diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 7a343300b287985de4e87377613fb7aca3aadf52..6dab861fec441e86ec0687d7f4885ef95a6cfd1a 100644 --- a/analysis/current/analysis-report-hybrid.json +++ b/analysis/current/analysis-report-hybrid.json @@ -1,38 +1,707 @@ { "schema_version": "1.0", "repo": "huggingface/transformers", - "snapshot_id": "20260714T120113Z", - "generated_at": "2026-07-14T12:13:49Z", + "snapshot_id": "20260714T180038Z", + "generated_at": "2026-07-14T18:14:57Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ { - "cluster_id": "cluster-46376-13", - "summary": "Cluster of 13 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-43451-14", + "summary": "Cluster of 1 issues and 13 PRs centered on issue #47321.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 46658, - "issue_numbers": [], + "confidence": 0.6, + "canonical_issue_number": 47321, + "canonical_pr_number": 47322, + "issue_numbers": [ + 47321 + ], "pr_numbers": [ + 43451, 46376, 46509, 46658, + 46836, 47143, 47163, - 47166, 47181, - 47184, - 47198, 47216, 47292, 47301, - 47311 + 47311, + 47322 ], "evidence_types": [ + "closing_reference", "soft_similarity" ], "pr_comparisons": [ + { + "left_pr_number": 43451, + "right_pr_number": 46376, + "code_similarity": 0.069, + "size_similarity": 0.215, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.176, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "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": 0.293, + "size_similarity": 0.836, + "file_overlap": 0.073, + "area_overlap": 0.17, + "patch_similarity": 0.296, + "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, + 822 + ] + ] + }, + { + "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": 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 + ], + [ + 494, + 501 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47143, + "code_similarity": 0.155, + "size_similarity": 0.475, + "file_overlap": 0.1, + "area_overlap": 0.0, + "patch_similarity": 0.197, + "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": 47181, + "code_similarity": 0.285, + "size_similarity": 0.907, + "file_overlap": 0.147, + "area_overlap": 0.021, + "patch_similarity": 0.347, + "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" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 1375, + 1382 + ] + ], + "right_ranges": [ + [ + 1249, + 1256 + ] + ] + }, + { + "filename": "src/transformers/conversion_mapping.py", + "left_ranges": [ + [ + 737, + 769 + ] + ], + "right_ranges": [ + [ + 1633, + 1653 + ] + ] + }, + { + "filename": "src/transformers/models/__init__.py", + "left_ranges": [ + [ + 290, + 296 + ] + ], + "right_ranges": [ + [ + 78, + 84 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 97, + 105 + ], + [ + 1053, + 1059 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47216, + "code_similarity": 0.042, + "size_similarity": 0.11, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.131, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47292, + "code_similarity": 0.002, + "size_similarity": 0.01, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47301, + "code_similarity": 0.007, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.039, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.01, + "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": 46376, "right_pr_number": 46509, @@ -55,6 +724,17 @@ "shared_filenames": [], "shared_file_areas": [] }, + { + "left_pr_number": 46376, + "right_pr_number": 46836, + "code_similarity": 0.073, + "size_similarity": 0.233, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.176, + "shared_filenames": [], + "shared_file_areas": [] + }, { "left_pr_number": 46376, "right_pr_number": 47143, @@ -77,17 +757,6 @@ "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 46376, - "right_pr_number": 47166, - "code_similarity": 0.011, - "size_similarity": 0.043, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.017, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 46376, "right_pr_number": 47181, @@ -99,28 +768,6 @@ "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 46376, - "right_pr_number": 47184, - "code_similarity": 0.007, - "size_similarity": 0.022, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.014, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46376, - "right_pr_number": 47198, - "code_similarity": 0.012, - "size_similarity": 0.053, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.01, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 46376, "right_pr_number": 47216, @@ -165,6 +812,17 @@ "shared_filenames": [], "shared_file_areas": [] }, + { + "left_pr_number": 46376, + "right_pr_number": 47322, + "code_similarity": 0.003, + "size_similarity": 0.01, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, { "left_pr_number": 46509, "right_pr_number": 46658, @@ -238,6 +896,107 @@ } ] }, + { + "left_pr_number": 46509, + "right_pr_number": 46836, + "code_similarity": 0.147, + "size_similarity": 0.41, + "file_overlap": 0.103, + "area_overlap": 0.0, + "patch_similarity": 0.232, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py", + "utils/check_config_attributes.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 795, + 802 + ] + ], + "right_ranges": [ + [ + 893, + 900 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ] + }, + { + "filename": "utils/check_config_attributes.py", + "left_ranges": [ + [ + 42, + 50 + ] + ], + "right_ranges": [ + [ + 237, + 243 + ] + ] + } + ] + }, { "left_pr_number": 46509, "right_pr_number": 47143, @@ -310,17 +1069,6 @@ "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 46509, - "right_pr_number": 47166, - "code_similarity": 0.007, - "size_similarity": 0.024, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.014, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 46509, "right_pr_number": 47181, @@ -410,28 +1158,6 @@ } ] }, - { - "left_pr_number": 46509, - "right_pr_number": 47184, - "code_similarity": 0.004, - "size_similarity": 0.013, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.01, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46509, - "right_pr_number": 47198, - "code_similarity": 0.008, - "size_similarity": 0.03, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.011, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 46509, "right_pr_number": 47216, @@ -476,6 +1202,118 @@ "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": 46658, + "right_pr_number": 46836, + "code_similarity": 0.22, + "size_similarity": 0.771, + "file_overlap": 0.06, + "area_overlap": 0.0, + "patch_similarity": 0.32, + "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": [ + [ + 739, + 822 + ] + ], + "right_ranges": [ + [ + 127, + 133 + ], + [ + 1495, + 1545 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ], + "right_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1085, + 1091 + ] + ], + "right_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ] + } + ] + }, { "left_pr_number": 46658, "right_pr_number": 47143, @@ -560,17 +1398,6 @@ "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 46658, - "right_pr_number": 47166, - "code_similarity": 0.003, - "size_similarity": 0.008, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.008, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 46658, "right_pr_number": 47181, @@ -674,107 +1501,351 @@ }, { "left_pr_number": 46658, - "right_pr_number": 47184, + "right_pr_number": 47216, + "code_similarity": 0.038, + "size_similarity": 0.092, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.129, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47292, "code_similarity": 0.002, + "size_similarity": 0.008, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47301, + "code_similarity": 0.005, "size_similarity": 0.004, "file_overlap": 0.0, "area_overlap": 0.0, + "patch_similarity": 0.03, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, "patch_similarity": 0.007, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 46658, - "right_pr_number": 47198, - "code_similarity": 0.014, - "size_similarity": 0.01, - "file_overlap": 0.036, + "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": 46836, + "right_pr_number": 47143, + "code_similarity": 0.155, + "size_similarity": 0.515, + "file_overlap": 0.077, + "area_overlap": 0.0, + "patch_similarity": 0.19, + "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": [ + [ + 453, + 459 + ] + ], + "right_ranges": [ + [ + 305, + 311 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] + }, + { + "left_pr_number": 46836, + "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": 46836, + "right_pr_number": 47181, + "code_similarity": 0.278, + "size_similarity": 0.983, + "file_overlap": 0.116, "area_overlap": 0.0, - "patch_similarity": 0.009, + "patch_similarity": 0.31, "shared_filenames": [ - "src/transformers/conversion_mapping.py" + "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" ], "shared_file_areas": [ { - "filename": "src/transformers/conversion_mapping.py", + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 893, + 900 + ] + ], + "right_ranges": [ + [ + 1249, + 1256 + ] + ] + }, + { + "filename": "src/transformers/conversion_mapping.py", + "left_ranges": [ + [ + 127, + 133 + ], + [ + 1495, + 1545 + ] + ], + "right_ranges": [ + [ + 1633, + 1653 + ] + ] + }, + { + "filename": "src/transformers/models/__init__.py", + "left_ranges": [ + [ + 453, + 459 + ] + ], + "right_ranges": [ + [ + 78, + 84 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", "left_ranges": [ [ - 739, - 822 + 509, + 515 + ], + [ + 1119, + 1125 ] ], "right_ranges": [ [ - 140, - 148 + 97, + 105 + ], + [ + 1053, + 1059 ] ] } ] }, { - "left_pr_number": 46658, + "left_pr_number": 46836, "right_pr_number": 47216, - "code_similarity": 0.038, - "size_similarity": 0.092, + "code_similarity": 0.043, + "size_similarity": 0.119, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.129, + "patch_similarity": 0.127, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, + "left_pr_number": 46836, "right_pr_number": 47292, - "code_similarity": 0.002, - "size_similarity": 0.008, + "code_similarity": 0.003, + "size_similarity": 0.011, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.003, + "patch_similarity": 0.004, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, + "left_pr_number": 46836, "right_pr_number": 47301, - "code_similarity": 0.005, - "size_similarity": 0.004, + "code_similarity": 0.006, + "size_similarity": 0.005, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.03, + "patch_similarity": 0.035, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46658, + "left_pr_number": 46836, "right_pr_number": 47311, "code_similarity": 0.002, - "size_similarity": 0.002, + "size_similarity": 0.003, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.007, + "patch_similarity": 0.008, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47143, - "right_pr_number": 47163, - "code_similarity": 0.004, - "size_similarity": 0.009, + "left_pr_number": 46836, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.002, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.013, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, { "left_pr_number": 47143, - "right_pr_number": 47166, - "code_similarity": 0.005, - "size_similarity": 0.019, + "right_pr_number": 47163, + "code_similarity": 0.004, + "size_similarity": 0.009, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.009, + "patch_similarity": 0.013, "shared_filenames": [], "shared_file_areas": [] }, @@ -867,28 +1938,6 @@ } ] }, - { - "left_pr_number": 47143, - "right_pr_number": 47184, - "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": 47143, - "right_pr_number": 47198, - "code_similarity": 0.006, - "size_similarity": 0.024, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.005, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 47143, "right_pr_number": 47216, @@ -934,13 +1983,13 @@ "shared_file_areas": [] }, { - "left_pr_number": 47163, - "right_pr_number": 47166, - "code_similarity": 0.104, - "size_similarity": 0.483, + "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": 0.047, + "patch_similarity": 0.002, "shared_filenames": [], "shared_file_areas": [] }, @@ -955,28 +2004,6 @@ "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 47163, - "right_pr_number": 47184, - "code_similarity": 0.19, - "size_similarity": 0.933, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.022, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47163, - "right_pr_number": 47198, - "code_similarity": 0.078, - "size_similarity": 0.389, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.0, - "shared_filenames": [], - "shared_file_areas": [] - }, { "left_pr_number": 47163, "right_pr_number": 47216, @@ -1022,122 +2049,16 @@ "shared_file_areas": [] }, { - "left_pr_number": 47166, - "right_pr_number": 47181, - "code_similarity": 0.004, - "size_similarity": 0.01, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.011, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47166, - "right_pr_number": 47184, - "code_similarity": 0.11, - "size_similarity": 0.517, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.041, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47166, - "right_pr_number": 47198, - "code_similarity": 0.163, - "size_similarity": 0.806, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.013, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47166, - "right_pr_number": 47216, - "code_similarity": 0.022, - "size_similarity": 0.084, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.034, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47166, - "right_pr_number": 47292, - "code_similarity": 0.189, - "size_similarity": 0.935, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.01, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47166, - "right_pr_number": 47301, - "code_similarity": 0.108, - "size_similarity": 0.517, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.034, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47166, - "right_pr_number": 47311, - "code_similarity": 0.062, - "size_similarity": 0.293, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.021, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47181, - "right_pr_number": 47184, - "code_similarity": 0.003, - "size_similarity": 0.005, + "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": 0.01, + "patch_similarity": 0.008, "shared_filenames": [], "shared_file_areas": [] }, - { - "left_pr_number": 47181, - "right_pr_number": 47198, - "code_similarity": 0.017, - "size_similarity": 0.013, - "file_overlap": 0.043, - "area_overlap": 0.0, - "patch_similarity": 0.011, - "shared_filenames": [ - "src/transformers/conversion_mapping.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/conversion_mapping.py", - "left_ranges": [ - [ - 1633, - 1653 - ] - ], - "right_ranges": [ - [ - 140, - 148 - ] - ] - } - ] - }, { "left_pr_number": 47181, "right_pr_number": 47216, @@ -1178,106 +2099,18 @@ "size_similarity": 0.003, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.008, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47184, - "right_pr_number": 47198, - "code_similarity": 0.083, - "size_similarity": 0.417, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.0, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47184, - "right_pr_number": 47216, - "code_similarity": 0.012, - "size_similarity": 0.043, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.02, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47184, - "right_pr_number": 47292, - "code_similarity": 0.097, - "size_similarity": 0.484, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.0, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47184, - "right_pr_number": 47301, - "code_similarity": 0.207, - "size_similarity": 1.0, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.046, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47184, - "right_pr_number": 47311, - "code_similarity": 0.119, - "size_similarity": 0.567, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.038, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47198, - "right_pr_number": 47216, - "code_similarity": 0.023, - "size_similarity": 0.104, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.018, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47198, - "right_pr_number": 47292, - "code_similarity": 0.172, - "size_similarity": 0.861, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.0, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47198, - "right_pr_number": 47301, - "code_similarity": 0.087, - "size_similarity": 0.417, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.023, + "patch_similarity": 0.008, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 47198, - "right_pr_number": 47311, - "code_similarity": 0.047, - "size_similarity": 0.236, + "left_pr_number": 47181, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.002, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.0, + "patch_similarity": 0.004, "shared_filenames": [], "shared_file_areas": [] }, @@ -1314,6 +2147,55 @@ "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": 47292, "right_pr_number": 47301, @@ -1336,6 +2218,17 @@ "shared_filenames": [], "shared_file_areas": [] }, + { + "left_pr_number": 47292, + "right_pr_number": 47322, + "code_similarity": 0.045, + "size_similarity": 0.226, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.0, + "shared_filenames": [], + "shared_file_areas": [] + }, { "left_pr_number": 47301, "right_pr_number": 47311, @@ -1346,6 +2239,28 @@ "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, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.026, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47311, + "right_pr_number": 47322, + "code_similarity": 0.165, + "size_similarity": 0.824, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.0, + "shared_filenames": [], + "shared_file_areas": [] } ] }, @@ -2314,260 +3229,81 @@ 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-43451-2", - "summary": "Cluster of 2 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 - ], - "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": [ + 580, + 598 + ], [ - 331, - 338 + 601, + 612 ], [ - 1094, - 1100 + 634, + 640 ], [ - 1913, - 1919 - ] - ], - "right_ranges": [ + 646, + 653 + ], [ - 509, - 515 + 658, + 663 ], [ - 1119, - 1125 - ] - ] - }, - { - "filename": "utils/check_repo.py", - "left_ranges": [ + 666, + 674 + ], [ - 250, - 258 + 717, + 722 ], [ - 337, - 344 - ] - ], - "right_ranges": [ + 738, + 744 + ], [ - 267, - 274 + 749, + 772 ], [ - 494, - 501 + 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": [] } ] }, @@ -3058,7 +3794,7 @@ "status": "open", "confidence": 0.6, "canonical_issue_number": 47138, - "canonical_pr_number": 47140, + "canonical_pr_number": 47303, "issue_numbers": [ 47138, 47139 @@ -3222,7 +3958,7 @@ "cluster_id": "cluster-43240-3", "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", "status": "open", - "confidence": 0.75, + "confidence": 0.55, "canonical_issue_number": 43240, "canonical_pr_number": 43251, "issue_numbers": [ @@ -3233,8 +3969,7 @@ 43254 ], "evidence_types": [ - "closing_reference", - "shared_issue_target" + "closing_reference" ], "pr_comparisons": [ { @@ -3788,7 +4523,7 @@ "cluster_id": "cluster-46305-3", "summary": "Cluster of 1 issues and 2 PRs centered on issue #46305.", "status": "open", - "confidence": 0.8, + "confidence": 0.55, "canonical_issue_number": 46305, "canonical_pr_number": 47162, "issue_numbers": [ @@ -3799,9 +4534,7 @@ 47162 ], "evidence_types": [ - "closing_reference", - "shared_issue_target", - "soft_similarity" + "closing_reference" ], "pr_comparisons": [ { @@ -3968,185 +4701,6 @@ } ] }, - { - "cluster_id": "cluster-47122-3", - "summary": "Cluster of 3 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 47150, - "issue_numbers": [], - "pr_numbers": [ - 47122, - 47150, - 47224 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 47122, - "right_pr_number": 47150, - "code_similarity": 0.085, - "size_similarity": 0.423, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.0, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47122, - "right_pr_number": 47224, - "code_similarity": 0.474, - "size_similarity": 0.302, - "file_overlap": 0.4, - "area_overlap": 0.818, - "patch_similarity": 0.051, - "shared_filenames": [ - "src/transformers/models/edgetam/modeling_edgetam.py", - "src/transformers/models/edgetam/modular_edgetam.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/edgetam/modeling_edgetam.py", - "left_ranges": [ - [ - 464, - 472 - ] - ], - "right_ranges": [ - [ - 464, - 474 - ] - ] - }, - { - "filename": "src/transformers/models/edgetam/modular_edgetam.py", - "left_ranges": [ - [ - 210, - 218 - ] - ], - "right_ranges": [ - [ - 210, - 220 - ] - ] - } - ] - }, - { - "left_pr_number": 47150, - "right_pr_number": 47224, - "code_similarity": 0.442, - "size_similarity": 0.128, - "file_overlap": 0.2, - "area_overlap": 1.0, - "patch_similarity": 0.04, - "shared_filenames": [ - "src/transformers/models/whisper/generation_whisper.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/whisper/generation_whisper.py", - "left_ranges": [ - [ - 1937, - 1943 - ] - ], - "right_ranges": [ - [ - 1937, - 1943 - ] - ] - } - ] - } - ] - }, - { - "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.", @@ -4509,25 +5063,27 @@ 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-43451-2", - "canonical_pr_number": 43451, - "duplicate_pr_numbers": [ - 46836 - ], - "target_issue_number": null, - "reason": "PRs in cluster-43451-2 are treated as duplicates because they share soft_similarity evidence." + "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference evidence." }, { - "cluster_id": "cluster-43656-2", - "canonical_pr_number": 43656, + "cluster_id": "cluster-43451-14", + "canonical_pr_number": 47322, "duplicate_pr_numbers": [ - 43842 + 43451, + 46376, + 46509, + 46658, + 46836, + 47143, + 47163, + 47181, + 47216, + 47292, + 47301, + 47311 ], - "target_issue_number": null, - "reason": "PRs in cluster-43656-2 are treated as duplicates because they share soft_similarity evidence." + "target_issue_number": 47321, + "reason": "PRs in cluster-43451-14 are treated as duplicates because they converge on issue #47321 with closing_reference, soft_similarity evidence." }, { "cluster_id": "cluster-43698-3", @@ -4607,27 +5163,7 @@ 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-46376-13", - "canonical_pr_number": 46658, - "duplicate_pr_numbers": [ - 46376, - 46509, - 47143, - 47163, - 47166, - 47181, - 47184, - 47198, - 47216, - 47292, - 47301, - 47311 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46376-13 are treated as duplicates because they share soft_similarity evidence." + "reason": "PRs in cluster-46305-3 are treated as duplicates because they converge on issue #46305 with closing_reference evidence." }, { "cluster_id": "cluster-46693-3", @@ -4658,16 +5194,6 @@ "target_issue_number": 46894, "reason": "PRs in cluster-46894-5 are treated as duplicates because they converge on issue #46894 with closing_reference, shared_issue_target, soft_similarity evidence." }, - { - "cluster_id": "cluster-47122-3", - "canonical_pr_number": 47150, - "duplicate_pr_numbers": [ - 47122, - 47224 - ], - "target_issue_number": null, - "reason": "PRs in cluster-47122-3 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-47123-2", "canonical_pr_number": 47123, @@ -4679,24 +5205,24 @@ }, { "cluster_id": "cluster-47138-4", - "canonical_pr_number": 47140, + "canonical_pr_number": 47303, "duplicate_pr_numbers": [ - 47303 + 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." } ], "best_issue": { - "cluster_id": "cluster-46894-5", - "issue_number": 46894, - "reason": "Issue #46894 is the strongest global issue candidate because it is open, belongs to a cluster with 5 artifacts, and carries 1 discussion comments plus 10 inbound references.", - "score": 162.0 + "cluster_id": "cluster-43451-14", + "issue_number": 47321, + "reason": "Issue #47321 is the strongest global issue candidate because it is open, belongs to a cluster with 14 artifacts, and carries 1 discussion comments plus 2 inbound references.", + "score": 244.0 }, "best_pr": { - "cluster_id": "cluster-46376-13", - "pr_number": 46658, - "reason": "PR #46658 is the strongest global PR candidate because it is open, belongs to a cluster with 13 artifacts, links to 0 issues, and carries 96 review/discussion events.", - "score": 434.442 + "cluster_id": "cluster-43451-14", + "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 } } diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index 51731656c1a71e9533ff0a73e077261945e15518..791c2f77439b385e96c8821cc1a5e028fae75734 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": "20260714T120113Z", - "generated_at": "2026-07-14T12:13:49Z", + "snapshot_id": "20260714T180038Z", + "generated_at": "2026-07-14T18:14:57Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -16,7 +16,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:24643", "issue:30277", @@ -131,397 +131,409 @@ "issue:45237|issue:46015", "issue:43039|issue:44805" ], - "prepared_review_unit_hash": "ddc64a7e49b2fc83679c6bfef181b009775c3b6a379f29cd1724f461eea0ce94", + "prepared_review_unit_hash": "cfec7e21f83a9b79e252126344e54f65642b52412576616ed7d1c7318463b481", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 62, - "serialized_chars": 46274, - "estimated_input_tokens": 11569, - "estimated_eval_tokens": 23394 + "serialized_chars": 46278, + "estimated_input_tokens": 11570, + "estimated_eval_tokens": 23396 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous, but it does contain a few tight duplicate families: the strongest is the DeepSpeed ZeRO-3 Bert index-out-of-bounds pair, plus a strong Qwen3_5MoeTextRotaryEmbedding offload/materialization pair, and a torch 2.4.1 incompatibility pair that looks like the same version-compatibility bug rather than an exact title match.", - "confidence": 0.77, - "canonical_issue_reason": "issue:43596 is the clearest representative of the strongest duplicate family: it names the Bert + DeepSpeed ZeRO-3 index-out-of-bounds path directly, and 43638 describes the same failure mode with very similar context.", + "summary": "This cluster is mostly a broad soft-similarity bucket, not a true duplicate set. The only clear duplicate pair is the two DeepSpeed ZeRO-3/BERT IndexError reports; the rest span unrelated model, docs, trainer, and device-mapping bugs.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 43596 is the best representative because it matches the only clearly duplicate issue (43638) and its title is the more general form of the shared DeepSpeed ZeRO-3/BERT IndexError.", "canonical_pr_reason": null, - "best_issue_reason": "Best overall cluster representative because it is a concrete, reproducible bug report with the strongest duplicate evidence attached to it.", + "best_issue_reason": "Issue 43596 is the strongest canonical choice for the duplicate subcluster; it is the clearest, most general statement of the shared bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different Qwen3.5 failures in different workflows; no clear shared code path." + "reason": "Different Qwen3.5 problems: missing config field vs streaming chat/completions API failure." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Different runtime/device errors on unrelated execution paths." + "reason": "Unrelated failures: AMD GPU support for GPT-OSS-20B vs an MPS histogram implementation error in MoE." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Model-specific dtype mismatch vs AMD GPU model failure; not the same bug." + "reason": "Different models and failure modes: image dtype mismatch vs GPT-OSS on AMD GPUs." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Both mention device/runtime issues, but the model and failure modes differ." + "reason": "Different bugs: autocast dtype mismatch in Phi-tiny-MoE vs GPT-OSS AMD GPU failure." }, { "left": "issue:45237", "right": "issue:46468", "accept": false, - "reason": "Broken example vs GPU model failure; unrelated." + "reason": "Different model/example issues with no shared concrete code path." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model/runtime bugs with no shared root cause." + "reason": "Different MoE/GPU problems; not the same bug." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "Distributed collective mismatch vs ONNX export error; different code paths." + "reason": "DeepSpeed collective mismatch is unrelated to ModernBERT ONNX export." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "ONNX export issue vs Qwen3Moe FP8 failure; not the same bug." + "reason": "Different model/export issue vs Qwen3Moe FP8 failure." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "Different subsystems and different failures; not duplicates." + "reason": "ONNX export error is unrelated to Qwen3 auto device-map cuda assert." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Task crash vs docs/example cleanup; unrelated." + "reason": "Table QA crash and outdated RAG docs are unrelated." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "Export issue vs import failure in a different module; not the same code path." + "reason": "ONNX export vs import-path breakage are different problems." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions, but they target different removed items and pages." + "reason": "Both are v5 docs cleanup items, but they target different removed features/pages and are not clearly the same bug." }, { "left": "issue:44918", "right": "issue:46165", "accept": false, - "reason": "Qwen embedding unpacking bug vs resume-training/remote-code issue." + "reason": "TRl SFT embedding unpacking and training-resume/remote-code issues are unrelated." }, { "left": "issue:46165", "right": "issue:46869", "accept": false, - "reason": "Different model/workflow failures; no common underlying bug." + "reason": "Training resume with remote code is unrelated to multi-GPU YOLO device-map auto failure." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Mask/tensor shape mismatch vs device-map placement mismatch." + "reason": "Mask/tensor shape mismatch and Gemma4 PLE device mismatch are different failures." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Superficially similar IndexErrors, but different model paths and failure surfaces." - }, - { - "left": "issue:43638", - "right": "issue:46165", - "accept": false, - "reason": "DeepSpeed Bert index error vs training-resume remote-code issue." + "reason": "Both mention index errors, but one is DeepSpeed ZeRO-3/BERT initialization and the other is a mask-shape mismatch." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Different model bugs; only the exception class is similar." + "reason": "Same broad IndexError theme only; different underlying code paths and triggers." }, { "left": "issue:44805", "right": "issue:46294", "accept": false, - "reason": "Mask shape mismatch vs video token/template mismatch; different causes." + "reason": "Mask-shape indexing error is unrelated to Qwen3.5 video token/feature mismatch." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Different model/runtime failures and different symptoms." + "reason": "Different runtime failures: mask/index mismatch vs GPT-OSS MoE tensor-parallel issue." }, { "left": "issue:46165", "right": "issue:46468", "accept": false, - "reason": "Resume-training issue vs broken official example; unrelated." + "reason": "Training-resume issue is unrelated to the broken Nemotron example." }, { "left": "issue:44560", "right": "issue:46165", "accept": false, - "reason": "Video StopIteration issue vs resume-training issue; unrelated." + "reason": "Video StopIteration and training resume/remote code are unrelated." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "Processor voice preset bug vs ONNX export issue." + "reason": "BarkProcessor voice_preset bug is unrelated to ModernBERT ONNX export." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Collective mismatch vs Bert ZeRO-3 index error; different failures." + "reason": "Different DeepSpeed-related failures: collective mismatch vs BERT zero-3 index error." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same DeepSpeed ZeRO-3 Bert index-out-of-bounds bug family." + "reason": "Same underlying DeepSpeed ZeRO-3/BERT initialization bug causing the same IndexError; these are clear duplicates." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger cross-entropy call issue vs GPT-OSS tensor-parallel issue; unrelated." + "reason": "Liger Kernel cross_entropy call and GPT-OSS tensor-parallel issue are unrelated." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Detokenization bug vs LoRA merge/save mismatch; different problems." + "reason": "Different problems: detokenization bug vs LoRA save/reload mismatch." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Different model/runtime failures with unrelated symptoms." + "reason": "Qwen-3-NVFP4 runtime error and LoRA save/reload mismatch are unrelated." }, { "left": "issue:44841", "right": "issue:46661", "accept": false, - "reason": "Processor failure vs remote-code config_class import error." + "reason": "Processor failure for Voxtral is unrelated to remote-code model_class/config_class errors." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Training weight shape error vs collective mismatch; different root causes." + "reason": "Different DeepSpeed training errors with different symptoms and likely different root causes." }, { "left": "issue:43716", "right": "issue:46869", "accept": false, - "reason": "Different device/runtime bugs on different model families." + "reason": "Different device-mapping failures for different model families." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Activation checkpointing/FSDP mismatch vs mask indexing error; not duplicates." + "reason": "Activation-checkpointing/FSDP tensor-size recompute mismatch is unrelated to mask-shape indexing error." }, { "left": "issue:45685", "right": "issue:46869", "accept": false, - "reason": "MPS histogram op failure vs multi-GPU device-map issue." + "reason": "MPS MoE histogram error and YOLO multi-GPU device-map failure are unrelated." }, { "left": "issue:45859", "right": "issue:45902", - "accept": true, - "reason": "Same Qwen3_5MoeTextRotaryEmbedding offload/materialization bug family." + "accept": false, + "reason": "Same Qwen3.5 MoE rotary-embedding area, but the reported failures are different enough that they do not clearly look like one duplicate bug." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "FSDP checkpointing bug vs trainer API signature mismatch." + "reason": "FSDP activation-checkpointing mismatch and CustomTrainer compute_loss signature breakage are unrelated." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "DeepSpeed Bert index error vs GPT-OSS TP issue; different model paths." + "reason": "BERT zero-3 index error and GPT-OSS MoE tensor-parallel issue are unrelated." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "Different model/runtime bug; no shared fix." + "reason": "Different models and different failure modes." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Different model-specific failures." + "reason": "Dtype mismatch in Mistral is unrelated to GraniteMoEHybrid invalid method call." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Different models, environments, and failure types." + "reason": "Different models and runtime environments; not the same bug." }, { "left": "issue:45874", "right": "issue:46294", "accept": false, - "reason": "Reverse embedding lookup issue vs video feature/token mismatch." + "reason": "Gemma4 reverse-embedding lookup and Qwen3.5 video chat-template mismatch are unrelated." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Gemma3 config attribute error vs tokenizer regex patch crash." + "reason": "Gemma3TextConfig v4.53/vLLM attribute issue and Mistral regex patch crash are unrelated." }, { "left": "issue:45800", "right": "issue:46515", - "accept": true, - "reason": "Same torch 2.4.1 incompatibility family, even if the titles are not identical." + "accept": false, + "reason": "Both mention torch 2.4.1 compatibility, but they are separate incompatibility reports with no shared concrete bug." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Device-map mismatch vs expensive reverse embedding lookup." + "reason": "Different Gemma4 failures: device mismatch vs expensive reverse embedding lookup." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "Different distributed/device placement bugs." + "reason": "FSDP activation-checkpointing mismatch is unrelated to Qwen3 auto device-map cuda assert." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Version incompatibility vs model-specific inference failure." + "reason": "General Torch incompatibility and Qwen-3-8B NVFP4 runtime error are unrelated." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Autocast dtype mismatch vs GPT-OSS tensor-parallel issue." + "reason": "Different model families and different failure modes." }, { "left": "issue:43541", "right": "issue:43716", "accept": false, - "reason": "Torch dynamo tracing error vs multimodal dtype mismatch; not the same bug." + "reason": "Mixtral grouped_mm tracing error is unrelated to Mistral image-preprocessor dtype mismatch." }, { "left": "issue:43828", "right": "issue:44292", "accept": false, - "reason": "Different models and different failure surfaces." + "reason": "Autocast dtype mismatch in Phi-tiny-MoE and Qwen-3-8B NVFP4 runtime error are unrelated." }, { "left": "issue:43716", "right": "issue:45685", "accept": false, - "reason": "Different device/runtime bugs with no common code path." + "reason": "Different models and hardware/runtime paths." }, { "left": "issue:45507", "right": "issue:45685", "accept": false, - "reason": "Invalid method call vs unsupported MPS op; unrelated." + "reason": "Invalid method call in GraniteMoEHybrid is unrelated to MPS histogram not implemented." }, { "left": "issue:44292", "right": "issue:44918", "accept": false, - "reason": "Different Qwen model bugs with different failure modes." + "reason": "Qwen NVFP4 runtime error and Qwen3.5 embedding unpacking with TRL are different bugs." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Different model/runtime failures." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT unpacking issue is unrelated to GPT-OSS tensor-parallel failure." }, { "left": "issue:43039", "right": "issue:44918", "accept": false, - "reason": "Liger kernel cross-entropy issue vs Qwen3.5 embedding unpacking." + "reason": "Liger Kernel cross_entropy call is unrelated to TRL SFT embedding unpacking." }, { "left": "issue:41762", "right": "issue:43638", "accept": false, - "reason": "Both are DeepSpeed ZeRO-3 index errors, but the model/code paths differ." + "reason": "Gemma3 ZeRO-3 loading index error and BERT ZeRO-3 index error are similar only superficially; different model-specific roots." }, { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions, but they concern different removed features." + "reason": "Both are v5 docs issues, but one is about pipeline tasks and the other about the transformers run command; not clearly one duplicate change." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Different device/runtime failures; no common underlying bug." + "reason": "Phi-tiny-MoE autocast mismatch and MPS histogram error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Liger Kernel cross_entropy usage and Phi-tiny-MoE autocast mismatch are unrelated." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "Distributed collective mismatch vs trainer API change." + "reason": "DeepSpeed collective mismatch and trainer loss-signature change are unrelated." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "Activation checkpointing/FSDP issue vs import failure." + "reason": "FSDP activation-checkpointing issue is unrelated to the GenerationMixin import breakage." }, { "left": "issue:45237", "right": "issue:46015", "accept": false, - "reason": "GPU model failure vs RAG docs issue." + "reason": "GPT-OSS AMD GPU failure and RAG docs examples are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Different model/runtime failures; not duplicates." + "reason": "Liger Kernel cross_entropy issue is unrelated to tensor mask shape mismatch." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster-level picture is mostly grounded, but the summary overstates one pair: calling 45800/46515 an \"exact\" torch 2.4.1 incompatibility pair is stronger than the packet supports (their titles are similar, but not identical, and there\u2019s no deterministic evidence). The DeepSpeed ZeRO-3 Bert pair and the Qwen3_5MoeTextRotaryEmbedding pair are plausible, but should be described more conservatively as likely/strong duplicate families rather than exact matches." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The canonical choice of issue:43596 over issue:43638 is reasonable given the nearly identical DeepSpeed ZeRO-3/BERT IndexError titles. The soft-edge verdicts are mostly cautious false negatives for clearly different bug families, and I did not see an overstatement that would require rejection." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -537,7 +549,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36010", "issue:39290", @@ -605,8 +617,8 @@ "issue:44863|issue:46015", "issue:45020|issue:46304", "issue:45812|issue:45920", - "issue:44292|issue:46015", "issue:43866|issue:44863", + "issue:44292|issue:46015", "issue:43872|issue:46304", "issue:39290|issue:43531", "issue:41093|issue:43638", @@ -631,286 +643,286 @@ "issue:41093|issue:41762", "issue:45399|issue:46661" ], - "prepared_review_unit_hash": "3c023c2ce450dac5631406f8f2683a26f81dcce908aaa717709ba73860b4864b", + "prepared_review_unit_hash": "61c6bae1e435fbe0cde27bb16809303b6b556479392ab196c9cc5280f8aca99d", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 41, - "serialized_chars": 39733, - "estimated_input_tokens": 9934, - "estimated_eval_tokens": 20124 + "serialized_chars": 39738, + "estimated_input_tokens": 9935, + "estimated_eval_tokens": 20126 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Almost all candidates are distinct bugs or model-specific failures. The only clear duplicate-like pair is the AutoTokenizer v5 regression affecting multiple model families, where one report is a narrow Granite-specific instance and the other is the broader umbrella issue.", - "confidence": 0.9, - "canonical_issue_reason": "Choose issue 45920 as canonical: it describes the broad v5 AutoTokenizer token-ID regression across multiple models, which subsumes the narrower Granite-only report in 45812.", + "summary": "Mostly unrelated issue reports. The only strong duplicate signal is the AutoTokenizer wrong-token-IDs regression pair (Granite-specific vs broader multi-model report); the rest are different model, device-mapping, docs, or backend-compatibility bugs.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 45920 is the broadest and most representative report of the tokenizer regression, covering multiple model families rather than a single Granite-only case.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45920 is the best representative because it is the most general, highest-signal description of the underlying regression and covers the widest set of affected models.", + "best_issue_reason": "Issue 45920 is the best global representative because it captures the shared v5 AutoTokenizer regression with the widest impact and clearest duplicate cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45800", "right": "issue:46304", "accept": false, - "reason": "Different failures: torch/Transformers version incompatibility vs USE_DISTRIBUTED=0 breakage." + "reason": "Both mention version/torch incompatibility, but they describe different failures and code paths." }, { "left": "issue:43296", "right": "issue:46304", "accept": false, - "reason": "PaddleOCR-VL/vLLM load failure is unrelated to PyTorch distributed support." + "reason": "PaddleOCR-VL loading in vLLM is unrelated to the USE_DISTRIBUTED=0 PyTorch support regression." }, { "left": "issue:43866", "right": "issue:46402", "accept": false, - "reason": "Different models and symptoms: corrupted Ovis2 checkpoint vs Sam3 text-encoder load issue." + "reason": "Checkpoint corruption and missing text-encoder weights are different defects." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Gemma3 sliding_window_pattern attr error and Qwen3 device-map cuda assert are different bugs." + "reason": "Gemma3 sliding_window_pattern missing in vLLM is unrelated to Qwen3 auto device-map CUDA asserts." }, { "left": "issue:41762", "right": "issue:43596", "accept": false, - "reason": "Both mention ZeRO-3 IndexError, but they involve different models and likely separate root causes." + "reason": "Both are ZeRO-3 index errors, but the models and likely failure points differ; not enough evidence of the same bug." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "Two unrelated documentation issues with different APIs and behavior changes." + "reason": "One is a pipeline docs mismatch, the other is obsolete run-command docs; different documentation issues." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "AMD GPU support problem vs table-question-answering crash; no shared code path." + "reason": "AMD GPU model failure and table-question-answering crash are unrelated." }, { "left": "issue:44589", "right": "issue:46304", "accept": false, - "reason": "Float8 storage type error is unrelated to the distributed build/support regression." + "reason": "Float8 storage lookup failure is a different backend/runtime issue from the distributed-build regression." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger cross-entropy routing and Qwen3-VL video StopIteration are different issues." + "reason": "Liger Kernel cross-entropy routing and Qwen3-vl video StopIteration are different code paths." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Different multimodal model failures: Voxtral processor vs Gemma4 device mismatch." + "reason": "Voxtral processor failure and Gemma4 device-mapping mismatch are unrelated." }, { "left": "issue:45823", "right": "issue:46294", "accept": false, - "reason": "Both are multimodal, but one is device placement and the other is chat-template video token mismatch." + "reason": "Both involve model loading/input handling, but they are distinct model-specific bugs." }, { "left": "issue:43975", "right": "issue:46015", "accept": false, - "reason": "Detokenization bug is unrelated to outdated RAG docs." + "reason": "Incorrect detokenization and outdated RAG docs are unrelated." }, { "left": "issue:43866", "right": "issue:46772", "accept": false, - "reason": "Checkpoint corruption and ModelOpt compatibility are unrelated." + "reason": "Checkpoint corruption is unrelated to ModelOpt compatibility." }, { "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "NemotronH checkpoint loading and RAG docs are different problem areas." + "reason": "NemotronH checkpoint loading and outdated RAG docs do not match." }, { "left": "issue:45020", "right": "issue:46304", "accept": false, - "reason": "Remote-code breakage is broader but not the same concrete issue as USE_DISTRIBUTED=0 support." + "reason": "Remote-code model breakage and USE_DISTRIBUTED=0 support issues are different regressions." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Same underlying AutoTokenizer v5 regression; 45920 is the broader umbrella report and 45812 is a specific Granite instance." + "reason": "Both report the same v5 AutoTokenizer wrong-token-IDs regression; 45920 is the broader multi-model manifestation." }, { - "left": "issue:44292", - "right": "issue:46015", + "left": "issue:43866", + "right": "issue:44863", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error is unrelated to RAG documentation." + "reason": "Corrupted checkpoint versus model implementation loading failure are different problems." }, { - "left": "issue:43866", - "right": "issue:44863", + "left": "issue:44292", + "right": "issue:46015", "accept": false, - "reason": "Different model-loading failures with no shared symptom or fix path." + "reason": "Qwen-3 NVFP4 runtime failure is unrelated to RAG docs." }, { "left": "issue:43872", "right": "issue:46304", "accept": false, - "reason": "bitsandbytes Int8Params incompatibility is not the same as PyTorch distributed build support." + "reason": "bitsandbytes Int8Params incompatibility is unrelated to the distributed build regression." }, { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding_window, but the reported failures and model/config contexts differ." + "reason": "Both mention sliding_window/Qwen3, but one is Gemma3/vLLM config missing and the other is a Qwen3-MoE issue." