", + "accept": false, + "reason": "GPT-Neo and LongT5 are unrelated output-tracing refactors." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBERT and MPT are unrelated model code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-Neo and SpeechT5 are different concrete fixes." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBERT and VisualBERT are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44074", + "accept": false, + "reason": "RegNet and TextNet are unrelated model implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44018", + "accept": false, + "reason": "SqueezeBERT and GPT-Neo are different concrete changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44129", + "accept": false, + "reason": "RegNet and SpeechT5 are unrelated model-specific refactors." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44129", + "accept": false, + "reason": "MPT and SpeechT5 are different model code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44161", + "accept": false, + "reason": "CVT/FNet and LongT5 are unrelated changes." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", + "accept": false, + "reason": "CVT/FNet and SegFormer are different model families." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBERT and DeBERTa v2 are unrelated refactors." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa v2 and TextNet are different code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing change with overlapping files; 44722 is the later, broader reroll." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both are GPT-J output-tracing refactors on the same file family, and 44722 expands the same change." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are unrelated model-specific patches." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "CVT/FNet and ResNet/RT-DETR ResNet are different model families." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are unrelated code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa v2 and SpeechT5 are different concrete fixes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa v2 and ViTS are unrelated model refactors." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet standardized outputs and ImageGPT hook capture are different model-specific changes." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: most PRs are separate model-specific output-tracing refactors, and the GPT-Neo/GPT-J/CodeGen subgroup is the only clearly duplicate-looking cluster. The conservative false verdicts are supported by distinct model families/code paths, and the true verdicts are limited to closely overlapping same-family changes with shared files or the same issue target. Best representative PR 44722 is also a reasonable choice because it is the later, broader GPT-J/CodeGen reroll." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 31, + "review_unit_index": 2, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43997", + "pull_request:44010", + "pull_request:44056", + "pull_request:44073", + "pull_request:44074", + "pull_request:44129" + ], + "soft_pairs": [ + "pull_request:44073|pull_request:44074", + "pull_request:44056|pull_request:44073", + "pull_request:44056|pull_request:44074", + "pull_request:43997|pull_request:44010", + "pull_request:44010|pull_request:44129" + ], + "prepared_review_unit_hash": "36d27060a1eb0fff7a5bfb737a60e51fcd7b19a67031c701f557fc7d31b1bcd1", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 5, + "serialized_chars": 13069, + "estimated_input_tokens": 3268, + "estimated_eval_tokens": 6792 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are all model-specific PRs to migrate output tracing/capture to a standardized decorator pattern. They are similar in theme, but they do not fix the same concrete code path, so they should not be merged as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44129 is the strongest representative: it has the clearest explicit issue link, a substantial tracing refactor, and the diff shows a concrete output-capture migration in a specific model path.", + "best_issue_reason": null, + "best_pr_reason": "PR 44129 is the best single PR for the cluster because it is the most explicit and complete example of the shared refactor pattern; the others are narrower, model-specific variants of the same migration theme.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Both are output-tracing refactors, but they touch different model code paths (VisualBert vs TextNet) and are not the same underlying change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "These are separate model migrations (MPNet vs VisualBert). Same refactor theme, but not the same bug or change and not plausibly one PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "MPNet and TextNet are unrelated model implementations; the similar decorator migration does not make them duplicates." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are different model paths with different output-handling code; this is a shared refactor pattern, not one concrete fix." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBert and SpeechT5 change different forward/output paths, so they should remain separate PRs despite similar standardized tracing goals." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43998-13", + "nodes": [ + "pull_request:43998", + "pull_request:43999", + "pull_request:44000", + "pull_request:44001", + "pull_request:44002", + "pull_request:44003", + "pull_request:44004", + "pull_request:44025", + "pull_request:44026", + "pull_request:44027", + "pull_request:44028", + "pull_request:44029", + "pull_request:44030" + ], + "soft_pairs": [ + "pull_request:43998|pull_request:44002", + "pull_request:43998|pull_request:44000", + "pull_request:43998|pull_request:44001", + "pull_request:43998|pull_request:44003", + "pull_request:43998|pull_request:44029", + "pull_request:43999|pull_request:44001", + "pull_request:43999|pull_request:44002", + "pull_request:43999|pull_request:44025", + "pull_request:43999|pull_request:44026", + "pull_request:43999|pull_request:44027", + "pull_request:44000|pull_request:44001", + "pull_request:44000|pull_request:44002", + "pull_request:44000|pull_request:44003", + "pull_request:44000|pull_request:44025", + "pull_request:44001|pull_request:44002", + "pull_request:44001|pull_request:44003", + "pull_request:44001|pull_request:44025", + "pull_request:44002|pull_request:44029", + "pull_request:44002|pull_request:44003", + "pull_request:44003|pull_request:44004", + "pull_request:43998|pull_request:44004", + "pull_request:44001|pull_request:44004", + "pull_request:44002|pull_request:44004", + "pull_request:44000|pull_request:44004", + "pull_request:44025|pull_request:44026", + "pull_request:44025|pull_request:44027", + "pull_request:44025|pull_request:44028", + "pull_request:44025|pull_request:44029", + "pull_request:44025|pull_request:44030", + "pull_request:44026|pull_request:44029", + "pull_request:44026|pull_request:44027", + "pull_request:44026|pull_request:44028", + "pull_request:44026|pull_request:44030", + "pull_request:44027|pull_request:44028", + "pull_request:44027|pull_request:44029", + "pull_request:44027|pull_request:44030", + "pull_request:44028|pull_request:44029", + "pull_request:44028|pull_request:44030", + "pull_request:44029|pull_request:44030" + ], + "prepared_review_unit_hash": "0df79e68744390b1ddfa4e868b8b44ac32662d66d6842d683bc2fdd8ff9415dc", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 39, + "serialized_chars": 39789, + "estimated_input_tokens": 9948, + "estimated_eval_tokens": 20152 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 39, + "serialized_chars": 39789, + "estimated_input_tokens": 9948, + "estimated_eval_tokens": 20152 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs share a common output-tracing/refactor theme, but they target different model implementations and code paths, so they are not true duplicates of one another.