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "Similar IndexError text, but different shapes/models and not enough evidence of one bug." + "reason": "Mask shape mismatch and ZeRO-3 zero-size index errors are different failure modes." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper run failure and table QA crash are unrelated." + "reason": "Whisper loading failure and table-question-answering crash are unrelated." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "GLM-4.7-Flash load failure is unrelated to Ovis2 checkpoint corruption." + "reason": "Model load failure in tests and corrupted checkpoint are not the same bug." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch version incompatibility and Pydantic PretrainedConfig regression are different issues." + "reason": "Torch 2.10 incompatibility and pydantic model-field breakage are unrelated." }, { "left": "issue:43541", "right": "issue:47228", "accept": false, - "reason": "Different models and compiler regressions: Mixtral grouped_mm tracing vs Sam3 bf16 compilation." + "reason": "Different models, different compiler/tracing failures, no shared concrete code path." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Detokenization bug and table QA crash are unrelated." + "reason": "Detokenization bug and table-QA crash are unrelated." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper runtime failure is unrelated to the Pydantic model regression." + "reason": "Whisper load failure and PretrainedConfig/pydantic regression are unrelated." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both involve Qwen3.5, but one is transformers chat crash and the other is streaming chat/completions API failure." + "reason": "Both involve Qwen3.5, but one is chat crash and the other is streaming chat/completions API failure; not the same bug." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "Similar IndexError wording but different model/loading paths and different root causes." + "reason": "ZeRO-3 BertModel index error and torch/version incompatibility are different issues." }, { "left": "issue:43541", "right": "issue:46869", "accept": false, - "reason": "Mixtral torch-dynamo tracing and YOLO multi-GPU auto-device-map failures are unrelated." + "reason": "Torch Dynamo grouped_mm tracing failure and YOLO auto device-map failure are unrelated." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "NVFP4 runtime error does not match table QA crash semantics." + "reason": "NVFP4 runtime failure and table-question-answering crash do not share a concrete code path." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Gemma3 config attribute error is unrelated to Gemma4 PLE device mismatch." + "reason": "Different models and symptoms; one is missing config attr, the other is device mismatch." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Streaming inference failure and MPS histogram-op error are different code paths." + "reason": "Chat/completions streaming failure is unrelated to the MPS histogram_int bug." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to Gemma3 sliding_window_pattern." + "reason": "ImportError for GenerationMixin and Gemma3 sliding_window_pattern error are unrelated." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Torch dynamo grouped_mm tracing failure and MPS histogram op issue are unrelated." + "reason": "Grouped_mm tracing failure and MPS histogram_int error are different backend issues." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Different model/dtype issues: Mixtral tracing vs Phi-tiny-MoE autocast mismatch." + "reason": "Grouped_mm tracing failure and autocast dtype mismatch are not the same bug." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Processor failure and template-node compilation error are different problems." + "reason": "Processor failure and non-template-node compile error are unrelated." }, { "left": "issue:45507", "right": "issue:46869", "accept": false, - "reason": "Invalid method call in GraniteMoEHybrid is not the same as YOLO multi-GPU auto device-map failure." + "reason": "GraniteMoEHybrid invalid method call and YOLO multi-GPU auto device-map failure are different issues." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Mistral3 weight tying bug is unrelated to Ovis2 checkpoint corruption." + "reason": "LM-head weight tying bug and checkpoint corruption are unrelated." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are IndexErrors, but the tensor shapes, models, and loading contexts differ." + "reason": "Mask/tensor shape mismatch and ZeRO-3 loading index error are different failures." }, { "left": "issue:45399", "right": "issue:46661", "accept": false, - "reason": "Flash-attn fallback gating and remote-code config_class AttributeError are unrelated." + "reason": "Flash-Attn fallback gating and remote-code config_class AttributeError are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet: issue 45920 is clearly the broader v5 AutoTokenizer regression, and 45812 is a narrower Granite-specific instance. The soft-edge verdicts are conservative overall and do not overclaim duplicates; the pair reasoning stays aligned with the issue titles and available metadata." + "feedback": "The summary is well grounded in the packet. The only clearly supported duplicate signal is the AutoTokenizer v5 regression pair (45812/45920), and choosing 45920 as the broader canonical issue is reasonable from the titles alone. The other soft-edge rejections stay conservative and do not over-merge distinct titles into the same bug." }, "evaluator_used": true, "retried": false, @@ -929,7 +941,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:39186", "issue:40990", @@ -990,8 +1002,8 @@ "issue:43828|issue:45464", "issue:43366|issue:45464", "issue:44291|issue:45070", - "issue:44387|issue:45005", "issue:43531|issue:45070", + "issue:44387|issue:45005", "issue:46869|issue:47228", "issue:45507|issue:47228", "issue:41093|issue:43596", @@ -1017,254 +1029,251 @@ "issue:44918|issue:45507", "issue:45183|issue:46463" ], - "prepared_review_unit_hash": "ea4cbb319017ac37c6eadc418c5cd517f36c994ef0656ccdaf0978a81d68a87a", + "prepared_review_unit_hash": "e2a62ed698009476ed068f2f1b270cd6b75c007bd740d66f652d4ebefd8280ef", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38034, - "estimated_input_tokens": 9509, - "estimated_eval_tokens": 19274 + "serialized_chars": 38037, + "estimated_input_tokens": 9510, + "estimated_eval_tokens": 19276 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated issue reports. A few soft duplicates appear around shared validation, remote-code loading, SigLIP2 behavior, and Qwen video/template regressions; the rest are distinct bugs or regressions.", - "confidence": 0.68, - "canonical_issue_reason": "issue:45020 is the best canonical issue here because it is a clear, broadly scoped regression report about recent transformers versions breaking remote-code models, with an explicit affected surface and strong generality.", + "summary": "These are mostly unrelated issue reports; none of the soft pairs look like true duplicates. The only loose thematic overlap is around remote-code/model-loading regressions, but the concrete failures differ too much to merge.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45020 is the broadest, most umbrella-style report in the set: a recent-transformers regression affecting remote-code models. It best serves as the representative issue for the closest thematic subcluster without overfitting to one model or one error string.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the strongest standalone issue to anchor the cluster: it is concrete, broadly applicable, and likely to subsume related remote-code loading regressions if more duplicates surface.", + "best_issue_reason": "issue:45020 is the most generally useful and central issue among the candidates, with broad impact across custom/remote-code model loading paths and the clearest chance of subsuming nearby regressions.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Both are config/field regressions in v5, but one is about StableLmConfig.pad_token_idx and the other is a broader PretrainedConfig pydantic-field break; not the same concrete bug." + "reason": "Different bugs: missing pad_token_idx after the 5.0 update vs pydantic field breakage in v5.4.0." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "One is a torch/transformers version incompatibility; the other is a Gemma4 device mismatch under device_map=auto. Different failure modes and code paths." + "reason": "Both are compatibility/device-related, but one is a torch/transformers version incompatibility and the other is a device mismatch in Gemma4 PLE forward." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss architecture support are unrelated model/loading issues." + "reason": "Different failures: PaddleOCR-VL loading with transformers 4.53.0 vs GGUF support for gpt-oss architecture." }, { "left": "issue:43872", "right": "issue:47228", "accept": false, - "reason": "bitsandbytes _is_hf_initialized incompatibility and a Sam3 bf16 compilation regression are separate problems." + "reason": "Different incompatibilities: bitsandbytes _is_hf_initialized constructor error vs a Sam3 bf16 compilation regression." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Both mention shape problems, but one is a mask/indexing error and the other is an SFT-trainer embedding unpacking failure; not enough to be the same bug." + "reason": "Both show shape/indexing errors, but the contexts differ: mask/tensor length mismatch vs Qwen3.5 embedding unpacking with TRL SFT." }, { "left": "issue:43716", "right": "issue:47228", "accept": false, - "reason": "Mistral-3 image-preprocessor dtype mismatch is unrelated to the Sam3 bf16 compilation regression." + "reason": "Different models and root causes: image preprocessor dtype mismatch vs a bf16 compilation regression." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "A torch.autocast dtype mismatch in Phi-tiny-MoE is not the same as a Qwen3.5 streaming chat/completions failure." + "reason": "Autocast dtype mismatch for Phi-tiny-MoE is not the same as Qwen3.5 streaming chat/completions failure." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF gpt-oss support and Qwen3.5 streaming inference are different model/runtime issues." + "reason": "GGUF gpt-oss support is unrelated to Qwen3.5 streaming inference." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "The _is_hf_initialized TypeError from init_empty_weights is distinct from the pydantic PretrainedConfig field regression." + "reason": "Both involve init/model loading regressions, but one is the unexpected _is_hf_initialized argument and the other is a pydantic PretrainedConfig field breakage." }, { - "left": "issue:44387", - "right": "issue:45005", + "left": "issue:43531", + "right": "issue:45070", "accept": false, - "reason": "CUDA memory growth under int4 quantization and tied-weights regressions are different code paths." + "reason": "Qwen3-MoE sliding_window behavior is a different model-specific issue from the v5.4 PretrainedConfig/pydantic problem." }, { - "left": "issue:43531", - "right": "issue:45070", + "left": "issue:44387", + "right": "issue:45005", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to the PretrainedConfig field/pydantic breakage." + "reason": "Int4 quantization OOM is not the same as tied-weights issues in translation models." }, { "left": "issue:46869", "right": "issue:47228", "accept": false, - "reason": "YOLO multi-GPU device_map issues and Sam3 bf16 compilation regression are not the same bug." + "reason": "Multi-GPU device map auto for YOLO is unrelated to Sam3 bf16 compilation." }, { "left": "issue:45507", "right": "issue:47228", "accept": false, - "reason": "GraniteMoEHybrid invalid method calls do not match the Sam3 compilation regression." + "reason": "GraniteMoEHybrid invalid method calls are a different bug than a compilation regression." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "These are both IndexErrors, but one is a mask-length mismatch and the other is a zero-sized tensor indexing failure under deepspeed zero3; different underlying causes." + "reason": "Both are IndexErrors, but the concrete shapes and code paths differ; no evidence they are the same bug." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code model breakage and torch/transformers incompatibility are too broad and distinct to merge as one bug." + "reason": "A remote-code regression and a torch/transformers version incompatibility are too broad and not the same concrete failure." }, { "left": "issue:43716", "right": "issue:46294", "accept": false, - "reason": "Image-preprocessor dtype mismatch is not the same as Qwen3.5 video token/feature mismatch." + "reason": "Both touch multimodal/video-ish paths, but one is image dtype mismatch and the other is a video feature/token count mismatch." }, { "left": "issue:44479", "right": "issue:46294", - "accept": true, - "reason": "Both point to the same Qwen3.5/Qwen3.5-MoE video-input regression path; the latter is a specific apply_chat_template manifestation of the broader video regression." + "accept": false, + "reason": "Both concern Qwen multimodal/video models, but one is a v5.3 video input regression and the other is a specific apply_chat_template feature/token mismatch." }, { "left": "issue:43054", "right": "issue:43994", - "accept": true, - "reason": "Both report SigLIP2 producing bad outputs through the model/pipeline path; the second is a more concrete symptom of the same incorrect behavior." + "accept": false, + "reason": "Siglip2 text embeddings being worse than Siglip is a model-quality report, not the same as AutoModel/pipeline nonsensical outputs." }, { "left": "issue:46468", "right": "issue:46869", "accept": false, - "reason": "The Nemotron example breakage is unrelated to YOLO multi-GPU device_map behavior." + "reason": "Official Nemotron example broken is unrelated to YOLO multi-GPU device map behavior." }, { "left": "issue:43596", "right": "issue:46165", "accept": false, - "reason": "Zero3 initialization indexing failure and remote-code resume/training issues are different problems." + "reason": "Different loading flows: deepspeed zero3/BertModel index error vs resuming training with local model and trust_remote_code." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite Qwen config unwrapping and uninitialized inv_freq after to_empty are different model initialization bugs." + "reason": "Both involve Qwen multimodal configs, but one is AutoModel.from_config text_config unwrapping and the other is uninitialized inv_freq after to_empty materialization." }, { "left": "issue:46640", "right": "issue:46661", "accept": false, - "reason": "Flash Attention 2 issues and remote-code model_class/config_class AttributeError are unrelated." + "reason": "Flash Attention 2 issue and remote-code model_class.config_class AttributeError are different code paths." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP weight-shape runtime error and Qwen3.5 TRL unpacking failure are different stack traces and code paths." + "reason": "FSDP 2-D weight RuntimeError is unrelated to Qwen3.5 embedding unpacking with TRL." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Mistral-3 dtype mismatch and Qwen3.5 embedding unpacking are unrelated." + "reason": "Image preprocessor dtype mismatch is not the same bug as Qwen3.5 embedding unpacking failures." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Custom-model initialization in Jupyter and flash-attn2 fallback blocking are separate regressions." + "reason": "Jupyter custom model initialization breakage is different from flash-attn2 fallback being blocked by other checks." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "High perplexity on gpt-oss-20b and a GLM5 issue are not the same underlying bug." + "reason": "High perplexity on gpt-oss is a model behavior issue, not the same as a GLM5 model topic." }, { "left": "issue:45632", "right": "issue:45698", - "accept": true, - "reason": "Both describe remote-code/custom-module loading going wrong due to cache/path handling; they look like the same module-resolution bug from different angles." + "accept": false, + "reason": "Both involve trust_remote_code/custom modules, but one is cache-path collision and the other is loading the wrong module after save_pretrained." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "Qwen3MoE FineGrainedFP8Config failure and flash-attn2 fallback checks do not share a concrete code-path problem." + "reason": "Qwen3Moe FineGrainedFP8Config failure is unrelated to flash-attn2 fallback logic." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "A specific Mistral3 lm_head tying bug is not the same as the broader translation-model tied-weights regression." + "reason": "Mistral3 tied lm_head bug is a model-specific weight-tying issue, not translation-model tied-weights across v5." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "DSA indexer/ReLU discussion and a docs-only run-command issue are unrelated." + "reason": "A DSA indexer discussion and docs mentioning a removed run command are different kinds of changes/problems." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization offloading behavior is unrelated to tied-weights regressions." + "reason": "Quantization offloading behavior is not the same as tied-weights regression in translation models." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A typo in cache strategies and stale docs about the removed run command are different documentation issues." + "reason": "A cache strategy typo and outdated docs about a removed command are unrelated." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache group-type crashes on Qwen3.5 do not match tied-weights translation-model regressions." + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weights behavior in translation models." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Qwen3.5 embedding unpacking and GraniteMoEHybrid invalid-method calls are different bugs." + "reason": "Qwen3.5 embedding unpacking and GraniteMoEHybrid invalid method calls are different failures." }, { "left": "issue:45183", "right": "issue:46463", - "accept": true, - "reason": "Both are validation/XOR logic problems producing wrong or misleading input_ids vs inputs_embeds errors; the T5 case looks like a model-specific manifestation of the same shared validation bug." + "accept": false, + "reason": "Both are validation/error-message bugs, but one is general XOR input validation and the other is T5 decoder input validation." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is well grounded in the packet, and the soft-edge decisions are mostly conservative. The accepted pairs are plausibly the same bug class (Qwen video regression, SigLIP2 behavior, remote-code/custom-module loading, and XOR validation messaging). The rejected pairs are also appropriately cautious given the distinct titles and failure modes. No clear overstatement stands out." - }, - "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 }, @@ -1279,7 +1288,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:24643", "issue:28530", @@ -1362,220 +1371,220 @@ "issue:43824|issue:43872", "issue:41553|issue:43883" ], - "prepared_review_unit_hash": "a60ef6bdd5bc4719a43f93d84770a2edaae31182457460ee92f0895fdd693041", + "prepared_review_unit_hash": "094e39ecfc0b9635a3af2b61177f6871281bee2dba307ec17651bc22393ab1b1", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 37012, - "estimated_input_tokens": 9253, - "estimated_eval_tokens": 18762 + "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 Transformers bug reports, but a few pairs are true duplicates or near-direct duplicates: the Gemma4 4-bit dtype-cast reports, the StaticCache.get_seq_length API regression, and the SigLIP2 implementation-vs-reference mismatch. Most other soft matches are only same-subsystem similarities, not the same bug.", - "confidence": 0.66, - "canonical_issue_reason": "issue:43950 is the clearest self-contained v5 regression in the set: it names the exact broken code path (`from_pretrained` and non-persistent buffers), has a concrete symptom, and is more actionable than the broader/less specific reports.", + "summary": "This cluster is mostly a loose bundle of unrelated Transformers issues, with only a few true duplicate subthreads. The clearest duplicate groups are around non-persistent buffer corruption, `get_seq_length()` returning a tensor instead of an int, and Gemma4 quantized embedder dtype casting.", + "confidence": 0.72, + "canonical_issue_reason": "#43950 is the strongest central representative of a concrete v5 regression: it states the non-persistent-buffer corruption bug directly and also anchors related reports that look like the same failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "Same as above: it is the most concrete, generalizable bug report in the cluster and a good anchor for related v5 regression triage.", + "best_issue_reason": "#43950 is the best standalone issue to keep as the hub because it names the bug clearly, is broad enough to cover nearby duplicates, and is more specific than the umbrella-style reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Unrelated failures: DeepSeek-Coder tokenizer detokenization vs Qwen3 NVFP4 runtime error." + "reason": "Different model-specific failures; no shared code path or concrete bug." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "Different bugs: NemotronH checkpoint loading vs tied-weights handling in translation models." + "reason": "Checkpoint loading failure vs tied-weights regression; too different to be duplicates." }, { "left": "issue:46489", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but on different model families and different concrete failure modes." + "reason": "Both are tokenizer regressions, but on different models with different symptoms." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "SigLIP2-related, but one is docs/example errors and quantization failure while the other is model quality/discrepancy; not the same bug." + "reason": "Docs/example breakage vs model quality discrepancy; not the same bug." }, { "left": "issue:43950", "right": "issue:44534", - "accept": false, - "reason": "Same broad v5 regression theme, but the titles point to different concrete bugs and code paths." + "accept": true, + "reason": "Same v5 non-persistent-buffer corruption regression described with nearly identical wording." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Different training/API problems: custom compute_loss signature vs DeepSpeed ZeRO-3 load failure." + "reason": "Trainer callback signature issue vs DeepSpeed ZeRO-3 loading error." }, { "left": "issue:46032", "right": "issue:47246", "accept": false, - "reason": "Related Mamba2 family, but one is incorrect use_cache behavior and the other is inter-chunk recurrence in a different implementation path." + "reason": "Related Mamba-family bugs, but different models and different recurrence paths." }, { - "left": "issue:43644", + "left": "issue:43643", "right": "issue:43950", "accept": false, - "reason": "Both mention regressions, but the underlying bugs are different: missing fields with trust_remote_code vs non-persistent buffer corruption." + "reason": "Remote-code config loading bug is distinct from buffer corruption." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Different incompatibilities: bitsandbytes Int8Params init vs missing Float8 storage object." + "reason": "Different bitsandbytes/storage compatibility failures." }, { "left": "issue:46899", "right": "issue:46932", "accept": true, - "reason": "Same Gemma4 quantized audio/vision dtype-casting bug; 46932 is explicitly the same issue and matches the failure described in 46899." + "reason": "Same Gemma4 quantized vision/audio dtype-casting bug; #46932 is essentially the duplicate fix report." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "Both discuss tokenization/embeddings, but one is a SigLIP2 model discrepancy and the other is a generic version-dependent tokenization change." + "reason": "SigLIP2 quality/implementation issue vs tokenization-version change." }, { "left": "issue:45920", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer regressions, but 45920 is a broad multi-model AutoTokenizer issue while 46489 is a DeepSeek-Coder-specific tokenizer bug." + "reason": "Broad tokenizer regression vs DeepSeek-Coder-specific tokenizer bug; not clearly the same defect." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Different Qwen3.5 problems: dtype mismatch under autocast vs gradient explosion from 3D position_ids/SDPA fallback." + "reason": "Different dtype/attention failures on different models and code paths." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Same model family only; one is unexpected zero probabilities, the other is embedding quality discrepancy." + "reason": "Both involve SigLIP2, but one is zero probabilities and the other is relative embedding quality; too different." }, { "left": "issue:45987", "right": "issue:46858", "accept": true, - "reason": "Same StaticCache.get_seq_length contract break: 46858 is the backend-specific manifestation of the same tensor-vs-int return bug." + "reason": "Same `cache.get_seq_length()` contract bug: it returns a tensor instead of an int and breaks downstream arithmetic." }, { "left": "issue:43950", "right": "issue:45902", - "accept": false, - "reason": "Different regression classes: non-persistent buffer corruption vs uninitialized inv_freq after meta->to_empty(cuda) materialization." + "accept": true, + "reason": "`inv_freq` uninitialized after materialization is a concrete manifestation of the same non-persistent-buffer corruption regression." }, { "left": "issue:43054", "right": "issue:43493", - "accept": true, - "reason": "Both describe the same SigLIP2 HF-vs-JAX implementation mismatch leading to degraded embeddings." + "accept": false, + "reason": "Likely related SigLIP2 implementation concerns, but not the same concrete failure." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Different Qwen3.5 bugs: flash-attention crash from 3D position_ids vs CPU-offload incompatibility in rotary embedding." + "reason": "Flash-Attn crash from 3D `position_ids` misinterpretation vs CPU-offload incompatibility." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Both are Gemma/Qwen-style model internals issues, but the code paths are unrelated: per-layer shared_kv_states rebuild vs uninitialized rotary frequency buffer." + "reason": "Different Gemma4/FSDP2 state handling issue vs non-persistent-buffer regression." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Different training/resource problems: ineffective gradient checkpointing with LoRA vs increased CUDA reserved memory under int4 quantization." + "reason": "Gradient-checkpointing/LoRA inefficiency vs int4 quantization OOM; distinct problems." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Different issues: generic load noise vs inability to load a specific model." + "reason": "Generic loading noise vs a specific tokenizer-load failure." }, { "left": "issue:24643", "right": "issue:28530", "accept": false, - "reason": "Unrelated: DeepSpeed training weight-shape RuntimeError vs early-stopping metric configuration warning." + "reason": "DeepSpeed training error vs early-stopping metric handling; unrelated." }, { "left": "issue:43975", "right": "issue:45237", "accept": false, - "reason": "Different model/runtime problems: tokenizer detokenization regression vs AMD GPU execution failure for GPT-OSS-20B." + "reason": "Tokenizer regression vs AMD GPU execution failure." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Both are loading-related, but one is noisy logging and the other is a concrete model load failure." + "reason": "Generic load noise vs a specific model loading failure." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Different model-specific load/runtime failures: Gemma3 ZeRO-3 load crash vs Qwen3Moe FP8 failure." + "reason": "ZeRO-3 loading crash vs FP8 config failure; unrelated." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "Both involve remote-code/config behavior, but the bugs are different: missing fields in AutoConfig return vs remote_code breakage in recent versions." + "reason": "#45020 is a broad umbrella for remote-code breakage, but #43643 is a specific AutoConfig missing-fields bug." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "Different performance/memory issues in different training paths; not the same bug." + "reason": "Different memory/performance issues in different training setups." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Unrelated: tokenizer load regression for a specific model vs flash-attn2 fallback gating logic." + "reason": "Tokenizer loading regression vs flash-attn2 fallback gating." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Different error classes and code paths: missing Qwen2.5 VL import vs bitsandbytes Int8Params constructor mismatch." + "reason": "Import symbol missing vs bitsandbytes parameter-constructor incompatibility." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad error message for AutoTokenizer loading Voxtral vs missing all_tied_weights_keys attribute in Molmo; not the same bug." + "reason": "Bad error message for tokenizer loading vs missing tied-weights key attribute." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The three accepted pairs are well-supported by the titles, and the rejected pairs mostly read as same-subsystem but different bugs. No overstatement stands out." + "feedback": "The summary is well grounded in the packet and stays conservative about duplicates. The three accepted subthreads are supported by the titles and pair evidence, and the rejected pairs generally look distinct enough to avoid overclaiming. No PR-mergeability issue applies here." }, "evaluator_used": true, "retried": false, @@ -1594,7 +1603,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:24643", "issue:30064", @@ -1675,208 +1684,208 @@ "issue:41093|issue:44805", "issue:44928|issue:45588" ], - "prepared_review_unit_hash": "7e37458049cfcb7862853b4ddf736a21601aedecdeb5099093659c266e134f23", + "prepared_review_unit_hash": "12b3656de958e58fb61d7822d7d7a391a315251f0515713e93a0d5de44e56e17", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36628, - "estimated_input_tokens": 9157, - "estimated_eval_tokens": 18570 + "serialized_chars": 36624, + "estimated_input_tokens": 9156, + "estimated_eval_tokens": 18568 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous model- and tokenizer-specific issue reports, not one cohesive duplicate cluster. I only see two strong duplicate threads: the docs/pipeline-removal reports and the mask-shape mismatch reports.", - "confidence": 0.61, - "canonical_issue_reason": "No single canonical issue fits the whole set; the best duplicate thread is the mask-shape mismatch pair (41093/44805), which is the clearest concrete bug report among the items.", + "summary": "This cluster is mostly heterogeneous model- and subsystem-specific bug reports. The only clear duplicate-like pair is the v5 docs cleanup around removed pipeline tasks; most other soft pairs are same-broad-area but different concrete failures and should stay separate.", + "confidence": 0.67, + "canonical_issue_reason": "issue:45020 is the broadest umbrella in the set, covering a cross-cutting remote_code regression rather than a single model-specific manifestation.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41093 is the best representative of the strongest duplicate thread because it states the concrete runtime failure clearly and matches another near-identical report.", + "best_issue_reason": "issue:45020 is the best global issue anchor because it is general enough to represent a broader regression class, while most other items are narrowly model- or subsystem-specific.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Both involve tokenizer/flash-attn adjacent code, but they describe different failures in different paths: Mistral regex patching vs flash-attn2 fallback checks." + "reason": "Tokenizer regex crash for Mistral loading vs flash-attn2 fallback gating; different code paths and failures." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "Different problems: an MPS histogram op limitation vs a device-mapping mismatch during forward." + "reason": "MPS histogram error in MoE code vs Gemma4 device-map forward mismatch; unrelated bugs." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Different model-loading failures with no clear shared code path; one is a Qwen3VL weight_only error, the other is a GLM5 loading issue." + "reason": "Qwen3VL weight_only load failure vs GLM5 loading issue; different models and symptoms." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Both mention model loading regressions, but one is about remote_code compatibility and the other is a Gemma4 device-map forward mismatch." + "reason": "Broad remote_code regression vs Gemma4 PLE device mismatch; not the same underlying bug." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "One is about add-new-model-like/tokenizer mapping handling; the other is a broad remote_code regression. Same area, not the same bug." + "reason": "TOKENIZER_MAPPING_NAMES add-new-model-like failure vs remote_code loading regression; different mechanisms." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Different concrete failures: autocast dtype mismatch vs device placement mismatch." + "reason": "Autocast dtype mismatch in Phi-tiny-MoE vs Gemma4 device placement issue; not duplicate." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Both are loading-related, but one is a tokenizer error message issue and the other is a model load failure for a different model family." + "reason": "Bad Voxtral AutoTokenizer error message vs glm-4v-9b load failure; different model-loading problems." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Remote-code loading breakage and Kimi-K2.5 tokenizer regression are unrelated code paths." + "reason": "Generic remote_code breakage vs Kimi-K2.5 tokenizer regression; different root causes." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Both mention LoRA/tie-word-embeddings, but they are different symptoms on different versions and models." + "reason": "LoRA compatibility break across HF versions vs a warning about tie_word_embeddings; related family, but 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." + "reason": "bitsandbytes Int8Params constructor incompatibility vs YOLO multi-GPU device-map failure; unrelated." }, { "left": "issue:45715", "right": "issue:46315", "accept": false, - "reason": "Both are tokenizer performance regressions, but they affect different conversion directions and different hot paths." + "reason": "convert_ids_to_tokens slowdown vs convert_tokens_to_ids slowdown; same subsystem but different hot paths and regressions." }, { "left": "issue:45759", "right": "issue:46661", "accept": false, - "reason": "Different loader bugs: composite config unwrapping vs remote-code model_class.config_class handling." + "reason": "Composite config unwrapping bug vs remote-code config_class AttributeError; similar area, but different concrete failures." }, { "left": "issue:45488", "right": "issue:46489", "accept": false, - "reason": "Both concern tokenizer regressions for DeepSeek-family models, but they point to different model classes and different tokenizer behavior changes." + "reason": "LlamaTokenizer pre-tokenizer override bug affecting DeepSeek V3/R1 vs DeepSeek-Coder v1 tokenizer regression; different underlying issues." }, { "left": "issue:45020", "right": "issue:46661", "accept": false, - "reason": "Both mention remote-code models, but the concrete failure modes are different enough that they should not be merged." + "reason": "Both concern remote_code loading, but one is a broad regression and the other a specific config_class error; not enough to treat as the same bug." }, { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "These are the same docs regression: stale references to removed pipeline tasks after v5 pipeline removals." + "reason": "Both report docs still referencing removed pipeline tasks after v5; same documentation cleanup." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Untied lm_head weights and non-persistent buffer corruption are different state-management bugs." + "reason": "Mistral3 lm_head tying bug vs silent corruption of non-persistent buffers in from_pretrained; different code paths." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Different model-loading problems for different model families; no shared concrete bug is evident." + "reason": "GLM-4v-9b loading failure vs missing all_tied_weights_keys in MolmoForCausalLM; distinct model-specific issues." }, { "left": "issue:46894", "right": "issue:46897", "accept": false, - "reason": "Very similar wording, but these are model-specific reports across different architectures; not enough evidence of one exact code-path bug." + "reason": "Same loss-shift pattern, but different model families (CohereASR vs Florence2); too model-specific to merge as duplicates." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Both are audio-related, but one is a Whisper run failure and the other is a low-level fbank feature issue." + "reason": "Whisper-large-v2 load failure vs a fbank feature-extraction issue; not clearly the same bug." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Both involve DeepSpeed, but one is weight conversion/loading and the other is gradient averaging with GAS." + "reason": "Qwen3 MOE weight-loading failure under ZeRO-3 vs gradient averaging under GAS with DeepSpeed; different failures." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Unexpected trainer kwarg vs mask/index shape mismatch are unrelated failures." + "reason": "CustomTrainer signature mismatch vs mask/tensor shape IndexError; unrelated." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Both mention gradient checkpointing, but one is ineffective LoRA checkpointing and the other is a missing flag on a remote-code variant." + "reason": "LoRA gradient checkpointing ineffective vs missing supports_gradient_checkpointing on a remote-code variant; related theme, different bug." }, { "left": "issue:43940", "right": "issue:45313", "accept": false, - "reason": "Both are Qwen3 ZeRO-3 loading failures, but the reported missing-weights failures are not clearly the same underlying bug." + "reason": "Both are ZeRO-3 load failures, but they affect different Qwen variants and describe different missing-weight patterns; not enough evidence of one underlying bug." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Different model families and different load failures; not the same bug." + "reason": "GLM-4.7-Flash unit-test loading issue vs NemotronH checkpoint loading failure; different models and causes." }, { "left": "issue:44479", "right": "issue:44560", "accept": false, - "reason": "Related video-input area, but the regression patterns and affected models differ enough that they should stay separate." + "reason": "Both are video-input regressions, but one is a broad v5.3.0 model regression and the other is a StopIteration in Qwen3-vl-embedding; not the same bug." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "Completely unrelated: DeepSpeed training weight-dimension error vs void segmentation map processing." + "reason": "DeepSpeed training RuntimeError about 2-D weights vs void segmentation map processing; completely different domains." }, { "left": "issue:41093", "right": "issue:44805", - "accept": true, - "reason": "Same underlying bug report: the same mask/tensor shape mismatch runtime error is described in both issues." + "accept": false, + "reason": "Both mention mask shape mismatch, but they are separate reports with different dimensions and contexts; not enough to unify." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "Different problems: Qwen3.5 RLHF gradient explosion from position_ids/SDPA behavior vs flash-attention s_aux=None crash." + "reason": "Qwen3.5 RLHF NaN/gradient explosion issue vs flash_attention sinkless-model AttributeError; unrelated failures." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The two accepted duplicate threads are well-supported by the titles: the docs/pipeline-removal pair (43827/44509) and the mask-shape mismatch pair (41093/44805). The rejected soft-edge pairs are also cautiously phrased and do not overstate equivalence." + "feedback": "The summary is grounded in the packet: issue:45020 is a reasonable broad anchor, the cluster is indeed mostly heterogeneous/model-specific, and the only clear duplicate-like pair identified (43827/44509) is a conservative same-bug docs cleanup match. The other soft-edge rejections stay cautious and do not overclaim duplication." }, "evaluator_used": true, "retried": false, @@ -1895,7 +1904,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36010", "issue:38175", @@ -1968,8 +1977,8 @@ "issue:43645|issue:43646", "issue:44560|issue:44918", "issue:36010|issue:42915", - "issue:44910|issue:46858", "issue:44683|issue:46515", + "issue:44910|issue:46858", "issue:43550|issue:44910", "issue:42947|issue:47254", "issue:45812|issue:46489", @@ -1981,20 +1990,20 @@ "issue:43638|issue:44661", "issue:43596|issue:44661" ], - "prepared_review_unit_hash": "2b948e958765dbd4b5cf8514eb880b955149befb84cc52bb8fbce449376db568", + "prepared_review_unit_hash": "8e342b78d56c95aded481e37859fc3ff9b75e02eb0aba88a1a8266f3f7f8e6d5", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 34, - "serialized_chars": 37341, + "serialized_chars": 37342, "estimated_input_tokens": 9336, "estimated_eval_tokens": 18928 }, @@ -2003,225 +2012,225 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated Transformers issues that were surfaced by soft similarity alone. The only clear duplicate-like pair is the two custom-model initialization regressions in Transformers 5.0.0 (issues 43645 and 43646).", - "confidence": 0.84, - "canonical_issue_reason": "Issue 43646 is the cleaner, broader statement of the same regression; 43645 is a notebook-specific phrasing of the same custom model initialization breakage.", + "summary": "Mostly heterogeneous cluster with one clear duplicate pair: the Transformers 5.0.0 custom model initialization reports (43645/43646). The SigLIP2 pair (38175/43994) looks related but not strong enough from the packet to treat as a definite duplicate.", + "confidence": 0.69, + "canonical_issue_reason": "Issue 43646 is the cleanest representative of the only clearly duplicated concrete bug: the Transformers 5.0.0 custom model initialization regression.", "canonical_pr_reason": null, - "best_issue_reason": "43646 is the best canonical issue because it captures the underlying bug without narrowing it to Jupyter notebooks.", + "best_issue_reason": "43646 is the best anchor because it states the underlying regression directly, without the notebook-specific repro detail from 43645.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43716", "right": "issue:46468", "accept": false, - "reason": "Different models and failures: image preprocessor/model dtype mismatch vs Nemotron example breakage." + "reason": "Different model families and different failure modes; only a broad loading/debugging theme overlaps." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "Broad GGUF architecture support request vs a specific autocast dtype mismatch on Phi-tiny-MoE." + "reason": "Unrelated bugs: GGUF architecture support versus an autocast dtype mismatch." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "Different problems: Qwen3.5 PagedAttentionCache linear_attention crash vs NemotronH checkpoint loading." + "reason": "PagedAttentionCache/group parsing crash and Nemotron checkpoint loading are different code paths." }, { "left": "issue:44863", "right": "issue:46402", "accept": false, - "reason": "NemotronH checkpoint loading issue vs Sam3Model text encoder weights not loading; different model families and code paths." + "reason": "Both are loading problems, but for different models and different breakages." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "Both mention Qwen3/Qwen3-MoE, but one is GGUF architecture support and the other is a sliding_window bug." + "reason": "Different feature areas: GGUF model support versus Qwen3-MoE sliding-window behavior." }, { "left": "issue:43596", "right": "issue:46515", "accept": false, - "reason": "Specific ZeRO-3/BertModel IndexError vs a generic torch 2.4.1 incompatibility claim." + "reason": "Specific ZeRO-3/Bert indexing failure versus a broad torch-version incompatibility claim." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Both are loading-related, but Voxtral tokenizer error messaging and EMBEDDIA est-roberta tokenizer failure are distinct issues." + "reason": "Both concern tokenizer/model loading, but the model and failure mechanism differ." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Different Qwen2.5-VL/Qwen3.5 rotary-position bugs with different symptoms and code paths." + "reason": "Different Qwen variants and different rotary/position-id bugs." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "GGUF support for gpt-oss is unrelated to a Float8 storage lookup error." + "reason": "Completely different issue classes: model support versus float8 storage lookup." }, { "left": "issue:43638", "right": "issue:46515", "accept": false, - "reason": "ZeRO-3 training on non-pretrained BertModel vs generic torch 2.4.1 incompatibility; not the same bug." + "reason": "The Bert/ZeRO-3 index error is not the same as a generic torch incompatibility report." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states bug vs Gemma4 PLE device mismatch under device_map='auto'." + "reason": "Same model family, but different bugs: shared_kv_states rebuild versus device mismatch." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "AudioFlamingo3 batched inference track leakage vs Whisper empty-transcription generation failure." + "reason": "Different audio/Whisper paths and different symptoms." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3Moe FP8 config failure vs meta-device loading regression; both involve loading but different defects." + "reason": "Different root causes: FP8 config failure versus meta-device loading regressions." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Mistral-3 dtype mismatch is unrelated to Qwen3-vl-embedding video StopIteration." + "reason": "Different model families and different crash modes." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Missing Qwen2.5-VL import vs meta-device loading breakage; different failure modes." + "reason": "Importing a missing class is not the same bug as meta-device loading corruption." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Non-persistent buffer corruption in from_pretrained vs Kosmos2.5 long OCR index error." + "reason": "Non-persistent buffer corruption and OCR index errors are unrelated." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "Not able to run 3d_parallel.py is unrelated to GGUF support for gpt-oss." + "reason": "Unrelated: a parallel script failure versus GGUF model support." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Voxtral AutoTokenizer error message issue vs sleng-bert loading regression; different models and symptoms." + "reason": "Both mention loading failures, but they affect different models and likely different code paths." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Qwen3-vl-embedding video StopIteration vs MPS histogram not implemented for Int in moe code." + "reason": "Different failures: Qwen video StopIteration versus an MPS histogram type issue." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same underlying Transformers 5.0.0 regression: custom model initialization breaks; 43645 is the notebook-specific phrasing." + "reason": "Same underlying Transformers 5.0.0 custom-model initialization regression; 43645 just adds a notebook-specific repro context." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Video StopIteration in Qwen3-vl-embedding vs Qwen3.5 embedding unpacking failure with TRL SFT trainer." + "reason": "Both are Qwen3.5-related, but the failure mechanisms are distinct." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import error vs Qwen3Moe FineGrainedFP8Config failure; unrelated code paths." + "reason": "Importing GenerationMixin and a FineGrainedFP8Config failure are unrelated problems." }, { - "left": "issue:44910", - "right": "issue:46858", + "left": "issue:44683", + "right": "issue:46515", "accept": false, - "reason": "Qwen3.5 FlashAttention position_ids bug vs cache.get_seq_length() returning a device tensor; different mechanisms." + "reason": "Torch-version compatibility is too broad here; the concrete bugs differ." }, { - "left": "issue:44683", - "right": "issue:46515", + "left": "issue:44910", + "right": "issue:46858", "accept": false, - "reason": "Compiled flex_attention on torch >= 2.9 vs a generic torch 2.4.1 incompatibility report." + "reason": "Both are attention/cache issues, but one is 3D position_ids on Qwen3.5 and the other is device-tensor seq-length arithmetic." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Bamba torch.compile + SDPA failure is unrelated to Qwen3.5 FlashAttention illegal memory access." + "reason": "Different models and different attention backends; not the same defect." }, { "left": "issue:42947", "right": "issue:47254", "accept": false, - "reason": "Both involve checkpointing/PEFT, but one is gradient checkpointing ineffective and the other is a CheckpointError with ZeRO-3; not the same bug." + "reason": "Both involve PEFT/checkpointing, but one is ineffective checkpointing and the other is a ZeRO-3 CheckpointError." }, { "left": "issue:45812", "right": "issue:46489", "accept": false, - "reason": "Granite AutoTokenizer wrong token IDs vs DeepSeek-Coder v1 tokenizer output regression; different models and tokenizer logic." + "reason": "Both are tokenizer regressions, but for different model families and likely different fixes." }, { "left": "issue:45823", "right": "issue:46515", "accept": false, - "reason": "Gemma4 PLE device mismatch vs generic torch 2.4.1 incompatibility; not the same underlying issue." + "reason": "A Gemma4 device-placement bug is not the same as a generic torch incompatibility." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "add-new-model-like tokenizer mapping bug vs Gemma4 PLE device mismatch; unrelated failures." + "reason": "Different problems: tokenizer-mapping validation versus Gemma4 device mismatch." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Qwen3-vl video StopIteration vs a mask/tensor shape mismatch IndexError; different symptoms and likely code paths." + "reason": "Different index-error contexts and different input pipelines." }, { "left": "issue:38175", "right": "issue:43994", "accept": false, - "reason": "Both concern siglip2-base-patch16-224, but one is zero probabilities and the other is nonsensical outputs; not enough evidence of the same bug." + "reason": "Likely related via SigLIP2 output corruption, but the packet only supports a cautious 'related' call, not a definite duplicate." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Strict config blocking granite_speech vs config validation preventing tiny-random glm4v loading; same general area, different models/issues." + "reason": "Both are config validation issues, but they involve different models and different validation changes." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "ZeRO-3 BertModel IndexError vs tokenizer mapping-name issue; unrelated." + "reason": "ZeRO-3 Bert indexing and tokenizer-mapping validation are unrelated." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "ZeRO-3 BertModel IndexError vs tokenizer mapping-name issue; different code paths and failure modes." + "reason": "Different bug class and different subsystem; only superficial error-text overlap." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only clearly duplicate-like pair is 43645/43646, and the rest of the soft-edge rejections are appropriately cautious rather than overclaiming matches. No mergeability issues are apparent for the accepted pair." + "accept": false, + "feedback": "The analysis is mostly cautious, but it overstates at least one soft-edge match. In particular, issue:38175 vs issue:43994 is accepted as the same underlying regression, which is stronger than the packet supports from titles alone; a more conservative call would be \"likely related\" rather than a definite duplicate. The rest of the negatives look reasonably grounded, but this positive edge is not conservative enough to accept the cluster as written." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -2237,7 +2246,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:38175", "issue:39692", @@ -2306,8 +2315,8 @@ "issue:46459|issue:46512", "issue:43644|issue:43645", "issue:39692|issue:43994", - "issue:44488|issue:44991", "issue:43257|issue:43854", + "issue:44488|issue:44991", "issue:43856|issue:43873", "issue:43854|issue:43931", "issue:42491|issue:45335", @@ -2319,14 +2328,14 @@ "issue:45488|issue:45812", "issue:45276|issue:45335" ], - "prepared_review_unit_hash": "40a88712ef7d39374fb26f528cb7b99215484fee7fc1100e757579cdeffd4753", + "prepared_review_unit_hash": "508260a5781cd771acf2bbe1333575fe7f2bc596c2e607fa79510ad2b5624325", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, @@ -2341,198 +2350,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose set of related-but-distinct Transformers issues, with only a couple of true duplicates/near-duplicates. The clearest duplicate anchor is the Transformers 5.0.0 non-persistent-buffer regression; a Gemma4 4-bit multimodal dtype-cast pair also looks like the same bug.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43644 is the strongest canonical issue candidate because it is an exact duplicate of 44534 and describes a clear, self-contained Transformers 5.0.0 regression.", + "summary": "This set is mostly heterogeneous; only a couple of soft pairs are true duplicates of the same concrete bug. The clearest duplicate cluster is the Transformers v5 non-persistent buffer regression, with one likely duplicate around `resize_token_embeddings` behavior.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43644 is the cleanest representative: it states the underlying v5 regression directly and has an exact duplicate in issue:44534.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43644 is the best overall issue representative because it is the cleanest duplicate anchor in the set and has the most directly mergeable counterpart.", + "best_issue_reason": "issue:43644 is the best issue to anchor on because it is broad, specific, and clearly reproducible; it also captures the shared bug more directly than the paired duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45672", "right": "issue:46899", "accept": true, - "reason": "Both describe Gemma4 4-bit multimodal inference breaking from unsafe dtype handling in the same embedders/audio-vision path." + "reason": "Same Gemma4 multimodal 4-bit dtype-cast failure; both describe the same unsafe `uint8`/NF4 embedders bug." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer-mapping registration failure and a mask/tensor shape mismatch are unrelated bugs." + "reason": "Both are failures, but one is tokenizer-mapping logic and the other is a mask/tensor shape error; different code paths and symptoms." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "Different code paths: Qwen3 MoE LoRA compatibility vs BitNet packed-weight loading." + "reason": "Different models and different failures: qwen3_moe LoRA compatibility vs BitNet packed-weight unpacking in accelerate loading." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "A device-map CUDA assert and a Kosmos2.5 OCR index error are separate failures." + "reason": "Unrelated regressions: `_is_hf_initialized` in Transformers v5 vs a Kosmos2.5 OCR long-input index error." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "Offload/dequantized loading bug vs Strix Halo mmap OOM; not the same underlying issue." + "reason": "Both involve loading/memory, but one is quantized model offload on device_map auto and the other is Strix Halo mmap OOM; not the same bug." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Same model family, but one is zero probabilities and the other is a docs/processor mismatch plus quantization failure." + "reason": "Same model family, but one is zero probabilities and the other is a broken docs example with processor mismatch/quantization failure; not the same concrete issue." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Auto device mapping CUDA assert and FineGrainedFP8Config failure are different problems." + "reason": "Different Qwen3 failure modes: CUDA assert on A800 vs FineGrainedFP8Config loading failure." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Exact duplicate: both report the same Transformers 5.0.0 non-persistent-buffer junk regression." + "reason": "Near-exact duplicate: both report Transformers v5 filling non-persistent buffers with junk." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "Qwen3Moe FP8 failure is unrelated to the Strix Halo mmap/OOM issue." + "reason": "Different subsystems and symptoms; Qwen3Moe FP8 loading vs Strix Halo mmap memory issues." }, { "left": "issue:42491", - "right": "issue:45335", + "right": "issue:45276", "accept": false, - "reason": "Different model families and different resize_token_embeddings symptoms." + "reason": "Both are model-specific regressions, but one is Qwen3 MOE LoRA loading and the other is Gemma4 token-embedding resizing." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "GLM5 is unrelated to Qwen3.5 flash-attention generation failure." + "reason": "Unrelated model families and failures: GLM5 vs Qwen3.5 flash-attention generation." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are load failures, but they concern different checkpoints with no shared concrete failure mode shown." + "reason": "Both are tokenizer/model loading reports, but they concern different checkpoints and the evidence is too weak to call them the same bug." }, { "left": "issue:44863", "right": "issue:46846", "accept": false, - "reason": "NemotronH checkpoint loading and Qwen3.5 TP-plan shape/OOM issues are unrelated." + "reason": "Different problems: NemotronH checkpoint loading vs Qwen3.5 TP-plan missing `linear_attn` entries." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "DeepSpeed ZeRO-3 weight loading and save_pretrained visual-encoder key regression are distinct." - }, - { - "left": "issue:45507", - "right": "issue:46468", - "accept": false, - "reason": "Invalid method call in GraniteMoEHybrid is unrelated to the Nemotron example breakage." + "reason": "Both are Qwen3.5 regressions, but one is ZeRO-3 weight loading and the other is incorrect visual-encoder keys on save; not the same bug." }, { "left": "issue:46459", "right": "issue:46512", "accept": false, - "reason": "One is dtype ignored on load; the other is config.dtype staying stale after from_config." + "reason": "Both concern dtype handling, but one is loading composite checkpoints and the other is `from_config(dtype=...)` leaving config stale; different APIs and failure modes." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "Same broad v5 context, but custom-model-in-Jupyter initialization is a different bug." + "reason": "They both relate to Transformers v5 regressions, but one is buffer initialization and the other is custom model initialization in notebooks." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2, but docs-example errors and nonsensical model outputs are not clearly the same code-path bug." + "reason": "Both mention SigLIP2, but one is a docs example and the other is bad inference outputs; not clearly the same underlying defect." }, { - "left": "issue:44488", - "right": "issue:44991", + "left": "issue:43257", + "right": "issue:43854", "accept": false, - "reason": "Different checkpoints and different tokenizer/loading symptoms; too little evidence of one root cause." + "reason": "Different model families and loading paths: Qwen3 MOE deepspeed/accelerate conversion vs GLM-4.7-Flash unit-test loading." }, { - "left": "issue:43257", - "right": "issue:43854", + "left": "issue:44488", + "right": "issue:44991", "accept": false, - "reason": "Qwen3 MOE weight conversion under accelerate/deepspeed is unrelated to GLM-4.7-Flash test loading." + "reason": "Both are tokenizer-loading failures for different models, but the evidence does not show a shared concrete root cause." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Training memory inefficiency and quantization offloading behavior are different issues." + "reason": "Different issues: Qwen3 MoE training memory inefficiency vs quantization offloading not working." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "GLM-4.7-Flash loading failure and Qwen3-VL weight-shape mismatch are unrelated." + "reason": "Both are model-loading errors, but the affected models and failure mechanisms differ (GLM-4.7-Flash vs Qwen3-VL shape mismatch)." }, { "left": "issue:42491", "right": "issue:45335", "accept": false, - "reason": "Different model families and different embedding-resize failures." + "reason": "Different model families and different failures; Qwen3 MOE LoRA loading is unrelated to t5gemma embedding resizing." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "Same API name, but the model/context and affected submodule differ; not the same concrete bug." + "reason": "Same as above: unrelated Qwen3 MOE loading issue vs generic resize-token-embeddings behavior." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both are memory-related, but one is Qwen3 MoE training inefficiency and the other is int4 quantization reserved-memory growth." + "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is int4 quantization reserved-memory/OOM in v5." }, { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Trainer resume/max_steps logic and SGD optimizer args not being used are unrelated." + "reason": "Trainer checkpoint-resume math bug vs optimizer args being ignored; different code paths." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "GLM5 and sinq save/load breakage are different bugs." + "reason": "Unrelated issues: GLM5 model work vs quantized save/load breakage after `sinq` quantization." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "extra_state save/load handling and unexpected _is_hf_initialized in init_empty_weights are distinct regressions." + "reason": "Both involve loading/saving, but one is extra_state handling and the other is `init_empty_weights` / `_is_hf_initialized`; not the same bug." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and different concrete tokenizer behavior." + "reason": "Both are tokenizer regressions, but one is LlamaTokenizer pre-tokenizer override and the other is wrong Granite token IDs; different concrete problems." }, { "left": "issue:45276", "right": "issue:45335", - "accept": false, - "reason": "Similar resize_token_embeddings symptom, but Gemma4 and T5Gemma affect different model-specific submodules and are not clearly the same bug." + "accept": true, + "reason": "Same underlying `resize_token_embeddings` regression: the resized embeddings are not propagated to model-specific output/decoder embeddings." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The two accepted soft-edge pairs are well supported by the titles/body themes (43644\u219444534 and 45672\u219446899), and the other rejections do not overreach. The canonical issue choice of 43644 is defensible from the duplicate wording with 44534." + "feedback": "The summary is grounded in the packet and stays conservative overall. The canonical choice of issue:43644 is well supported by the exact title match with issue:44534, and the accepted soft pairs are plausible same-bug matches based on the issue titles. No PR-mergeability concerns apply here because this is an issues-only set." }, "evaluator_used": true, "retried": false, @@ -2551,7 +2554,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36010", "issue:39404", @@ -2566,6 +2569,7 @@ "issue:43454", "issue:43645", "issue:43828", + "issue:43854", "issue:43873", "issue:43957", "issue:44079", @@ -2584,14 +2588,13 @@ "issue:44912", "issue:44928", "issue:44960", + "issue:45216", "issue:45276", "issue:45292", "issue:45362", "issue:45406", "issue:45588", "issue:45663", - "issue:45684", - "issue:45698", "issue:45704", "issue:45759", "issue:45823", @@ -2609,8 +2612,8 @@ "issue:44291|issue:44756", "issue:44402|issue:44488", "issue:45663|issue:45759", - "issue:41720|issue:44155", "issue:44910|issue:45588", + "issue:41720|issue:44155", "issue:43404|issue:43454", "issue:39692|issue:44960", "issue:43873|issue:44387", @@ -2632,222 +2635,222 @@ "issue:45588|issue:45663", "issue:43828|issue:43957", "issue:43645|issue:44164", - "issue:45684|issue:45698" + "issue:43854|issue:45216" ], - "prepared_review_unit_hash": "ec584ca1b21a319b03b91697f8688e8a04d31d6ae7bf19d131314f54e83cf04b", + "prepared_review_unit_hash": "ee9a77371c88832cf0202f5635bd0baef2f4ed9f43ed918fc1ec550b185ce47d", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36788, - "estimated_input_tokens": 9197, - "estimated_eval_tokens": 18650 + "serialized_chars": 36784, + "estimated_input_tokens": 9196, + "estimated_eval_tokens": 18648 }, "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. The only clear duplicate pair is the resize_token_embeddings/output_embeddings bug (45276/45292); the rest are similar by subsystem or symptom but not the same concrete issue.", - "confidence": 0.96, - "canonical_issue_reason": "issue:45276 is the best representative because it is the broader, more specific report of the resize_token_embeddings/output_embeddings failure, and it cleanly subsumes the more generic issue:45292.", + "summary": "This cluster is mostly heterogeneous. Only two soft pairs look like true duplicates: the `resize_token_embeddings` regression pair and the training-loss double-shift pair. The remaining pairs share broad subsystems or vocabulary but do not look like the same concrete bug/change.", + "confidence": 0.88, + "canonical_issue_reason": "Issue:45276 is the strongest representative because it states the broader `resize_token_embeddings` regression, and issue:45292 appears to be the same bug from a narrower angle. The rest of the cluster is too mixed to provide a better single canonical issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45276 is the most actionable issue here: it is concrete, well-scoped, and has a true duplicate that points to the same underlying bug.", + "best_issue_reason": "Issue:45276 is the best issue representative: it is specific, clearly described, and aligns with the only clearly duplicate issue in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46897", "right": "issue:46901", - "accept": false, - "reason": "Same high-level training-loss symptom, but different models/code paths; not clearly one bug." + "accept": true, + "reason": "Both describe the same training-loss double-shift problem: the model is training against shifted labels when it should not. Different models, but the underlying bug pattern matches." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different failures in different model families: chat crash vs missing processor attribute." + "reason": "Different failures in different paths: one is a Qwen chat crash, the other is a Gemma4 processor `_tokenizer` attribute error in serve." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Torch version compatibility and meta-device loading are separate problems." + "reason": "Both are version/loading regressions, but they are different root causes: Torch 2.10 incompatibility vs `meta`-device model loading failures." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "init_empty_weights argument handling and Strix Halo mmap/OOM are unrelated." + "reason": "One is an `init_empty_weights` / `_is_hf_initialized` constructor mismatch; the other is an OOM workaround for mmap on Strix Halo. Not the same bug." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Both are load-related, but the model/tokenizer failures are different and not the same bug." + "reason": "Both involve model loading, but the specific problems differ: vocab-size token placement vs a separate loading failure for `cjvt/sleng-bert`." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "FSDP2 shared_kv_states KeyError and Qwen config unwrapping are distinct issues." + "reason": "Different code paths: FSDP2 attention state rebuilding in Gemma4 vs `from_config` not unwrapping `text_config` in Qwen multimodal configs." }, { - "left": "issue:41720", - "right": "issue:44155", + "left": "issue:44910", + "right": "issue:45588", "accept": false, - "reason": "Different subsystems and different failure mechanisms." + "reason": "Both touch flash attention, but the failures are distinct: 3D `position_ids` packing misinterpretation vs `s_aux=None` AttributeError for sink-less models." }, { - "left": "issue:44910", - "right": "issue:45588", + "left": "issue:41720", + "right": "issue:44155", "accept": false, - "reason": "Both touch flash attention, but one is a position_ids interpretation bug and the other is an s_aux=None AttributeError." + "reason": "Unrelated issues: a Qwen3 device-mapping CUDA assert vs an AudioFlamingo3 batched inference embedding/token leak." }, { "left": "issue:43404", "right": "issue:43454", "accept": false, - "reason": "Both mention untied lm_head weights, but they are model-specific reports and not the same concrete code path." + "reason": "Very similar symptom, but they concern different model classes and likely separate fixes; not enough to treat as the same underlying bug with high confidence." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Documentation/example errors and a GLM5 issue are unrelated." + "reason": "Different models and different problem domains: SigLIP2 documentation errors vs a generic GLM5 issue with no matching bug description." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Quantization/offloading and int4 reserved-memory OOM are different bugs." + "reason": "Both mention quantization and OOM-like behavior, but one is about offloading behavior and the other about increased reserved memory under int4 quantization." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing with LoRA and an apex RMSNorm memory leak are unrelated." + "reason": "Completely different bugs: gradient checkpointing with PEFT LoRA vs a T5/apex RMSNorm memory leak." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Different model-specific bugs: Gemma4 FSDP2 vs Qwen3_5Moe CPU offload." + "reason": "Different model families and different failures: Gemma4 FSDP2 KeyError vs Qwen3.5 MoE rotary embedding CPU-offload incompatibility." }, { "left": "issue:44079", "right": "issue:46829", "accept": false, - "reason": "Both concern output fields, but the generic ModelOutput key issue is not the same as the VideoPrism hidden_states bug." + "reason": "Both concern output containers, but one is a generic `ModelOutput` key assignment bug and the other is a VideoPrism output field mix-up." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized and MXFP4 quantization fallback are unrelated." + "reason": "Both are quantization-related loading issues, but they involve different formats and symptoms (`_is_hf_initialized`/loading error vs MXFP4 fallback to bf16)." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading and disabling mmap on Strix Halo address different root causes." + "reason": "Different root causes and different layers: `meta`-device loading regressions vs an OOM avoidance change for mmap on Strix Halo." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "AudioFlamingo3 token leak and Gemma4 device mismatch are not the same bug." + "reason": "Distinct problems: AudioFlamingo3 track leakage in batched inference vs Gemma4 PLE device mismatch under `device_map=\"auto\"`." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "ImportError for GenerationMixin and Whisper return_language regression are unrelated." + "reason": "Unrelated: an import path break for `GenerationMixin` vs Whisper pipeline `return_language` regression." }, { "left": "issue:43335", "right": "issue:45663", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and Gemma4 FSDP2 KeyError are distinct." + "reason": "Different model/config bugs: sparse layer creation in SwitchTransformers vs Gemma4 FSDP2 attention-state rebuilding." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Both can OOM, but one is an int4 memory regression and the other is a platform-specific mmap issue." + "reason": "Both can lead to OOM, but one is reserved-memory growth under int4 quantization and the other is a platform-specific mmap workaround." }, { "left": "issue:44863", "right": "issue:46468", "accept": false, - "reason": "Same Nemotron family, but one is a checkpoint-loading incompatibility and the other is a broken example." + "reason": "Related family name, but not the same concrete bug: one cannot load NemotronH checkpoints, the other is a broken official example." }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Video-input regression and strict config loading are unrelated." + "reason": "Different regressions in different areas: video-input handling for Qwen VL variants vs strict config loading for granite_speech." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same underlying resize_token_embeddings/output_embeddings bug; 45276 is the broader specific report and 45292 is the generic duplicate." + "reason": "Same underlying regression: `resize_token_embeddings` fails to propagate correctly to output embeddings; 45276 just includes the additional `embed_tokens_per_layer` detail." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Gradient checkpointing in eval mode and Qwen3.5 position_ids-induced NaNs are unrelated." + "reason": "Different training-time issues: gradient checkpointing in eval mode vs Qwen3.5 RLHF instability from 3D `position_ids` and SDPA fallback." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "extra_state save/load handling and Strix Halo mmap/OOM are unrelated." + "reason": "Unrelated save/load extra-state handling vs a memory/OOM workaround." }, { "left": "issue:42796", "right": "issue:46326", "accept": false, - "reason": "Both involve tokenizer behavior, but BPE marker handling and continue_final_message are different failures." + "reason": "Both are tokenizer-related, but one is BPE marker handling and the other is `continue_final_message` behavior for `mistral_common` tokenizers." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Flash-attention s_aux=None crash and Gemma4 shared_kv_states rebuild are unrelated." + "reason": "Different subsystems and failures: flash attention `s_aux=None` crash vs Gemma4 FSDP2 shared KV state rebuilding." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Autocast dtype mismatch and meta-device loading are different issues." + "reason": "Different failures: autocast dtype mismatch for Phi-tiny-MoE vs `meta`-device loading breaks some models." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Custom-model Jupyter initialization and extra_state save/load are unrelated." + "reason": "One is a Jupyter/custom-model initialization regression in Transformers 5.0, the other is save/from_pretrained handling of extra state." }, { - "left": "issue:45684", - "right": "issue:45698", + "left": "issue:43854", + "right": "issue:45216", "accept": false, - "reason": "Both concern custom-model save/load, but one is read-only permissions propagation and the other is wrong module selection." + "reason": "Both are model-specific loading/save issues, but they involve different models and different APIs (`unit tests` load failure vs `save_pretrained` regression)." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft pair, 45276/45292, is well supported by the titles as the same resize_token_embeddings/output_embeddings bug, with 45276 being the more specific report. The other pairs are reasonably treated as non-duplicates based on differing models, subsystems, or failure modes." + "feedback": "The summary is grounded in the packet and stays conservative on the soft-edge calls. The two accepted pairs are well supported by the titles, and the canonical/best issue choice around the resize_token_embeddings regression is reasonable. No clear overstatement of evidence." }, "evaluator_used": true, "retried": false, @@ -2866,7 +2869,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36032", "issue:39404", @@ -2878,7 +2881,6 @@ "issue:43708", "issue:43716", "issue:43828", - "issue:43854", "issue:43901", "issue:43906", "issue:43994", @@ -2900,9 +2902,10 @@ "issue:44991", "issue:45081", "issue:45084", - "issue:45216", "issue:45292", "issue:45335", + "issue:45684", + "issue:45698", "issue:45701", "issue:45759", "issue:45812", @@ -2918,7 +2921,7 @@ "issue:47228" ], "soft_pairs": [ - "issue:43854|issue:45216", + "issue:45684|issue:45698", "issue:43425|issue:47228", "issue:45292|issue:45335", "issue:44589|issue:44756", @@ -2930,10 +2933,10 @@ "issue:45812|issue:46315", "issue:43425|issue:46869", "issue:44530|issue:45084", - "issue:39404|issue:45701", - "issue:43611|issue:43645", "issue:44683|issue:45823", + "issue:43611|issue:43645", "issue:43425|issue:43716", + "issue:39404|issue:45701", "issue:39404|issue:46665", "issue:44479|issue:44805", "issue:44589|issue:45084", @@ -2944,229 +2947,220 @@ "issue:44265|issue:44898", "issue:43906|issue:44530", "issue:46050|issue:46326", - "issue:43906|issue:46326", "issue:44991|issue:45081", + "issue:43906|issue:46326", "issue:44671|issue:44991", "issue:39404|issue:44912" ], - "prepared_review_unit_hash": "d60f6c9533c50bc03a235a7e816743b51f4fca95771acef5ddde774c263755fb", + "prepared_review_unit_hash": "22e69106b13eb217af433d24c3da1fe2b31f44766d0315c2dabcca3606fc317c", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36821, - "estimated_input_tokens": 9206, - "estimated_eval_tokens": 18668 + "serialized_chars": 36830, + "estimated_input_tokens": 9208, + "estimated_eval_tokens": 18672 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is mostly a loose bundle of unrelated Transformers regressions and docs issues, not a single coherent duplicate cluster. The only convincing near-duplicate subpair is `issue:45292` / `issue:45335` around `resize_token_embeddings` not updating tied embeddings. No PRs were provided.", - "confidence": 0.31, - "canonical_issue_reason": "`issue:45292` is the broadest version of the only plausible duplicate subtopic and states the core embedding-resize inconsistency without model-specific narrowing.", + "summary": "These items are mostly distinct bugs; the only strong thematic overlap is around Transformers v5 regressions, but the concrete failure modes differ. I would anchor the cluster on the broad custom-model loading regression, not merge any of the soft pairs.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43611 is the broadest, most central regression item here: it describes a concrete v5 loading breakage affecting models via base_model_prefix, and several nearby issues are narrower variants of v5 custom-model/init/load failures.", "canonical_pr_reason": null, - "best_issue_reason": "`issue:45292` is the best representative of the resize-token-embeddings bug subcluster; it is the most general statement of that failure mode.", + "best_issue_reason": "issue:43611 is the best representative issue because it captures a general, user-facing Transformers v5 model-loading regression rather than a model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43854", - "right": "issue:45216", + "left": "issue:45684", + "right": "issue:45698", "accept": false, - "reason": "Different failures: model loading/unit-test breakage vs save_pretrained checkpoint regression for Qwen3.5." + "reason": "Both concern custom-model save/load behavior, but one is about permission propagation during save_pretrained and the other about loading the wrong custom module after saving; different bugs and different code paths." }, { "left": "issue:43425", "right": "issue:47228", "accept": false, - "reason": "Both mention Torch, but one is general incompatibility and the other is a bf16 compilation regression; not the same bug." + "reason": "Unrelated problems: torch 2.10 compatibility versus a Sam3 bf16 compilation regression between releases." }, { "left": "issue:45292", "right": "issue:45335", - "accept": true, - "reason": "Same underlying bug class: `resize_token_embeddings` not propagating to tied output/decoder embeddings; `t5gemma` is just a model-specific instance." + "accept": false, + "reason": "Related API area, but one reports resize_token_embeddings not updating output_embeddings generally, while the other is a t5gemma-specific decoder.embed_tokens issue; not clearly the same concrete bug." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Unrelated runtime/storage error vs mmap/OOM workaround; no shared code-path bug." + "reason": "Completely different failures: missing Float8 storage type versus disabling mmap to avoid OOM on Strix Halo." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model init breakage vs Trainer resume/max_steps miscalculation; different subsystems and symptoms." + "reason": "Different regressions: custom model initialization in notebooks versus Trainer resume_from_checkpoint/max_steps miscalculation." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Both are tokenizer/model loading problems, but they concern different models and different root causes." + "reason": "Both are tokenizer-loading related, but one is an out-of-vocab special-token issue and the other is a specific model load failure for ScandiBERT; insufficient evidence of the same root cause." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Generic Torch 2.10 incompatibility is not the same as an autocast dtype mismatch on a specific Phi model." + "reason": "Both mention dtype/torch issues, but one is torch 2.10 incompatibility and the other is an autocast dtype mismatch for a specific model; distinct bugs." }, { "left": "issue:46894", "right": "issue:46901", "accept": false, - "reason": "Similar double-shift wording, but these are model-specific training-loss bugs with no evidence of a shared code path." + "reason": "Same broad double-shift pattern, but the reports are for different model families and may reflect separate implementations rather than one shared code-path bug." }, { "left": "issue:46315", "right": "issue:46491", "accept": false, - "reason": "Tokenizer conversion slowdown is unrelated to CodeLlama encode/decode spacing regression." + "reason": "Different tokenizer regressions: performance slowdown in convert_tokens_to_ids versus an encode-decode leading-space round-trip behavior change." }, { "left": "issue:45812", "right": "issue:46315", "accept": false, - "reason": "Wrong Granite token IDs and slow `convert_tokens_to_ids` are different tokenizer regressions." + "reason": "Wrong token IDs for Granite models versus a tokenizer performance regression; not the same issue." }, { "left": "issue:43425", "right": "issue:46869", "accept": false, - "reason": "Torch version incompatibility and YOLO multi-GPU device-map failure are unrelated." + "reason": "Torch version incompatibility and YOLO multi-GPU device_map failure are unrelated." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear-attention crash is unrelated to template compilation failure." + "reason": "A Qwen3.5 PagedAttentionCache crash is not the same as a template-compilation TypeError." }, { - "left": "issue:39404", - "right": "issue:45701", + "left": "issue:44683", + "right": "issue:45823", "accept": false, - "reason": "Whisper pipeline `return_language` regression and generic tokenization changes are not the same bug." + "reason": "Different failure modes and subsystems: compiled flex_attention on torch >=2.9 versus Gemma4 PLE device placement with device_map='auto'." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "`base_model_prefix` loading failure and Jupyter custom-model initialization breakage are distinct issues." + "reason": "Both are v5 custom-model related, but one is a base_model_prefix loading break and the other is notebook-specific model initialization behavior; not the same concrete bug." }, { - "left": "issue:44683", - "right": "issue:45823", + "left": "issue:43425", + "right": "issue:43716", "accept": false, - "reason": "Compiled flex_attention on newer Torch and Gemma4 device mismatch are different failures." + "reason": "Torch compatibility versus a Mistral-3 image preprocessor/model dtype mismatch." }, { - "left": "issue:43425", - "right": "issue:43716", + "left": "issue:39404", + "right": "issue:45701", "accept": false, - "reason": "Torch incompatibility and Mistral-3 image-preprocessor dtype mismatch are unrelated." + "reason": "Whisper pipeline return_language regression and a tokenizer-version-induced tokenization change are different issues." }, { "left": "issue:39404", "right": "issue:46665", "accept": false, - "reason": "Whisper pipeline regression and `from_pretrained` inside Python classes do not share the same concrete bug." + "reason": "Pipeline Whisper regression versus inconsistent from_pretrained behavior inside Python classes; distinct code paths." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Video-input regression and mask shape `IndexError` are different model/runtime problems." + "reason": "Video-input regression for Qwen models is not the same as a generic mask/index shape mismatch error." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to non-template compilation errors." + "reason": "Missing Float8 storage type and non-template-node compilation errors are unrelated." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Both are docs-related, but they cover different removed/changed pipeline behaviors and are not one underlying issue." + "reason": "Both are docs-related, but they cover different removed/changed pipeline behaviors and should not be merged as one bug." }, { "left": "issue:44164", "right": "issue:45759", "accept": false, - "reason": "Saving/loading extra state and `from_config` text_config unwrapping are unrelated API bugs." + "reason": "Save/from_pretrained extra_state handling and from_config text_config unwrapping are separate serialization/configuration bugs." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "Nonsensical model outputs and `ModelOutput` key assignment with `None` are different code paths." + "reason": "Wrong SigLIP2 model outputs and ModelOutput key assignment behavior are different problems." }, { "left": "issue:36032", "right": "issue:43452", "accept": false, - "reason": "Tokenizer special-token conflict and `gguf_file` loading breakage are unrelated tokenizer/load issues." + "reason": "T5 tokenizer special-token method conflict and gguf_file loading failures are unrelated tokenizer/model-loading issues." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "`torch.export.export` failing on `torch_compilable_check` models is not the same as Perceiver non-default-resolution failure." + "reason": "torch.export failures with torch_compilable_check are not the same as Perceiver non-default-resolution failures." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "An isolated reproduction of a previous issue is not the same as the Qwen3.5 linear-attention crash." + "reason": "An isolated reproduction of another issue is not evidence of the same underlying bug as the Qwen3.5 PagedAttentionCache crash." }, { "left": "issue:46050", "right": "issue:46326", "accept": false, - "reason": "Quantized save/load breakage and mistral_common `continue_final_message` bug are unrelated." - }, - { - "left": "issue:43906", - "right": "issue:46326", - "accept": false, - "reason": "No shared bug: isolated reproduction of one issue vs tokenizer message-continuation behavior." + "reason": "Quantized save/load corruption and a mistral_common continue_final_message bug are unrelated." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Different tokenizer loading failures for different model families and different root causes." + "reason": "Both are tokenizer-loading failures, but one is a generic tokenizer load break for a specific model and the other is a Mistral regex-patching crash; different root causes." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "Masked-LM prediction regression and tokenizer loading failure are unrelated." + "reason": "CamemBERT masked-LM prediction errors and tokenizer loading failure for EMBEDDIA/est-roberta are unrelated." }, { - "left": "issue:39404", + "left": "issue:43425", "right": "issue:44912", "accept": false, - "reason": "Whisper pipeline behavior and MXFP4 quantization fallback are different problems." + "reason": "Torch 2.10 incompatibility and MXFP4 quantization falling back to bf16 are different issues." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The writeup stays grounded in the packet and is conservative about the soft edges. The conclusion that this is mostly a loose collection of unrelated regressions, with only the 45292/45335 resize_token_embeddings pair looking like a real near-duplicate, is reasonable from the titles and metadata provided. No PRs were present, so there is no mergeability issue to evaluate." - }, - "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 }, @@ -3181,8 +3175,9 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ + "issue:32090", "issue:34567", "issue:36032", "issue:39404", @@ -3211,7 +3206,6 @@ "issue:44488", "issue:44530", "issue:44661", - "issue:44671", "issue:44683", "issue:44737", "issue:44756", @@ -3235,236 +3229,236 @@ "soft_pairs": [ "issue:45823|issue:46899", "issue:44164|issue:46821", - "issue:43856|issue:44756", "issue:43596|issue:44683", + "issue:43856|issue:44756", "issue:45479|issue:46894", "issue:44488|issue:44987", "issue:43632|issue:47050", "issue:44488|issue:45081", "issue:36032|issue:42915", - "issue:45479|issue:46897", "issue:43299|issue:43931", + "issue:45479|issue:46897", "issue:43611|issue:43644", "issue:42796|issue:43404", - "issue:44462|issue:46489", "issue:43716|issue:45084", + "issue:44462|issue:46489", "issue:46285|issue:46402", "issue:44661|issue:44683", "issue:43541|issue:45084", "issue:44451|issue:44987", - "issue:39404|issue:44671", "issue:42915|issue:44530", + "issue:32090|issue:39404", "issue:43643|issue:44530", "issue:44737|issue:45859", "issue:40990|issue:43421", - "issue:40444|issue:46326", - "issue:34567|issue:40444", "issue:36032|issue:43645", + "issue:34567|issue:40444", "issue:43644|issue:43646", + "issue:40444|issue:46326", "issue:42175|issue:46178" ], - "prepared_review_unit_hash": "ff0d318812d03c7b646e06997c6570a61c29ed78a42a0f90f8eb65d6e277e114", + "prepared_review_unit_hash": "9751e332e97fb7131eb6d6ea760ef6d91bbda11841835afaa89b100b5eb4cd2d", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36495, - "estimated_input_tokens": 9124, - "estimated_eval_tokens": 18504 + "serialized_chars": 36543, + "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 heterogeneous: the soft pairs mostly share broad themes (loading, tokenizers, v5 regressions, multimodal/device issues) but not the same concrete bug or code path. I would not merge any of the candidate pairs as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "No single canonical issue is justified here; the items are mostly unrelated regressions with only superficial title similarity.", + "summary": "This cluster is mostly a loose set of unrelated Transformers regressions, with many v5-related model/tokenizer loading or initialization bugs but no strong duplicate pairs. I would treat it as similarity clustering rather than a true duplicate set.", + "confidence": 0.66, + "canonical_issue_reason": "If one issue must represent the cluster, issue #43646 is the broadest and most generic v5 regression here: it captures the recurring custom-model initialization breakage that several other issues loosely resemble.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44462 is the strongest standalone issue in the set: it has a clear reproduction, broader loading impact, and inbound references, but it is still not a duplicate of the others.", + "best_issue_reason": "Issue #43646 is the best single representative because it is broad, user-facing, and closest to the repeated v5 initialization/loading failures across the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45823", "right": "issue:46899", "accept": false, - "reason": "Both concern Gemma4 multimodal behavior, but one is a device-placement bug and the other is an unsafe dtype cast in 4-bit embedders; different code paths." + "reason": "Both are Gemma4 multimodal failures, but one is a device mismatch in forward and the other is a 4-bit dtype-cast bug in embedders; different code paths." }, { "left": "issue:44164", "right": "issue:46821", "accept": false, - "reason": "Extra_state save/load failure and TimesFM window_size AttributeError are unrelated issues." + "reason": "Unrelated symptoms: extra_state save/load handling versus TimesFM window_size AttributeError." }, { - "left": "issue:43856", - "right": "issue:44756", + "left": "issue:43596", + "right": "issue:44683", "accept": false, - "reason": "Qwen3 MoE training memory usage and Strix Halo mmap OOM are different problems with different causes." + "reason": "Different problems in different areas: DeepSpeed ZeRO-3/BertModel init versus torch>=2.9 flex_attention compilation." }, { - "left": "issue:43596", - "right": "issue:44683", + "left": "issue:43856", + "right": "issue:44756", "accept": false, - "reason": "Zero3/BertModel init failure and compiled flex_attention torch>=2.9 failure are unrelated." + "reason": "Both mention memory pain, but one is Qwen3 MoE training memory inefficiency and the other is mmap/OOM on Strix Halo; not the same bug." }, { "left": "issue:45479", "right": "issue:46894", "accept": false, - "reason": "Zero-loss classification regression and training-loss double-shift bug are not the same defect." + "reason": "Different training-loss bugs: classification zero-loss degeneracy versus ASR double-shift behavior." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Both are model-loading complaints, but they target different models and likely different loader failures." + "reason": "Both are model loading failures, but for different checkpoints and likely different root causes." }, { "left": "issue:43632", "right": "issue:47050", "accept": false, - "reason": "_is_hf_initialized flag regression and fp16 ONNX attention dtype mismatch are unrelated." + "reason": "v5 _is_hf_initialized flag breakage is unrelated to Canine fp16 ONNX attention dtype mismatch." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Both involve tokenizer/model loading, but one is a missing model load and the other is a specific Mistral regex patch crash." + "reason": "Different tokenizer-loading issues affecting different models and different failure points." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Tokenizer special-token method conflict and Qwen3Moe FineGrainedFP8Config failure are unrelated." + "reason": "T5 tokenizer method-name conflict is unrelated to Qwen3Moe FP8 config failure." }, { - "left": "issue:45479", - "right": "issue:46897", + "left": "issue:43299", + "right": "issue:43931", "accept": false, - "reason": "These are both training-loss reports, but the model families and failure mechanisms differ." + "reason": "Closely related model family, but one is a broad Qwen3VL MoE load regression and the other is a specific checkpoint shape-mismatch load error; not clearly one concrete bug." }, { - "left": "issue:43299", - "right": "issue:43931", + "left": "issue:45479", + "right": "issue:46897", "accept": false, - "reason": "Both concern Qwen3-VL loading, but one is a version regression and the other is a shape-mismatch load error; not the same bug." + "reason": "Different model families and different loss bugs; only the broad theme of training-loss regressions is shared." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Both are v5 regressions, but base_model_prefix loading and non-persistent-buffer junk are distinct code paths." + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization junk." }, { "left": "issue:42796", "right": "issue:43404", "accept": false, - "reason": "Tokenizer BPE-marker handling and lm_head weight tying are separate bugs." + "reason": "Both involve Mistral-family artifacts, but tokenizer BPE markers and lm_head weight tying are separate issues." }, { - "left": "issue:44462", - "right": "issue:46489", + "left": "issue:43716", + "right": "issue:45084", "accept": false, - "reason": "Both are tokenizer-related, but one ignores tokenizer.json while the other is an incorrect-output regression for a specific model." + "reason": "Mistral image-preprocessor dtype mismatch is unrelated to template-compilation TypeError." }, { - "left": "issue:43716", - "right": "issue:45084", + "left": "issue:44462", + "right": "issue:46489", "accept": false, - "reason": "Image-preprocessor dtype mismatch and non-template-node compilation error are unrelated." + "reason": "Both concern tokenizer behavior, but one ignores tokenizer.json while the other is a DeepSeek-Coder output regression on v5+." }, { "left": "issue:46285", "right": "issue:46402", "accept": false, - "reason": "CLIP structure inconsistency and Sam3Model text encoder weight loading are different regressions." + "reason": "Different model-loading/structure problems for unrelated model families." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Tokenizer mapping lookup failure and flex_attention compilation failure are unrelated." + "reason": "Tokenizer-mapping model registration bug is unrelated to flex_attention compile failure." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Both involve tracing/compilation, but they fail in different compiler paths with different errors." + "reason": "Torch Dynamo grouped_mm tracing failure is unrelated to non-template-node compilation error." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Both are model-load failures, but the affected models and root causes are unrelated." + "reason": "Both are model-loading complaints, but for different repositories/models and likely different underlying causes." }, { - "left": "issue:39404", - "right": "issue:44671", + "left": "issue:42915", + "right": "issue:44530", "accept": false, - "reason": "Whisper pipeline language handling and CamemBERT masked-LM prediction regression are unrelated." + "reason": "Different Qwen3-family issues: FP8 training config versus PagedAttentionCache linear_attention crash." }, { - "left": "issue:42915", - "right": "issue:44530", + "left": "issue:32090", + "right": "issue:39404", "accept": false, - "reason": "Qwen3Moe FP8 config failure and PagedAttentionCache invalid group type are different issues." + "reason": "Trainer _gpu_broadcast_one NoneType error is unrelated to Whisper return_language pipeline behavior." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "AutoConfig trust_remote_code field loss and PagedAttentionCache crash are unrelated." + "reason": "AutoConfig trust_remote_code field loss is unrelated to Qwen3.5 PagedAttentionCache group-type crash." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both are device/offload-related, but XLNet CPU arange and Qwen3_5Moe rotary embedding offload incompatibility are different bugs." + "reason": "Both involve device/offload behavior, but XLNet CPU arange and Qwen3_5Moe rotary embedding offload are different code paths." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity regression on gpt-oss and runtime post-processor update bug are unrelated." + "reason": "Perplexity regression on gpt-oss-20b is unrelated to runtime special-token/post-processor update behavior." }, { - "left": "issue:40444", - "right": "issue:46326", + "left": "issue:36032", + "right": "issue:43645", "accept": false, - "reason": "Multi-image IterableDataset training failure and mistral_common continue_final_message bug are unrelated." + "reason": "Tokenizer method conflict and Jupyter custom-model initialization breakage are different bugs." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "TrainerState token accounting and Qwen2.5-VL multi-image finetuning failure are unrelated." + "reason": "TrainerState token counter update and iterable-dataset multi-image finetuning failure are unrelated." }, { - "left": "issue:36032", - "right": "issue:43645", + "left": "issue:43644", + "right": "issue:43646", "accept": false, - "reason": "Tokenizer method-name conflict and Jupyter custom-model initialization regression are unrelated." + "reason": "Both are v5 initialization regressions, but one is junk-filled non-persistent buffers and the other is custom model initialization; different concrete bugs." }, { - "left": "issue:43644", - "right": "issue:43646", + "left": "issue:40444", + "right": "issue:46326", "accept": false, - "reason": "Both are v5 custom-model regressions, but one is about non-persistent buffers and the other about initialization in notebooks/custom models." + "reason": "Different multimodal training/tokenizer issues: multi-image IterableDataset failure versus mistral_common continue_final_message behavior." }, { "left": "issue:42175", "right": "issue:46178", "accept": false, - "reason": "Both mention backend packaging, but they are different backend definitions and different failing code paths." + "reason": "One is missing tensorflow in the torch extra install, the other is an invalid backend name in BACKENDS_MAPPING." } ] }, @@ -3486,7 +3480,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28530", "issue:34567", @@ -3550,16 +3544,16 @@ "issue:43208|issue:47013", "issue:44403|issue:45081", "issue:43645|issue:46661", - "issue:43262|issue:43377", "issue:43792|issue:46869", + "issue:43262|issue:43377", "issue:42915|issue:45084", "issue:43646|issue:44534", "issue:44987|issue:45081", "issue:43377|issue:43819", "issue:44849|issue:45325", "issue:42915|issue:44291", - "issue:28530|issue:34567", "issue:43404|issue:44466", + "issue:28530|issue:34567", "issue:36032|issue:43742", "issue:44062|issue:44987", "issue:44451|issue:45081", @@ -3568,208 +3562,208 @@ "issue:42915|issue:44589", "issue:43531|issue:44877" ], - "prepared_review_unit_hash": "973e483f734144258c6708ae2c0c59154c3a6f0eaaba8bb03b4852dfe00c3c50", + "prepared_review_unit_hash": "b47cda437914c50f0ba0580bf061ff14d0df87182ace07156ca9cc31ead80318", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36485, - "estimated_input_tokens": 9122, - "estimated_eval_tokens": 18500 + "serialized_chars": 36483, + "estimated_input_tokens": 9121, + "estimated_eval_tokens": 18498 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly heterogeneous; I did not find any soft edge that clearly describes the same underlying bug/change. One issue stands out as the broadest representative of the v5 initialization regressions, but the rest look like separate model- or subsystem-specific bugs.", - "confidence": 0.43, - "canonical_issue_reason": "issue:43632 is the most central concrete regression in the set: a broad Transformers v5 initialization flag breakage that overlaps conceptually with several other loading/custom-model reports, and it has stronger discussion than the narrowly scoped model-specific issues.", + "summary": "This cluster is heterogeneous. Most items are distinct bug reports with only superficial overlap around transformers v5 loading/init regressions, tokenizer/audio/model-preprocessing paths, or specific model-family issues. I do not see any true duplicate pairs among the proposed soft edges.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Same as canonical: it is the clearest high-level representative of the cluster\u2019s Transformer v5 init/loading breakage theme, without being as narrowly tied to a single model or tokenizer edge case.", + "best_issue_reason": "43646 is the broadest representative of the recurring v5 custom-model-initialization regressions, but the cluster is not cohesive enough to treat it as a duplicate hub.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "Both are audio-processing bugs, but they point to different code paths and symptoms; nothing indicates the same underlying defect." + "reason": "Different audio bugs: an unsupported `center` argument in LASR feature extraction vs a failure in `_torch_extract_fbank_features()`." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Completely different areas: Apple Silicon allocator warmup vs. Transformer v5 initialization flag handling." + "reason": "Unrelated regressions: Apple Silicon allocator warmup TypeError vs `_is_hf_initialized`/model-init flag breakage." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are v5 loading failures, but for different repos/model types with no sign of a shared concrete bug." + "reason": "Both are load failures, but they affect different models and likely different code paths; not the same bug." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Unrelated issues: template compilation error vs. GraniteMoEHybrid method call bug." + "reason": "Different failures: template-node compilation error vs GraniteMoEHybrid calling an invalid method." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "Both involve GGUF loading, but one is a broad API break and the other is a specific Qwen2.5-GGUF failure; too little evidence of the same bug." + "reason": "Both touch GGUF loading, but one is generic `gguf_file` handling and the other is a Qwen2.5-GGUF v5 regression; too different to merge." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both are Whisper-related, but one is batch_decode token skipping and the other is empty-transcription generation after align_special_tokens; different bugs." + "reason": "Whisper decoding ignores `skip_special_tokens` vs empty-transcription failure after `align_special_tokens`; related area, different defects." }, { "left": "issue:43482", "right": "issue:47312", "accept": false, - "reason": "No meaningful relation: GGUF loading vs. kernel compatibility." + "reason": "Qwen2.5-GGUF loading failure vs a generic kernels 0.16 incompatibility; unrelated." }, { "left": "issue:43746", "right": "issue:46326", "accept": false, - "reason": "Different subsystems and symptoms: PEFT local checkpoint loading vs. mistral_common tokenizer message continuation." + "reason": "PEFT local-checkpoint loading problem vs `continue_final_message` behavior for mistral_common tokenizers." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "Both are loading failures, but one is device_map/offload dequantization and the other is Qwen2.5-GGUF; not the same bug." + "reason": "Different loading problems: device-map/offload dequantization failure vs Qwen2.5-GGUF v5 load failure." }, { "left": "issue:43208", "right": "issue:47013", "accept": false, - "reason": "Both concern xLSTM training, but one is a bundle of three bugs and the other is a specific return_last_states crash; not a single concrete issue." + "reason": "Both are xLSTM-related, but one is about three training blockers for small models and the other is a train-mode unpacking crash." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Generic loading noise vs. a Mistral regex-patching crash; different problems." + "reason": "Unnecessary load-time noise vs a Mistral regex patch crash; not the same issue." }, { "left": "issue:43645", "right": "issue:46661", "accept": false, - "reason": "Both involve custom/remote-code model initialization, but the specific failures differ and there is no clear shared code-path bug." - }, - { - "left": "issue:43262", - "right": "issue:43377", - "accept": false, - "reason": "Different audio bugs: apply_chat_template sampling-rate defaulting vs. MIMI batched/single-output mismatch from padding-mask handling." + "reason": "Both involve model initialization/loading, but one is notebook-defined custom models and the other is remote-code models missing `config_class`." }, { "left": "issue:43792", "right": "issue:46869", "accept": false, - "reason": "Completely unrelated: Whisper model loading/runtime vs. YOLO multi-GPU device_map auto." + "reason": "Whisper inference failure and YOLO multi-GPU `device_map=auto` failure are unrelated." }, { - "left": "issue:42915", - "right": "issue:45084", + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug vs MIMI batching/padding-mask mismatch; different code paths." + }, + { + "left": "issue:42915", + "right": "issue:45084", "accept": false, - "reason": "Unrelated: FineGrainedFP8Config failure in Qwen3Moe vs. a template-compilation TypeError." + "reason": "Qwen3MoE FP8 config failure vs template-node compilation error; unrelated." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Both are v5 regressions, but one is custom model initialization and the other is non-persistent buffer serialization; not the same defect." + "reason": "General v5 custom-model-init regression vs non-persistent buffer corruption; too broad to be one concrete duplicate." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Different failures in different model/tokenizer paths; no evidence of a shared Mistral regex bug." + "reason": "Different model-loading failures: `physical-intelligence/fast` vs Mistral tokenizer regex patch." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "Different audio-model behaviors: padding-mask batching mismatch vs. DAC.from_latents / missing STE divergence." + "reason": "Different model bugs: MIMI batching/padding-mask issue vs DAC latent/forward mismatch with missing STE." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Distinct Qwen model bugs with different mechanisms: hidden-state output vs. rope/position_id scaling." + "reason": "Different Qwen bugs: hidden-states behavior vs still-image temporal position ID scaling." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "Both mention initialization/FP8-ish loading problems, but one is Qwen3Moe quantization and the other is init_empty_weights/_is_hf_initialized argument handling." + "reason": "Qwen3MoE FineGrainedFP8 failure vs `init_empty_weights` / `_is_hf_initialized` argument error; unrelated." }, { - "left": "issue:28530", - "right": "issue:34567", + "left": "issue:43404", + "right": "issue:44466", "accept": false, - "reason": "Training callback/early-stopping logic vs. TrainerState token counter update; unrelated bugs." + "reason": "Both mention `lm_head.weight`, but one is untied weights in Mistral3 and the other is device-dependent serialization inconsistency." }, { - "left": "issue:43404", - "right": "issue:44466", + "left": "issue:28530", + "right": "issue:34567", "accept": false, - "reason": "Both involve lm_head/tied weights, but one is a missing tie in Mistral3ForConditionalGeneration and the other is device-dependent serialization; likely related area, not the same bug." + "reason": "Early stopping metric selection and token-count tracking are unrelated trainer issues." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "Tokenizer loading failures, but with different model families and different error modes." + "reason": "Tokenizer `add_special_tokens` conflict vs MobileLLM load key error; unrelated." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "Different tokenizer/model loading errors with no shared concrete failure." + "reason": "AddedToken keyword conflict vs a model loading failure; no shared bug." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Different models and error signatures; no evidence of the same tokenizer backend crash." + "reason": "ScandiBERT loading failure vs Mistral regex patch crash; unrelated." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Both are vision-related, but one is bad model/pipeline outputs and the other is processor image-size mismatch." + "reason": "Different vision issues: bad SigLIP2 outputs vs OmDet processor/model input-size mismatch." }, { "left": "issue:44008", "right": "issue:46899", "accept": false, - "reason": "Both are Gemma multimodal issues, but the reported bugs differ: variable-name collision in Gemma3nModel.forward vs. unsafe dtype cast in Gemma4 quantized embedders." + "reason": "Gemma 3n variable collision vs Gemma4 4-bit dtype-cast bug; different versions and code paths." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "Different low-level failures: Qwen3Moe FP8 config vs. missing Float8 storage type." + "reason": "Qwen3MoE FP8 config bug vs missing Float8 storage type; unrelated." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Completely different issues: Qwen3-MoE sliding window behavior vs. granite_speech config strictness." + "reason": "Qwen3-MoE sliding-window issue vs GraniteSpeech strict config loading problem; unrelated." } ] }, @@ -3791,7 +3785,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:33666", "issue:35707", @@ -3843,250 +3837,251 @@ ], "soft_pairs": [ "issue:45923|issue:47254", - "issue:36032|issue:41720", "issue:44464|issue:45987", + "issue:36032|issue:41720", "issue:44265|issue:44610", "issue:43632|issue:46821", - "issue:43322|issue:43540", "issue:44479|issue:45325", "issue:43089|issue:45987", + "issue:43322|issue:43540", "issue:43425|issue:44403", - "issue:43296|issue:43531", "issue:43883|issue:44488", + "issue:43296|issue:43531", "issue:40444|issue:45325", - "issue:45216|issue:45357", "issue:43872|issue:44291", - "issue:43611|issue:44534", "issue:44220|issue:44683", - "issue:44991|issue:46315", + "issue:45216|issue:45357", + "issue:43611|issue:44534", "issue:43716|issue:46899", + "issue:44991|issue:46315", "issue:44403|issue:44991", - "issue:42796|issue:43746", "issue:45812|issue:46491", + "issue:42796|issue:43746", "issue:43381|issue:45923", "issue:44291|issue:44589", - "issue:33666|issue:40444", "issue:43792|issue:44589", + "issue:33666|issue:40444", "issue:43632|issue:44756", "issue:40990|issue:44811", - "issue:42796|issue:46710", "issue:43883|issue:44991", - "issue:35707|issue:40444" + "issue:42796|issue:46710", + "issue:35707|issue:40444", + "issue:43531|issue:44291" ], - "prepared_review_unit_hash": "538846a2570c2fb486bf35da4aeb36b474195dde88445c0bc6f627a534cf029c", + "prepared_review_unit_hash": "ad7fba281069b7653e9f7390b9536f3d23119fd839e25fc55f2188d4da1023f0", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36070, - "estimated_input_tokens": 9018, - "estimated_eval_tokens": 18292 + "soft_pair_count": 31, + "serialized_chars": 36369, + "estimated_input_tokens": 9093, + "estimated_eval_tokens": 18442 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous and does not form one duplicate set. Only two soft-edge pairs look plausibly duplicate: the transformers v5 `_is_hf_initialized` loading regression family, and the Qwen3.5 `save_pretrained` regression family. Everything else appears to be separate bugs in different models/code paths.", - "confidence": 0.63, - "canonical_issue_reason": "If a single issue must represent the cluster, issue 44291 is the clearest concrete report in the only tight duplicate family here (`_is_hf_initialized` / v5 loading failures). The rest of the cluster is too mixed to justify a stronger canonical claim.", + "summary": "This cluster is mostly a mixed bag of unrelated model/tokenizer/runtime regressions. Only a couple of soft edges look like true duplicates: the `_is_hf_initialized` compatibility break and the Qwen3.5 `save_pretrained` checkpoint-key regression.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43872 is the strongest central report for the shared `_is_hf_initialized`/v5 compatibility regression: it has the most activity and inbound references, and issue:44291 is essentially the same failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44291 is the best overall issue anchor because it is the most specific root-cause report among the few genuinely related items, and it maps cleanly to the shared `_is_hf_initialized` regression family.", + "best_issue_reason": "issue:43872 is the best representative issue because it states the core constructor-arg incompatibility clearly and appears to be the broadest, most reusable fix target in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45923", "right": "issue:47254", "accept": false, - "reason": "Both involve training/checkpointing, but they describe different failure modes and code paths; not the same bug." + "reason": "Different bugs: missing gradient-checkpointing support on a trust_remote_code model vs a PEFT+DeepSpeed checkpointing runtime error." }, { - "left": "issue:36032", - "right": "issue:41720", + "left": "issue:44464", + "right": "issue:45987", "accept": false, - "reason": "Tokenizer method-name conflict vs CUDA assert in device mapping; unrelated failures." + "reason": "Unrelated generation problems: compiled-forward chunked decoding inconsistency vs `StaticCache.get_seq_length()` returning the wrong type/shape." }, { - "left": "issue:44464", - "right": "issue:45987", + "left": "issue:36032", + "right": "issue:41720", "accept": false, - "reason": "Both are generation/runtime internals, but one is inconsistent chunked generation and the other is a cache length type-contract issue." + "reason": "Tokenizer constructor conflict and a CUDA assert during Qwen3 device mapping are different code paths." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export failure with torch_compilable_check is unrelated to a processor size mismatch." + "reason": "`torch.export`/`torch_compilable_check` failure is unrelated to the OmDet processor image-size mismatch." }, { "left": "issue:43632", "right": "issue:46821", "accept": false, - "reason": "`_is_hf_initialized` regression and TimesFM `window_size` attribute error are different bugs." - }, - { - "left": "issue:43322", - "right": "issue:43540", - "accept": false, - "reason": "Segfault on Llava Next load vs Qwen3OmniMoe video ValueError; different models and symptoms." + "reason": "One is the `_is_hf_initialized` v5 regression; the other is a TimesFM `window_size` attribute bug." }, { "left": "issue:44479", "right": "issue:45325", "accept": false, - "reason": "Both touch Qwen VL, but one is a broad video-input regression and the other is a still-image rope-index scaling bug." + "reason": "Both involve Qwen vision/video behavior, but one is a broad video-input regression and the other is a specific rope-index scaling bug; not enough to treat as the same concrete issue." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation overhead/perf issue is not the same as a cache length return-type bug." + "reason": "Generation-performance overhead is a different problem from the cache length return-type bug." }, { - "left": "issue:43425", - "right": "issue:44403", + "left": "issue:43322", + "right": "issue:43540", "accept": false, - "reason": "Torch version incompatibility vs noisy logging on load; not the same defect." + "reason": "Llava Next model loading segfault vs Qwen3OmniMoe video-input `ValueError`; distinct failures." }, { - "left": "issue:43296", - "right": "issue:43531", + "left": "issue:43425", + "right": "issue:44403", "accept": false, - "reason": "PaddleOCR-VL load failure in vLLM and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "Torch version incompatibility and noisy transformer loading are unrelated." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Missing tied-weights metadata vs tokenizer load failure for a specific model; no shared code-path evidence." + "reason": "Missing `all_tied_weights_keys` on Molmo is unrelated to the sleng-bert tokenizer loading failure." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM and Qwen3-MoE sliding-window behavior are different bugs." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen2.5-VL related, but training with multiple images is not the same bug as rope-index scaling." + "reason": "Both are Qwen2.5-VL related, but training with multi-image iterable datasets is a different bug from temporal position-id scaling." }, { - "left": "issue:45216", - "right": "issue:45357", + "left": "issue:43872", + "right": "issue:44291", "accept": true, - "reason": "Both report the same Qwen3.5 `save_pretrained` regression producing incorrect checkpoint contents/keys." + "reason": "Same underlying regression: both report v5 passing an unexpected `_is_hf_initialized` kwarg into model-loading/init code, with bitsandbytes as one affected path." }, { - "left": "issue:43872", - "right": "issue:44291", + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction issue and compiled flex-attention failure are unrelated." + }, + { + "left": "issue:45216", + "right": "issue:45357", "accept": true, - "reason": "Same underlying transformers v5 `_is_hf_initialized` compatibility break, with one report surfacing through bitsandbytes and the other through `init_empty_weights`." + "reason": "Same Qwen3.5 save_pretrained regression: both describe incorrect checkpoint/visual-encoder key saving after v5.4/5.5 changes." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "Base-model-prefix loading issue and junk-filled non-persistent buffers are different regressions." + "reason": "`base_model_prefix` loading break and junk-filled non-persistent buffers are separate v5 regressions." }, { - "left": "issue:44220", - "right": "issue:44683", + "left": "issue:43716", + "right": "issue:46899", "accept": false, - "reason": "Audio fbank extraction and compiled flex_attention failures are unrelated." + "reason": "Both mention dtype issues in multimodal models, but they are different models and different failure mechanisms." }, { "left": "issue:44991", "right": "issue:46315", "accept": false, - "reason": "Tokenizer load breakage vs tokenizer performance regression; different problems." - }, - { - "left": "issue:43716", - "right": "issue:46899", - "accept": false, - "reason": "Both are multimodal dtype/quantization issues, but they affect different models and code paths." + "reason": "Tokenizer load failure vs `convert_tokens_to_ids` slowdown are different tokenizer issues." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Generic loading noise is not the same as tokenizer loading failure." + "reason": "Generic load noise and a tokenizer load regression are not the same bug." }, { - "left": "issue:42796", - "right": "issue:43746", + "left": "issue:45812", + "right": "issue:46491", "accept": false, - "reason": "Tokenizer BPE-marker handling and PEFT local-checkpoint loading are unrelated." + "reason": "Both are tokenizer regressions, but they affect different model families and different behavior (wrong IDs vs lost leading space)." }, { - "left": "issue:45812", - "right": "issue:46491", + "left": "issue:42796", + "right": "issue:43746", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and behaviors (token IDs vs leading-space round-trip)." + "reason": "Tokenizer BPE-marker handling and GraniteSpeech PEFT checkpoint loading are unrelated." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Eval-mode checkpointing restriction and missing gradient-checkpointing support flag are different issues." + "reason": "Gradient-checkpointing-in-eval and missing `supports_gradient_checkpointing` are different code-path problems." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "`_is_hf_initialized` argument handling is unrelated to missing Float8 storage support." + "reason": "Unexpected `_is_hf_initialized` kwarg and missing Float8 storage support are different compatibility failures." }, { - "left": "issue:33666", - "right": "issue:40444", + "left": "issue:43792", + "right": "issue:44589", "accept": false, - "reason": "Different Qwen VL training problems with no concrete shared failure mode." + "reason": "Whisper loading failure and Float8 storage lookup error are unrelated." }, { - "left": "issue:43792", - "right": "issue:44589", + "left": "issue:33666", + "right": "issue:40444", "accept": false, - "reason": "Whisper runtime failure and Float8 storage lookup failure are unrelated." + "reason": "Qwen2-VL multi-GPU training and Qwen2.5-VL multi-image iterable-dataset failure are not the same bug." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "Transformers v5 init-flag regression and Strix Halo mmap OOM are unrelated." + "reason": "A v5 `_is_hf_initialized` regression and Strix Halo mmap OOM are unrelated." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "High perplexity on gpt-oss and Whisper batch_decode behavior are completely separate bugs." + "reason": "High perplexity on a language-model eval and Whisper `batch_decode` skipping special tokens are different issues." }, { "left": "issue:42796", "right": "issue:46710", "accept": false, - "reason": "Tokenizer BPE-marker handling and DeepSeek accuracy regression do not share a concrete code path." + "reason": "Tokenizer BPE-marker handling and DeepSeek accuracy regression are unrelated." }, { - "left": "issue:43883", - "right": "issue:44991", + "left": "issue:35707", + "right": "issue:40444", "accept": false, - "reason": "Molmo tied-weights metadata and tokenizer loading failure are unrelated." + "reason": "Progressive generation with `inputs_embeds`/`past_key_values` is unrelated to Qwen2.5-VL multi-image iterable training." }, { - "left": "issue:35707", - "right": "issue:40444", + "left": "issue:43531", + "right": "issue:44291", "accept": false, - "reason": "Progressive generation with embeds/past KV is unrelated to Qwen2.5-VL multi-image iterable dataset finetuning." + "reason": "Qwen3-MoE sliding-window behavior and the `_is_hf_initialized` constructor-arg regression are different failures." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative. The two accepted soft-edge pairs are plausible duplicates based on the issue titles and described failure modes, and the rest of the rejections are appropriately cautious. The heterogeneous-cluster conclusion also matches the evidence." + "feedback": "The summary is grounded in the packet, and the two accepted soft edges are conservative clear matches: issue:43872\u2194issue:44291 share the same unexpected `_is_hf_initialized` regression, and issue:45216\u2194issue:45357 are both Qwen3.5 `save_pretrained` checkpoint/key-save regressions. The other pairs are appropriately rejected as distinct bugs." }, "evaluator_used": true, "retried": false, @@ -4105,7 +4100,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:40444", "issue:41720", @@ -4119,7 +4114,7 @@ "issue:43482", "issue:43493", "issue:43525", - "issue:43531", + "issue:43606", "issue:43611", "issue:43644", "issue:43646", @@ -4158,250 +4153,256 @@ ], "soft_pairs": [ "issue:44079|issue:45663", - "issue:43531|issue:44291", "issue:43482|issue:43525", "issue:41720|issue:45081", "issue:43421|issue:44568", "issue:42175|issue:45542", "issue:43296|issue:46821", - "issue:44462|issue:45812", "issue:42175|issue:43296", - "issue:44079|issue:44534", + "issue:44462|issue:45812", "issue:43482|issue:45538", "issue:44509|issue:46909", "issue:43317|issue:43856", + "issue:44079|issue:44534", "issue:43611|issue:43646", "issue:43377|issue:45491", + "issue:43818|issue:46402", "issue:44265|issue:45072", "issue:43644|issue:44079", - "issue:43818|issue:46402", - "issue:40444|issue:45381", "issue:45325|issue:45834", + "issue:40444|issue:45381", "issue:43818|issue:45357", "issue:43493|issue:43994", "issue:45325|issue:45381", "issue:44291|issue:44877", "issue:43454|issue:44466", - "issue:40444|issue:47217", "issue:44829|issue:45479", + "issue:40444|issue:47217", "issue:43873|issue:45701", "issue:43701|issue:45325", - "issue:42796|issue:45381" + "issue:42796|issue:45381", + "issue:43606|issue:44265" ], - "prepared_review_unit_hash": "fca553163abd2edc37c8e2f649d51276cc3fd5c258136efb3283100f95ba48ad", + "prepared_review_unit_hash": "cfeb5c69f0589d8699af65382269266a9d70c6ff49bd473c2d174a7171fc277c", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36997, - "estimated_input_tokens": 9250, - "estimated_eval_tokens": 18756 + "serialized_chars": 37034, + "estimated_input_tokens": 9259, + "estimated_eval_tokens": 18774 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: many items share broad areas like tokenizers, multimodal models, or v5 regressions, but usually describe different concrete bugs. The clearest duplicate subclusters are the AutoTokenizer/tokenization regression pair and the SigLIP2 output-discrepancy pair. No PRs are present.", - "confidence": 0.78, - "canonical_issue_reason": "issue:44462 is the broadest representative of the tokenizer regression subcluster: it describes AutoTokenizer choosing the wrong repository tokenizer path, which plausibly underlies the Granite token-ID failures.", + "summary": "The cluster is mostly heterogeneous. A few soft edges are real duplicates around Transformers v5 tokenizer/model regressions, but most pairs are unrelated broad-similarity matches and should stay separate.", + "confidence": 0.69, + "canonical_issue_reason": "Best anchor for the cluster is issue 44462: it is a concrete, high-impact v5 AutoTokenizer regression with a clear duplicate-like counterpart (45812) and broader applicability than the narrower follow-up reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44462 is the best issue representative because it is generic, user-facing, and matches the broad tokenizer-loading regression theme seen in the closest duplicate candidate.", + "best_issue_reason": "Issue 44462 is the strongest representative issue in this set because it captures a specific, user-visible regression in core tokenizer loading and generalizes beyond a single model family.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "Different bugs: ModelOutput key bookkeeping vs Gemma4 shared_kv_states/FSDP behavior." - }, - { - "left": "issue:43531", - "right": "issue:44291", - "accept": false, - "reason": "Unrelated regressions: Qwen3-MoE sliding_window behavior vs init_empty_weights unexpected argument handling." + "reason": "Different bugs: ModelOutput key assignment vs Gemma4 shared_kv_states/FSDP2 behavior." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Different failure modes and code paths: GGUF/v5 loading vs Llama4Config missing pad_token_id." + "reason": "Unrelated failures: GGUF loading in v5 vs missing pad_token_id on Llama4Config." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Distinct areas: Qwen3 auto device mapping cuda assert vs Mistral regex patch tokenizer crash." + "reason": "Different code paths and symptoms: Qwen3 device mapping CUDA assert vs Mistral tokenizer regex patch crash." }, { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Both involve special tokens, but one is runtime post-processor sync while the other is add_special_tokens behavior for a specific tokenizer." + "reason": "Both involve special tokens, but one is runtime post-processor updates while the other is add_special_tokens behavior for a specific tokenizer." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "Packaging/backends are related, but these are different dependency mistakes and error conditions." + "reason": "Both touch TF packaging/backend setup, but the reported failures are different and not clearly the same bug." }, { "left": "issue:43296", "right": "issue:46821", "accept": false, - "reason": "Completely different models and failures: PaddleOCR-VL loading vs TimesFM window_size attribute access." - }, - { - "left": "issue:44462", - "right": "issue:45812", - "accept": true, - "reason": "Same tokenizer regression family: AutoTokenizer is using the wrong tokenizer path, and Granite wrong token IDs look like a concrete manifestation of that same bug." + "reason": "Different models and failure modes: PaddleOCR-VL load error vs TimesFM window_size AttributeError." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Different topics: missing tensorflow in an install extra vs model loading failure in vLLM." + "reason": "No meaningful overlap beyond broad dependency/backend concerns." }, { - "left": "issue:44079", - "right": "issue:44534", - "accept": false, - "reason": "Different bugs despite similar wording: ModelOutput key assignment vs non-persistent buffer contents." + "left": "issue:44462", + "right": "issue:45812", + "accept": true, + "reason": "Same underlying regression: AutoTokenizer/tokenizer.json handling causing wrong Granite token IDs in v5." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "Unrelated tokenizer issues: GGUF/v5 loading failure vs CLIPTokenizer max length setting." + "reason": "Different issues: GGUF model loading vs CLIPTokenizer max length handling." }, { "left": "issue:44509", "right": "issue:46909", "accept": false, - "reason": "Both are documentation problems, but they concern different docs and different concrete fixes." + "reason": "Both are docs-related, but one is removed pipeline tasks and the other is a broken DIA example." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "Both mention offloading/memory, but one is dequantized loading and the other is Qwen3 MoE training memory usage." + "reason": "Different problems: quantized offload loading failure vs Qwen3 MoE training memory usage." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "Different bugs: ModelOutput key bookkeeping vs non-persistent buffer initialization." }, { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Related to v5 model loading, but the concrete bugs differ: base_model_prefix handling vs custom model initialization." + "reason": "Both are v5 loading regressions, but the concrete causes differ: base_model_prefix handling vs custom model initialization." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Different batching/mask problems in different models; not the same underlying bug." + "reason": "Different models and root causes: padding-mask support in MIMI vs Gemma3 NaNs with sliding-window attention." + }, + { + "left": "issue:43818", + "right": "issue:46402", + "accept": false, + "reason": "Unrelated model-specific failures: Video-LLaVA architecture issue vs Sam3 text encoder weights not loading." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "No shared code-path: torch.export / compilable checks vs bf16 dtype mismatch in inference." + "reason": "Different subsystems: torch.export/torch_compilable_check vs bfloat16 dtype mismatch inference." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "One is buffer serialization state, the other is ModelOutput field bookkeeping." + "reason": "Non-persistent buffer junk vs ModelOutput key assignment are separate bugs." }, { - "left": "issue:43818", - "right": "issue:46402", + "left": "issue:45325", + "right": "issue:45834", "accept": false, - "reason": "Different multimodal/model-weight issues: Video-LLaVA tower weights vs Sam3 text encoder loading." + "reason": "Both are vision/model-specific, but the actual failures are unrelated (Qwen2.5-VL RoPE indexing vs Kosmos2.5 long OCR index error)." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL multimodal issues, but one is iterable-dataset multi-image finetuning while the other is video vision_position_ids." - }, - { - "left": "issue:45325", - "right": "issue:45834", - "accept": false, - "reason": "Different models and failures: Qwen2.5-VL rope/position-id scaling vs Kosmos2.5 OCR index error." + "reason": "Both mention Qwen2.5-VL/image positions, but one is finetuning with multiple images and the other is video vision_position_ids." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Not the same bug: missing temporal attention / shared weights vs incorrect visual encoder key saving." + "reason": "Different issues in different models: Video-LLaVA weights/temporal attention vs Qwen3.5 save_pretrained key serialization." }, { "left": "issue:43493", "right": "issue:43994", "accept": true, - "reason": "Likely the same SigLIP2 implementation problem: one compares HF vs JAX behavior, the other reports nonsense outputs from the same model in AutoModel/pipeline." + "reason": "Same underlying SigLIP2 bug: HF implementation diverges from the original and yields bad/nonsensical outputs." }, { "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Both involve Qwen2.5-VL positions, but the exact bugs differ between still-image scaling and video input handling." + "reason": "Same model family, but still-image temporal position_ids and video vision_position_ids are distinct bug reports." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Different v5 loading regressions: init_empty_weights signature mismatch vs strict config validation." + "reason": "Different load-time failures: init_empty_weights unexpected argument vs strict config blocking granite_speech." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "Both mention lm_head weights, but one is weight tying causing generation issues and the other is serialization varying by device." + "reason": "Both involve lm_head/tied weights, but one is generation quality and the other is save serialization differing by device." }, { - "left": "issue:40444", - "right": "issue:47217", + "left": "issue:44829", + "right": "issue:45479", "accept": false, - "reason": "Similar multimodal symptoms, but not enough evidence they are the same concrete bug." + "reason": "Different training regressions with different triggers and model families." }, { - "left": "issue:44829", - "right": "issue:45479", + "left": "issue:40444", + "right": "issue:47217", "accept": false, - "reason": "Different classification/training regressions with different triggers and model families." + "reason": "Both are VL placeholder-related, but the failing scenarios are different: multi-image finetuning vs placeholder placement in user text." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Offloading/quantization is unrelated to tokenizer-version behavior changes." + "reason": "Offloading/quantization behavior and tokenization-version changes are unrelated." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL rope-index scaling." + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL RoPE indexing." }, { "left": "issue:42796", "right": "issue:45381", "accept": false, - "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL vision-position IDs are different bugs." + "reason": "Different tokenizer/model bugs: BPE marker handling vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "Different bugs: CPU offload device mismatch vs torch.export compatibility." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-identical duplicate reports of the same non-persistent buffer junk regression in v5." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary stays grounded in the packet, correctly notes the cluster\u2019s heterogeneity, and the two accepted soft-edge pairs are plausible and phrased conservatively. The rejected pairs also avoid overmerging clearly distinct bugs. No PR-pair mergeability issue is present." + "feedback": "The summary is well grounded in the packet and stays conservative overall. The two accepted soft edges are plausible same-bug duplicates from the titles, while the rejected pairs are mostly explained by clearly different symptoms, models, or subsystems. No obvious overstatement stands out." }, "evaluator_used": true, "retried": false, @@ -4420,7 +4421,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:34634", "issue:35707", @@ -4471,259 +4472,252 @@ "issue:46821" ], "soft_pairs": [ - "issue:43606|issue:44265", "issue:44479|issue:45823", - "issue:40990|issue:45584", "issue:43828|issue:46821", + "issue:40990|issue:45584", "issue:43262|issue:43746", "issue:34634|issue:35707", "issue:35707|issue:41720", - "issue:41720|issue:45823", "issue:44186|issue:44265", + "issue:41720|issue:45823", "issue:43606|issue:43819", "issue:43454|issue:43550", "issue:43257|issue:45923", "issue:45823|issue:46821", "issue:42617|issue:42915", "issue:40990|issue:42947", - "issue:42796|issue:46209", "issue:43454|issue:45072", - "issue:44568|issue:45920", "issue:43377|issue:44610", + "issue:42796|issue:46209", + "issue:44568|issue:45920", "issue:41553|issue:43582", "issue:36331|issue:42617", - "issue:43749|issue:44534", "issue:43317|issue:43873", + "issue:43749|issue:44534", "issue:43454|issue:43720", "issue:43377|issue:44898", - "issue:41720|issue:43322", "issue:43572|issue:45759", + "issue:41720|issue:43322", "issue:46305|issue:46315", "issue:36331|issue:42915", "issue:44877|issue:44991", "issue:43322|issue:46050", "issue:41720|issue:44977" ], - "prepared_review_unit_hash": "2562ebf4937545f2b55c52e503053306febbb8bd31b57a2e309b344c00876863", + "prepared_review_unit_hash": "0828ba4d2fb3b8e23c37e8d8f960e5c4f9493601c7f3b161f73cc97cb7676d9c", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36774, - "estimated_input_tokens": 9194, - "estimated_eval_tokens": 18644 + "soft_pair_count": 31, + "serialized_chars": 36473, + "estimated_input_tokens": 9119, + "estimated_eval_tokens": 18494 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated model-specific bug reports. The closest thematic overlap is around offloading/quantization/loading, but even those reports describe different failure modes and code paths, so none of the soft pairs should be merged.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43873 is the broadest open item and the closest umbrella for the small offloading/quantization subset, so it is the best single issue to anchor triage.", + "summary": "The cluster is very heterogeneous: most items are distinct transformer regressions across tokenization, generation, loading/offloading, audio, and export paths. I found no soft pair that looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43873 is the broadest umbrella-style report in the set and best represents the offloading/quantization failure class, with multiple inbound references and discussion. The cluster is not truly duplicate-heavy, but this is the most suitable representative issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the best global issue candidate because it is open, broadly scoped, and closest to the only meaningful subtheme in the cluster; the rest are narrower or clearly unrelated.", + "best_issue_reason": "43873 is the best global issue candidate because it describes the core system-level regression rather than a narrow model-specific symptom, making it the most useful triage anchor for related reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43606", - "right": "issue:44265", - "accept": false, - "reason": "Both are loading/runtime failures, but one is a CPU offload device mismatch and the other is a torch.export check failure; different code paths and symptoms." - }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Both involve multimodal models, but one is a video input regression and the other is a device mismatch in PLE forward; not the same bug." + "reason": "Both are multimodal/model-loading regressions, but one is a video-input regression in Qwen VL variants and the other is a Gemma4 device mismatch during forward. Different code paths and symptoms." }, { - "left": "issue:40990", - "right": "issue:45584", + "left": "issue:43828", + "right": "issue:46821", "accept": false, - "reason": "High perplexity on a model is unrelated to Whisper empty-transcription generation failure after token alignment." + "reason": "Different models and failures: autocast dtype mismatch in Phi-tiny-MoE versus an AttributeError for TimesFM window_size." }, { - "left": "issue:43828", - "right": "issue:46821", + "left": "issue:40990", + "right": "issue:45584", "accept": false, - "reason": "A torch.autocast dtype mismatch and a TimesFM window_size AttributeError are unrelated failures." + "reason": "Perplexity regression on gpt-oss-20b is unrelated to Whisper failing on empty transcription after align_special_tokens." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech PEFT checkpoint loading are distinct processor vs model-loading issues." + "reason": "Audio-related issues, but one is chat template sampling-rate defaulting and the other is PEFT adapter checkpoint loading. Not the same bug." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "BarkProcessor voice_preset not working and progressive generation with inputs_embeds/past_key_values are different features and failure modes." + "reason": "BarkProcessor voice_preset and progressive generation with inputs_embeds/past_key_values are unrelated bugs." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Progressive generation with cache inputs and Qwen3 auto device-map CUDA assert are unrelated code paths." + "reason": "Generation cache behavior and Qwen3 auto device mapping/cuda assert are different failure modes in different paths." }, { - "left": "issue:41720", - "right": "issue:45823", + "left": "issue:44186", + "right": "issue:44265", "accept": false, - "reason": "Both mention device-related failures, but one is a CUDA assert during Qwen3 loading and the other is a PLE forward mismatch in Gemma4; not duplicates." + "reason": "Tokenizer crash on NER/batching is unrelated to torch.export failing on torch_compilable_check." }, { - "left": "issue:44186", - "right": "issue:44265", + "left": "issue:41720", + "right": "issue:45823", "accept": false, - "reason": "Tokenizer padding/truncation crashes and torch.export torch_compilable_check failures are separate subsystems." + "reason": "Both involve device placement, but one is a cudaErrorAssert on Qwen3 auto device mapping and the other is a Gemma4 PLE device mismatch during forward." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "CPU offload device mismatch and DAC.from_latents/STE inconsistency do not describe the same underlying bug." + "reason": "Bark-small CPU offload device mismatch and DAC.from_latents mismatch with missing STE are different models and different bugs." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "A missing lm_head tie in AyaVision and a torch.compile SDPA failure in Bamba are different model-specific issues." + "reason": "AyaVision weight-tying failure and Bamba torch.compile/SDPA failure are unrelated." }, { "left": "issue:43257", "right": "issue:45923", "accept": false, - "reason": "MoE weight conversion with accelerate+deepspeed and a missing supports_gradient_checkpointing flag are unrelated loading/config problems." + "reason": "Both touch Qwen3/MoE-ish configs, but one is weight conversion under accelerate+deepspeed and the other is a missing supports_gradient_checkpointing flag." }, { "left": "issue:45823", "right": "issue:46821", "accept": false, - "reason": "Gemma4 device-mapping mismatch and TimesFM window_size AttributeError are different models and different bugs." + "reason": "Gemma4 device mismatch and TimesFM window_size AttributeError are unrelated." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "A script execution failure and a FineGrainedFP8Config model-loading failure are not the same issue." + "reason": "3d_parallel.py runtime failure is not the same as Qwen3Moe failing with FineGrainedFP8Config." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "Perplexity evaluation on a model and gradient checkpointing/LoRA behavior are unrelated." + "reason": "High perplexity on a model is not the same bug as gradient checkpointing being ineffective with PEFT LoRA." }, { - "left": "issue:42796", - "right": "issue:46209", + "left": "issue:43454", + "right": "issue:45072", "accept": false, - "reason": "Tokenizer BPE marker handling and FP8 inference on B200 are distinct issues with different components." + "reason": "Weight tying for AyaVision and dtype mismatches in SwitchTransformers/TimmWrapperModel are different issues." }, { - "left": "issue:43454", - "right": "issue:45072", + "left": "issue:43377", + "right": "issue:44610", "accept": false, - "reason": "AyaVision lm_head tying and dtype mismatches in SwitchTransformers/TimmWrapperModel are separate bugs." + "reason": "MIMI padding-mask inconsistency and OmDet-Turbo image-size mismatch are unrelated processor/model bugs." }, { - "left": "issue:44568", - "right": "issue:45920", + "left": "issue:42796", + "right": "issue:46209", "accept": false, - "reason": "Missing BOS/EOS insertion for one tokenizer and wrong token IDs across many tokenizers are related only at a very broad tokenizer-regression level, not the same concrete bug." + "reason": "Tokenizer BPE marker handling and FP8 inference failure on B200 are different bugs." }, { - "left": "issue:43377", - "right": "issue:44610", + "left": "issue:44568", + "right": "issue:45920", "accept": false, - "reason": "MIMI padding-mask batch inconsistency and OmDet-Turbo processor size mismatch are unrelated processor/model bugs." + "reason": "add_special_tokens BOS/EOS regression in one tokenizer is not the same as broad wrong-token-id regressions across many tokenizers." }, { "left": "issue:41553", "right": "issue:43582", "accept": false, - "reason": "Bad tokenizer error messaging for Voxtral and an Apple Silicon TypeError in caching_allocator_warmup are unrelated." + "reason": "Bad AutoTokenizer error message for Voxtral is unrelated to caching_allocator_warmup raising TypeError on Apple Silicon." }, { "left": "issue:36331", "right": "issue:42617", "accept": false, - "reason": "Trainer compute_loss signature mismatch and a 3d_parallel.py runtime failure are different layers of the stack." + "reason": "CustomTrainer compute_loss signature breakage and 3d_parallel.py startup failure are different problems." }, { - "left": "issue:43749", - "right": "issue:44534", + "left": "issue:43317", + "right": "issue:43873", "accept": false, - "reason": "Broken FSDP CPU RAM efficient loading and junk-filled non-persistent buffers are different regression classes." + "reason": "Both involve offloading/quantization, but one is dequantized model loading with device_map=auto and the other is a broader offloading regression. Not enough evidence of the same concrete code path." }, { - "left": "issue:43317", - "right": "issue:43873", + "left": "issue:43749", + "right": "issue:44534", "accept": false, - "reason": "These are the closest pair in the cluster, but one is specifically failure to load dequantized models with gpu+cpu offload while the other is a broader open report about offloading with quantization; similar area, not the same concrete bug." + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk in v5 are distinct regressions." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "AyaVision weight-tying failure and BitNet packed-weight unpacking failure are unrelated model implementation issues." + "reason": "AyaVision lm_head tying and BitNet packed-weight unpacking during accelerate loading are unrelated." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "MIMI batching/padding-mask mismatch and Perceiver non-default resolution interpolation failure are separate vision/audio model problems." + "reason": "MIMI batched padding-mask bug and Perceiver non-default resolution failure are different model-specific issues." }, { - "left": "issue:41720", - "right": "issue:43322", + "left": "issue:43572", + "right": "issue:45759", "accept": false, - "reason": "A CUDA assert in Qwen3 auto device mapping and a segmentation fault loading Llava Next are different failure modes and models." + "reason": "Missing pad_token_idx in StableLmConfig and failure to unwrap text_config for composite Qwen multimodal configs are different config-loading bugs." }, { - "left": "issue:43572", - "right": "issue:45759", + "left": "issue:41720", + "right": "issue:43322", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and not unwrapping text_config for composite Qwen multimodal configs are distinct config-loading issues." + "reason": "Qwen3 auto device mapping cuda assert and Llava Next segmentation fault during loading are unrelated failures." }, { "left": "issue:46305", "right": "issue:46315", "accept": false, - "reason": "An off-by-one in sliding window masking and a tokenizer performance regression are unrelated." + "reason": "A sliding-window off-by-one bug and a slow convert_tokens_to_ids regression are not the same issue." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Trainer API signature breakage and Qwen3Moe FineGrainedFP8Config loading failure are different components and symptoms." + "reason": "CustomTrainer compute_loss keyword incompatibility and Qwen3Moe FineGrainedFP8Config failure are unrelated." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Strict config blocking granite_speech and tokenizer loading failure for est-roberta are different model/tokenizer compatibility problems." + "reason": "Strict granite_speech config loading and tokenizer loading failure for est-roberta are different code paths." }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "A Llava Next loading segfault and sinq quantized model save/load breakage are not the same underlying bug." + "reason": "Llava Next load-time segfault and quantized-model save/load breakage after sinq are different bugs." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Qwen3 auto device-map CUDA assert and Qwen3.5 flash-attention generation issues are related only by family name, not by concrete code path." + "reason": "Both are Qwen3-family issues, but one is a device-mapping cuda assert and the other is a flash-attention generation regression. Not the same underlying bug." } ] }, @@ -4745,7 +4739,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:26413", "issue:28530", @@ -4801,15 +4795,15 @@ "issue:40990|issue:45701", "issue:41720|issue:42617", "issue:43450|issue:45701", - "issue:46897|issue:47317", "issue:44466|issue:45357", + "issue:46897|issue:47317", "issue:43873|issue:44756", "issue:26413|issue:28530", - "issue:44186|issue:45072", "issue:45685|issue:46821", + "issue:44186|issue:45072", "issue:43295|issue:45701", - "issue:44451|issue:47020", "issue:43065|issue:46402", + "issue:44451|issue:47020", "issue:36010|issue:42617", "issue:46612|issue:46858", "issue:43646|issue:46665", @@ -4818,223 +4812,223 @@ "issue:43550|issue:43606", "issue:44186|issue:44610", "issue:43404|issue:45357", + "issue:43881|issue:44488", "issue:42915|issue:43317", "issue:46821|issue:46869", - "issue:43881|issue:44488", - "issue:41720|issue:45161", "issue:44610|issue:44898", + "issue:41720|issue:45161", "issue:28530|issue:33666", "issue:44008|issue:45823", "issue:44423|issue:45406" ], - "prepared_review_unit_hash": "511efc4a964b31d5a480c1861e8726046b86533aee8e56a8c6b30af63c2c2612", + "prepared_review_unit_hash": "e09ab4085d731ab4f803df10085c438d1402bb20107dfaaf2840796ebf292fc6", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36434, - "estimated_input_tokens": 9109, - "estimated_eval_tokens": 18474 + "serialized_chars": 36441, + "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 loose semantic bundle of unrelated Transformers issues; the apparent matches are all false positives, so no soft edges are accepted and there is no defensible single canonical issue.", - "confidence": 0.96, - "canonical_issue_reason": "No clear canonical issue: the cluster spans unrelated bugs across training, loading, quantization, tokenization, serving, and device placement, and none of the soft pairs look like true duplicates.", + "summary": "This is a loose similarity bucket of largely unrelated Hugging Face issues spanning Trainer resume logic, tokenizer/model loading, multimodal processors, quantization/offload, and serving/runtime failures. None of the proposed soft pairs look like the same underlying bug, so all soft edges are rejected.", + "confidence": 0.69, + "canonical_issue_reason": "issue:43873 is the most central anchor in this loose bucket: it has the most discussion/inbound reference signal and a broad, concrete failure mode that makes it the best representative issue.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a safe dedupe target; the strongest-looking candidates are still too broad and do not explain the rest of the cluster.", + "best_issue_reason": "issue:43873 is the strongest single issue to represent the cluster because it is well-discussed, externally referenced, and specific enough to summarize the general offloading/quantization theme without overcommitting to a false duplicate merge.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Int4 quantization OOM/reserved-memory growth is unrelated to an Apex RMSNorm memory leak in T5." + "reason": "Both mention memory problems, but one is int4 quantization/OOM and the other is an apex FusedRMSNorm leak in T5; different root causes." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "GPT-OSS perplexity regression and tokenizer changes are different symptoms with different likely causes." + "reason": "One is perplexity quality on gpt-oss-20b, the other is tokenizer behavior changing across versions; not the same bug." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "Qwen3 auto device-map CUDA assert and 3d_parallel.py failure are different code paths." + "reason": "Qwen3 auto device-map CUDA assert vs a 3d_parallel.py runtime failure; unrelated models/code paths." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Batched video processor shape mismatch is unrelated to a tokenizer/version behavior issue." + "reason": "Video processor output-shape bug vs tokenization changes; different components and symptoms." }, { - "left": "issue:46897", - "right": "issue:47317", + "left": "issue:44466", + "right": "issue:45357", "accept": false, - "reason": "Both mention training loss, but they involve different models and no shared concrete bug path." + "reason": "Both concern saving weights, but one is lm_head serialization by device and the other is incorrect visual encoder keys in Qwen3.5." }, { - "left": "issue:44466", - "right": "issue:45357", + "left": "issue:46897", + "right": "issue:47317", "accept": false, - "reason": "lm_head tying/serialization differs from incorrect visual-encoder save keys." + "reason": "Both are training-loss issues, but for different models and likely different loss code paths; too generic to merge." }, { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Quantization offloading failure is not the same as Strix Halo mmap/OOM behavior." + "reason": "Quantization/offload bug vs Strix Halo mmap OOM; related theme but not the same underlying failure." }, { "left": "issue:26413", "right": "issue:28530", "accept": false, - "reason": "Checkpoint resume epoch-iterator failure and early-stopping metric lookup are distinct Trainer bugs." + "reason": "Checkpoint resume sample-iteration failure vs early-stopping metric lookup; separate Trainer behaviors." }, { - "left": "issue:44186", - "right": "issue:45072", + "left": "issue:45685", + "right": "issue:46821", "accept": false, - "reason": "Tokenizer padding/truncation crash and dtype mismatch are different failures." + "reason": "MPS histogram/int error in MoE vs TimesFM window_size attribute access; unrelated subsystems." }, { - "left": "issue:45685", - "right": "issue:46821", + "left": "issue:44186", + "right": "issue:45072", "accept": false, - "reason": "MPS Int histogram error is unrelated to TimesFM's window_size attribute error." + "reason": "Tokenizer padding/truncation crash vs bfloat16 dtype mismatch in models; different failure modes." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Processor/tokenizer regression is not the same bug as version-dependent tokenization changes." + "reason": "Processor.tokenizer regression vs tokenization changes; one is API regression, the other text output drift." }, { - "left": "issue:44451", - "right": "issue:47020", + "left": "issue:43065", + "right": "issue:46402", "accept": false, - "reason": "Model load failure for ScandiBERT and XLMRobertaTokenizer/Unigram init bug are not the same issue." + "reason": "Sam3 pixel decoder dummy Conv2d vs text-encoder weights not loading; distinct Sam3 bugs." }, { - "left": "issue:43065", - "right": "issue:46402", + "left": "issue:44451", + "right": "issue:47020", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and missing Sam3 text-encoder weights affect different parts of the model." + "reason": "Model-loading failure for ScandiBERT vs Unigram/vocab type mismatch in XLMRobertaTokenizer; separate tokenizer/model issues." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "GenerationMixin import error and 3d_parallel.py runtime failure are unrelated." + "reason": "GenerationMixin import error vs 3d_parallel.py execution failure; unrelated." }, { "left": "issue:46612", "right": "issue:46858", "accept": false, - "reason": "Beam-search reorder omission and static-cache seq_length arithmetic are separate generation bugs." + "reason": "Beam-search cache reorder skipped in several models vs static-cache seq-length device tensor arithmetic; different generation bugs." }, { "left": "issue:43646", "right": "issue:46665", "accept": false, - "reason": "Custom model initialization breakage and from_pretrained-in-class behavior are too broad and not the same concrete defect." + "reason": "Custom model initialization breakage vs inconsistent from_pretrained inside Python classes; overlapping area but not the same concrete bug." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to BitNet packed-weight unpacking during loading." + "reason": "torch.compile+SDPA failure in Bamba vs BitNet packed-weight loading bug; different model-specific paths." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "BLIP2 dtype propagation and resume_from_checkpoint max_steps calculation are different bugs." + "reason": "Blip2 dtype not propagating vs max_steps miscalculated on checkpoint resume; unrelated loading vs Trainer logic." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Bamba compile/SDPA failure and Bark CPU-offload device mismatch are different models and code paths." + "reason": "Bamba compile/SDPA issue vs Bark CPU-offload device mismatch; different models and triggers." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "LayoutLMv2 tokenizer crash and OmDet-Turbo image-size mismatch are different preprocessing problems." + "reason": "LayoutLMv2 NER/tokenizer crash vs OmDet-Turbo image-size mismatch; different modalities and root causes." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "Mistral3 lm_head tying bug and Qwen3.5 visual-key serialization bug are unrelated." + "reason": "lm_head weight tying bug in Mistral3 vs incorrect visual encoder keys in Qwen3.5 save_pretrained; separate serialization problems." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "glm-4v-9b loading failure vs ScandiBERT loading failure; same symptom class, but different models and likely causes." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "FineGrainedFP8Config failure and dequantized gpu+cpu offload loading failure are distinct quantization paths." + "reason": "FineGrainedFP8Config failure vs dequantized model offload load failure; distinct quantization/offload paths." }, { "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 AttributeError vs YOLO multi-GPU device_map loading; unrelated." }, { - "left": "issue:43881", - "right": "issue:44488", + "left": "issue:44610", + "right": "issue:44898", "accept": false, - "reason": "Two different model-loading failures with no shared root cause shown." + "reason": "Both are vision-resolution issues, but OmDet-Turbo has a processor/model size mismatch while Perceiver has interpolation behavior failing; not identical." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Qwen3 CUDA assert and GPT-OSS tensor-parallelism failure are different distributed-inference bugs." - }, - { - "left": "issue:44610", - "right": "issue:44898", - "accept": false, - "reason": "Both touch vision sizes, but one is a processor/model size mismatch and the other is Perceiver interpolation handling." + "reason": "Qwen3 auto device-map assert vs GPT-OSS tensor parallelism issue; different parallelization bugs." }, { "left": "issue:28530", "right": "issue:33666", "accept": false, - "reason": "Early stopping metric lookup and Qwen2-VL multi-GPU training are unrelated." + "reason": "Early-stopping metric handling vs Qwen2-VL multi-GPU training; unrelated." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Gemma3n variable-name collision and Gemma4 PLE device mismatch are different model-forward bugs." + "reason": "Gemma3n forward variable collision vs Gemma4 PLE device mismatch; different model families and code paths." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Serve crash in continuous batching and Gemma4Processor _tokenizer absence are different serving failures." + "reason": "Continuous-batching multimodal crash with str.to vs Gemma4Processor _tokenizer missing; similar serving context but distinct attribute errors." }, { - "left": "issue:44387", - "right": "issue:45704", + "left": "issue:44451", + "right": "issue:47020", "accept": false, - "reason": "Int4 quantization OOM/reserved-memory growth is unrelated to an Apex RMSNorm memory leak in T5." + "reason": "Model loading failure for a specific BERT vs tokenizer constructor passing dict to Unigram; different failure stages." } ] }, @@ -5056,7 +5050,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28530", "issue:29159", @@ -5108,16 +5102,16 @@ "issue:47317" ], "soft_pairs": [ - "issue:42915|issue:46291", "issue:43531|issue:43572", + "issue:42915|issue:46291", "issue:36331|issue:41720", "issue:44734|issue:46858", "issue:45479|issue:47317", - "issue:44671|issue:44743", "issue:44464|issue:46858", + "issue:44671|issue:44743", "issue:44186|issue:44361", - "issue:45632|issue:45684", "issue:43716|issue:46821", + "issue:45632|issue:45684", "issue:28530|issue:30064", "issue:46894|issue:47317", "issue:43316|issue:44913", @@ -5126,206 +5120,206 @@ "issue:30064|issue:34634", "issue:42915|issue:43322", "issue:43452|issue:47020", - "issue:33357|issue:35707", "issue:43881|issue:44991", + "issue:33357|issue:35707", "issue:43873|issue:45704", - "issue:43540|issue:43782", "issue:43531|issue:45759", + "issue:43540|issue:43782", "issue:43526|issue:43883", + "issue:44492|issue:44509", "issue:44534|issue:46665", - "issue:41720|issue:43746", - "issue:44492|issue:44509" + "issue:41720|issue:43746" ], - "prepared_review_unit_hash": "c5e63347f8f4d7f9adfa9aeb42248cd61345e57ad2946807c33325e2004ffcbf", + "prepared_review_unit_hash": "2792a65cd59705c937407b9f507ebe7af449e4ef7f30b5504e28eef411fa01ff", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35730, - "estimated_input_tokens": 8933, - "estimated_eval_tokens": 18122 + "serialized_chars": 35726, + "estimated_input_tokens": 8932, + "estimated_eval_tokens": 18120 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated bug reports across different subsystems; none of the soft pairs look like the same underlying defect or change.", - "confidence": 0.88, - "canonical_issue_reason": "If one issue must serve as the anchor, issue:43873 is the strongest candidate because it is open, has inbound references, and describes a broad quantization/offloading regression rather than a narrowly model-specific symptom.", + "summary": "These are mostly unrelated bug reports across different components; none of the soft pairs look like true duplicates. I picked the most active, broadly scoped issue as the representative canonical issue.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 43873 is the most centrally connected and broadly impactful report here, with substantial discussion and inbound references; it is the best single representative of the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the most suitable standalone issue to surface for triage: it is active, externally referenced, and has the broadest user impact among this set.", + "best_issue_reason": "Issue 43873 is a strong standalone issue: active, well-scoped, and likely to matter to multiple users. It is the best issue to anchor follow-up triage.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:42915", - "right": "issue:46291", + "left": "issue:43531", + "right": "issue:43572", "accept": false, - "reason": "Different failures: Qwen3MoE FP8 loading versus a pipeline import dependency/version requirement." + "reason": "Different bugs: Qwen3-MoE sliding-window handling vs StableLmConfig missing pad_token_idx after v5 update." }, { - "left": "issue:43531", - "right": "issue:43572", + "left": "issue:42915", + "right": "issue:46291", "accept": false, - "reason": "Separate components and root causes: Qwen3-MoE sliding-window behavior versus StableLmConfig pad-token metadata." + "reason": "Different failure modes: Qwen3Moe FP8 loading vs pipelines import breakage from kernels version/revision requirements." }, { "left": "issue:36331", "right": "issue:41720", "accept": false, - "reason": "Unrelated problems: trainer loss-signature API change versus device-mapping CUDA assert on a Qwen3 model." + "reason": "Unrelated code paths: trainer compute_loss signature change vs Qwen3 auto device-mapping CUDA assert." }, { "left": "issue:44734", "right": "issue:46858", "accept": false, - "reason": "Both involve cache logic, but the concrete bugs differ: KV-cache continuation indexing in serve versus seq-length tensor arithmetic in prefill/static cache." + "reason": "Both involve cache logic, but the concrete failures differ: serve KV continuation indexing vs static-cache slice arithmetic on non-CUDA backends." }, { "left": "issue:45479", "right": "issue:47317", "accept": false, - "reason": "Both are training-loss bugs, but they affect different models and likely different loss code paths." + "reason": "Both are training-loss complaints, but they concern different models and no shared concrete bug pattern is established." }, { - "left": "issue:44671", - "right": "issue:44743", + "left": "issue:44464", + "right": "issue:46858", "accept": false, - "reason": "Different model families and mechanisms: CamemBERT MLM prediction bug versus recurrent-state reset with cache in Qwen3 modular code." + "reason": "Different problems in generation: compiled-forward inconsistency vs cache length tensor type breaking arithmetic." }, { - "left": "issue:44464", - "right": "issue:46858", + "left": "issue:44671", + "right": "issue:44743", "accept": false, - "reason": "Chunked generation with compiled forward is a different failure mode than static-cache seq-length slice arithmetic." + "reason": "Different model-specific behaviors: CamemBERT masked LM predictions vs Qwen3 recurrent state reset with cache." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Tokenizer crashes, but on different tokenizers and different task behaviors; not the same defect." + "reason": "Both are tokenizer/task failures, but they affect different tokenizers and likely different root causes." }, { - "left": "issue:45632", - "right": "issue:45684", + "left": "issue:43716", + "right": "issue:46821", "accept": false, - "reason": "Cache-path collisions and read-only permission propagation are unrelated filesystem issues." + "reason": "Different domains: multimodal dtype mismatch vs TimesFM window_size attribute error." }, { - "left": "issue:43716", - "right": "issue:46821", + "left": "issue:45632", + "right": "issue:45684", "accept": false, - "reason": "Different domains and symptoms: image preprocessor dtype mismatch versus TimesFM window_size attribute handling." + "reason": "Filesystem/cache-path collision is unrelated to read-only permission propagation in save_pretrained." }, { "left": "issue:28530", "right": "issue:30064", "accept": false, - "reason": "Early-stopping metric selection is unrelated to image processor segmentation-map handling." + "reason": "Unrelated: early stopping metric resolution vs image processor failure on void segmentation maps." }, { "left": "issue:46894", "right": "issue:47317", "accept": false, - "reason": "Both mention training loss, but they are model-specific reports with different concrete bugs; too broad to merge." + "reason": "Both mention training loss, but there is no evidence they share the same underlying bug or model path." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Config API discrepancy and rotary_pct reload persistence are different issues." + "reason": "Different configuration issues: Gemma3TextConfig API discrepancy vs GPTNeoX rotary_pct not persisting on reload." }, { "left": "issue:45701", "right": "issue:47145", "accept": false, - "reason": "Tokenizer/version behavior changes are unrelated to FP8 checkpoint weight_scale loading." + "reason": "Version-dependent tokenization changes are not the same as FP8 weight_scale being skipped on load." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "Slow/fast tokenizer inconsistency is a different bug from the T5 add_special_tokens name conflict." + "reason": "Tokenizer inconsistency vs T5Tokenizer add_special_tokens name conflict are different concrete bugs." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Different processors and failure modes; only both are input-processing issues at a high level." + "reason": "Different processors and behaviors: segmentation map processing vs BarkProcessor voice_preset handling." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "FP8/Qwen3MoE loading failure is not the same as Llava Next segmentation faults." + "reason": "FP8 quantization loading failure and Llava Next segmentation fault are distinct model-loading bugs." }, { "left": "issue:43452", "right": "issue:47020", "accept": false, - "reason": "Different tokenizer/model-loading bugs with different code paths and root causes." + "reason": "Different tokenizer/model loading issues affecting different APIs and root causes." }, { - "left": "issue:33357", - "right": "issue:35707", + "left": "issue:43881", + "right": "issue:44991", "accept": false, - "reason": "A macOS CLIP bus error is unrelated to progressive generation with inputs_embeds and past_key_values." + "reason": "Distinct regressions: glm-4v-9b load failure vs tokenizer loading breakage for a specific model." }, { - "left": "issue:43881", - "right": "issue:44991", + "left": "issue:33357", + "right": "issue:35707", "accept": false, - "reason": "Different loading failures: GLM-4V model load versus tokenizer load for a specific roberta model." + "reason": "MacOS bus error on CLIP loading is unrelated to progressive generation with inputs_embeds/past_key_values." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantization/offloading behavior and apex FusedRMSNorm memory leak are unrelated." + "reason": "Quantization offloading behavior is unrelated to T5's apex FusedRMSNorm memory leak." }, { - "left": "issue:43540", - "right": "issue:43782", + "left": "issue:43531", + "right": "issue:45759", "accept": false, - "reason": "Both are Qwen3 multimodal issues, but they affect different modalities and failure points." + "reason": "Qwen3-MoE sliding-window behavior is different from from_config not unwrapping text_config in composite multimodal configs." }, { - "left": "issue:43531", - "right": "issue:45759", + "left": "issue:43540", + "right": "issue:43782", "accept": false, - "reason": "Sliding-window handling in Qwen3-MoE is not the same bug as unwrapping text_config for composite multimodal configs." + "reason": "Both are Qwen multimodal issues, but one is video input processing and the other is from_pretrained weight_only loading." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "ImageProcessor label reduction and Molmo tied-weights metadata are unrelated." + "reason": "BeitImageProcessorFast reduce_labels bug is unrelated to Molmo tied-weights attribute failure." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A typo in cache strategy docs is not the same as stale docs still listing removed pipeline tasks." }, { "left": "issue:44534", "right": "issue:46665", "accept": false, - "reason": "Non-persistent buffer initialization and from_pretrained behavior inside Python classes are different regressions." + "reason": "Non-persistent buffer corruption in v5 is unrelated to inconsistent from_pretrained behavior inside Python classes." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Different model-specific failures: CUDA assert during auto device mapping versus local checkpoint loading with PEFT adapters." - }, - { - "left": "issue:44492", - "right": "issue:44509", - "accept": false, - "reason": "A cache-strategy typo and outdated pipeline-task docs are not the same issue." + "reason": "Qwen3 device-mapping CUDA assert and GraniteSpeech PEFT local-checkpoint loading are different code paths and failures." } ] }, @@ -5347,7 +5341,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:34567", "issue:34634", @@ -5403,8 +5397,8 @@ "issue:43257|issue:47254", "issue:43606|issue:43720", "issue:42490|issue:43329", - "issue:34567|issue:35707", "issue:43531|issue:46821", + "issue:34567|issue:35707", "issue:45406|issue:45464", "issue:34634|issue:41720", "issue:43531|issue:44589", @@ -5421,27 +5415,27 @@ "issue:44898|issue:45072", "issue:44625|issue:45479", "issue:41720|issue:45704", - "issue:39692|issue:44610", "issue:44918|issue:44936", + "issue:39692|issue:44610", "issue:43381|issue:43844", + "issue:42886|issue:44843", "issue:43381|issue:45704", - "issue:43746|issue:46050", - "issue:42886|issue:44843" + "issue:43746|issue:46050" ], - "prepared_review_unit_hash": "530d457f010e5077ab7a4589f3a4cfb69cd7c7ad3f82d4e914a7a988cacd9b1e", + "prepared_review_unit_hash": "cbabfffc2d24f75c24342af85ef6175f8f73d1fb8b6343b05e419535ce5d3821", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36310, + "serialized_chars": 36312, "estimated_input_tokens": 9078, "estimated_eval_tokens": 18412 }, @@ -5450,190 +5444,187 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated transformer issues. The only soft duplicate that looks like the same bug is the HF_HUB_OFFLINE tokenizer/cache failure pair; the rest are different models, subsystems, or root causes.", - "confidence": 0.86, - "canonical_issue_reason": "issue:44843 is the clearest root-cause report: it names the specific bad call to `model_info()` inside `_patch_mistral_regex` that breaks offline loading.", + "summary": "This set is a grab-bag of unrelated Transformers bugs across tokenizers, multimodal processors, training/runtime, and model-specific loading paths. The soft pairs mostly share subsystem vocabulary but not the same concrete failure mode, so none should be merged as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44843 is the cleanest library-level regression in the set: it has a precise failure path (`AutoTokenizer.from_pretrained` -> `_patch_mistral_regex` -> unconditional `model_info()`), a clear offline-mode impact, and broader applicability than the more model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44843 is the best representative of the only accepted duplicate group because it describes the concrete failure path and likely fix location.", + "best_issue_reason": "issue:44843 is the best representative issue because it is well-scoped, reproducible, and touches a core loading path rather than a single model edge case.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "Different models and unrelated bugs: SDPA/torch.compile vs tokenizer special-token registration." + "reason": "Different bugs: one is a torch.compile/SDPA failure for Bamba, the other is a tokenizer special-token registration bug for BigBird." }, { "left": "issue:43257", "right": "issue:47254", "accept": false, - "reason": "Both involve training infrastructure, but one is Qwen3 MoE weight conversion and the other is PEFT/ZeRO-3 checkpointing; not the same code path." + "reason": "Both involve DeepSpeed/accelerate, but one is MoE weight conversion on load and the other is a checkpointing/PEFT ZeRO-3 error; different code paths and fixes." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "Both touch accelerate/loading, but they are different model-specific failures with different underlying causes." + "reason": "CPU offload device mismatch in Bark is unrelated to BitNet packed-weight unpacking during accelerate loading." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Import-path slash warning is unrelated to multimodal token counting/video-branch undefined names." + "reason": "Path-normalization import warning is unrelated to a multimodal video token counting bug." }, { - "left": "issue:34567", - "right": "issue:35707", + "left": "issue:43531", + "right": "issue:46821", "accept": false, - "reason": "Trainer token-count tracking and progressive generation with embeds/past key values are separate problems." + "reason": "Qwen3-MoE sliding-window behavior and TimesFM `window_size` AttributeError are separate model-specific API failures." }, { - "left": "issue:43531", - "right": "issue:46821", + "left": "issue:34567", + "right": "issue:35707", "accept": false, - "reason": "Similar wording around window size, but different models and APIs; not the same bug." + "reason": "TrainerState token counting and progressive generation with embeds/past KV are different training/inference bugs." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Gemma4Processor tokenizer attribute failure and Qwen3.5 streaming API crash are distinct serving/processor issues." + "reason": "Gemma4Processor missing `_tokenizer` in serve is not the same as Qwen3.5 streaming chat/completions failure." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "BarkProcessor voice preset and Qwen3 auto device-mapping CUDA assert are unrelated." + "reason": "BarkProcessor voice preset handling and Qwen3 auto device mapping cuda assert are unrelated." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Qwen3-MoE sliding-window handling and Float8 storage lookup errors do not share a concrete failure path." + "reason": "Sliding-window logic for Qwen3-MoE is not the same as the Float8 storage type lookup error." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Bark CPU-offload device mismatch and BigBirdTokenizer special-token registration are unrelated." + "reason": "Device offload mismatch and BigBird mask-token decode failure affect different subsystems and require different fixes." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Different task areas: distributed metric gathering vs SAM3 text/bbox behavior." + "reason": "Metric gathering tuple truncation and SAM3 PCS text/bbox behavior are unrelated data-processing bugs." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Both are Qwen3.5-related, but one is config unwrapping and the other is CPU-offload compatibility; different bugs." + "reason": "Config unwrapping for composite Qwen 3.5/3.6 multimodal models is distinct from CPU-offload incompatibility in rotary embeddings." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-load failures, but the affected models and reported failure modes are different." + "reason": "Both are tokenizer loading problems, but they target different failure modes and different model repos; not the same bug." }, { "left": "issue:43540", "right": "issue:46858", "accept": false, - "reason": "Qwen3OmniMoe video-input error and cache seq-length tensor arithmetic are unrelated." + "reason": "Video-input processing in Qwen3OmniMoe is unrelated to static-cache slice arithmetic on non-CUDA backends." }, { "left": "issue:45357", "right": "issue:46402", "accept": false, - "reason": "Both are multimodal weight-loading complaints, but one is save_pretrained key mapping and the other is missing text encoder weights." + "reason": "Qwen3.5 save_pretrained key regression and Sam3Model text-encoder loading are separate serialization/loading 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": "The patchtsmixer post_init policy change is unrelated to Qwen3.5 num_labels propagation." }, { "left": "issue:43644", "right": "issue:46665", "accept": false, - "reason": "Non-persistent buffer corruption and from_pretrained behavior inside Python classes are too broad and not the same concrete bug." + "reason": "Non-persistent buffer junk initialization and inconsistent from_pretrained behavior inside Python classes are different bugs." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "A model-load regression and an image-processor torchvision dependency issue are different." + "reason": "A model-loading regression for `physical-intelligence/fast` is not the same as PIL image processors incorrectly requiring torchvision." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "AddedToken constructor misuse and ScandiBERT load failure are not the same underlying error." + "reason": "AddedToken keyword collision and loading `vesteinn/ScandiBERT` are not the same underlying issue." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver interpolation failure and dtype mismatch in inference are separate issues." + "reason": "Perceiver resize failure and dtype mismatches in bfloat16 inference are distinct model/runtime problems." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Both involve classification labels, but one is config propagation and the other is loss computation for `num_labels=1`." + "reason": "Missing `num_labels` propagation in Qwen3.5 config is not the same as the broader single-label zero-loss classification bug." }, { "left": "issue:41720", "right": "issue:45704", "accept": false, - "reason": "CUDA assert in Qwen3 auto device mapping is unrelated to T5/apex RMSNorm memory leak." + "reason": "Qwen3 auto device-map cuda assert and T5/apex FusedRMSNorm memory leak are unrelated runtime issues." }, { - "left": "issue:39692", - "right": "issue:44610", + "left": "issue:44918", + "right": "issue:44936", "accept": false, - "reason": "Documentation/model-processor mismatch and OmDet-Turbo resolution mismatch are different processor-level problems." + "reason": "TRL SFT unpacking failure and `trainer.evaluate()` after `train()` are different Trainer integration bugs." }, { - "left": "issue:44918", - "right": "issue:44936", + "left": "issue:39692", + "right": "issue:44610", "accept": false, - "reason": "TRl SFT embedding unpacking and trainer.evaluate() after train() are not the same failure." + "reason": "SigLIP2 doc example errors and OmDet-Turbo processor/model resolution mismatch are separate vision-model issues." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Gradient checkpointing in eval mode and ZeRO-3 gradient blow-up are unrelated training bugs." + "reason": "Gradient checkpointing in eval mode and abnormal gradient growth under ZeRO-3 are not the same failure." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "HF_HUB_OFFLINE cache loading and `_patch_mistral_regex` unconditionally calling `model_info()` are related only at a broad offline/loading level, but the concrete code paths differ." }, { "left": "issue:43381", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing restriction and T5 RMSNorm memory leak have no shared code path." + "reason": "Gradient checkpointing eval-mode failure and T5/apex memory leak are unrelated." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "Local PEFT adapter loading and sinq quantized model save/load are different persistence problems." - }, - { - "left": "issue:42886", - "right": "issue:44843", - "accept": true, - "reason": "Both describe HF_HUB_OFFLINE tokenizer loading failures; 44843 gives the concrete root cause in `_patch_mistral_regex` calling `model_info()` unconditionally." + "reason": "PEFT adapters failing from local checkpoints and sinq quantization breaking after save/load are different serialization paths." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The assessment is grounded in the packet and stays conservative. The only accepted soft pair, issue:42886 vs issue:44843, is well supported by the titles as the same HF_HUB_OFFLINE tokenizer-loading failure, with 44843 providing the more specific root-cause framing. The other pairwise rejections read appropriately cautious and do not overclaim duplicate status." - }, - "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 }, @@ -5648,13 +5639,14 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:30064", "issue:36010", "issue:39692", "issue:43295", "issue:43421", + "issue:43493", "issue:43526", "issue:43531", "issue:43540", @@ -5686,7 +5678,6 @@ "issue:45356", "issue:45357", "issue:45381", - "issue:45759", "issue:45834", "issue:45907", "issue:45923", @@ -5700,14 +5691,14 @@ "issue:47050" ], "soft_pairs": [ - "issue:44514|issue:44545", "issue:44368|issue:45357", + "issue:44514|issue:44545", "issue:43421|issue:43927", "issue:46655|issue:46693", + "issue:43531|issue:45923", "issue:43746|issue:43873", - "issue:45759|issue:45923", "issue:30064|issue:36010", - "issue:39692|issue:43873", + "issue:39692|issue:43493", "issue:43421|issue:43653", "issue:43994|issue:44671", "issue:44464|issue:47050", @@ -5718,247 +5709,212 @@ "issue:43782|issue:45381", "issue:43421|issue:44186", "issue:45356|issue:45907", + "issue:39692|issue:43873", "issue:39692|issue:47050", "issue:43526|issue:43881", "issue:43720|issue:43819", - "issue:43295|issue:45834", "issue:43596|issue:44220", - "issue:43540|issue:46294", + "issue:43295|issue:45834", "issue:43550|issue:43927", + "issue:43540|issue:46294", "issue:43550|issue:43819", "issue:44062|issue:44488", - "issue:44625|issue:47050", "issue:43653|issue:43720", + "issue:44625|issue:47050", "issue:43531|issue:43643", "issue:45381|issue:46693", "issue:44488|issue:47020" ], - "prepared_review_unit_hash": "80dc9b41298bc5b742865adcc3c23930b7130ec781f265b930b3e20cc08444dd", + "prepared_review_unit_hash": "a0c26a9b92b7dc444ac7a947c216d81b650263717850c0c1133979ece486cdf7", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 37724, - "estimated_input_tokens": 9431, - "estimated_eval_tokens": 19118 + "soft_pair_count": 33, + "serialized_chars": 37979, + "estimated_input_tokens": 9495, + "estimated_eval_tokens": 19246 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "One clear duplicate issue pair appears in the set: the two Qwen2_5_VLProcessor.apply_chat_template padding=False reports. The rest are distinct bugs across unrelated models/components and should stay separate.", - "confidence": 0.93, - "canonical_issue_reason": "issue:44514 is the earlier, more detailed report of the exact same batched-input padding=False crash as issue:44545.", + "summary": "This cluster is mostly a loose collection of unrelated Transformers issues. The only clear duplicate pair is the two Qwen2.5-VLProcessor batched apply_chat_template crashes with padding=False.", + "confidence": 0.96, + "canonical_issue_reason": "issue:44545 is the clearest representative of the duplicate Qwen2.5-VL batched apply_chat_template crash: it names the exact method, the batched-input condition, and the padding=False trigger.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44514 is the best canonical anchor for the duplicate pair because it is the earlier report and states the failure mode clearly.", + "best_issue_reason": "issue:44545 is the best global issue to keep because it is the most specific and reproducible statement of the only true duplicate pair in the set.", "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 and described failure match." - }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "Both involve Qwen3.5, but one is a tie_word_embeddings warning and the other is incorrect visual encoder key saving; different bugs." + "reason": "Different bugs: one is a Qwen3.5 LoRA warning about tie_word_embeddings, the other is incorrect visual encoder key serialization." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same underlying bug: Qwen2.5_VLProcessor.apply_chat_template crashes on batched input when padding=False." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Both are tokenizer/config persistence issues, but one is runtime special-token post-processing and the other is DiaConfig losing custom token IDs; not the same bug." + "reason": "Both relate to tokens/config, but one is runtime post-processor updates and the other is save/load loss of custom token IDs; not the same concrete failure." }, { "left": "issue:46655", "right": "issue:46693", "accept": false, - "reason": "Video padding in SmolVLM vs FA2 long-context performance regression are different code paths and symptoms." + "reason": "Video padding calculation error vs long-context generation performance regression are different code paths and symptoms." }, { - "left": "issue:43746", - "right": "issue:43873", + "left": "issue:43531", + "right": "issue:45923", "accept": false, - "reason": "PEFT/local checkpoint loading failure is unrelated to offloading/quantization behavior." + "reason": "Sliding-window handling in Qwen3-MoE is unrelated to a missing supports_gradient_checkpointing flag on a trust_remote_code model." }, { - "left": "issue:45759", - "right": "issue:45923", + "left": "issue:43746", + "right": "issue:43873", "accept": false, - "reason": "One is from_config text_config unwrapping for multimodal configs; the other is a missing supports_gradient_checkpointing flag." + "reason": "PEFT local checkpoint loading and quantization offloading are separate loading/dispatch problems." }, { "left": "issue:30064", "right": "issue:36010", "accept": false, - "reason": "ImageProcessor segmentation-map handling and a GenerationMixin ImportError are unrelated." + "reason": "Void segmentation map processing and a GenerationMixin import error are unrelated." }, { "left": "issue:39692", - "right": "issue:43873", + "right": "issue:43493", "accept": false, - "reason": "Documentation example errors and quantized offloading behavior are not the same underlying bug." + "reason": "Documentation example errors are not the same issue as a SigLIP2 implementation discrepancy against JAX." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Both touch special tokens, but one is runtime post-processor updates and the other is a BigBirdTokenizer registration/decoding issue." + "reason": "One is about updating post-processors after special-token mutation; the other is a BigBird tokenizer special-token registration bug." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different models and failures: SigLIP2 incorrect outputs vs CamemBERT masked-LM prediction regression." + "reason": "Different model families and failures: SigLIP2 nonsensical outputs vs CamemBERT masked LM regression." }, { "left": "issue:44464", "right": "issue:47050", "accept": false, - "reason": "Chunked generation with compiled forward is unrelated to fp16 ONNX attention dtype mismatch." + "reason": "Compiled chunked generation inconsistency and ONNX fp16 dtype mismatch are different problems." }, { "left": "issue:44995", "right": "issue:46858", "accept": false, - "reason": "Both cache-related, but one is a stale indexer bug and the other is a device-tensor seq_length arithmetic bug; different fixes." + "reason": "Stale indexer cache on second forward is not the same as device-tensor seq_length breaking static-cache arithmetic under compile." }, { "left": "issue:46032", "right": "issue:46463", "accept": false, - "reason": "use_cache correctness in Mamba2Mixer is unrelated to T5 decoder input validation error messaging." + "reason": "Incorrect use_cache behavior and an incorrect T5 decoder input validation error message are unrelated." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both mention fix_mistral_regex, but one is an AttributeError in regex patching and the other is a Kimi-K2.5 codec/warning regression; not clearly the same defect." + "reason": "Both mention tokenizer regex/fix_mistral_regex, but one crashes on backend_tokenizer access while the other is a codec-handling regression with a misleading warning." }, { "left": "issue:44857", "right": "issue:47050", "accept": false, - "reason": "AMP/CUDA loss crash and fp16 ONNX attention dtype mismatch are different export/runtime problems." + "reason": "Different model/export paths: LwDetrImageLoss AMP crash vs CanineModel ONNX attention dtype mismatch." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Qwen3VL weight_only loading and Qwen2.5-VL video position IDs are different multimodal bugs." - }, - { - "left": "issue:43421", - "right": "issue:44186", - "accept": false, - "reason": "Special-token post-processor update logic is unrelated to LayoutLMv2 NER/padding crashes." - }, - { - "left": "issue:45356", - "right": "issue:45907", - "accept": false, - "reason": "Tokenizer regression around fix_mistral_regex is unrelated to list_repo_files revision forwarding." - }, - { - "left": "issue:39692", - "right": "issue:47050", - "accept": false, - "reason": "SigLIP2 docs/model mismatch and Canine fp16 ONNX export dtype mismatch are unrelated." - }, - { - "left": "issue:43526", - "right": "issue:43881", - "accept": false, - "reason": "Beit reduce_labels bug and glm-4v-9b loading failure are not the same issue." - }, - { - "left": "issue:43720", - "right": "issue:43819", - "accept": false, - "reason": "Packed-weight accelerate loading and DAC.from_latents STE mismatch are separate model-loading/training bugs." - }, - { - "left": "issue:43295", - "right": "issue:45834", - "accept": false, - "reason": "Processor.tokenizer regression and Kosmos2.5 long OCR index error are different multimodal bugs." + "reason": "Qwen3VL weight_only loading failure is unrelated to Qwen2.5-VL batched chat-template padding behavior." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "deepspeed zero3/BertModel indexing and audio fbank feature extraction are unrelated." + "reason": "Deepspeed ZeRO-3/BertModel indexing and _torch_extract_fbank_features are unrelated." }, { - "left": "issue:43540", - "right": "issue:46294", + "left": "issue:43295", + "right": "issue:45834", "accept": false, - "reason": "Both are Qwen video-related, but one is a ValueError in Qwen3OmniMoe video processing and the other is apply_chat_template token/feature mismatch in Qwen3.5-9B." + "reason": "Processor.tokenizer regression with images is not the same as a Kosmos2.5 long-OCR index error." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "torch.compile SDPA failure for Bamba is unrelated to DiaConfig custom token ID persistence." + "reason": "torch.compile + SDPA failure in Bamba is unrelated to DiaConfig losing custom token IDs after save/load." }, { - "left": "issue:43550", - "right": "issue:43819", + "left": "issue:43540", + "right": "issue:46294", "accept": false, - "reason": "Bamba torch.compile/SDPA and DAC STE mismatch are different code paths." + "reason": "These are different Qwen video-processing failures: one is a Qwen3OmniMoe ValueError, the other is a Qwen3.5-VL video feature/token mismatch." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "AddedToken special-argument duplication and sleng-bert loading failure are different tokenizer issues." + "reason": "AddedToken special-argument duplication and inability to load cjvt/sleng-bert are different tokenizer bugs." }, { - "left": "issue:44625", - "right": "issue:47050", + "left": "issue:43653", + "right": "issue:43720", "accept": false, - "reason": "Qwen3.5 num_labels propagation and Canine fp16 ONNX export dtype mismatch are unrelated." + "reason": "BigBird special-token registration and BitNet packed-weight loading are unrelated." }, { - "left": "issue:43653", - "right": "issue:43720", + "left": "issue:44625", + "right": "issue:47050", "accept": false, - "reason": "Tokenizer special-token registration and BitNet packed-weight loading are different bugs." + "reason": "Qwen3.5 num_labels propagation and CanineModel ONNX dtype mismatch are separate issues." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Sliding_window behavior in Qwen3-MoE is unrelated to trust_remote_code config field loss." + "reason": "Sliding-window behavior and missing fields in AutoConfig with trust_remote_code are unrelated." }, { "left": "issue:45381", "right": "issue:46693", "accept": false, - "reason": "Qwen2.5-VL video position IDs and FA2 long-context performance regression are different problems." + "reason": "Wrong Qwen2.5-VL vision_position_ids for video input is not the same as an FA2 long-context performance regression." }, { "left": "issue:44488", "right": "issue:47020", "accept": false, - "reason": "Tokenizer loading failure for sleng-bert and XLMRoberta Unigram init type mismatch are different issues." + "reason": "A model loading failure and an XLMRobertaTokenizer/Unigram constructor bug are different problems." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the single accepted duplicate pair (issue:44514 / issue:44545) is well-supported by the nearly identical titles and failure description. The rest of the soft-edge rejections stay conservative and do not overclaim sameness." + "feedback": "The summary is well grounded in the packet and stays conservative. The one accepted duplicate pair, issue:44514 and issue:44545, is strongly supported by the titles and looks mergeable as a single bugfix. The other soft-edge verdicts are appropriately cautious and do not overstate evidence." }, "evaluator_used": true, "retried": false, @@ -5977,7 +5933,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:30064", "issue:34634", @@ -5987,8 +5943,8 @@ "issue:41720", "issue:43322", "issue:43335", - "issue:43421", "issue:43540", + "issue:43606", "issue:43611", "issue:43618", "issue:43643", @@ -6005,7 +5961,6 @@ "issue:44661", "issue:44671", "issue:44805", - "issue:44811", "issue:44821", "issue:44843", "issue:44877", @@ -6013,6 +5968,7 @@ "issue:44936", "issue:44991", "issue:45020", + "issue:45072", "issue:45081", "issue:45084", "issue:45584", @@ -6029,223 +5985,223 @@ ], "soft_pairs": [ "issue:46829|issue:46832", - "issue:43611|issue:46665", + "issue:44220|issue:44661", "issue:41720|issue:44991", "issue:30064|issue:36331", - "issue:44220|issue:44661", + "issue:43611|issue:46665", "issue:44821|issue:45084", - "issue:44843|issue:45081", "issue:45987|issue:46829", "issue:44186|issue:44898", "issue:43653|issue:43819", - "issue:43540|issue:44008", + "issue:44843|issue:45081", "issue:43335|issue:43756", - "issue:34634|issue:36010", + "issue:43540|issue:44008", "issue:44008|issue:46858", + "issue:34634|issue:36010", "issue:43643|issue:44877", "issue:44991|issue:47020", "issue:45987|issue:46932", "issue:39692|issue:45834", "issue:43618|issue:43697", "issue:44561|issue:45020", - "issue:44448|issue:44671", "issue:44186|issue:44442", + "issue:44448|issue:44671", "issue:45584|issue:45834", - "issue:44991|issue:46291", "issue:44805|issue:44936", + "issue:44991|issue:46291", "issue:43322|issue:45685", - "issue:43421|issue:44811" + "issue:43606|issue:45072" ], - "prepared_review_unit_hash": "58db549dca433a5dccb3c208b59475d6c034d439869ee26e6db055fb517885e4", + "prepared_review_unit_hash": "30c5fe24bec1c1c6c644f3a10257c7514858b39cd30233693ba96f529aa5e20d", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35418, - "estimated_input_tokens": 8855, - "estimated_eval_tokens": 17966 + "serialized_chars": 35401, + "estimated_input_tokens": 8851, + "estimated_eval_tokens": 17958 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated Transformers bug reports across tokenizers, processors, model loading, and training/runtime regressions. I don\u2019t see any truly duplicate pairs; the only near-neighbors share a subsystem or helper name but not the same concrete failure mode.", - "confidence": 0.84, - "canonical_issue_reason": "Issue #45020 is the broadest, most representative regression report in the set: it captures the general class of recent-version / remote_code breakages that several other issues resemble, even though it is not a duplicate of them.", - "canonical_pr_reason": null, - "best_issue_reason": "#45020 is the best single issue to anchor this cluster because it is broad, user-impacting, and closest to the recurring v5 compatibility regressions appearing elsewhere in the set.", - "best_pr_reason": null, + "summary": "This cluster is highly heterogeneous: most pairs are clearly unrelated and should not be merged. The only plausible umbrella-like issue is the broad remote_code regression report, but none of the soft pairs look like true duplicates.", + "confidence": 0.41, + "canonical_issue_reason": "issue:45020 is the broadest umbrella in the set: it describes a general regression in recent transformers versions breaking `remote_code` models, which is the closest thing to a cluster-level theme. The other issues are narrower, model-specific, or entirely unrelated.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "issue:45020 best represents the cluster because it is the most general and centrally worded regression report among otherwise unrelated issues.", + "best_pr_reason": "No pull requests are present in this cluster.", "soft_edge_verdicts": [ { "left": "issue:46829", "right": "issue:46832", "accept": false, - "reason": "Different models and different bugs: output-field mismatch in VideoPrism vs feature-level mismatch in RT-DETR." + "reason": "Different failures in different model families: one is an output-field bug, the other is a forward-pass shape/type error." }, { - "left": "issue:43611", - "right": "issue:46665", + "left": "issue:44220", + "right": "issue:44661", "accept": false, - "reason": "Both concern loading behavior, but one is a base_model_prefix regression and the other is inconsistent from_pretrained behavior inside Python classes; not the same bug." + "reason": "Unrelated subsystems: audio feature extraction vs tokenizer mapping/name-generation tooling." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Completely different failures: Qwen3 auto device mapping CUDA assert vs tokenizer loading for est-roberta." + "reason": "Different models and symptoms: CUDA assert during device mapping vs tokenizer loading failure." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Image processor segmentation-map handling and CustomTrainer compute_loss signature are unrelated." + "reason": "Completely different code paths: image segmentation map processing vs Trainer loss signature incompatibility." }, { - "left": "issue:44220", - "right": "issue:44661", + "left": "issue:43611", + "right": "issue:46665", "accept": false, - "reason": "Audio feature extraction failure vs tokenizer mapping-name failure; different code paths and root causes." + "reason": "Both involve loading behavior, but the concrete bugs differ: `base_model_prefix` handling vs inconsistent `from_pretrained` use inside Python classes." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "Unrelated issues: loading AutoImageProcessor from URL vs template compilation error." - }, - { - "left": "issue:44843", - "right": "issue:45081", - "accept": false, - "reason": "Both hit _patch_mistral_regex, but one is an offline model_info() call and the other is a backend_tokenizer attribute crash; distinct bugs." + "reason": "No shared underlying bug: image processor loading from URL vs template compilation failure." }, { "left": "issue:45987", "right": "issue:46829", "accept": false, - "reason": "StaticCache sequence-length tensor type bug vs VideoPrism hidden_states output bug." + "reason": "Static cache length contract bug is unrelated to VideoPrism output packaging." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Tokenizer batching/NER crash and Perceiver interpolation failure are different model-specific issues." + "reason": "Different tokenizer/model families and different failure modes." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Tokenizer special-token registration bug vs DAC from_latents/forward mismatch are unrelated." + "reason": "Tokenizer special-token registration bug vs DAC latent reconstruction mismatch; not the same code-path." }, { - "left": "issue:43540", - "right": "issue:44008", + "left": "issue:44843", + "right": "issue:45081", "accept": false, - "reason": "Video input processing failure in Qwen3OmniMoe vs Gemma3n forward variable-name collision; different code paths." + "reason": "Both touch `_patch_mistral_regex`, but one is an offline-network call issue and the other is an attribute error on tokenizer internals." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "SwitchTransformers sparse-layer creation bug and Smollm3 RoPE layer-drop bug are unrelated model-configuration problems." + "reason": "Different model config bugs: sparse-layer creation vs RoPE-layer dropping." }, { - "left": "issue:34634", - "right": "issue:36010", + "left": "issue:43540", + "right": "issue:44008", "accept": false, - "reason": "BarkProcessor voice_preset bug and GenerationMixin import error are not the same issue." + "reason": "Different multimodal models and unrelated forward-pass bugs." }, { "left": "issue:44008", "right": "issue:46858", "accept": false, - "reason": "Gemma3n attribute collision and cache.get_seq_length device-tensor behavior are different regressions." + "reason": "Variable-name collision in Gemma3n is unrelated to static cache sequence-length behavior." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice preset handling is unrelated to the GenerationMixin import error." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "One is missing fields from trust_remote_code AutoConfig loading; the other is strict config blocking granite_speech config loading." + "reason": "Both are config-loading issues, but one is `trust_remote_code` field loss and the other is strict config validation for a specific model." }, { "left": "issue:44991", "right": "issue:47020", "accept": false, - "reason": "Both are tokenizer-load failures, but for different models and different root causes." + "reason": "Both are tokenizer-load failures, but they affect different tokenizers and stem from different initialization bugs." }, { "left": "issue:45987", "right": "issue:46932", "accept": false, - "reason": "StaticCache return-type bug and Gemma4 quantized embedder dtype-casting bug are unrelated." + "reason": "Static cache return-type bug is unrelated to Gemma4 quantized embedder dtype casting." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "Documentation example mistakes vs long-OCR index error are not the same underlying bug." + "reason": "Documentation/example mismatch and quantization failure are unrelated to long OCR-input indexing." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "CLIPOutput attentions regression and RTDetrV2 output differences are different model-output issues." + "reason": "Different regressions in different model outputs; no shared concrete code-path." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Both involve remote_code regressions, but one is specifically about is_torch_fx_available removal while the other is a broad umbrella report; too generic to treat as the same bug." + "reason": "Both mention `remote_code` breakage, but one is specifically about removing `is_torch_fx_available` while the other is a broader regression report; not the same concrete bug." }, { - "left": "issue:44448", - "right": "issue:44671", + "left": "issue:44186", + "right": "issue:44442", "accept": false, - "reason": "Different models and different output regressions: Pegasus generation vs CamemBERT masked LM predictions." + "reason": "Different tokenizer bugs affecting different models and input shapes." }, { - "left": "issue:44186", - "right": "issue:44442", + "left": "issue:44448", + "right": "issue:44671", "accept": false, - "reason": "LayoutLMv2 tokenizer crash and FastSpeech2Conformer tokenizer load failure are unrelated tokenizer bugs." + "reason": "Both are version-v4/v5 output discrepancies, but the model-specific failure modes are unrelated." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation issue and Kosmos2.5 long OCR index error are different failure modes." + "reason": "Whisper empty-transcription failure is unrelated to Kosmos2.5 long-input indexing." }, { - "left": "issue:44991", - "right": "issue:46291", + "left": "issue:44805", + "right": "issue:44936", "accept": false, - "reason": "Tokenizer loading regression and kernels v0.15 import-time dependency error are unrelated." + "reason": "Mask-shape IndexError and post-train evaluate failure are different trainer/runtime problems." }, { - "left": "issue:44805", - "right": "issue:44936", + "left": "issue:44991", + "right": "issue:46291", "accept": false, - "reason": "Mask-shape IndexError and trainer.evaluate after train are different failures." + "reason": "Tokenizer loading failure is unrelated to pipeline import breakage from kernels version/revision requirements." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Llava Next segfault and MPS histogram_int error are unrelated runtime issues." + "reason": "Llava Next loading crash and MPS histogram dtype limitation are unrelated backends/problems." }, { - "left": "issue:43421", - "right": "issue:44811", + "left": "issue:43606", + "right": "issue:45072", "accept": false, - "reason": "Special-token runtime update bug and Whisper batch_decode skip_special_tokens bug are adjacent tokenizer concerns but not the same concrete bug." + "reason": "CPU offload device mismatch and bfloat16 dtype mismatch are different bugs in different models/code paths." } ] }, @@ -6267,7 +6223,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28218", "issue:29159", @@ -6275,11 +6231,11 @@ "issue:36331", "issue:42175", "issue:43388", + "issue:43421", "issue:43540", "issue:43550", "issue:43577", "issue:43582", - "issue:43606", "issue:43644", "issue:43645", "issue:43716", @@ -6297,12 +6253,12 @@ "issue:44479", "issue:44610", "issue:44671", + "issue:44811", "issue:44977", "issue:45020", "issue:45072", "issue:45081", "issue:45183", - "issue:45325", "issue:45381", "issue:45479", "issue:45563", @@ -6319,20 +6275,20 @@ "issue:47276" ], "soft_pairs": [ - "issue:43606|issue:45072", - "issue:45183|issue:45479", + "issue:43421|issue:44811", "issue:44479|issue:45834", - "issue:43550|issue:45072", - "issue:46294|issue:46463", "issue:43577|issue:43818", + "issue:43550|issue:45072", + "issue:45183|issue:45479", "issue:43644|issue:43749", + "issue:46294|issue:46463", "issue:43782|issue:44977", "issue:43716|issue:44008", "issue:43582|issue:46512", "issue:45563|issue:47276", "issue:43825|issue:43994", - "issue:28218|issue:29159", "issue:43931|issue:46846", + "issue:28218|issue:29159", "issue:43388|issue:46032", "issue:43540|issue:45381", "issue:44610|issue:45072", @@ -6345,212 +6301,202 @@ "issue:43577|issue:43582", "issue:46697|issue:46728", "issue:43825|issue:43827", - "issue:43742|issue:45020", - "issue:43782|issue:45325" + "issue:43742|issue:45020" ], - "prepared_review_unit_hash": "42edb96d59e329bed0c1f80ba71e192fda246f798f95cc7a93e712a3d35fffc7", + "prepared_review_unit_hash": "61a07c097fd6e3ab966d124b74221d8fa8b68513b53b56580b95b6b509ec462f", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36548, - "estimated_input_tokens": 9137, - "estimated_eval_tokens": 18530 + "soft_pair_count": 27, + "serialized_chars": 36239, + "estimated_input_tokens": 9060, + "estimated_eval_tokens": 18376 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a heterogeneous cluster of unrelated bugs and regressions (tokenization, multimodal/video preprocessing, loading/dtype issues, trainer behavior, docs/warnings). None of the soft pairs look like true duplicates of the same concrete bug/change, so there is no safe canonical artifact for the cluster.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "Most pairs are clearly different bugs or different layers of the stack. The only credible duplicate-style match is the tokenizer pair 28218/29159, where the broader slow-vs-fast inconsistency report plausibly covers the added-token spacing symptom.", + "confidence": 0.67, + "canonical_issue_reason": "29159 is the broader tokenizer report and can reasonably subsume the more specific added-token spacing bug in 28218.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "29159 is the most representative issue here: it is broad, tokenizer-focused, and the only pair that looks like a plausible duplicate relationship.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43606", - "right": "issue:45072", + "left": "issue:43421", + "right": "issue:44811", "accept": false, - "reason": "Both are dtype/device-related, but they affect different models and different code paths; not the same bug." + "reason": "Different tokenizer/processor behaviors: post-processor updating special tokens vs batch_decode skip_special_tokens handling." }, { - "left": "issue:45183", - "right": "issue:45479", + "left": "issue:44479", + "right": "issue:45834", "accept": false, - "reason": "One is about input-validation error messaging, the other about degenerate zero loss in classification." + "reason": "Different models and different failures: video regression in Qwen VL vs OCR index error in Kosmos2.5." }, { - "left": "issue:44479", - "right": "issue:45834", + "left": "issue:43577", + "right": "issue:43818", "accept": false, - "reason": "Both are vision/video adjacent, but the failures are different (video regression vs OCR index error) and not the same bug." + "reason": "Unrelated model-loading dtype issue vs Video-LLaVA architecture/weight-sharing bug." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "torch.compile+SDPA failure is unrelated to the dtype mismatch issue." + "reason": "Different code paths and symptoms: torch.compile + SDPA failure vs bfloat16 dtype mismatch in wrappers." }, { - "left": "issue:46294", - "right": "issue:46463", + "left": "issue:45183", + "right": "issue:45479", "accept": false, - "reason": "Different subsystems: Qwen video chat-template mismatch vs T5 decoder-input validation." + "reason": "Input validation error messaging is not the same as the single-label-classification zero-loss bug." }, { - "left": "issue:43577", - "right": "issue:43818", + "left": "issue:43644", + "right": "issue:43749", "accept": false, - "reason": "Blip2 dtype staleness and Video-LLaVA temporal-attention/weight-sharing are unrelated loading/model-architecture bugs." + "reason": "Different loading regressions: non-persistent buffer initialization vs FSDP CPU RAM efficient loading." }, { - "left": "issue:43644", - "right": "issue:43749", + "left": "issue:46294", + "right": "issue:46463", "accept": false, - "reason": "Non-persistent buffer junk and FSDP CPU RAM efficient loading are different loading defects." + "reason": "Different model families and different problems: Qwen video chat-template mismatch vs T5 decoder input validation message." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Different Qwen3.5 failure modes: weight_only load error vs flash-attention generation issue." + "reason": "Different failures: weight_only loading error vs flash-attention generation regression." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Different root causes: dtype mismatch in Mistral-3 preprocessing vs Gemma3n variable-name collision." + "reason": "Different bugs in different models: image preprocessor dtype mismatch vs variable-name collision in Gemma3n forward." }, { "left": "issue:43582", "right": "issue:46512", "accept": false, - "reason": "Apple Silicon warmup TypeError is unrelated to stale config.dtype after from_config." + "reason": "Apple Silicon allocator TypeError is unrelated to stale config.dtype after from_config." }, { "left": "issue:45563", "right": "issue:47276", "accept": false, - "reason": "Both are warning-message issues, but they concern different warnings and different generation paths." + "reason": "Both are warning-message issues, but they concern different warnings in different generation paths." }, { "left": "issue:43825", - "right": "issue:43994", - "accept": false, - "reason": "One is a misleading pipeline error message; the other is a model output quality problem." - }, - { - "left": "issue:28218", - "right": "issue:29159", + "right": "issue:43827", "accept": false, - "reason": "Both mention tokenizer inconsistency, but one is a specific added-token spacing bug and the other is broader slow/fast tokenizer divergence." + "reason": "One is a runtime error-message bug, the other is docs lagging behind v5 removals; not the same code-path problem." }, { "left": "issue:43931", "right": "issue:46846", "accept": false, - "reason": "Different shape-mismatch bugs: direct Qwen3-VL load failure vs TP-plan missing entries at TP>1." + "reason": "Different Qwen3-VL loading/shape problems; TP-plan omissions are not the same as a model weight-shape mismatch report." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": true, + "reason": "The broader slow-vs-fast tokenizer inconsistency report plausibly includes the added-token extra-space symptom." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Different failures in different components: gather_for_metrics label handling vs Mamba2 use_cache correctness." + "reason": "Different logic errors: last-batch label dropping in gather_for_metrics vs incorrect use_cache results in Mamba2Mixer." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both are video-input bugs, but they affect different models and different position/feature handling logic." + "reason": "Both are video-related, but they affect different models and different bugs (processing error vs vision_position_ids bug)." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Image-size mismatch in OmDet-Turbo is unrelated to the dtype mismatch issue." + "reason": "Image-size preprocessing mismatch is unrelated to bfloat16 dtype mismatch." }, { "left": "issue:43645", "right": "issue:46665", "accept": false, - "reason": "Both involve model initialization, but one is a notebook/custom-model regression and the other is from_pretrained behavior inside Python classes." + "reason": "Custom model initialization in notebooks is a different issue from from_pretrained behavior inside Python classes." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP RAM-efficient loading and init_empty_weights/_is_hf_initialized are different loading mechanisms and failures." + "reason": "Different loading mechanisms and different failure modes: FSDP RAM-efficient loading vs init_empty_weights unexpected argument." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss signature mismatch are unrelated." + "reason": "BarkProcessor voice_preset is unrelated to CustomTrainer.compute_loss signature breakage." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "Packed-weight accelerate loading failure is not the same as a bfloat16 dtype mismatch." + "reason": "Packed-weight unpacking during accelerate loading is a different bug from dtype mismatch in inference wrappers." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "TensorFlow backend packaging and Mistral tokenizer regex patch crash are unrelated." + "reason": "Backend installation dependency issue is unrelated to the Mistral regex patch AttributeError." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "CamemBERT MLM prediction regression is not the same as general tokenization changes." + "reason": "The CamemBERT MLM regression and the generic tokenization-change report are too broad and not clearly the same underlying bug." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "dtype metadata staleness and Apple Silicon warmup TypeError are different issues." + "reason": "Different dtype/loading problems in different subsystems; no shared concrete code path." }, { "left": "issue:46697", "right": "issue:46728", "accept": false, - "reason": "Both are image sizing/padding bugs, but they arise from different calculations and models." - }, - { - "left": "issue:43825", - "right": "issue:43827", - "accept": false, - "reason": "Runtime error-message wording and docs still referencing pipeline() are related thematically but not the same bug." + "reason": "Both are image-processing edge cases, but one is over-padding and the other is incorrect patch counting." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "Specific key error on one model is not the same as broad remote_code breakage." - }, - { - "left": "issue:43782", - "right": "issue:45325", - "accept": false, - "reason": "Qwen3VL weight_only load error and Qwen2.5-VL rope-index scaling bug are distinct." + "reason": "Key error loading one model is not the same as the broad remote_code breakage regression." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted duplicate-style pairing, 28218/29159, is reasonably supported by the issue titles and the broader/narrower relationship. The other soft-edge rejections also read as cautious and consistent with the listed topics." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -6565,7 +6511,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36010", "issue:36331", @@ -6580,11 +6526,11 @@ "issue:43550", "issue:43582", "issue:43638", - "issue:43643", "issue:43645", "issue:43653", "issue:43697", "issue:43746", + "issue:43782", "issue:43819", "issue:43854", "issue:43906", @@ -6602,10 +6548,10 @@ "issue:45042", "issue:45081", "issue:45084", + "issue:45325", "issue:45372", "issue:45406", "issue:45538", - "issue:45759", "issue:45859", "issue:46175", "issue:46209", @@ -6617,224 +6563,224 @@ "issue:47217" ], "soft_pairs": [ + "issue:43782|issue:45325", "issue:44589|issue:45042", - "issue:43262|issue:47217", "issue:43262|issue:43819", - "issue:43582|issue:44857", - "issue:43653|issue:44912", "issue:46175|issue:46932", + "issue:43653|issue:44912", + "issue:43582|issue:44857", + "issue:43262|issue:47217", "issue:45003|issue:46285", - "issue:36010|issue:43645", "issue:45372|issue:46326", + "issue:36010|issue:43645", "issue:36010|issue:36331", "issue:43550|issue:44912", "issue:43531|issue:45042", "issue:43638|issue:44936", "issue:44060|issue:44368", - "issue:43257|issue:43329", "issue:42175|issue:46291", - "issue:43452|issue:44991", - "issue:43906|issue:46209", + "issue:43257|issue:43329", "issue:43526|issue:43697", - "issue:43819|issue:43927", + "issue:43452|issue:44991", "issue:45081|issue:45406", + "issue:43819|issue:43927", + "issue:43906|issue:46209", + "issue:45084|issue:45538", "issue:43854|issue:44977", "issue:43746|issue:44811", - "issue:45084|issue:45538", - "issue:45859|issue:47050", "issue:42490|issue:43257", - "issue:43643|issue:45759" + "issue:45859|issue:47050" ], - "prepared_review_unit_hash": "0581263dcc661b4e6f1ea8c525dde83d6c8b8df57e20023f0a0ab213a4cc4620", + "prepared_review_unit_hash": "562d2cad220aa480eba5a3651a8d3f8aadcbef13469180dfb1f51864b892d4da", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36335, - "estimated_input_tokens": 9084, - "estimated_eval_tokens": 18424 + "serialized_chars": 36307, + "estimated_input_tokens": 9077, + "estimated_eval_tokens": 18410 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous; these issues mostly cover unrelated tokenizer, processor, model-loading, training, and quantization regressions. I found no soft pair that looks like the same underlying bug/change, so all candidate edges are rejected. No PRs are present.", - "confidence": 0.84, - "canonical_issue_reason": "Issue 45081 is the best anchor only in the sense that it is a concrete, narrowly scoped loading crash with relatively strong discussion/inbound signal; the rest of the set is too diverse to form a true duplicate hub.", + "summary": "This cluster is largely heterogeneous: the soft-similarity pairs share superficial keywords or the same broad subsystem, but they do not describe the same concrete bug/change. I would not merge any of the proposed soft edges.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue stands out; the items are mostly unrelated bug reports with only a few superficial thematic overlaps.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45081 is the most representative issue to use as a stand-in canonical because it has a specific failure mode and enough surrounding activity to be triage-worthy, even though it does not unify the cluster semantically.", + "best_issue_reason": "Issue 44060 is the closest to a representative item only because it has the strongest topical overlap with another issue in the list, but overall the cluster is too mixed to have a true canonical issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44589", - "right": "issue:45042", + "left": "issue:43782", + "right": "issue:45325", "accept": false, - "reason": "Different subsystems and failures: float8 storage lookup vs PIL image processor torchvision dependency." + "reason": "Both are Qwen/VL-adjacent, but one is a weight_only loading error and the other is a rope-indexing bug; different code paths." }, { - "left": "issue:43262", - "right": "issue:47217", + "left": "issue:44589", + "right": "issue:45042", "accept": false, - "reason": "Audio chat-template sampling-rate default bug is unrelated to VL image-placeholder placement failures." + "reason": "FP8 storage lookup failure vs PIL image processor requiring torchvision; unrelated failures." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Sampling-rate defaulting in audio processors is unrelated to DAC forward/from_latents mismatch." + "reason": "Audio chat-template sampling-rate default vs DAC latent/STE mismatch; different components and symptoms." }, { - "left": "issue:43582", - "right": "issue:44857", + "left": "issue:46175", + "right": "issue:46932", "accept": false, - "reason": "Apple Silicon allocator TypeError and CUDA float16 AMP loss crash are different code paths and platforms." + "reason": "Both are Gemma4-related, but one is a huge one-hot position embedding allocation and the other is dtype casting for quantized embedders." }, { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "Tokenizer special-token registration bug is unrelated to MXFP4 quantization fallback during model loading." + "reason": "Tokenizer special-token registration bug vs MXFP4 quantization fallback; no shared bug path." }, { - "left": "issue:46175", - "right": "issue:46932", + "left": "issue:43582", + "right": "issue:44857", "accept": false, - "reason": "Both are Gemma4-related, but one is a memory-heavy patch embedder implementation issue and the other is dtype casting for quantized embedders." + "reason": "Apple Silicon warmup TypeError and CUDA float16 loss crash are unrelated runtime issues." }, { - "left": "issue:45003", - "right": "issue:46285", + "left": "issue:43262", + "right": "issue:47217", "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils is not the same as the CLIP text-model structural inconsistency." + "reason": "Audio processor sampling-rate default is not the same as VL image-placeholder handling in user text." }, { - "left": "issue:36010", - "right": "issue:43645", + "left": "issue:45003", + "right": "issue:46285", "accept": false, - "reason": "GenerationMixin import failure is a different regression from Jupyter custom-model initialization in Transformers 5.0." + "reason": "Unsafe sys.modules access is unrelated to CLIP text-model structural changes." }, { "left": "issue:45372", "right": "issue:46326", "accept": false, - "reason": "Mistral_common import breakage for Gemma4 processor loading is distinct from continue_final_message behavior in a tokenizer path." + "reason": "Both mention mistral_common, but one is an import error for ReasoningEffort and the other is continue_final_message behavior; different problems." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "GenerationMixin import failure and Jupyter custom-model initialization breakage are distinct issues." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "Importing GenerationMixin and CustomTrainer.compute_loss signature mismatch are unrelated bugs." + "reason": "Importing GenerationMixin and CustomTrainer.compute_loss signature mismatch are unrelated." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "torch.compile+SDPA failure in Bamba is unrelated to MXFP4 quantization fallback." + "reason": "torch.compile/SDPA failure in Bamba is not the same as MXFP4 quantization fallback in git-oss-20b." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and PIL-backend torchvision requirement are different issues." + "reason": "Qwen3-MoE sliding_window bug and PIL/torchvision dependency regression do not share a concrete code path." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Deepspeed ZeRO-3 index error with a non-pretrained Bert model is not the same as train()-then-evaluate() failure." + "reason": "DeepSpeed ZeRO-3 with non-pretrained Bert IndexError is not the same as evaluate() failing after train()." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Both mention tied-weights warnings around Qwen variants, but the concrete symptoms and contexts differ enough that they do not look like the same bug." - }, - { - "left": "issue:43257", - "right": "issue:43329", - "accept": false, - "reason": "Qwen3 MOE weight conversion under accelerate/deepspeed is unrelated to undefined video-token counting variables." + "reason": "Both mention tied-weights warnings in Qwen models, but one is a specific wrong tie warning in Qwen3-Next and the other is a separate fine-tuning warning/report; not enough evidence of one bug." }, { "left": "issue:42175", "right": "issue:46291", "accept": false, - "reason": "Package/backend installation issue and kernels dependency/version validation failure are different failure modes." + "reason": "Package/backend dependency issues, but one is missing TensorFlow in a torch extra and the other is kernels version/revision requirements." }, { - "left": "issue:43452", - "right": "issue:44991", - "accept": false, - "reason": "gguf_file handling breaks from_pretrained loaders, but the est-roberta issue is a separate tokenizer-loading regression." - }, - { - "left": "issue:43906", - "right": "issue:46209", + "left": "issue:43257", + "right": "issue:43329", "accept": false, - "reason": "Reproduction of an older issue and FP8 inference failure on B200 are unrelated." + "reason": "Qwen3-MoE weight conversion/loading and a multimodal token-count video-branch undefined-variable bug are different failures." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug and RTDetrV2 output drift are different model/processor regressions." + "reason": "BeitImageProcessorFast label reduction and RTDetrV2 output differences are unrelated model issues." }, { - "left": "issue:43819", - "right": "issue:43927", + "left": "issue:43452", + "right": "issue:44991", "accept": false, - "reason": "DAC STE mismatch and DiaConfig save/load custom-token loss are unrelated." + "reason": "gguf_file handling for tokenizer/model loading is unrelated to the EMBEDDIA tokenizer load regression." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Mistral regex patch crash during tokenizer loading and Gemma4Processor missing _tokenizer in serve are different processor/tokenizer failures." + "reason": "Mistral tokenizer regex patch crash and Gemma4Processor missing _tokenizer are separate attribute/path bugs." }, { - "left": "issue:43854", - "right": "issue:44977", + "left": "issue:43819", + "right": "issue:43927", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading failure and Qwen3.5 flash-attention generation issue are not the same code-path problem." + "reason": "DAC latent conversion mismatch and DiaConfig losing custom token IDs are different model/config bugs." }, { - "left": "issue:43746", - "right": "issue:44811", + "left": "issue:43906", + "right": "issue:46209", "accept": false, - "reason": "PEFT checkpoint loading for GraniteSpeech is unrelated to Whisper batch_decode skipping special tokens." + "reason": "An isolated reproduction of another issue is not the same as an FP8 inference failure on B200." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template compilation error and CLIPTokenizer model_max_length bug are unrelated." + "reason": "Template-node compilation error and CLIPTokenizer max_length regression are unrelated." }, { - "left": "issue:45859", - "right": "issue:47050", + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "GLM-4.7-Flash loading test failure and Qwen3.5 flash-attention generation failure are model-specific but not the same bug." + }, + { + "left": "issue:43746", + "right": "issue:44811", "accept": false, - "reason": "CPU offload incompatibility in Qwen3_5MoeTextRotaryEmbedding is not the same as ONNX fp16 attention dtype mismatch in Canine." + "reason": "GraniteSpeech PEFT checkpoint loading and Whisper batch_decode skip_special_tokens behavior are different code paths." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Trailing-slash local model import warning and Qwen3 MOE weight conversion under deepspeed are unrelated." + "reason": "Local-path slash import warning and Qwen3 MOE weight conversion are unrelated." }, { - "left": "issue:43643", - "right": "issue:45759", + "left": "issue:45859", + "right": "issue:47050", "accept": false, - "reason": "trust_remote_code config-field loss and AutoModelForCausalLM.from_config unwrapping text_config are different config-loading bugs." + "reason": "Qwen3_5MoE rotary embedding CPU-offload incompatibility and Canine fp16 ONNX dtype mismatch are distinct issues." } ] }, @@ -6856,13 +6802,12 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28282", "issue:29159", "issue:29870", "issue:34567", - "issue:34689", "issue:38175", "issue:39692", "issue:40990", @@ -6874,6 +6819,7 @@ "issue:43450", "issue:43525", "issue:43526", + "issue:43643", "issue:43650", "issue:43696", "issue:43742", @@ -6897,6 +6843,7 @@ "issue:45459", "issue:45538", "issue:45600", + "issue:45759", "issue:45803", "issue:45859", "issue:46291", @@ -6907,231 +6854,231 @@ "issue:47254" ], "soft_pairs": [ + "issue:43643|issue:45759", "issue:43257|issue:43696", - "issue:43329|issue:45600", - "issue:43262|issue:44811", "issue:43526|issue:44554", + "issue:43329|issue:45600", "issue:38175|issue:44610", + "issue:43262|issue:44811", "issue:45200|issue:47254", - "issue:39692|issue:44912", "issue:42915|issue:44821", - "issue:43329|issue:44479", + "issue:39692|issue:44912", "issue:42915|issue:43742", + "issue:43329|issue:44479", "issue:44291|issue:45042", "issue:43329|issue:43450", - "issue:34567|issue:40990", "issue:43257|issue:43931", + "issue:34567|issue:40990", "issue:44387|issue:44617", "issue:29159|issue:34567", "issue:44912|issue:47145", "issue:45459|issue:45803", "issue:45081|issue:47020", "issue:44987|issue:45538", - "issue:43650|issue:43906", "issue:43525|issue:46291", "issue:43425|issue:46291", - "issue:45859|issue:47254", "issue:46421|issue:46424", + "issue:45859|issue:47254", "issue:28282|issue:29870", "issue:43749|issue:44756", - "issue:28282|issue:34689" + "issue:43650|issue:43906" ], - "prepared_review_unit_hash": "9fd40466fbd80e41b3621b981872e4bf4da9c0dbc591c577ab985ed3d5203293", + "prepared_review_unit_hash": "1451aee02cd631e6da896464ec7e4c29e1bd610e5a6b8b67807f3e5ab7c8ecf0", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35595, - "estimated_input_tokens": 8899, - "estimated_eval_tokens": 18054 + "serialized_chars": 36242, + "estimated_input_tokens": 9061, + "estimated_eval_tokens": 18378 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the soft-similar pairs mostly share broad subsystems (tokenizers, video processors, loading, quantization) but not the same concrete bug or fix. I would not merge any of the soft edges as duplicates; if a representative issue is needed, issue:43329 is the loosest hub because it is the most connected of the set.", - "confidence": 0.74, - "canonical_issue_reason": "Issue:43329 is the most central item in this set by soft-link connectivity and is a concrete, well-scoped bug report; however, the cluster overall is still too mixed to treat it as a true duplicate canonical.", + "summary": "None of the proposed soft pairs look like true duplicates; they mostly share broad subsystems or vague symptoms (video, quantization, loading, OOM) but describe different concrete failures. I\u2019d keep the issues separate rather than merging them into duplicate clusters.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43329 is the clearest, most concrete bug report in the set: it names an exact failing helper, the bad video code path, and the undefined variables involved. It reads like a precise canonical example for this collection.