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "Pick PR 44025 as the representative anchor: it applies the shared output-tracing refactor to two closely related model variants and is one of the clearest, self-contained examples in the set.", + "best_issue_reason": null, + "best_pr_reason": "PR 44025 is the best cluster representative because it is broad enough to reflect the refactor pattern without being a catch-all, and it touches a closely related pair of model files.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Same refactor theme, but different model families and forward paths; not the same concrete change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Both are output-tracing refactors, but they affect unrelated model implementations and APIs." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Different models and code paths; similarity is only at the boilerplate refactor level." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Timm backbone vs. Mamba/Falcon Mamba are unrelated fixes, despite shared refactor wording." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures and output plumbing; not mergeable as one concrete bug fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "Different model code paths; shared output-tracing mechanics are too broad to treat as duplicate." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "MobileNetV1 and UperNet are separate implementations with distinct forward behavior." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "Unrelated model families; same refactor pattern does not imply the same underlying change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Different model architectures and files; only the tracing refactor motif is shared." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "Separate encoder-decoder stack versus MobileNetV1; not the same concrete code-path problem." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Vision-text dual encoder and UnivNet have unrelated forward paths; too broad to dedupe." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Same pattern, different model subsystems; no shared bug or identical change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "These are distinct model implementations with different output contracts." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families; the shared refactor theme is not enough for a duplicate match." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "UnivNet and UperNet are unrelated code paths; this is only a similar refactor style." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "No shared concrete bug or change; just a common tracing refactor pattern." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families and output behaviors; not a plausible single merged PR." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are unrelated implementations; same refactor theme is insufficient." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "These touch different subsystems and do not fix the same underlying issue." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Mamba/Falcon Mamba output tracing and CodeGen cache/output handling are different concrete changes." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Different models and different forward semantics; only superficially similar refactor language." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet and CodeGen are separate code paths with distinct outputs and kwargs handling." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are unrelated; not the same bug or fix." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model code paths and output signatures; too broad to merge as duplicates." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Depth Anything family versus vision-encoder-decoder are different implementations, not one shared fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "Different model families and code paths; shared refactor wording is not enough." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Depth Anything and SuperPoint are unrelated forward implementations." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures and output handling; not the same underlying bug/change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "Depth Anything and DPR are separate model changes with no concrete overlap." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Vision-encoder-decoder and RWKV do not share the same code path; only the refactor theme matches." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Encoder-decoder vision and speech wrappers are distinct implementations, not duplicates." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families and output APIs; no single merged PR would make sense." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "Separate encoder-decoder wrapper versus DPR encoder; not the same concrete fix." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "Speech encoder-decoder and SuperPoint are unrelated code paths." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and output contracts; no duplicate relationship." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "Speech encoder-decoder and DPR are separate changes with no shared bug." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV are unrelated implementations; similarity is only in the refactor pattern." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model subsystems and output behavior; not mergeable as one change." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR touch different code paths and fix different output-tracing details." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 14, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44662-2", + "nodes": [ + "pull_request:44662", + "pull_request:45401" + ], + "soft_pairs": [ + "pull_request:44662|pull_request:45401" + ], + "prepared_review_unit_hash": "dbe1daf59cd115c68f572826b93fde0591ee457d6abd6ffb8d87cdfac269dd79", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6551, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6551, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate model-addition PRs for different architectures (PenguinVL vs Voxtral TTS). They share common Transformers scaffolding files, but the underlying codepaths, model types, and goals are distinct, so they are not duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45401 is the better canonical PR only in the sense that it has an explicit tracking issue and a narrower, well-scoped target (Voxtral TTS support). PR 44662 is a different model implementation entirely.", + "best_issue_reason": null, + "best_pr_reason": "PR 45401 is the strongest representative of its own change set because it is tied to a specific issue and focused on one model family. PR 44662 is not a better cluster representative since it targets a different model with different files and behavior.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Reject: they add support for different models (PenguinVL vs Voxtral TTS). Shared auto-registration/docs files are generic boilerplate for new model integrations and do not indicate the same underlying bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 15, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44979-2", + "nodes": [ + "pull_request:44979", + "pull_request:45363" + ], + "soft_pairs": [ + "pull_request:44979|pull_request:45363" + ], + "prepared_review_unit_hash": "e6ee7e54dbfa9103a6b7855fc6d2441a68cdeb2503c74b0cb62abaac983a4f78", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs overlap on fusion-related infrastructure, but they target different changes: #44979 introduces a generic Module Fusion API, while #45363 implements a specific KernelConfig-based n-to-1 kernel fusion path plus integration updates. They are related work, not duplicates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: #44979 is a general module-fusion API scaffold, while #45363 is a separate concrete kernel-fusion feature with different files and scope.", + "best_issue_reason": null, + "best_pr_reason": "#45363 is the better representative PR because it is the more complete, user-facing implementation of a concrete fusion mechanism rather than just the foundational API layer.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Reject: they are related but not the same underlying change. #44979 adds a generic Module Fusion API, whereas #45363 implements KernelConfig-based n-to-1 fusion and integration plumbing. They would not plausibly be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 16, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45086-2", + "nodes": [ + "pull_request:45086", + "pull_request:45317" + ], + "soft_pairs": [ + "pull_request:45086|pull_request:45317" + ], + "prepared_review_unit_hash": "3a132a0c3c394290e317652dcc6e0170be3740bd078458f019cf884689857920", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs fix the same AttributeError in `_patch_mistral_regex` by switching `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer` in the same file; 45317 is the more complete version because it adds tests.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same code fix as 45086 plus test coverage, making it the more complete change set.", + "best_issue_reason": null, + "best_pr_reason": "PR 45317 is the best representative: same concrete fix as 45086, same target issue, and it includes validation tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying bug and same fix in the same code path (`_patch_mistral_regex`); 45317 simply extends 45086 with tests, so they are duplicates/overlapping changes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 17, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45115-2", + "nodes": [ + "pull_request:45115", + "pull_request:45569" + ], + "soft_pairs": [ + "pull_request:45115|pull_request:45569" + ], + "prepared_review_unit_hash": "060ac854cebefb825878af0ae7b05b81a1c7539c46dd06a2674d52211bc0f0f7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6066, + "estimated_input_tokens": 1517, + "estimated_eval_tokens": 3290 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6066, + "estimated_input_tokens": 1517, + "estimated_eval_tokens": 3290 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related NemotronH-area pull requests, but they are not duplicates: one is a focused inheritance/conversion refactor for NemotronH, while the other splits the architecture into dedicated dense/sparse model variants and updates docs/tests.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45569 is the broader, more representative NemotronH change set because it introduces the new dense/sparse model structure and associated public-facing docs/model wiring.", + "best_issue_reason": null, + "best_pr_reason": "PR 45569 is the best overall candidate: it contains the substantive model API/architecture work, not just internal conversion-mapping refactors.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "They share NemotronH core files, but the underlying changes differ: 45115 is about conversion-mapping and inheritance refactoring, while 45569 adds separate NemotronHDense/NemotronHSparse implementations plus docs/tests. They are related, not the same mergeable fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 18, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45415-2", + "nodes": [ + "pull_request:45415", + "pull_request:45425" + ], + "soft_pairs": [ + "pull_request:45415|pull_request:45425" + ], + "prepared_review_unit_hash": "4fd001073c0d5241f02cd0724950eaedebb2d91c2c28d4becbc588ca39e828e9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are related by typing work in transformers, but they target different changes: one is a broad type-checking sweep across many files, while the other adds shared typing protocols/aliases and fixes specific typing integration points. They are not the same underlying change.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45425 is the more focused, self-contained typing-infrastructure change; #45415 is a broader type-checking sweep and does not subsume it.", + "best_issue_reason": null, + "best_pr_reason": "PR #45425 is the better representative of the cluster because it has a clearer, narrower purpose and concrete typing API additions, but it is still not a duplicate of #45415.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Different change sets: #45415 broadly enables type checking across many source files and adjusts pyproject ignores, while #45425 adds typing aliases/Protocols in _typing.py and related helpers. They do not appear to fix the same concrete code-path problem." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 19, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45548-2", + "nodes": [ + "pull_request:45548", + "pull_request:45662" + ], + "soft_pairs": [ + "pull_request:45548|pull_request:45662" + ], + "prepared_review_unit_hash": "6aed6f583f811074649c481b405a82a20c6c2882ab2cfd1fb800da52fced3ad0", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4507, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4507, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are about expert-parallel loading, but they fix different failure modes: one addresses EP + ZeRO-3/accelerate config handling, the other fixes EP + FSDP2 expert tensors being overwritten during shard loading. They are related, not duplicates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "No clear duplicate canonical PR: the two PRs target different concrete code-path bugs, despite overlapping in EP/loading infrastructure.", + "best_issue_reason": null, + "best_pr_reason": "PR 45662 is the stronger representative of the cluster because it fixes a concrete, reproducible overwrite bug in the loading path and spans the broader EP/FSDP2 remediation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Different underlying bugs: 45548 focuses on EP + DeepSpeed ZeRO-3/accelerate config serialization, while 45662 fixes experts being overwritten during distributed shard loading for FSDP2. Shared EP context is not enough to merge them as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 20, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45639-2", + "nodes": [ + "pull_request:45639", + "pull_request:45645" + ], + "soft_pairs": [ + "pull_request:45639|pull_request:45645" + ], + "prepared_review_unit_hash": "24fb3b1b490be06bc0f1b3b500da0a59116f6e25e3b7a0a61a4d59978e69cb63", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4672, + "estimated_input_tokens": 1168, + "estimated_eval_tokens": 2592 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4672, + "estimated_input_tokens": 1168, + "estimated_eval_tokens": 2592 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests address the same xdist collision around captured_info output, with 45645 extending the code fix in 45639 and adding the CI-side log collection needed for the new per-worker files.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45645 is the broader, more complete fix: it includes the xdist-safe testing_utils change plus the workflow update to collect all captured_info files, while 45639 only covers the library-side logging change.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the best representative because it solves the underlying xdist debug-log collision end to end, not just the file-writing half of it.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying xdist captured_info collision fix; 45645 builds on the same testing_utils change and adds the necessary CI artifact handling, so they could plausibly be merged together." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 21, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45737-2", + "nodes": [ + "pull_request:45737", + "pull_request:45738" + ], + "soft_pairs": [ + "pull_request:45737|pull_request:45738" + ], + "prepared_review_unit_hash": "cc2ade73a49ec6a8883a7a24c433988645588bcd049e8aad2bce56b1b24daa1c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs address the same MusicgenMelody cache initialization bug in `forward()`: replacing `EncoderDecoderCache(DynamicCache, DynamicCache)` with a single `DynamicCache`. PR #45737 is the more complete duplicate because it bundles the fix with a regression test, while #45738 is the fix-only variant.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45737 is the best canonical PR because it contains the same code fix as #45738 plus a regression test, making it the most complete representation of the bug fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #45737 is the strongest overall candidate: it fixes the concrete cache bug and adds a targeted regression test, whereas #45738 only changes the implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45737", + "right": "pull_request:45738", + "accept": true, + "reason": "Same underlying code-path bug in `musicgen_melody` cache setup; the diffs match on the core fix and the test-only addition in #45737 could plausibly live in the same PR as #45738." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet: both PRs make the same `musicgen_melody` cache initialization change, and #45737 only adds a regression test on top of that fix. The soft-edge accept is conservative enough because the diffs are mergeable into one PR for the same bugfix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46126-2", + "nodes": [ + "pull_request:46126", + "pull_request:46269" + ], + "soft_pairs": [ + "pull_request:46126|pull_request:46269" + ], + "prepared_review_unit_hash": "bf0592aeb9df3a25bb1879c6722ecc7f081a0ee47c01e59c9ccb27b194689c44", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5526, + "estimated_input_tokens": 1382, + "estimated_eval_tokens": 3020 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5526, + "estimated_input_tokens": 1382, + "estimated_eval_tokens": 3020 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs touch distributed training code, but they are not the same change: #46126 is a focused EP/Trainer integration, while #46269 is a broad distributed branch base spanning FSDP, integrations, model loading, CI, and multiple subsystems.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "#46126 is the more focused, self-contained PR centered on expert-parallel routing and Trainer integration, making it the better canonical representative of a single concrete change.", + "best_issue_reason": null, + "best_pr_reason": "#46126 is the best representative because it targets one specific feature path with a clear technical scope; #46269 is too broad and mixes many unrelated distributed changes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46126", + "right": "pull_request:46269", + "accept": false, + "reason": "Shared distributed filenames are not enough here: #46126 is a specific EP/Trainer feature, while #46269 is a large base branch touching many areas and is not plausibly mergeable as the same PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 23, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46138-10", + "nodes": [ + "pull_request:46138", + "pull_request:46432", + "pull_request:46501", + "pull_request:46502", + "pull_request:46503", + "pull_request:46504", + "pull_request:46554", + "pull_request:46558", + "pull_request:46563", + "pull_request:46716" + ], + "soft_pairs": [ + "pull_request:46138|pull_request:46554", + "pull_request:46138|pull_request:46558", + "pull_request:46138|pull_request:46563", + "pull_request:46138|pull_request:46716", + "pull_request:46138|pull_request:46504", + "pull_request:46432|pull_request:46503", + "pull_request:46432|pull_request:46501", + "pull_request:46432|pull_request:46502", + "pull_request:46432|pull_request:46554", + "pull_request:46432|pull_request:46558", + "pull_request:46501|pull_request:46503", + "pull_request:46501|pull_request:46502", + "pull_request:46501|pull_request:46504", + "pull_request:46501|pull_request:46554", + "pull_request:46502|pull_request:46503", + "pull_request:46502|pull_request:46504", + "pull_request:46502|pull_request:46554", + "pull_request:46503|pull_request:46504", + "pull_request:46503|pull_request:46554", + "pull_request:46504|pull_request:46554", + "pull_request:46504|pull_request:46558", + "pull_request:46554|pull_request:46558", + "pull_request:46554|pull_request:46563", + "pull_request:46554|pull_request:46716", + "pull_request:46558|pull_request:46563", + "pull_request:46558|pull_request:46716", + "pull_request:46563|pull_request:46716", + "pull_request:46504|pull_request:46563", + "pull_request:46432|pull_request:46563", + "pull_request:46504|pull_request:46716", + "pull_request:46432|pull_request:46716" + ], + "prepared_review_unit_hash": "8e8978aa7527355477e072582a5e02480bcd879954420f857ca74fa7bfef7027", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 31, + "serialized_chars": 26679, + "estimated_input_tokens": 6670, + "estimated_eval_tokens": 13596 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 31, + "serialized_chars": 26679, + "estimated_input_tokens": 6670, + "estimated_eval_tokens": 13596 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated workflow-maintenance PRs. The only tight subgroup is several edits to pr-ci-security-gate.yml, but they change different behaviors (checkout/diff checks, permissions, reusable-workflow comments, secrets handling), so they are not duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: the cluster spans separate workflow-only changes across different files, and the pr-ci-security-gate PRs are about different safeguards rather than the same fix.", + "best_issue_reason": null, + "best_pr_reason": "If forced to pick one representative PR, 46554 is the most substantive change in the only coherent subgroup (pr-ci-security-gate), since it alters the core security-check logic.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46138", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow files and different changes; one updates self-comment CI stdin handling, the other changes pr-ci-security-gate logic." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46558", + "accept": false, + "reason": "Unrelated workflow files and different behavior; no shared code-path or same bug fix." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46563", + "accept": false, + "reason": "Unrelated workflow files; self-comment CI input handling is not the same change as pr-ci-security-gate comments/guardrails." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow file and different purpose; not the same underlying change." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files and unrelated fixes; no duplicate relationship." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflow files and distinct changes; ssh-runner hardening is unrelated to benchmark dispatch behavior." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46501", + "accept": false, + "reason": "Different workflow files and different fixes; ssh-runner changes are not the same as anti-slop permission scoping." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46502", + "accept": false, + "reason": "Different workflow files and unrelated behavior; no shared concrete bug." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow files and different concerns; ssh-runner hardening is not the same as pr-ci-security-gate changes." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow files and unrelated fixes; not the same change." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46503", + "accept": false, + "reason": "Both are workflow tweaks, but they address different workflows and different behavior, so they are not duplicates." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46502", + "accept": false, + "reason": "Same general area of CI workflows, but the files and fixes differ; not the same underlying issue." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files and unrelated changes; no common code-path problem." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46554", + "accept": false, + "reason": "anti-slop permission scoping is unrelated to pr-ci-security-gate workflow logic." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflow files and distinct fixes; not mergeable as one duplicate change." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files and unrelated behavior changes." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46554", + "accept": false, + "reason": "Assign-reviewers checkout ref change is unrelated to pr-ci-security-gate security checks." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files and different CI concerns; not the same bug or change." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46554", + "accept": false, + "reason": "Benchmark dispatch support is unrelated to pr-ci-security-gate workflow changes." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow files and different behavior; not the same underlying fix." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow files and unrelated CI logic changes." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46558", + "accept": false, + "reason": "Same workflow file, but different concrete changes: security-check/diff scanning versus permissions/approval gating." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46563", + "accept": false, + "reason": "Same workflow file, but one adds security checks and the other only adds safety comments about the reusable workflow." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46716", + "accept": false, + "reason": "Same workflow file, but different safeguards: checkout/diff-based checks versus removing inherited secrets." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46563", + "accept": false, + "reason": "Same workflow file, but different changes: permissions/approval condition versus documentation comments." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46716", + "accept": false, + "reason": "Same workflow file, but one adjusts job permissions/approval flow and the other changes secrets passing; not the same fix." + }, + { + "left": "pull_request:46563", + "right": "pull_request:46716", + "accept": false, + "reason": "Same workflow file, but comments-only clarification is not the same as changing secrets inheritance." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow files and unrelated fixes; no duplicate overlap beyond generic CI maintenance." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow files and unrelated purposes; not the same underlying change." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow files and different CI concerns; not a duplicate." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow files and unrelated fixes; no shared concrete issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 24, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46174-2", + "nodes": [ + "pull_request:46174", + "pull_request:46780" + ], + "soft_pairs": [ + "pull_request:46174|pull_request:46780" + ], + "prepared_review_unit_hash": "a48ce4ca162f555682f79411d7229ac1ccc08d7822dfab654bb881639fccddbc", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6030, + "estimated_input_tokens": 1508, + "estimated_eval_tokens": 3272 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6030, + "estimated_input_tokens": 1508, + "estimated_eval_tokens": 3272 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two PRs target the same reported MPS crash in the sinusoidal position embedding helper, but they implement different fixes and are not safe to treat as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46174 is the broader, more complete fix for the float64-on-MPS crash: it updates the helper across all affected model files and keeps the internal math on CPU float64 before casting back.", + "best_issue_reason": null, + "best_pr_reason": "PR 46174 is the best representative of the cluster because it is the original, more comprehensive fix and has more discussion/review context than the narrower follow-up.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46174", + "right": "pull_request:46780", + "accept": false, + "reason": "They share the same issue target and helper, but the patches are different solutions: 46174 avoids float64 on MPS by doing the arithmetic on CPU first, while 46780 switches the helper to use dtype for internal arithmetic. Those are alternative implementations, not one mergeable duplicate PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 25, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46310-2", + "nodes": [ + "pull_request:46310", + "pull_request:47162" + ], + "soft_pairs": [ + "pull_request:46310|pull_request:47162" + ], + "prepared_review_unit_hash": "c78619d4333d6a6498f73c61a48a777174a21ae9b55cc0bd783a287ab8fd2e6f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4883, + "estimated_input_tokens": 1221, + "estimated_eval_tokens": 2698 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4883, + "estimated_input_tokens": 1221, + "estimated_eval_tokens": 2698 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs fix the same off-by-one error in Moonshine Streaming\u2019s sliding-window right boundary; PR 47162 is the more complete superset because it also covers Gemma4 and adds tests.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47162 is the better canonical PR because it covers the same Moonshine Streaming fix as 46310, extends the same boundary correction to Gemma4, and includes a test file, making it the more complete change set.", + "best_issue_reason": null, + "best_pr_reason": "PR 47162 is the best representative of the cluster because it contains the shared bug fix plus additional coverage, whereas 46310 is a narrower subset of the same fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46310", + "right": "pull_request:47162", + "accept": true, + "reason": "Accept: both PRs correct the same off-by-one right-window mask bug (`-dist <= right_window_size`) in the same Moonshine Streaming code path, and 47162 is a superset that plausibly subsumes 46310." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 26, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46574-2", + "nodes": [ + "pull_request:46574", + "pull_request:46580" + ], + "soft_pairs": [ + "pull_request:46574|pull_request:46580" + ], + "prepared_review_unit_hash": "0d69c2492508fc47d07ff807d649b6a5a2d4862d92d15554f9aed69fe18a4d49", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4410, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4410, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated pull requests: one fixes CodeLlama decoding of real leading spaces, and the other fixes DeepSeek-Coder/DeepSeek-LLM tokenizer class selection in auto-loading. They share only a vague tokenizer theme, not the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:46574", + "right": "pull_request:46580", + "accept": false, + "reason": "Different code paths and failures: CodeLlama decoder stripping vs. auto tokenizer class overrides for DeepSeek checkpoints. Not the same fix and would not be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 27, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46773-2", + "nodes": [ + "pull_request:46773", + "pull_request:47087" + ], + "soft_pairs": [ + "pull_request:46773|pull_request:47087" + ], + "prepared_review_unit_hash": "71c1148e891f219c27aefe1909893bdc3958e76bc032fb8aa955d1aa928992d0", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5821, + "estimated_input_tokens": 1456, + "estimated_eval_tokens": 3168 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5821, + "estimated_input_tokens": 1456, + "estimated_eval_tokens": 3168 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two pull requests in the same masking area, but they fix different bugs: one restores padding masking for batch-size-1 linear-attention forwards, the other fixes recurrent masking being skipped on continued/cache forwards. They are related conceptually, but not duplicates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47087 is the better canonical PR: it targets the shared masking utility at the source, has an explicit issue target, and covers the recurrent/cached-forward masking regression in a more centralized way.", + "best_issue_reason": null, + "best_pr_reason": "PR 47087 is the strongest standalone change for the cluster because it fixes a concrete code-path bug in masking_utils and propagates the fix across affected recurrent models with tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46773", + "right": "pull_request:47087", + "accept": false, + "reason": "Both concern skipped padding masks, but 46773 fixes a batch-size-1 condition in linear-attention model code paths while 47087 fixes masking on continued forwards/chunked prefill via masking_utils. Different triggers and different bugs, so not duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 28, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-47074-3", + "nodes": [ + "pull_request:47074", + "pull_request:47076", + "pull_request:47195" + ], + "soft_pairs": [ + "pull_request:47074|pull_request:47076", + "pull_request:47074|pull_request:47195" + ], + "prepared_review_unit_hash": "df63fc7596a26857ade3149c3470f9e111dec61cc9cef50cae19dde8fb9844f1", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 6433, + "estimated_input_tokens": 1609, + "estimated_eval_tokens": 3474 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 6433, + "estimated_input_tokens": 1609, + "estimated_eval_tokens": 3474 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "No duplicates here: the three PRs fix different code paths in different subsystems, so none of the soft edges should be merged.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: 47074 fixes AnyToAnyPipeline input key normalization, 47076 fixes Gemma4 processor placeholder validation, and 47195 fixes candidate decoding length capping. They are separate bugs with no shared code path.", + "best_issue_reason": null, + "best_pr_reason": "47074 is the best standalone representative only because it is the clearest user-facing bugfix, but it should not be treated as a duplicate of the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47074", + "right": "pull_request:47076", + "accept": false, + "reason": "Different components and behaviors: multimodal pipeline input-key mapping vs Gemma4 audio/text placeholder construction. Shared multimodal context is too broad to indicate the same bug or mergeable PR." + }, + { + "left": "pull_request:47074", + "right": "pull_request:47195", + "accept": false, + "reason": "Unrelated fixes in different code paths: pipeline input normalization vs generation candidate length capping. No evidence they address the same underlying defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 29, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-47122-10", + "nodes": [ + "pull_request:47122", + "pull_request:47123", + "pull_request:47124", + "pull_request:47224", + "pull_request:47266", + "pull_request:47280", + "pull_request:47281", + "pull_request:47282", + "pull_request:47310", + "pull_request:47380" + ], + "soft_pairs": [ + "pull_request:47122|pull_request:47281", + "pull_request:47122|pull_request:47123", + "pull_request:47122|pull_request:47280", + "pull_request:47122|pull_request:47224", + "pull_request:47122|pull_request:47310", + "pull_request:47123|pull_request:47280", + "pull_request:47123|pull_request:47266", + "pull_request:47123|pull_request:47281", + "pull_request:47123|pull_request:47124", + "pull_request:47124|pull_request:47310", + "pull_request:47124|pull_request:47282", + "pull_request:47124|pull_request:47380", + "pull_request:47124|pull_request:47281", + "pull_request:47266|pull_request:47280", + "pull_request:47266|pull_request:47380", + "pull_request:47266|pull_request:47281", + "pull_request:47124|pull_request:47266", + "pull_request:47280|pull_request:47281", + "pull_request:47280|pull_request:47310", + "pull_request:47280|pull_request:47380", + "pull_request:47281|pull_request:47282", + "pull_request:47282|pull_request:47380", + "pull_request:47282|pull_request:47310", + "pull_request:47280|pull_request:47282", + "pull_request:47310|pull_request:47380" + ], + "prepared_review_unit_hash": "8f61d10c20f34d9a3d576b9b437e8bd48034a78f9bdaeee49c3ccce654f5dab6", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 25, + "serialized_chars": 28954, + "estimated_input_tokens": 7239, + "estimated_eval_tokens": 14734 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 25, + "serialized_chars": 28954, + "estimated_input_tokens": 7239, + "estimated_eval_tokens": 14734 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated model-specific integration-test fixes that only look superficially similar. I found two true duplicate pairs: the MiniMax-M3-VL load_error/offload-folder fixes and the Mamba2 OOM fixes. The EdgeTAM PRs and the rest are false positives or broader multi-bug umbrellas.", + "confidence": 0.84, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47280 is the cleanest representative of a true duplicate fix: it directly addresses the MiniMax-M3-VL load_error by adding a temporary offload folder, with the same root cause as 47123 but a slightly more complete patch.", + "best_issue_reason": null, + "best_pr_reason": "PR 47280 is the best overall representative among the actual duplicates because it is a direct, local fix for a concrete load_error, rather than a broad umbrella or a cross-model edit.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47122", + "right": "pull_request:47281", + "accept": false, + "reason": "Different models and different problems: EdgeTAM import_or_config kwarg filtering vs GLM46V modular/model refactor work." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47123", + "accept": false, + "reason": "Different models and failure modes: EdgeTAM backbone kwarg handling vs MiniMax-M3-VL load_error/offload handling." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47280", + "accept": false, + "reason": "Different models and unrelated code paths; this is EdgeTAM kwarg filtering vs MiniMax-M3-VL load_error mitigation." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47224", + "accept": false, + "reason": "47224 is a multi-bug umbrella touching tokenizer/whisper/other fixes in addition to EdgeTAM, so it is not the same single change as 47122." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47310", + "accept": false, + "reason": "Different models and unrelated fixes: EdgeTAM kwarg filtering vs Gemma3 generation cache behavior." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47280", + "accept": true, + "reason": "Same underlying MiniMax-M3-VL load_error on model loading; both add an offload folder for accelerate/device_map behavior, with 47280 being a more complete version." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47266", + "accept": false, + "reason": "Both are load_error fixes, but for different models and different code paths, so they are not the same bug." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47281", + "accept": false, + "reason": "MiniMax-M3-VL load_error fix vs GLM46V model refactor; different artifact and change type." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47124", + "accept": false, + "reason": "Different models and failure modes: MiniMax-M3-VL load_error vs Falcon-Mamba output_mismatch." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47310", + "accept": false, + "reason": "Both are output mismatches, but for different models and different generation code paths; too broad to be the same change." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47266", + "accept": false, + "reason": "Different model-specific failures and fixes: Falcon-Mamba output mismatch vs DeepSeek-V3.2 load_error." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47282", + "accept": false, + "reason": "Different models and different root causes: Falcon-Mamba output mismatch vs Mamba2 OOM." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47380", + "accept": false, + "reason": "Different models and different failure modes; no shared concrete code-path problem." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47281", + "accept": false, + "reason": "Different models and different kinds of changes; this is not the same bug." + }, + { + "left": "pull_request:47266", + "right": "pull_request:47280", + "accept": false, + "reason": "Different models and different failure modes: DeepSeek-V3.2 load_error vs MiniMax-M3-VL load_error." + }, + { + "left": "pull_request:47266", + "right": "pull_request:47281", + "accept": false, + "reason": "Different model families and unrelated fixes; the similarity is only superficial integration-test cleanup." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47281", + "accept": false, + "reason": "MiniMax-M3-VL load_error fix vs GLM46V model refactor; not the same underlying bug." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47282", + "accept": false, + "reason": "Different models and different failures: MiniMax-M3-VL load_error vs Mamba2 OOM." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47310", + "accept": false, + "reason": "Different models and unrelated fixes; no shared code path." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47380", + "accept": false, + "reason": "Different models and different failure modes; MiniMax-M3-VL load_error mitigation is unrelated to Mamba2 OOM." + }, + { + "left": "pull_request:47281", + "right": "pull_request:47282", + "accept": false, + "reason": "Different models and different fix types; GLM46V refactor/update vs Mamba2 OOM workaround." + }, + { + "left": "pull_request:47282", + "right": "pull_request:47380", + "accept": true, + "reason": "Same Mamba2 integration-test OOM problem in the same test file; both are plausible variants of the same memory-pressure fix and could be merged into one PR." + }, + { + "left": "pull_request:47282", + "right": "pull_request:47310", + "accept": false, + "reason": "Different models and unrelated failures: Mamba2 OOM vs Gemma3 output mismatch." + }, + { + "left": "pull_request:47310", + "right": "pull_request:47380", + "accept": false, + "reason": "Different models and different concrete issues; cache default handling in Gemma3 is unrelated to Mamba2 OOM." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 30, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-47140-2", + "nodes": [ + "pull_request:47140", + "pull_request:47303" + ], + "soft_pairs": [ + "pull_request:47140|pull_request:47303" + ], + "prepared_review_unit_hash": "71bda56b95be761b35c09dd75f0d74633192a4e1af0250880651c2a08a5d4f26", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4409, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4409, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs fix the same chat CLI management-endpoint URL bug: requests were being sent relative to the wrong base instead of the service root. The differences are mostly implementation style and scope, not a separate underlying change.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #47303 is the better canonical PR: it directly frames the fix as routing chat management calls to the service root, has an explicit tracking issue, and uses a shared helper to normalize management URLs.", + "best_issue_reason": null, + "best_pr_reason": "PR #47303 is the strongest representative of the cluster because it captures the general service-root routing problem more clearly than the /v1-specific variant and appears to cover the same chat management code path.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47140", + "right": "pull_request:47303", + "accept": true, + "reason": "Both changes address the same underlying chat CLI management URL bug in src/transformers/cli/chat.py, redirecting management requests from a /v1-style API base to the service root." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both PRs touch the same chat CLI management URL path in src/transformers/cli/chat.py and the titles/diffs support the shared service-root vs /v1-base correction. The soft-edge verdict is also conservative enough, since it says the changes appear to address the same underlying bug rather than claiming they are identical." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 31, + "component_count": 31, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-47261-2", + "nodes": [ + "pull_request:47261", + "pull_request:47262" + ], + "soft_pairs": [ + "pull_request:47261|pull_request:47262" + ], + "prepared_review_unit_hash": "57a03110bcb78e2d550970d665196eaafb63311ba36411283d1dac107d8c9309", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4419, + "estimated_input_tokens": 1105, + "estimated_eval_tokens": 2466 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4419, + "estimated_input_tokens": 1105, + "estimated_eval_tokens": 2466 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two PRs are unrelated: one fixes loss scaling with label smoothing during gradient accumulation in Trainer, and the other changes TopHLogitsWarper to honor min_tokens_to_keep. They touch different subsystems and different concrete code paths, so this is not a duplicate pair.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #47261 is the better canonical representative only in the sense that it is a concrete, self-contained bug fix with an explicit issue target; PR #47262 is a separate generation feature/validation change in a different path.", + "best_issue_reason": null, + "best_pr_reason": "PR #47261 is the most representative standalone PR here because it addresses a specific correctness bug with clear reproduction context, whereas PR #47262 is not the same underlying change and should not be clustered with it.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47261", + "right": "pull_request:47262", + "accept": false, + "reason": "Different bugs in different modules: Trainer label-smoothing/gradient-accumulation loss scaling vs TopHLogitsWarper token-retention handling. They are not the same underlying change and should not be merged." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/manifest.json b/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..5f784dee5cd85feeba4ac26470b03e2cfb59fc10 --- /dev/null +++ b/snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260719t060027z", + "artifacts": { + "hybrid": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-07-19T06:10:38Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260719T060027Z", + "variant": "hybrid" +} diff --git a/snapshots/20260719T060027Z/manifest.json b/snapshots/20260719T060027Z/manifest.json index 59e61a0230a9c1791da4612a998ed35f68eda5d6..9b434a32583a69885392418061c6fe640353c5bc 100644 --- a/snapshots/20260719T060027Z/manifest.json +++ b/snapshots/20260719T060027Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 265 }, "extracted_at": "2026-07-19T06:00:27Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260719t060027z", + "runs": { + "hybrid-model-20260719t060027z": { + "analysis_id": "hybrid-model-20260719t060027z", + "artifacts": { + "hybrid": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260719T060027Z/analysis-runs/hybrid-model-20260719t060027z/manifest.json", + "model": null, + "published_at": "2026-07-19T06:10:38Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260719T060027Z", "watermark": {