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:43329 is the best available representative only in a loose triage sense: it has the most related soft links and a specific code-path failure, but it should not be taken as proof the cluster is one duplicate family.", + "best_issue_reason": "issue:43329 is the strongest standalone issue candidate because it is specific, actionable, and tied to a clearly identified code path. It is more precise than the more generic loading/OOM reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43257", - "right": "issue:43696", + "left": "issue:43643", + "right": "issue:45759", "accept": false, - "reason": "Different problems: MOE weight conversion/loading with accelerate+deepspeed vs a GPT-oss CUDA OOM. No shared code path." + "reason": "Both concern config/model-loading behavior around composite Qwen-style configs, but one is about missing fields from AutoConfig with trust_remote_code and the other is about AutoModelForCausalLM not unwrapping text_config. Different APIs and failures." }, { - "left": "issue:43329", - "right": "issue:45600", + "left": "issue:43257", + "right": "issue:43696", "accept": false, - "reason": "One is a video token-counting undefined-variable bug; the other is stale auto_mappings config references. Same area, different fix." + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is unrelated to a GPT-oss CUDA OOM. Different models, mechanisms, and symptoms." }, { - "left": "issue:43262", - "right": "issue:44811", + "left": "issue:43526", + "right": "issue:44554", "accept": false, - "reason": "Both are audio/processor related, but one is apply_chat_template sampling-rate defaulting and the other is Whisper batch_decode skip_special_tokens handling." + "reason": "BeitImageProcessorFast label reduction is a processor bug; the MPS issue is an attention correctness problem. No shared code path." }, { - "left": "issue:43526", - "right": "issue:44554", + "left": "issue:43329", + "right": "issue:45600", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels bug vs an MPS attention correctness issue; unrelated code paths." + "reason": "Video token counting with undefined variables is not the same as stale auto_mappings references to removed Sam3LiteText configs. Different files and failures." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Both involve vision models/processors, but the symptoms and likely fixes differ: zero probabilities vs processor/model input-size mismatch." + "reason": "Unexpected zero probabilities in SigLIP2 and an OmDet-Turbo input-size mismatch are different model/processor issues." }, { - "left": "issue:45200", - "right": "issue:47254", + "left": "issue:43262", + "right": "issue:44811", "accept": false, - "reason": "Gemma 4 mm_token_type_ids defaulting vs a PEFT+DeepSpeed checkpointing error are unrelated loading/training failures." + "reason": "Audio chat-template sample-rate defaulting and Whisper batch_decode ignoring skip_special_tokens are unrelated tokenizer/processor behaviors." }, { - "left": "issue:39692", - "right": "issue:44912", + "left": "issue:45200", + "right": "issue:47254", "accept": false, - "reason": "SigLIP2 doc-example errors and gpt-oss MXFP4 fallback are separate issues with different models and failure modes." + "reason": "Gemma 4 mm_token_type_ids defaulting and a PEFT/DeepSpeed checkpoint error are separate training/loading problems." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8 loading failure vs AutoImageProcessor-from-URL failure; no concrete overlap." + "reason": "FineGrainedFP8Config failure on Qwen3Moe and loading AutoImageProcessor from URL are different paths with no concrete overlap." }, { - "left": "issue:43329", - "right": "issue:44479", + "left": "issue:39692", + "right": "issue:44912", "accept": false, - "reason": "Both touch video VLMs, but one is undefined variables in multimodal token counting and the other is a broader v5.3.0 video-input regression." + "reason": "A broken SigLIP2 documentation example is not the same bug as MXFP4 loading falling back to bf16." }, { "left": "issue:42915", "right": "issue:43742", "accept": false, - "reason": "FineGrainedFP8 Qwen3Moe failure vs MobileLLM key error are unrelated model-loading bugs." + "reason": "Qwen3Moe FineGrainedFP8Config failure and MobileLLM key errors are distinct loading bugs." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Both involve video, but one is an undefined-variable bug in multimodal token counting while the other is a broader v5.3.0 regression across several video models. Not enough to treat as the same concrete defect." }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized argument mismatch is unrelated to PIL backend processors requiring torchvision." + "reason": "Unexpected _is_hf_initialized handling in init_empty_weights is unrelated to PIL backend processors requiring torchvision." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Both concern video processing, but one is multimodal token counting and the other is batched processor output shape. Not the same bug." + "reason": "A bad video token-count helper and batched video processor shape bugs are different issues in the same broad area, but not the same underlying bug." }, { - "left": "issue:34567", - "right": "issue:40990", + "left": "issue:43257", + "right": "issue:43931", "accept": false, - "reason": "TrainerState token counters and GPT-oss perplexity are unrelated; no shared fix." + "reason": "Qwen3 MOE weight conversion and Qwen3-VL weight-shape mismatches are separate model-loading failures." }, { - "left": "issue:43257", - "right": "issue:43931", + "left": "issue:34567", + "right": "issue:40990", "accept": false, - "reason": "Both mention Qwen3, but one is MOE weight conversion with distributed loading and the other is a VL weight-shape mismatch. Different failure classes." + "reason": "TrainerState token counting not updating is unrelated to high perplexity on GPT-oss-20b." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Both are OOM reports, but one is int4 quantization reserved-memory growth and the other is Sam3Video memory usage. Too generic to merge." + "reason": "Both mention OOM, but one is int4 quantization memory regression in Transformers 5.x and the other is a Sam3Video model-specific OOM. Different root causes." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Tokenizer slow/fast inconsistency is unrelated to TrainerState num_input_tokens_seen not updating." + "reason": "Tokenizer slow/fast inconsistency is not the same as TrainerState token accounting not updating." }, { "left": "issue:44912", "right": "issue:47145", "accept": false, - "reason": "Both are quantization/loading bugs, but one is MXFP4 falling back to bf16 and the other is FP8 frozen checkpoints missing weight_scale. Distinct mechanisms." + "reason": "Both concern quantized model loading, but one is MXFP4 fallback to bf16 and the other is silent corruption from missing weight_scale application on frozen FP8 checkpoints. Different formats and bugs." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve exception handling, but the code paths are different and the affected APIs are unrelated." + "reason": "One is exception handling around protobuf decode errors; the other is a bare except swallowing KeyboardInterrupt in Fuyu tensor conversion. Different code and behavior." }, { "left": "issue:45081", "right": "issue:47020", "accept": false, - "reason": "Mistral regex patch crash vs XLMRobertaTokenizer Unigram sequence/dict mismatch are separate tokenizer bugs." + "reason": "Mistral regex patch crashing on tokenizer internals and XLMRobertaTokenizer passing a dict to Unigram are unrelated tokenizer initialization bugs." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "Model loading failure for physical-intelligence/fast and CLIPTokenizer model_max_length misuse do not describe the same defect." - }, - { - "left": "issue:43650", - "right": "issue:43906", - "accept": false, - "reason": "'ADD THE DATA' is too underspecified and does not match an isolated reproduction of issue 38071." + "reason": "Loading physical-intelligence/fast on transformers>=5.1.0 and CLIPTokenizer model_max_length being absurdly large are separate issues." }, { "left": "issue:43525", "right": "issue:46291", "accept": false, - "reason": "Llama4Config missing pad_token_id and kernels import-time version/revision enforcement are unrelated." + "reason": "Llama4Config missing pad_token_id is unrelated to the kernels package version/revision requirement when importing pipelines." }, { "left": "issue:43425", "right": "issue:46291", "accept": false, - "reason": "Torch 2.10 incompatibility and kernels import requirements are different dependency problems." + "reason": "Torch 2.10 incompatibility is a dependency-version issue, but not the same as kernels import failure." }, { - "left": "issue:45859", - "right": "issue:47254", + "left": "issue:46421", + "right": "issue:46424", "accept": false, - "reason": "CPU offload incompatibility in Qwen3_5Moe rotary embedding vs a checkpointing error under PEFT+ZeRO-3 are different code paths." + "reason": "Lazy cache initialization causing spurious recompilation and max_cache_length only applying to Quantized Cache are different cache-behavior bugs." }, { - "left": "issue:46421", - "right": "issue:46424", + "left": "issue:45859", + "right": "issue:47254", "accept": false, - "reason": "Both are cache-related, but one is lazy cache init causing recompilation and the other is max_cache_length only applying to Quantized Cache." + "reason": "CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding is unrelated to PEFT/DeepSpeed checkpoint errors." }, { "left": "issue:28282", "right": "issue:29870", "accept": false, - "reason": "PyTorch missing for AutoModel import and Bart generate decoder_attention_mask unbound local are unrelated." + "reason": "PyTorch-missing AutoModel ImportError and Bart generate decoder_attention_mask unbound variable are unrelated failures." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "FSDP CPU RAM efficient loading breakage and Strix Halo mmap OOM are both memory/loading issues but have different root causes and fixes." + "reason": "FSDP CPU RAM efficient loading broken and disabling mmap on Strix Halo to avoid OOM are different memory-related issues." }, { - "left": "issue:28282", - "right": "issue:34689", + "left": "issue:43650", + "right": "issue:43906", "accept": false, - "reason": "AutoModel import missing PyTorch and Llama 3.2 Vision loading regression are not the same bug." + "reason": "'ADD THE DATA' is not a substantive bug report, and the reproduction issue is clearly separate." } ] }, @@ -7153,7 +7100,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:24643", "issue:27957", @@ -7162,7 +7109,7 @@ "issue:29159", "issue:33357", "issue:34567", - "issue:36331", + "issue:34689", "issue:39379", "issue:41720", "issue:42175", @@ -7181,7 +7128,6 @@ "issue:44062", "issue:44291", "issue:44479", - "issue:44488", "issue:44554", "issue:44560", "issue:44568", @@ -7205,6 +7151,7 @@ "issue:47254" ], "soft_pairs": [ + "issue:28282|issue:34689", "issue:43540|issue:46701", "issue:43454|issue:43577", "issue:44568|issue:45356", @@ -7214,17 +7161,17 @@ "issue:43742|issue:44291", "issue:44936|issue:46821", "issue:43697|issue:45357", - "issue:28282|issue:29127", "issue:45137|issue:45405", - "issue:44479|issue:46693", + "issue:28282|issue:29127", "issue:39379|issue:44554", + "issue:44479|issue:46693", "issue:43582|issue:47254", - "issue:44291|issue:44821", "issue:42915|issue:45538", + "issue:44291|issue:44821", "issue:44857|issue:46832", + "issue:45538|issue:46869", "issue:24643|issue:29159", "issue:44062|issue:44991", - "issue:45538|issue:46869", "issue:42915|issue:43749", "issue:43452|issue:43582", "issue:27957|issue:28282", @@ -7232,225 +7179,210 @@ "issue:41720|issue:43540", "issue:43540|issue:44560", "issue:43577|issue:46512", - "issue:43531|issue:44821", - "issue:43540|issue:44479", - "issue:42175|issue:44488", - "issue:34567|issue:36331" + "issue:43531|issue:44821" ], - "prepared_review_unit_hash": "bfa3131928e6b253ac5f11d3018783efb4cbed8dc2ae5b260b1d10a4c4736e47", + "prepared_review_unit_hash": "045a9258f53eb5f174d6f49ca8d54291af3f24902dc8752e7ad9a04d59b98da5", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 37007, - "estimated_input_tokens": 9252, - "estimated_eval_tokens": 18760 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35835, + "estimated_input_tokens": 8959, + "estimated_eval_tokens": 18174 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a heterogeneous mix of unrelated bugs across tokenization, multimodal/video, training, loading, dtype, and device-mapping paths. None of the suggested soft edges look like the same underlying defect, so all are rejected.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue is defensible here; the candidate pairs do not form one duplicate cluster.", + "summary": "This cluster is highly heterogeneous: most soft pairs only share a broad subsystem (tokenizers, loading, video, training, distributed) and do not look like the same underlying bug. I reject all soft edges. If a single anchor is needed, issue:45488 is the best representative because it is the most active open regression report and the closest to the tokenizer-regression subset, but it does not truly cover the whole cluster.", + "confidence": 0.71, + "canonical_issue_reason": "No single issue cleanly represents this mixed cluster; issue:45488 is the least-bad anchor because it is open, active, has inbound references, and matches the tokenizer-regression theme better than the rest.", "canonical_pr_reason": null, - "best_issue_reason": "No best issue candidate stands out because the items are not one shared bug family.", + "best_issue_reason": "issue:45488 is the strongest standalone issue here: recent activity, multiple references, and a broad regression description make it the best issue to keep visible, even though the overall set is not a true duplicate cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Both are model-loading failures, but one is a missing-PyTorch import error and the other is a specific Llama vision loading regression; different root causes." + }, { "left": "issue:43540", "right": "issue:46701", "accept": false, - "reason": "Different multimodal failures: Qwen3OmniMoe video input handling vs DiffusionGemma image-text pipeline output." + "reason": "Both involve multimodal/video pipelines, but the model families and failure modes differ too much to treat as one bug." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "One is missing lm_head tying in AyaVision/Mistral3; the other is dtype propagation in BLIP2." + "reason": "One is about tied lm_head weights and garbage generation; the other is about dtype propagation in BLIP2. Shared loading theme only." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but on different models with different failure modes and fixes." + "reason": "Both are tokenizer regressions, but they point to different behaviors and likely different code paths." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "Mac bus error on CLIP loading is unrelated to TrainerState token-count tracking." + "reason": "Completely unrelated: MacOS bus error in CLIP loading vs. a TrainerState counter not updating." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Missing lm_head tying in Mistral3 is not the same bug as LoRA-merge collapse with extended vocab." + "reason": "Both mention tied embeddings, but one is a missing tie in Mistral3 and the other is a LoRA-merge regression after vocab extension; not the same bug." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Packaging/backend dependency issue is unrelated to a missing Qwen2.5-VL class import." + "reason": "Packaging/backend availability issue vs. a missing model class import error; unrelated." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "Different loading failures: key error for one model vs unexpected _is_hf_initialized arg in init_empty_weights." + "reason": "Different loading failures with different triggers: a MobileLLM key error versus init_empty_weights passing an unexpected argument." }, { "left": "issue:44936", "right": "issue:46821", "accept": false, - "reason": "Trainer.evaluate() after train() is a trainer lifecycle issue, not the TimesFM window_size AttributeError." + "reason": "Trainer lifecycle bug vs. TimesFM window_size AttributeError; no shared underlying defect." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RTDetrV2 output regression and Qwen3.5 save_pretrained key mapping are separate code paths." + "reason": "RT-DETR output drift and Qwen3.5 save_pretrained key handling are different regressions." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed deque error and a minimum-version bump issue are unrelated." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Missing PyTorch backend import and LayoutLMv3 box-validation messaging are unrelated." + "reason": "One is missing PyTorch; the other is a LayoutLMv3 error-message clarity issue. Not duplicates." }, { - "left": "issue:45137", - "right": "issue:45405", + "left": "issue:39379", + "right": "issue:44554", "accept": false, - "reason": "DeepSpeed ZeRO-3 deque error is unrelated to a PEFT version bump being unreleased." + "reason": "Different model/runtime correctness problems: mllama masking logic versus an MPS attention issue." }, { "left": "issue:44479", "right": "issue:46693", "accept": false, - "reason": "Video-input regression and FA2 long-context performance regression are distinct issues." + "reason": "Both are v5 / generation regressions, but one is video-input handling across Qwen VL variants and the other is FA2 long-context performance." }, { - "left": "issue:39379", - "right": "issue:44554", + "left": "issue:43582", + "right": "issue:47254", "accept": false, - "reason": "mllama mask handling and MPS attention correctness with mismatched head dims are different bugs." + "reason": "Apple Silicon allocator warmup TypeError and a PEFT+ZeRO-3 checkpoint error are unrelated." }, { - "left": "issue:43582", - "right": "issue:47254", + "left": "issue:42915", + "right": "issue:45538", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError is unrelated to PEFT+DeepSpeed checkpointing failures." + "reason": "Qwen3Moe FP8 loading and CLIPTokenizer max length are unrelated subsystems." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "init_empty_weights compatibility and loading AutoImageProcessor from URL are not the same defect." + "reason": "init_empty_weights argument handling and AutoImageProcessor URL loading are different bugs." }, { - "left": "issue:42915", - "right": "issue:45538", + "left": "issue:44857", + "right": "issue:46832", "accept": false, - "reason": "FineGrainedFP8Config failure is unrelated to CLIPTokenizer model_max_length behavior." + "reason": "Different RT-DETR-related failures in separate code paths; one is loss/AMP crashing, the other is forward logic with feature-level mismatch." }, { - "left": "issue:44857", - "right": "issue:46832", + "left": "issue:45538", + "right": "issue:46869", "accept": false, - "reason": "LwDetr AMP crash and RTDetr feature-level TypeError are separate detection bugs." + "reason": "Tokenizer max-length behavior and YOLO multi-GPU device mapping are unrelated." }, { "left": "issue:24643", "right": "issue:29159", "accept": false, - "reason": "DeepSpeed training weight-shape error is unrelated to tokenizer slow/fast inconsistency." + "reason": "DeepSpeed weight-shape runtime error and slow/fast tokenizer inconsistency are not the same issue." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "AddedToken special-kwarg duplication is not the same as a model-specific tokenizer load failure." - }, - { - "left": "issue:45538", - "right": "issue:46869", - "accept": false, - "reason": "Tokenizer model_max_length bug and YOLO multi-GPU device-map failure are unrelated." + "reason": "This is the closest pair, but the evidence is still too thin: one is a specific AddedToken TypeError, the other a model-specific tokenizer load failure. Could be related, but not enough to merge." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Qwen3MoE FP8 config failure is unrelated to FSDP CPU RAM efficient loading." + "reason": "Both are training/loading-related in broad terms, but they concern different subsystems and failure modes." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "gguf_file loading breakage and allocator warmup TypeError do not share a concrete code path." + "reason": "Tokenizer/from_pretrained gguf handling and Apple Silicon allocator warmup are unrelated." }, { "left": "issue:27957", "right": "issue:28282", "accept": false, - "reason": "XLMRoberta Flash Attention 2 support and missing PyTorch import are unrelated." + "reason": "XLMRoberta with Flash Attention 2 and missing-PyTorch AutoModel import are unrelated." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both are tokenizer regressions, but one drops BOS/EOS tokens while the other overrides tokenizer.json pre-tokenization." + "reason": "Both are tokenizer regressions, but they affect different models and different behaviors; shared broad theme only." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Qwen3 auto device-map CUDA assert and Qwen3OmniMoe video ValueError are different failure modes." + "reason": "Different Qwen-family device-map versus video-input failures; not the same underlying bug." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Both are Qwen3-family video bugs, but they affect different models and code paths." + "reason": "Both involve video inputs, but they are different models and different exceptions." }, { "left": "issue:43577", "right": "issue:46512", "accept": false, - "reason": "BLIP2 dtype staying float32 is not the same as from_config leaving config.dtype stale." + "reason": "One is BLIP2 dtype loading behavior; the other is stale config.dtype after from_config(dtype=...). Similar area, different bug." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding_window issue is unrelated to AutoImageProcessor URL loading." - }, - { - "left": "issue:43540", - "right": "issue:44479", - "accept": false, - "reason": "Qwen3OmniMoe video crash and broader v5.3.0 video regression are related by modality only, not by a concrete shared bug." - }, - { - "left": "issue:42175", - "right": "issue:44488", - "accept": false, - "reason": "Dependency/backend installation issue is unrelated to loading a specific tokenizer/model." - }, - { - "left": "issue:34567", - "right": "issue:36331", - "accept": false, - "reason": "TrainerState input-token counting and CustomTrainer.compute_loss signature compatibility are separate trainer issues." + "reason": "Sliding_window behavior in Qwen3-MoE and AutoImageProcessor URL loading are unrelated." } ] }, @@ -7472,14 +7404,16 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:24643", "issue:28282", "issue:29870", "issue:32090", + "issue:34567", + "issue:36331", "issue:38175", - "issue:42796", + "issue:42175", "issue:43257", "issue:43452", "issue:43540", @@ -7489,7 +7423,6 @@ "issue:43756", "issue:43906", "issue:44062", - "issue:44466", "issue:44479", "issue:44488", "issue:44568", @@ -7502,7 +7435,6 @@ "issue:44898", "issue:44964", "issue:44987", - "issue:45005", "issue:45042", "issue:45072", "issue:45081", @@ -7525,6 +7457,9 @@ ], "soft_pairs": [ "issue:44849|issue:45701", + "issue:43540|issue:44479", + "issue:42175|issue:44488", + "issue:34567|issue:36331", "issue:44479|issue:45381", "issue:44062|issue:45081", "issue:44987|issue:45907", @@ -7535,230 +7470,231 @@ "issue:43257|issue:43756", "issue:43756|issue:45325", "issue:43756|issue:46399", - "issue:46463|issue:47276", "issue:45672|issue:45674", "issue:44737|issue:46463", + "issue:46463|issue:47276", "issue:43716|issue:46701", "issue:45823|issue:46463", "issue:45405|issue:45600", - "issue:43906|issue:46693", - "issue:44568|issue:46491", "issue:43577|issue:45072", + "issue:44568|issue:46491", + "issue:43906|issue:46693", "issue:24643|issue:29870", "issue:44625|issue:44877", "issue:43540|issue:46693", "issue:43452|issue:44488", "issue:28282|issue:32090", - "issue:42796|issue:43906", - "issue:44466|issue:45005", "issue:46463|issue:46899" ], - "prepared_review_unit_hash": "7bbe853a0377e8f3d0293029af9d88108ea76393da3f767867ea0d6f87e0ceda", + "prepared_review_unit_hash": "597e44665c281439e5717a81dc2cf025fd01633003efe358f83b6fb5223c6232", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36444, - "estimated_input_tokens": 9111, - "estimated_eval_tokens": 18478 + "soft_pair_count": 29, + "serialized_chars": 36713, + "estimated_input_tokens": 9179, + "estimated_eval_tokens": 18614 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Conservative read: the cluster is mostly unrelated false positives, with only the Qwen video-position-id pair and the tied-weights serialization pair looking like the same underlying bug. The Gemma4 4-bit pair issue:45672 \u2194 issue:45674 should not be merged; they share only a broad 4-bit/quantization theme, but the titles point to distinct failures in different code paths.", - "confidence": 0.8, - "canonical_issue_reason": "issue:44479 is the broadest concrete regression report here, covering multiple Qwen VL variants and the video-input position-id path.", + "summary": "This cluster is a grab bag of mostly unrelated Transformers issues across trainers, tokenizers, multimodal models, configs, and packaging. I don't see any soft pair that clearly shares the same underlying bug or change.", + "confidence": 0.91, + "canonical_issue_reason": "issue:46693 is the strongest single anchor here: it has the clearest regression framing, broad impact, and the most discussion/inbound references among otherwise unrelated reports.", "canonical_pr_reason": null, - "best_issue_reason": "Same as canonical: it is the most representative single issue to keep as the cluster anchor.", + "best_issue_reason": "issue:46693 is the best representative issue because it is a concrete, high-signal performance regression with strong evidence of user impact and active follow-up.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 hidden-state behavior vs a generic tokenization-change report; different subsystem and symptom." + "reason": "Qwen3.5 hidden-state behavior vs tokenizer/version-dependent tokenization changes; different subsystems and symptoms." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related, but one is a Qwen3OmniMoe input-processing failure and the other is a broader v5.3.0 video regression across Qwen VL models." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Packaging/backend install issue vs a specific model-loading failure; no shared code path." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "TrainerState token counter update bug vs CustomTrainer compute_loss signature breakage; separate trainer issues." }, { "left": "issue:44479", "right": "issue:45381", - "accept": true, - "reason": "Both describe the same Qwen VL video-input position-id regression; 45381 looks like a narrower manifestation of 44479." + "accept": false, + "reason": "Both mention Qwen video regressions, but the affected model families and failure modes differ enough that this is not the same bug." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "AddedToken argument conflict vs a Mistral regex patch crash; unrelated failure modes." + "reason": "Tokenizer AddedToken argument conflict vs Mistral regex patch crash; unrelated tokenizer internals." }, { "left": "issue:44987", "right": "issue:45907", "accept": false, - "reason": "Generic model-loading failure vs a CI bug about forwarding revision in list_repo_files; different code paths." + "reason": "Model loading failure vs list_repo_files not forwarding revision; different APIs and fixes." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image processor torchvision dependency regression vs tokenizer regex crash; not the same bug." + "reason": "Image-processor torchvision dependency regression vs a tokenizer regex AttributeError; unrelated areas." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "SigLIP zero-probabilities issue vs Perceiver non-default-resolution failure; different models and mechanics." + "reason": "Different image-model failures with no common concrete mechanism." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Janus test failure vs Phi-4 multimodal loading failure; too different to merge." + "reason": "Janus test failure vs Phi-4 multimodal loading failure; same broad domain, different bugs." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "MobileLLM key error vs missing Float8 storage type error; unrelated load/runtime issues." + "reason": "MobileLLM key error vs Float8 storage lookup error; unrelated runtime/load failures." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Accelerate+DeepSpeed MoE weight conversion bug vs Smollm3 RoPE-layer drop; different model families and code paths." + "reason": "Qwen3 MoE deepspeed conversion problem vs SmolLM3 RoPE layer omission; different model families and root causes." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Smollm3 RoPE-layer count bug vs Qwen2.5-VL temporal position scaling bug; different architectures and symptoms." + "reason": "SmolLM3 architecture bug vs Qwen2.5-VL temporal position scaling; not the same code path." }, { "left": "issue:43756", "right": "issue:46399", "accept": false, - "reason": "Smollm3 RoPE bug vs Qwen3.5 kernel registration crash; unrelated." - }, - { - "left": "issue:46463", - "right": "issue:47276", - "accept": false, - "reason": "T5 decoder-input validation/message issue vs a synced_gpus warning; not the same defect." + "reason": "SmolLM3 RoPE issue vs Qwen3.5 kernel registration crash; unrelated fixes." }, { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Gemma4 multimodal uint8 dtype-cast bug vs BitsAndBytes lm_head exclusion regression; both 4-bit-related, but they affect different mechanisms and code paths." + "reason": "Both are 4-bit inference issues, but one is Gemma4 uint8/torch.finfo handling and the other is BitsAndBytes skip-module exclusion logic; different root causes." }, { "left": "issue:44737", "right": "issue:46463", "accept": false, - "reason": "XLNet CPU arange bug vs T5 decoder validation issue; no overlap in bug or fix." + "reason": "XLNet CPU-device bug vs T5 decoder-input validation message; unrelated model paths." + }, + { + "left": "issue:46463", + "right": "issue:47276", + "accept": false, + "reason": "Both are diagnostics/warnings, but the affected checks and code paths differ." }, { "left": "issue:43716", "right": "issue:46701", "accept": false, - "reason": "Mistral image-preprocessor dtype mismatch vs DiffusionGemma pipeline failure; different model paths." + "reason": "Mistral-3 image-preprocessor dtype mismatch vs DiffusionGemma pipeline failure; different multimodal components." }, { "left": "issue:45823", "right": "issue:46463", "accept": false, - "reason": "Gemma4 device_map auto mismatch vs T5 decoder validation issue; unrelated." + "reason": "Gemma4 device_map forward mismatch vs T5 validation error text; no shared bug." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "PEFT version-bump issue vs removed Sam3LiteText auto-mapping references; one is dependency metadata, the other CI/config cleanup." + "reason": "PEFT version pin issue vs stale auto_mappings reference; separate packaging/CI problems." }, { - "left": "issue:43906", - "right": "issue:46693", + "left": "issue:43577", + "right": "issue:45072", "accept": false, - "reason": "Reproduction of another issue vs FA2 long-context performance regression; not the same bug." + "reason": "Both involve dtype mismatches, but on different model stacks and with different failure mechanisms." }, { "left": "issue:44568", "right": "issue:46491", "accept": false, - "reason": "mdeberta special-token insertion regression vs CodeLlama leading-space round-trip regression; different tokenizers and behaviors." + "reason": "Tokenizer special-token regression vs CodeLlama round-trip spacing regression; distinct tokenizer behaviors." }, { - "left": "issue:43577", - "right": "issue:45072", + "left": "issue:43906", + "right": "issue:46693", "accept": false, - "reason": "Blip2 dtype staying float32 vs SwitchTransformers/TimmWrapper bfloat16 mismatch; both dtype issues but not the same code path." + "reason": "One is a reproduction of an earlier issue, the other is a long-context FA2 performance regression; not the same bug." }, { "left": "issue:24643", "right": "issue:29870", "accept": false, - "reason": "DeepSpeed training weight-shape error vs Bart generate() decoder_attention_mask UnboundLocalError; unrelated." + "reason": "DeepSpeed training weight-shape error vs BART generate mask initialization bug; unrelated failures." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation bug vs strict config blocking granite_speech load; different configuration problems." + "reason": "Qwen3.5 num_labels propagation vs strict config loading for granite_speech; different config plumbing." }, { "left": "issue:43540", "right": "issue:46693", "accept": false, - "reason": "Qwen3OmniMoe video-input ValueError vs FA2 performance regression; unrelated." + "reason": "Video-input processing in Qwen3OmniMoe vs FA2 long-context performance regression; unrelated subsystems." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "Generic gguf_file/from_pretrained breakage vs a specific model-load failure; too little evidence they share the same concrete bug." + "reason": "GGUF tokenizer/model loading regression vs a specific sleng-bert load failure; too different to merge." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "PyTorch missing import error vs NoneType passed to _gpu_broadcast_one; different runtime failures." - }, - { - "left": "issue:42796", - "right": "issue:43906", - "accept": false, - "reason": "Tokenizer BPE-marker handling bug vs an isolated reproduction of another issue; not the same underlying defect." - }, - { - "left": "issue:44466", - "right": "issue:45005", - "accept": true, - "reason": "Both report tied-weights/lm_head serialization regressions in v5, with device-dependent behavior; they look like the same underlying change." + "reason": "Missing PyTorch dependency error vs unsupported NoneType in `_gpu_broadcast_one`; unrelated issues." }, { - "left": "issue:46463", + "left": "issue:45672", "right": "issue:46899", "accept": false, - "reason": "T5 decoder validation/message issue vs Gemma4 4-bit multimodal processing bug; unrelated." + "reason": "Both concern Gemma4 multimodal 4-bit inference, but the reported roots differ: unsafe dtype casting in embedders vs torch.finfo on uint8 audio weights." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The cluster anchor and the two Qwen/tied-weights soft edges are grounded in the packet, but the acceptance of issue:45672 \u2194 issue:45674 is too aggressive. Their titles point to different 4-bit bugs in Gemma4: one is an unsafe dtype cast in multimodal/audio embedders, the other is a BitsAndBytesConfig lm_head exclusion regression. They are related only at a very high level, so the summary overstates evidence. The overall evaluation should be more conservative and not treat that pair as the same bug family." - }, - "evaluator_used": true, - "retried": true, - "accepted_nontrivial_soft_edge": true, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -7773,7 +7709,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:29159", "issue:33357", @@ -7782,19 +7718,19 @@ "issue:41762", "issue:43295", "issue:43452", - "issue:43525", "issue:43531", "issue:43643", - "issue:43646", "issue:43653", "issue:43688", "issue:43697", "issue:43720", "issue:43742", "issue:43749", + "issue:43906", "issue:43994", "issue:44361", "issue:44442", + "issue:44466", "issue:44493", "issue:44737", "issue:44811", @@ -7805,8 +7741,8 @@ "issue:44913", "issue:44936", "issue:44991", + "issue:45005", "issue:45216", - "issue:45356", "issue:45357", "issue:45372", "issue:45375", @@ -7818,19 +7754,21 @@ "issue:46050", "issue:46121", "issue:46170", - "issue:46701", + "issue:46468", "issue:46821", "issue:46829", "issue:46846", "issue:47000" ], "soft_pairs": [ + "issue:44466|issue:45005", + "issue:43906|issue:46468", "issue:45663|issue:46846", - "issue:43452|issue:44843", "issue:44913|issue:46121", - "issue:44737|issue:45588", - "issue:43697|issue:46829", "issue:41762|issue:44936", + "issue:43697|issue:46829", + "issue:43452|issue:44843", + "issue:44737|issue:45588", "issue:43697|issue:45216", "issue:43720|issue:44898", "issue:45357|issue:45920", @@ -7849,216 +7787,204 @@ "issue:45372|issue:46170", "issue:44857|issue:47000", "issue:43688|issue:45704", - "issue:43688|issue:45701", - "issue:43525|issue:46701", - "issue:43646|issue:43749", - "issue:43295|issue:45356" + "issue:43688|issue:45701" ], - "prepared_review_unit_hash": "1881077c69ae979969f9271fb0ec7fdcdef20d919a2e3d3f1b937178f92438c5", + "prepared_review_unit_hash": "9fad21d4dad710328697ca96907059847437fb3fc059356fbafa103a09e2e111", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36402, - "estimated_input_tokens": 9101, - "estimated_eval_tokens": 18458 + "soft_pair_count": 27, + "serialized_chars": 36096, + "estimated_input_tokens": 9024, + "estimated_eval_tokens": 18304 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a collection of distinct regression reports across different models, configs, and runtimes. Issue 45920 is still the best umbrella issue, but no soft-edge pair is convincing as a true duplicate; in particular, 45372 vs 46170 are unrelated failures and should be rejected. Overall the cluster should be treated as heterogeneous rather than merged on duplicates.", - "confidence": 0.88, - "canonical_issue_reason": "Issue 45920 is the broadest and most central tokenizer regression here, with multiple model families affected and the clearest umbrella scope among the set.", + "summary": "No strong duplicate group emerges from these issues; the soft pairs mostly share only broad subsystem similarity (tokenizers, configs, loading, vision models) but describe different failures and code paths.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45920 is the best representative: it describes a broad v5 tokenizer regression affecting multiple model families, making it the most general and cluster-like issue among otherwise mostly model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45920 is the best global representative because it captures a wide tokenizer-loading/token-ID regression pattern rather than a model-specific edge case.", + "best_issue_reason": "Same reason: it is the broadest and most reusable issue title in the set, with a clear regression framing and multiple impacted models.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45663", - "right": "issue:46846", + "left": "issue:44466", + "right": "issue:45005", "accept": false, - "reason": "Different models and different failure modes: Gemma-4 FSDP2 KeyError vs Qwen3.5 TP planning/OOM-shape issues." + "reason": "Both mention tied weights in v5, but one is a device-dependent lm_head serialization bug and the other is a broader translation-model tied-weights issue; not the same concrete failure." }, { - "left": "issue:43452", - "right": "issue:44843", + "left": "issue:43906", + "right": "issue:46468", + "accept": false, + "reason": "One is an isolated reproduction of an older issue, the other is an example-specific Nemotron breakage; different artifacts and likely different root causes." + }, + { + "left": "issue:45663", + "right": "issue:46846", "accept": false, - "reason": "Both involve tokenizer loading, but one is gguf_file handling and the other is an offline-mode model_info call in mistral regex patching." + "reason": "Different models and different failure modes: Gemma-4 FSDP2 KeyError vs Qwen3.5 TP plan OOM/shape errors." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "Separate config bugs: GPTNeoX rotary_pct reload regression vs rope-validation TypeError on JSON-loaded lists." + "reason": "Both involve rope/config persistence, but one is a silent default reset on reload and the other is a TypeError from a JSON-loaded list; not the same bug." }, { - "left": "issue:44737", - "right": "issue:45588", + "left": "issue:41762", + "right": "issue:44936", "accept": false, - "reason": "Unrelated code paths: XLNet device placement bug vs flash_attention crash on s_aux=None." + "reason": "Loading Gemma3 with ZeRO-3 and trainer.evaluate() after train are unrelated execution paths with different symptoms." }, { "left": "issue:43697", "right": "issue:46829", "accept": false, - "reason": "Different models and outputs; no shared concrete bug beyond generic v5 regression symptoms." + "reason": "Different models and outputs: RTDetrV2 regression vs VideoPrism output field mix-up." }, { - "left": "issue:41762", - "right": "issue:44936", + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both are tokenizer-loading problems, but gguf_file breaks model/tokenizer loading generally while the other is an offline-mode regex patch bug." + }, + { + "left": "issue:44737", + "right": "issue:45588", "accept": false, - "reason": "Different subsystems: DeepSpeed ZeRO-3 loading failure vs trainer evaluate-after-train failure." + "reason": "XLNet CPU-device inefficiency and flash_attention s_aux=None crash are unrelated code paths." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Different regressions: RTDetrV2 output drift vs Qwen3.5 save_pretrained checkpoint corruption." + "reason": "Different Qwen/RTDetr concerns: one is changed outputs at inference, the other is incorrect save_pretrained checkpoint serialization." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "BitNet packed-weight loading is unrelated to Perceiver resolution/interpolation failure." + "reason": "BitNet accelerate-loading unpacking failure and Perceiver non-default-resolution failure are different model-specific bugs." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 save_pretrained visual-encoder key bug is distinct from the broad tokenizer token-ID regression." + "reason": "Qwen3.5 visual encoder key serialization is a model save_pretrained bug, while the other is a broad tokenizer-ID regression across many models." }, { "left": "issue:44493", "right": "issue:46050", "accept": false, - "reason": "Unexpected state-dict key issue is not the same as sinq quantized model save/load breakage." + "reason": "Unexpected position-id keys and sinq quantization save/load breakage affect different serialization surfaces." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "Both are tokenizer-related, but BigBird special-token registration and MLuke task AttributeError are different bugs." + "reason": "Both are tokenizer-related, but BigBird special-token registration and MLuke task AttributeError are distinct bugs." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "Different tokenizer failures: MLuke task AttributeError vs AutoTokenizer failing on FastSpeech2ConformerTokenizer." + "reason": "MLuke task failures and AutoTokenizer loading FastSpeech2ConformerTokenizer are unrelated tokenizer issues." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Siglip2 zero-probability output issue is unrelated to processor.tokenizer/image-passing regression." + "reason": "Siglip2 zero-probability outputs and processor.tokenizer/image-passing regression are different model/processor failures." }, { "left": "issue:43688", "right": "issue:43749", "accept": false, - "reason": "Aux-loss normalization bug does not match the FSDP CPU RAM efficient loading regression." + "reason": "Aux-loss normalization in MoE models and FSDP CPU RAM efficient loading are unrelated subsystems." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both mention missing config fields, but the first is a generic trust_remote_code load issue and the second is a specific strict-drop field regression; not clearly the same bug." + "reason": "Both mention missing config fields, but one is a generic trust_remote_code loading issue and the other is a specific Qwen vision config field dropped by strict parsing." }, { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to MobileLLM key errors." + "reason": "Qwen3-MoE sliding_window behavior and MobileLLM-125M load key error are different model-specific issues." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Different vision-model issues: nonsensical SigLIP2 outputs vs Perceiver non-default resolution failure." + "reason": "Siglip2 nonsensical outputs and Perceiver interpolation failure are different vision-model regressions." }, { "left": "issue:44821", "right": "issue:46821", "accept": false, - "reason": "AutoImageProcessor-from-URL loading is unrelated to the TimesFM window_size AttributeError." + "reason": "AutoImageProcessor URL loading and TimesFM window_size AttributeError are unrelated loading vs model-argument bugs." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer loading regressions, but one is offline-mode mistral-regex networking behavior and the other is a model-specific tokenizer load failure." + "reason": "Offline HF hub lookup in mistral regex patch and tokenizer loading for EMBEDDIA/est-roberta are different causes." }, { "left": "issue:40444", "right": "issue:44811", "accept": false, - "reason": "Multiple-images iterable finetuning failure is unrelated to Whisper batch_decode skip_special_tokens behavior." + "reason": "IterableDataset multi-image finetuning failure and Whisper batch_decode skip_special_tokens bug are different code paths." }, { "left": "issue:29159", "right": "issue:33357", "accept": false, - "reason": "Tokenizer slow/fast inconsistency and CLIP Mac bus error are not the same failure." + "reason": "Tokenizer slow/fast inconsistency and MacOS bus error on CLIP model are unrelated." }, { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Gemma 4 processor loading via mistral_common and Windows/ROCm torch.distributed import/type errors are unrelated code paths and not a single mergeable fix." + "reason": "Gemma 4 processor import failure from mistral_common and Windows ROCm torch.distributed dependency issues are different platform/library problems." }, { "left": "issue:44857", "right": "issue:47000", "accept": false, - "reason": "Different detection losses and different crash causes; not mergeable as one fix." + "reason": "LwDetrImageLoss AMP crash and RTDetrLoss infeasible cost-matrix error are different loss-function bugs." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Auxiliary-loss normalization bug is unrelated to T5 apex.FusedRMSNorm memory leak." + "reason": "MoE auxiliary-loss normalization and T5 apex RMSNorm memory leak are unrelated." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Different problem class: aux-loss normalization vs version-dependent tokenization changes." - }, - { - "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:43646", - "right": "issue:43749", - "accept": false, - "reason": "Custom model initialization regression is unrelated to FSDP_CPU_RAM_EFFICIENT_LOADING." - }, - { - "left": "issue:43295", - "right": "issue:45356", - "accept": false, - "reason": "Different tokenizer regressions: processor.tokenizer/image handling vs Kimi-K2.5 codec handling and mistral-regex warning." + "reason": "Aux-loss normalization in MoE models and generic tokenization changes across versions are not the same underlying bug." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The output is mostly conservative, but one soft-edge verdict is overstated: issue:45372 and issue:46170 do not look like the same bug or a single mergeable fix. Their titles point to unrelated failures (Gemma 4 processor loading via mistral_common vs Windows/ROCm torch.distributed import/type errors), so marking them as a true duplicate is not grounded in the packet. The rest of the summary is broadly reasonable, but this false positive means the cluster should be rejected." - }, - "evaluator_used": true, - "retried": true, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -8074,7 +8000,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:32090", "issue:33666", @@ -8083,15 +8009,17 @@ "issue:41720", "issue:42898", "issue:43278", + "issue:43295", "issue:43452", "issue:43504", + "issue:43525", "issue:43550", "issue:43577", - "issue:43606", "issue:43643", + "issue:43646", "issue:43653", "issue:43688", - "issue:43720", + "issue:43749", "issue:43756", "issue:43761", "issue:43782", @@ -8105,9 +8033,9 @@ "issue:44568", "issue:44625", "issue:44704", - "issue:44734", "issue:44743", "issue:44977", + "issue:45356", "issue:45405", "issue:45440", "issue:45542", @@ -8115,21 +8043,23 @@ "issue:46178", "issue:46402", "issue:46491", - "issue:46612", "issue:46693", + "issue:46701", "issue:46829", "issue:46832", "issue:46932", "issue:47013", - "issue:47254", - "issue:47276" + "issue:47254" ], "soft_pairs": [ + "issue:43525|issue:46701", + "issue:43646|issue:43749", "issue:45542|issue:46178", - "issue:43653|issue:44568", + "issue:43295|issue:45356", "issue:43756|issue:46402", - "issue:43761|issue:46829", + "issue:43653|issue:44568", "issue:46832|issue:46932", + "issue:43761|issue:46829", "issue:43761|issue:47013", "issue:43653|issue:43927", "issue:33666|issue:36010", @@ -8147,230 +8077,213 @@ "issue:43688|issue:45440", "issue:43504|issue:43550", "issue:44484|issue:44704", - "issue:43782|issue:46693", "issue:43278|issue:43688", - "issue:43643|issue:44561", "issue:45405|issue:47254", - "issue:44734|issue:46612", - "issue:43504|issue:43720", - "issue:44625|issue:47276", - "issue:43720|issue:43927", - "issue:43606|issue:43927" + "issue:43782|issue:46693", + "issue:43643|issue:44561" ], - "prepared_review_unit_hash": "50f99cd14e95ad7d840d37206c52de4d512461c49d79df4419c4be04445679d7", + "prepared_review_unit_hash": "9882fe2ec05e433c7b9f65425e286ec48ff57b383edde344ee639c410e2929cb", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 36711, - "estimated_input_tokens": 9178, - "estimated_eval_tokens": 18612 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36638, + "estimated_input_tokens": 9160, + "estimated_eval_tokens": 18576 }, "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, tokenizer, loading, or training bugs, so the soft links read as broad semantic similarity rather than true duplicates. No pull requests are present.", - "confidence": 0.79, - "canonical_issue_reason": "If a single anchor is required, issue 43761 is the cleanest concrete regression report in the set, but the cluster is too mixed to treat it as a true canonical duplicate target.", + "summary": "This cluster is mostly a grab bag of model-specific regressions and API breakages. The similarity scores seem driven by shared transformer/regression vocabulary rather than true duplicate bugs, so I would not merge any of the soft-edge pairs.", + "confidence": 0.24, + "canonical_issue_reason": "Issue #43646 is the broadest representative of the set because it describes a general v5/custom-model initialization regression, but the overall cluster is too heterogeneous for a strong true canonical.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43761 is the most self-contained, reproducible bug report here and has a clear regression shape, so it is the best representative issue overall.", + "best_issue_reason": "Issue #43646 is the best cluster anchor: it is generic, regression-oriented, and closest to the common theme of user-facing breakage across custom/model-loading paths.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43525", + "right": "issue:46701", + "accept": false, + "reason": "Different model families and failure modes: missing pad_token_id vs a broken image-text pipeline." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Both are regressions, but one is custom model init while the other is FSDP CPU RAM loading." + }, { "left": "issue:45542", "right": "issue:46178", "accept": false, - "reason": "Both are backend/TF-related failures, but they describe different problems: tensorboard-without-TensorFlow backend detection vs an invalid BACKENDS_MAPPING entry for tensorflow_text." + "reason": "Both mention backend/import issues, but one is TensorFlow absence handling and the other is tensorflow_text mapping." }, { - "left": "issue:43653", - "right": "issue:44568", + "left": "issue:43295", + "right": "issue:45356", "accept": false, - "reason": "Both involve tokenizer special-token behavior, but one is BigBird mask-token registration and the other is mdeberta-v3 BOS/EOS handling; not the same bug." + "reason": "Both are tokenizer-related regressions, but they concern unrelated code paths and different symptoms." }, { "left": "issue:43756", "right": "issue:46402", "accept": false, - "reason": "Different model loading issues: Smollm3 RoPE-layer dropping versus Sam3 text encoder weights not loading." + "reason": "Different models and bugs: RoPE layer dropping vs text encoder weights not loading." }, { - "left": "issue:43761", - "right": "issue:46829", + "left": "issue:43653", + "right": "issue:44568", "accept": false, - "reason": "Both concern output fields, but one is CLIPVision hidden_states regression and the other is VideoPrism returning last_hidden_state in the hidden_states slot; not one concrete code-path." + "reason": "Both involve special tokens, but the affected tokenizers and exact failures differ." }, { "left": "issue:46832", "right": "issue:46932", "accept": false, - "reason": "Unrelated failures: RT-DETR feature-level mismatch versus Gemma4 quantized embedder dtype casting." + "reason": "RT-DETR feature-level mismatch and Gemma4 quantized embedder casting are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:46829", + "accept": false, + "reason": "Both touch hidden_states output, but they are different models with different output-contract bugs." }, { "left": "issue:43761", "right": "issue:47013", "accept": false, - "reason": "Both mention missing/incorrect intermediate states, but they affect different model families and different execution paths." + "reason": "Similar tensor/output shape theme, but CLIPVision hidden_states and xLSTM state unpacking are different failures." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "Both are serialization/metadata issues, but one is tokenizer special-token registration and the other is DiaConfig losing custom token IDs after save/load." + "reason": "BigBird special-token registration and DiaConfig token-id persistence are different tokenizer/config bugs." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Multi-GPU training setup versus an import failure for GenerationMixin; no shared bug." + "reason": "Multi-GPU Qwen2-VL training and GenerationMixin import failure are unrelated." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Both touch generation/cache behavior, but one is compiled-forward chunked generation inconsistency and the other is recurrent state reset in modular_qwen3_5." + "reason": "Different code paths: compiled chunked generation vs recurrent-state reset in modular_qwen3_5." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Same Qwen3.5 area, but one is a video-input regression and the other is num_labels propagation; different defects." + "reason": "Video-input regression and num_labels propagation are unrelated Qwen3.5 issues." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "Both are loading-related, but one is gguf_file handling in Auto* APIs and the other is dtype casting for BLIP2 submodules." + "reason": "GGUF loading failures and BLIP2 dtype retention are separate model-loading bugs." }, { "left": "issue:43550", "right": "issue:44361", "accept": false, - "reason": "Different subsystems: torch.compile/SDPA model execution versus tokenizer AttributeError on tasks." + "reason": "torch.compile SDPA failure for Bamba is not the same as MLukeTokenizer task AttributeError." }, { "left": "issue:42898", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer behavior regressions, but one is cleanup-space semantics in v5 and the other is CodeLlama leading-space round-trip loss." + "reason": "Both are tokenizer behavior issues, but one is a global v5 cleanup option change and the other is a CodeLlama round-trip regression." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Both mention training/distributed use, but one is a Trainer broadcast NoneType crash and the other is Qwen2-VL multi-GPU training generally." + "reason": "Trainer GPU broadcast NoneType error and Qwen2-VL multi-GPU training are not the same bug." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Both are save/load serialization bugs, but one is tied lm_head serialization and the other is sinq quantization persistence." + "reason": "Both involve save/load behavior, but tied-weight serialization and sinq quantization persistence are distinct problems." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Device-mapping CUDA assert in Qwen3 is unrelated to compiled-forward chunked generation inconsistency." + "reason": "CUDA assert with auto device mapping is unrelated to compiled-forward chunked generation inconsistencies." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Qwen2-VL multi-GPU training and CustomTrainer.compute_loss signature mismatch are unrelated training issues." + "reason": "Multi-GPU Qwen2-VL training and compute_loss signature mismatch are different Trainer issues." }, { "left": "issue:44977", "right": "issue:46693", "accept": false, - "reason": "Both involve attention/generation, but one is a Qwen3.5 correctness bug and the other is a FA2 performance regression." + "reason": "Both are generation regressions, but flash-attention failure and long-context FA2 performance regression are not the same defect." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Different generation failures: compiled-forward inconsistency versus flash-attention behavior on Qwen3.5." + "reason": "Both affect generation, but compiled-forward inconsistency and flash-attention malfunction have different root causes." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Both are MoE-related, but one is auxiliary-loss normalization and the other is DeepseekV3MoE implementation divergence." + "reason": "Auxiliary-loss normalization and DeepSeekV3MoE implementation divergence are different model/math issues." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "Legacy field loading for BEiT segmentation is unrelated to Bamba torch.compile/SDPA failure." + "reason": "Pretrained preset loading with a legacy field is unrelated to torch.compile SDPA behavior." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "Max shard size default and AutoProcessor kwargs forwarding are unrelated API issues." - }, - { - "left": "issue:43782", - "right": "issue:46693", - "accept": false, - "reason": "Weight-only Qwen3VL loading error is unrelated to FA2 long-context performance regression." + "reason": "max_shard_size semantics and AutoProcessor cached_file kwargs are separate API issues." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "Both mention dtype/behavior regressions in model internals, but one is BF16-to-FP32 embedding drift and the other is MoE auxiliary-loss normalization." - }, - { - "left": "issue:43643", - "right": "issue:44561", - "accept": false, - "reason": "Both involve trust_remote_code and config/import behavior, but one is missing fields from AutoConfig and the other is v5 removal of is_torch_fx_available breaking remote-code models." + "reason": "Embedding dtype drift during eval and auxiliary-loss normalization are unrelated training bugs." }, { "left": "issue:45405", "right": "issue:47254", "accept": false, - "reason": "PEFT version pinning and DeepSpeed checkpointing are related only by dependency name, not by the same bug." - }, - { - "left": "issue:44734", - "right": "issue:46612", - "accept": false, - "reason": "Both are cache-related generation bugs, but one is serve/KV-cache continuation indexing and the other is beam-search cache reorder skipping for specific architectures." - }, - { - "left": "issue:43504", - "right": "issue:43720", - "accept": false, - "reason": "Both concern model loading, but one is a legacy config field and the other is unpacking packed weights during accelerate loading." - }, - { - "left": "issue:44625", - "right": "issue:47276", - "accept": false, - "reason": "Qwen3.5 label propagation and a continuous-batching warning message are unrelated." + "reason": "A version pin not yet on PyPI and a PEFT/DeepSpeed checkpoint error are not duplicates." }, { - "left": "issue:43720", - "right": "issue:43927", + "left": "issue:43782", + "right": "issue:46693", "accept": false, - "reason": "Packed-weight loading and config save/load field loss are different bugs." + "reason": "Qwen3VL weight_only load failure and FA2 long-context performance regression are different failure modes." }, { - "left": "issue:43606", - "right": "issue:43927", + "left": "issue:43643", + "right": "issue:44561", "accept": false, - "reason": "CPU offload device mismatch for bark-small is unrelated to DiaConfig losing custom token IDs." + "reason": "Both involve trust_remote_code, but one is missing config fields and the other is a removed helper import." } ] }, @@ -8392,11 +8305,10 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28218", "issue:33357", - "issue:33666", "issue:39692", "issue:42796", "issue:43278", @@ -8406,8 +8318,8 @@ "issue:43504", "issue:43525", "issue:43572", - "issue:43575", "issue:43577", + "issue:43606", "issue:43611", "issue:43646", "issue:43701", @@ -8416,7 +8328,6 @@ "issue:43844", "issue:43927", "issue:43994", - "issue:44291", "issue:44361", "issue:44466", "issue:44488", @@ -8424,6 +8335,8 @@ "issue:44554", "issue:44568", "issue:44610", + "issue:44625", + "issue:44734", "issue:44843", "issue:44964", "issue:44977", @@ -8439,18 +8352,23 @@ "issue:46209", "issue:46365", "issue:46489", - "issue:46710", - "issue:46846" + "issue:46612", + "issue:46846", + "issue:47276" ], "soft_pairs": [ - "issue:39692|issue:43493", + "issue:44734|issue:46612", + "issue:43504|issue:43720", + "issue:44625|issue:47276", + "issue:43720|issue:43927", + "issue:43606|issue:43927", "issue:43504|issue:44610", "issue:43577|issue:44554", "issue:43611|issue:46846", - "issue:43720|issue:44361", "issue:43525|issue:43572", - "issue:44514|issue:45290", + "issue:43720|issue:44361", "issue:44964|issue:45405", + "issue:44514|issue:45290", "issue:43493|issue:43825", "issue:28218|issue:33357", "issue:44977|issue:46209", @@ -8458,165 +8376,191 @@ "issue:45491|issue:46002", "issue:44554|issue:46365", "issue:39692|issue:43295", - "issue:44488|issue:44843", "issue:43994|issue:46192", + "issue:44488|issue:44843", "issue:42796|issue:44977", "issue:43334|issue:45759", - "issue:43646|issue:44466", "issue:43278|issue:43844", - "issue:43278|issue:43701", + "issue:43646|issue:44466", "issue:43334|issue:43572", + "issue:43278|issue:43701", "issue:44568|issue:46489", - "issue:43493|issue:43844", "issue:44568|issue:45812", "issue:43927|issue:44361", - "issue:46209|issue:46710", - "issue:43575|issue:45405", - "issue:43572|issue:44291", - "issue:33357|issue:33666" + "issue:43493|issue:43844" ], - "prepared_review_unit_hash": "f649ab9a44c5ae76ec1addcef5c27903e0ee7cabb74ca14a65bab135030970dd", + "prepared_review_unit_hash": "809b7fc3b6207bdc562f5789afadc309f6ee0d82835e28d3cb926375c2efd769", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36728, - "estimated_input_tokens": 9182, - "estimated_eval_tokens": 18620 + "serialized_chars": 37321, + "estimated_input_tokens": 9331, + "estimated_eval_tokens": 18918 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the soft pairs share broad themes like tokenizers, configs, or loading failures, but they do not look like the same underlying bug/change. I would reject all soft edges and treat the items as separate issues.", - "confidence": 0.87, - "canonical_issue_reason": "Issue 45812 is the broadest, highest-signal regression here: it reports a silent v4\u2192v5 tokenizer break affecting all Granite models and has multiple inbound references, making it the most representative issue if one must be chosen.", + "summary": "The cluster is thematically broad and the soft pairs mostly share only subsystem-level similarity (tokenizers, configs, multimodal loading, generation, dtype/loading regressions). I don\u2019t see any true duplicate pairs here, and there are no PRs in the set.", + "confidence": 0.95, + "canonical_issue_reason": "Issue 45812 is the best representative anchor: it describes a concrete, high-impact v4\u2192v5 tokenizer regression with clear user-visible wrong outputs and strong reporting context, making it the most canonical issue in this mixed set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45812 is the strongest standalone representative because it describes a clearly scoped, user-facing regression with broad impact and strong evidence compared with the more narrow or model-specific reports.", + "best_issue_reason": "Issue 45812 is the strongest overall triage candidate because it is specific, reproducible, and clearly tied to a concrete code-path regression rather than a vague platform symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:39692", - "right": "issue:43493", + "left": "issue:44734", + "right": "issue:46612", + "accept": false, + "reason": "Both involve generation/cache behavior, but 44734 is a serve KV-continuation tensor-indexing crash while 46612 is a beam-search cache reorder omission for specific model classes. Different code paths and fixes." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "43504 is a Beit legacy-field load failure; 43720 is BitNet packed-weight unpacking during accelerate loading. Different models and root causes." + }, + { + "left": "issue:44625", + "right": "issue:47276", + "accept": false, + "reason": "44625 is missing num_labels propagation in Qwen3.5 configs; 47276 is a misleading warning message in continuous batching. No shared bug." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "BitNet packed-weight loading vs DiaConfig save/load losing custom token IDs. Same broad loading theme, but unrelated failures." + }, + { + "left": "issue:43606", + "right": "issue:43927", "accept": false, - "reason": "Both are about SigLIP2, but one is a broken docs example and the other is an implementation mismatch with JAX; different bugs." + "reason": "CPU offload device mismatch for bark-small vs config token-ID loss on save/load. Different subsystems and symptoms." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "BEiT legacy-field loading failure vs OmDet-Turbo image-size mismatch; unrelated code paths." + "reason": "Beit legacy-field loading and OmDet-Turbo processor image-size mismatch are unrelated bugs despite both being model-loading issues." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "BLIP2 dtype propagation issue and MPS attention correctness bug are different problems." + "reason": "Blip2 dtype not propagating and an MPS attention correctness bug are different failures in different execution paths." }, { "left": "issue:43611", "right": "issue:46846", "accept": false, - "reason": "Base-model-prefix loading regression is unrelated to Qwen3.5 tensor-parallel plan OOM/shape errors." - }, - { - "left": "issue:43720", - "right": "issue:44361", - "accept": false, - "reason": "BitNet packed-weight loading failure and MLukeTokenizer task AttributeError are unrelated." + "reason": "Base-model-prefix loading regression vs Qwen3.5 tensor-parallel plan missing linear_attn entries. Different APIs and failure modes." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Both involve missing token ID attributes, but they affect different configs and fields; not the same bug." + "reason": "Both are missing token-id-style config fields, but they affect different model configs and different attributes (pad_token_id vs pad_token_idx). Not the same bug." }, { - "left": "issue:44514", - "right": "issue:45290", + "left": "issue:43720", + "right": "issue:44361", "accept": false, - "reason": "Different apply_chat_template crashes with different triggers and inputs." + "reason": "Packed-weight unpacking in BitNet and MLukeTokenizer task AttributeError are unrelated." }, { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "Model-loading regression and an unreleased PEFT version bump are unrelated." + "reason": "Phi-4 multimodal load failure and an unreleased PEFT version bump are not the same underlying issue." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both crash in apply_chat_template, but one is batched input with padding=False and the other is assistant tool-call messages with no content. Different triggers and likely different fixes." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 implementation discrepancy and pipeline translation-message regression do not share a concrete code path." + "reason": "SigLIP2 HF-vs-JAX discrepancy is unrelated to a pipeline warning about translation support in v5." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer spacing bug vs MacOS bus error in CLIP loading; unrelated." + "reason": "Tokenizer added-space behavior and a MacOS bus error in CLIP loading are unrelated." }, { "left": "issue:44977", "right": "issue:46209", "accept": false, - "reason": "Flash-attention generation bug and FP8 B200 inference failure are distinct issues." + "reason": "Qwen3.5 flash-attention generation failure and FP8 inference failure on B200 are different hardware/model-path issues." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Device-map forward mismatch vs quantized save/load corruption; different failure modes." + "reason": "Gemma4 device-map mismatch and sinq quantization save/load breakage are unrelated regressions." }, { "left": "issue:45491", "right": "issue:46002", "accept": false, - "reason": "Both mention NaNs, but one is a sliding-window attention edge case and the other is a checkout/init classification-head issue." + "reason": "Both mention NaNs, but one is Gemma3 mixed-length sliding-window attention and the other is classification-head init/checkout. Different code paths and causes." }, { "left": "issue:44554", "right": "issue:46365", "accept": false, - "reason": "MPS attention correctness bug and Trainer import dependency issue are unrelated." + "reason": "An MPS attention correctness issue is unrelated to importing Trainer being blocked by datasets." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "Docs-example errors and processor/tokenizer API breakage are different regressions." + "reason": "SigLIP2 docs example errors/quantization failure are not the same as the processor.tokenizer/image-passing regression." }, { - "left": "issue:44488", - "right": "issue:44843", + "left": "issue:43994", + "right": "issue:46192", "accept": false, - "reason": "Model-loading failure for a specific checkpoint vs unconditional model_info call breaking offline mode." + "reason": "Wrong SigLIP2 outputs with AutoModel/pipeline and return_tensors being ignored with text_kwargs are different bugs." }, { - "left": "issue:43994", - "right": "issue:46192", + "left": "issue:44488", + "right": "issue:44843", "accept": false, - "reason": "SigLIP2 incorrect inference outputs and return_tensors being ignored are different tokenizer/pipeline problems." + "reason": "Both are tokenizer-loading related, but one is a specific model load failure and the other is an offline-mode regression in _patch_mistral_regex. Not the same issue." }, { "left": "issue:42796", "right": "issue:44977", "accept": false, - "reason": "Tokenizer BPE-marker handling and flash-attention generation failure are not the same underlying bug." + "reason": "Ministral BPE-marker handling and Qwen3.5 flash-attention generation are unrelated." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_id on Qwen3-VL and text_config unwrapping for composite Qwen configs are separate loading issues." + "reason": "Both involve Qwen multimodal configs, but one is a missing pad_token_id on text config and the other is failure to unwrap text_config in AutoModelForCausalLM.from_config. Different concrete bugs." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype drift in evaluate and gradient growth under ZeRO-3 are distinct training/runtime problems." }, { "left": "issue:43646", @@ -8625,58 +8569,40 @@ "reason": "Custom model initialization breakage and inconsistent lm_head.weight serialization are different v5 regressions." }, { - "left": "issue:43278", - "right": "issue:43844", + "left": "issue:43334", + "right": "issue:43572", "accept": false, - "reason": "Embedding dtype mismatch during evaluation and gradient explosion under ZeRO-3 are unrelated." + "reason": "Both are missing config token fields, but they affect different model families and different attribute names; not a duplicate." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "Embedding dtype regression and resume_from_checkpoint key mismatch are separate problems." + "reason": "Dtype mismatch in evaluation and resume_from_checkpoint key mismatch are unrelated." }, { - "left": "issue:43334", - "right": "issue:43572", + "left": "issue:44568", + "right": "issue:46489", "accept": false, - "reason": "Different model families and different missing config fields; not a duplicate." + "reason": "Both are tokenizer regressions, but one is add_special_tokens not adding BOS/EOS for mdeberta-v3 and the other is DeepSeek-Coder v1 BPE-marker output differences. Different concrete failures." }, { "left": "issue:44568", - "right": "issue:46489", + "right": "issue:45812", "accept": false, - "reason": "mdeberta-v3 special-token omission and DeepSeek-Coder tokenizer regression are both tokenizer issues but not the same concrete bug." + "reason": "Both are tokenizer regressions in v5, but they affect different tokenizers and different behaviors: special-token insertion vs wrong token IDs for Granite." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "DiaConfig save/load token ID loss and MLukeTokenizer task AttributeError are unrelated." - }, - { - "left": "issue:46209", - "right": "issue:46710", - "accept": false, - "reason": "FP8 inference on B200 and accuracy regression after a version upgrade are different issues." + "reason": "DiaConfig save/load custom token-ID loss and MLukeTokenizer task AttributeError are unrelated." }, { - "left": "issue:43575", - "right": "issue:45405", - "accept": false, - "reason": "Tensor-parallel OOM and PEFT version bump are unrelated." - }, - { - "left": "issue:43572", - "right": "issue:44291", - "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and unexpected _is_hf_initialized argument are different config/load regressions." - }, - { - "left": "issue:33357", - "right": "issue:33666", + "left": "issue:43493", + "right": "issue:43844", "accept": false, - "reason": "MacOS bus error in CLIP loading and Qwen2-VL multi-GPU training are unrelated." + "reason": "SigLIP2 implementation discrepancy vs a gradient anomaly under HfDeepSpeedConfig/ZeRO-3 are not the same underlying bug." } ] }, @@ -8698,44 +8624,44 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:15354", "issue:28218", "issue:29159", "issue:32090", + "issue:33357", + "issue:33666", "issue:34567", "issue:38175", "issue:42490", "issue:42947", "issue:43381", - "issue:43493", "issue:43504", + "issue:43572", + "issue:43575", "issue:43606", "issue:43653", "issue:43696", - "issue:43701", "issue:43824", "issue:43844", "issue:43873", "issue:43994", "issue:44112", - "issue:44423", + "issue:44291", "issue:44464", "issue:44493", "issue:44589", - "issue:44734", "issue:44849", "issue:44964", "issue:45072", "issue:45084", - "issue:45200", "issue:45229", "issue:45245", "issue:45357", + "issue:45405", "issue:45520", "issue:45584", - "issue:45588", "issue:45823", "issue:46050", "issue:46209", @@ -8750,6 +8676,10 @@ "issue:47254" ], "soft_pairs": [ + "issue:43572|issue:44291", + "issue:43575|issue:45405", + "issue:46209|issue:46710", + "issue:33357|issue:33666", "issue:43824|issue:44589", "issue:44589|issue:45520", "issue:43504|issue:43606", @@ -8760,228 +8690,224 @@ "issue:45245|issue:46710", "issue:43504|issue:43653", "issue:43606|issue:44112", - "issue:43994|issue:44493", "issue:45072|issue:46832", + "issue:43994|issue:44493", "issue:44464|issue:45584", - "issue:43873|issue:46050", "issue:42490|issue:43696", + "issue:43873|issue:46050", "issue:15354|issue:38175", "issue:43844|issue:44964", "issue:44849|issue:47145", "issue:42947|issue:43381", "issue:46399|issue:46846", - "issue:29159|issue:32090", "issue:47013|issue:47254", + "issue:29159|issue:32090", "issue:45229|issue:45823", "issue:45084|issue:45520", - "issue:28218|issue:34567", - "issue:45200|issue:45229", - "issue:45588|issue:46050", - "issue:44423|issue:44734", - "issue:43493|issue:43701" + "issue:28218|issue:34567" ], - "prepared_review_unit_hash": "9540792547a910ab8af758f9b409ac3b1f744e9b3f8f3783d1fcdb462bd741ca", + "prepared_review_unit_hash": "bb8889c19e7ca70f509b048cc3d1eb9938078535a66eb7a923e1b327dd622122", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36537, - "estimated_input_tokens": 9135, - "estimated_eval_tokens": 18526 + "serialized_chars": 36405, + "estimated_input_tokens": 9102, + "estimated_eval_tokens": 18460 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose collection of unrelated issue reports; the soft pairs are mostly lexical near-matches rather than true duplicates.", - "confidence": 0.97, - "canonical_issue_reason": "There is no real duplicate cluster here. If one issue must anchor the set, issue 43824 is the most self-contained and well-scoped report, but it is not a duplicate of the others.", + "summary": "No soft edge looks like a true duplicate or mergeable PR match; the candidate pairs mostly share only broad subsystem similarity (tokenizer, training, loading, OOM, etc.) while describing different concrete failures. This cluster appears heterogeneous rather than a single duplicate set.", + "confidence": 0.79, + "canonical_issue_reason": "No clear canonical issue across the set: the items span many unrelated bugs/regressions, so none safely subsumes the rest.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43824 is the best anchor only because it is a concrete, reproducible bug report with some discussion/references; the rest of the items span unrelated models and failure modes.", + "best_issue_reason": "issue:29159 is the broadest tokenizer-focused report and the most plausible representative if one issue must be chosen, but it still does not explain the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Both are config/loading failures, but one is a missing `pad_token_idx` issue and the other is an unexpected `_is_hf_initialized` argument with `init_empty_weights`; different code paths." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Both mention version-related loading problems, but one is TP/OOM on a specific Qwen model and the other is an unmet PEFT minimum-version pin." + }, + { + "left": "issue:46209", + "right": "issue:46710", + "accept": false, + "reason": "Both are model-quality/performance complaints, but one is FP8 inference support on B200 and the other is an accuracy regression after a Transformers upgrade." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "One is a MacOS bus error with a community CLIP model; the other is a Qwen2-VL multi-GPU training question. Same broad area, different problem." + }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "Different bugs: missing Qwen2.5-VL import vs FP8 storage-object lookup failure." + "reason": "Both are import/runtime type failures, but they refer to different missing symbols/storage types and different affected code paths." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Unrelated root causes: storage registration error vs flash_attn import KeyError." + "reason": "These are unrelated import/type errors: missing Float8 storage object versus a `flash_attn` KeyError on Python 3.13." }, { "left": "issue:43504", "right": "issue:43606", "accept": false, - "reason": "BEiT legacy-field loading issue is unrelated to Bark CPU offload device mismatch." + "reason": "Both involve model loading/CI, but one is a legacy field in BEiT semantic segmentation and the other is a device mismatch with CPU offload on bark-small." }, { "left": "issue:46821", "right": "issue:46832", "accept": false, - "reason": "TimesFM window_size attribute error and RT-DETR feature-level mismatch are different code paths." + "reason": "Both are model forward/runtime errors, but one is a TimesFM `window_size` attribute issue and the other is RT-DETR feature-level mismatch." }, { "left": "issue:44849", "right": "issue:46209", "accept": false, - "reason": "Qwen3.5 hidden-states bug is unrelated to FP8 inference support on B200." + "reason": "One is a Qwen3.5 hidden-state output bug; the other is an FP8 inference failure. No shared concrete code path." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Runtime hidden-states issue vs save_pretrained visual-key regression are distinct problems." + "reason": "Both concern Qwen3.5, but one is `output_hidden_states=True` and the other is incorrect visual-encoder keys in `save_pretrained`; different failures." }, { "left": "issue:44493", "right": "issue:46285", "accept": false, - "reason": "Generic position-id key warning is not the same as the CLIP text-model structural API change." + "reason": "Both mention model structure/key mismatches, but one is a generic position-id key warning and the other is a CLIP class structure inconsistency." }, { "left": "issue:45245", "right": "issue:46710", "accept": false, - "reason": "Category-limit runtime error and model accuracy regression are different failures." + "reason": "A category-count limit error is not the same as an accuracy regression; these are different failure modes." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "BEiT pretrained preset loading and BigBirdTokenizer special-token decode are unrelated." + "reason": "One is a BEiT legacy-field load failure; the other is BigBirdTokenizer mask-token registration/empty decode behavior." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "Device-mismatch bug in Bark offload is not the same as a stale GraniteSpeech CI test." + "reason": "Both are CI/device related, but one is CPU offload mismatch in model loading and the other is a stale test failure. Not the same bug." }, { - "left": "issue:43994", - "right": "issue:44493", + "left": "issue:45072", + "right": "issue:46832", "accept": false, - "reason": "SigLIP2 nonsensical outputs are not the same as a broad unexpected-key warning." + "reason": "Different model families and different dtype/forward-shape problems; too broad to be one underlying bug." }, { - "left": "issue:45072", - "right": "issue:46832", + "left": "issue:43994", + "right": "issue:44493", "accept": false, - "reason": "Different models and failures: bfloat16 dtype mismatch vs RT-DETR forward type error." + "reason": "One is a SigLIP2 inference quality issue; the other is an unexpected key warning around position ids. No concrete duplicate evidence." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is unrelated to Whisper empty-transcription handling." + "reason": "Chunked generation with compiled forward is unrelated to Whisper empty transcription after `align_special_tokens`." }, { - "left": "issue:43873", - "right": "issue:46050", + "left": "issue:42490", + "right": "issue:43696", "accept": false, - "reason": "Quantization offloading behavior and sinq save/load corruption are different bugs." + "reason": "Local-path import warnings and CUDA OOM on GPT-oss-20b are unrelated." }, { - "left": "issue:42490", - "right": "issue:43696", + "left": "issue:43873", + "right": "issue:46050", "accept": false, - "reason": "Import warning on path handling is unrelated to CUDA out-of-memory." + "reason": "Quantization/offloading concerns do not make these the same issue: one is offloading not working as expected, the other is broken save/load after sinq quantization." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "torch.jit.script GeneratorExp support is unrelated to SigLIP2 zero-probability outputs." + "reason": "A TorchScript `GeneratorExp` scripting limitation and unexpected zero probabilities in SigLIP2 are unrelated." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "DeepSpeed gradient instability does not match Phi-4 multimodal loading failure." + "reason": "Both are model execution failures, but one is gradient explosion with DeepSpeed/ZeRO-3 and the other is a model-loading import failure for Phi-4 multimodal." }, { "left": "issue:44849", "right": "issue:47145", "accept": false, - "reason": "Qwen3.5 hidden-states bug and FP8 weight_scale omission are different regressions." + "reason": "Qwen3.5 hidden-state output bug versus FP8 checkpoint loading missing `weight_scale`; different subsystems and symptoms." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "LoRA gradient-checkpointing ineffectiveness is distinct from checkpointing being invalid in eval mode." + "reason": "Both mention gradient checkpointing, but one is ineffective with PEFT LoRA and the other says it cannot be used in eval mode; not the same concrete bug." }, { "left": "issue:46399", "right": "issue:46846", "accept": false, - "reason": "Qwen3.5 rotary-registration crash and missing tp-plan entries are separate issues." + "reason": "Both are Qwen3.5-related, but one is a missing decorator for `apply_rotary_pos_emb` registration and the other is an incomplete TP plan causing OOM/shape errors." }, { - "left": "issue:29159", - "right": "issue:32090", + "left": "issue:47013", + "right": "issue:47254", "accept": false, - "reason": "Tokenizer slow/fast inconsistency is unrelated to Trainer _gpu_broadcast_one NoneType errors." + "reason": "Both involve checkpointing/state handling, but one is an xLSTM train-mode unpacking crash and the other is a PEFT + DeepSpeed ZeRO-3 checkpoint error." }, { - "left": "issue:47013", - "right": "issue:47254", + "left": "issue:29159", + "right": "issue:32090", "accept": false, - "reason": "xLSTM return_last_states unpacking crash and PEFT/ZeRO-3 checkpoint errors are different code paths." + "reason": "Tokenizer slow/fast inconsistency is unrelated to a Trainer `_gpu_broadcast_one` NoneType type error." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Gemma4 multi-GPU OOM is not the same as a device-map forward mismatch in PLE." + "reason": "Both are large-model inference/multi-device problems, but one is CUDA OOM for Gemma4 and the other is a device mismatch in Gemma4 PLE with `device_map=\"auto\"`." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Template-node compilation failure and flash_attn import mapping error are unrelated." + "reason": "Compilation of non-template nodes and a `flash_attn` import KeyError are different failure classes." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Extra-space tokenizer behavior does not duplicate a TrainerState token counter bug." - }, - { - "left": "issue:45200", - "right": "issue:45229", - "accept": false, - "reason": "Missing mm_token_type_ids default for fine-tuning is unrelated to Gemma4 multi-GPU OOM." - }, - { - "left": "issue:45588", - "right": "issue:46050", - "accept": false, - "reason": "flash_attention s_aux=None crash and sinq persistence corruption affect different subsystems." - }, - { - "left": "issue:44423", - "right": "issue:44734", - "accept": false, - "reason": "Continuous-batching multimodal crash is separate from KV-cache continuation indexing." - }, - { - "left": "issue:43493", - "right": "issue:43701", - "accept": false, - "reason": "SigLIP2 HF/JAX discrepancy is unrelated to resume_from_checkpoint key mismatch." + "reason": "Tokenizer added-space behavior and TrainerState token-count tracking are unrelated issues." } ] }, @@ -9003,7 +8929,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:29870", "issue:29942", @@ -9015,35 +8941,36 @@ "issue:42915", "issue:43065", "issue:43295", - "issue:43334", "issue:43450", "issue:43493", - "issue:43525", "issue:43540", "issue:43550", "issue:43572", "issue:43575", + "issue:43701", "issue:43824", "issue:43844", "issue:43873", "issue:43937", "issue:43957", "issue:44186", - "issue:44521", + "issue:44423", "issue:44554", "issue:44610", + "issue:44734", "issue:44869", "issue:44977", "issue:45072", "issue:45198", "issue:45200", - "issue:45290", + "issue:45229", "issue:45406", "issue:45446", "issue:45520", "issue:45588", "issue:45759", "issue:45823", + "issue:46050", "issue:46121", "issue:46178", "issue:46192", @@ -9054,11 +8981,15 @@ "issue:46821" ], "soft_pairs": [ + "issue:45200|issue:45229", + "issue:45588|issue:46050", + "issue:44423|issue:44734", + "issue:43493|issue:43701", "issue:43937|issue:46701", "issue:41720|issue:46710", - "issue:44869|issue:46752", "issue:44554|issue:45446", "issue:37428|issue:43957", + "issue:44869|issue:46752", "issue:43065|issue:46369", "issue:43295|issue:43450", "issue:44977|issue:46701", @@ -9067,14 +8998,14 @@ "issue:45072|issue:45198", "issue:45588|issue:45759", "issue:45406|issue:46121", - "issue:43295|issue:43873", "issue:43572|issue:46821", + "issue:43295|issue:43873", "issue:43540|issue:45200", - "issue:45823|issue:46192", "issue:44186|issue:46192", + "issue:45823|issue:46192", "issue:43295|issue:44610", - "issue:43450|issue:43575", "issue:43572|issue:43957", + "issue:43450|issue:43575", "issue:43550|issue:43844", "issue:37428|issue:43824", "issue:45520|issue:46178", @@ -9083,230 +9014,240 @@ "issue:42490|issue:43493", "issue:43295|issue:43493", "issue:39692|issue:46192", - "issue:42915|issue:43824", - "issue:44521|issue:45290", - "issue:43334|issue:43525" + "issue:42915|issue:43824" ], - "prepared_review_unit_hash": "e0f8c30388d8d8c4afbe70044c85d9742595791781e330beff99f656817c85db", + "prepared_review_unit_hash": "86d9086a7c49d5de2b1ba484ebf91d289352ea872b98193318690fbb3454d05d", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36827, - "estimated_input_tokens": 9207, - "estimated_eval_tokens": 18670 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37960, + "estimated_input_tokens": 9490, + "estimated_eval_tokens": 19236 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These soft links are mostly false positives: the titles describe different bugs, regressions, or model-specific failures rather than the same underlying change. No PRs are present. If a single issue must anchor the set, issue 43873 is the most active/reference-heavy, but this is not a true duplicate cluster.", - "confidence": 0.93, - "canonical_issue_reason": "Issue 43873 is the strongest anchor by engagement and inbound references, but the overall set is too heterogeneous to treat as one duplicate cluster.", + "summary": "This cluster is heterogeneous and the soft-similarity pairs do not look like true duplicates. I found no safe issue-to-issue merges, so the cluster should be split rather than consolidated.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43295 is the most representative anchor because it describes a concrete regression in processor/tokenizer behavior with broader impact than the other items, which are mostly unrelated model-, backend-, or pipeline-specific bugs.", "canonical_pr_reason": null, - "best_issue_reason": "43873 is the best single issue to keep as an anchor if forced, because it has the most discussion/reference signal; otherwise these should be split by topic.", + "best_issue_reason": "issue:43295 is the best single issue to keep as the cluster representative, but only as an anchor for triage; it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Different Gemma4 problems: text-only fine-tuning defaults vs multi-GPU inference OOM." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Different failures: flash-attention sinkless-model crash vs sinq save/load breakage." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve/runtime crashes, but they hit different code paths and symptoms." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "Unrelated: SigLIP2 implementation discrepancy vs checkpoint resume key mismatch." + }, { "left": "issue:43937", "right": "issue:46701", "accept": false, - "reason": "Different failures: GLM generation config validation vs a DiffusionGemma pipeline output issue." + "reason": "Different models and issues: invalid GenerationConfig vs image-text pipeline failure." }, { "left": "issue:41720", "right": "issue:46710", "accept": false, - "reason": "Unrelated problems: Qwen3 auto device mapping CUDA assert vs a DeepSeek accuracy regression after upgrade." - }, - { - "left": "issue:44869", - "right": "issue:46752", - "accept": false, - "reason": "Whisper timestamp decoding crash is unrelated to apply_chat_template on an empty list." + "reason": "Different model-specific regressions: A800 auto-device-map crash vs DeepSeek accuracy regression." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "Both touch attention code, but one is an MPS correctness bug and the other is a PyTorch version check import bug." + "reason": "Different backend bugs: MPS attention correctness vs flex_attention import version check." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Different code paths: flash-attention import error vs meta-device model loading breakage." + "reason": "Both are loading/import issues, but they involve different APIs and failure modes." + }, + { + "left": "issue:44869", + "right": "issue:46752", + "accept": false, + "reason": "Different text-processing crashes: Whisper timestamp decoding vs empty chat template input." }, { "left": "issue:43065", "right": "issue:46369", "accept": false, - "reason": "Same model family, but dummy Conv2d setup and memory overflow are different concrete bugs." + "reason": "Different SAM3 issues: dummy layer implementation vs memory overflow in a text variant." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Both are processor issues, but one is tokenizer/image handling and the other is batched video shape handling." + "reason": "Both involve processors, but one is a tokenizer regression and the other is batched video shape handling." }, { "left": "issue:44977", "right": "issue:46701", "accept": false, - "reason": "Qwen flash-attention generation failure is unrelated to the DiffusionGemma pipeline issue." + "reason": "Different multimodal failures affecting different models and pipelines." }, { "left": "issue:29942", "right": "issue:30333", "accept": false, - "reason": "FlashAttention test failures are unrelated to MLflow reporting after training failure." + "reason": "No overlap: Flash Attention 2 test failures vs MLFlow reporting on training failure." }, { "left": "issue:44977", "right": "issue:46710", "accept": false, - "reason": "Different models and symptoms: flash-attention generation bug vs accuracy regression." + "reason": "Different regression targets: flash-attention generation bug vs accuracy drop after version upgrade." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "bfloat16 dtype mismatches are unrelated to Wav2Vec2 save_pretrained/tokenization failures." + "reason": "Different subsystem bugs: dtype mismatch inference vs save/tokenization failure." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Flash-attention sink-less model crash is unrelated to composite Qwen config unwrapping." + "reason": "Different root causes: flash_attention sink handling vs multimodal config unwrapping." }, { "left": "issue:45406", "right": "issue:46121", "accept": false, - "reason": "Gemma4Processor _tokenizer crash and rope-validation list TypeError are separate bugs." + "reason": "Different areas: Gemma4 processor private tokenizer attr vs RoPE validation list handling." }, { - "left": "issue:43295", - "right": "issue:43873", + "left": "issue:43572", + "right": "issue:46821", "accept": false, - "reason": "Processor/tokenizer regression is unrelated to offloading with quantization." + "reason": "Different config/argument bugs in unrelated models and code paths." }, { - "left": "issue:43572", - "right": "issue:46821", + "left": "issue:43295", + "right": "issue:43873", "accept": false, - "reason": "Both are missing-attribute reports, but for different configs and different fields." + "reason": "Processor/tokenizer regression is unrelated to quantization offloading behavior." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Qwen3OmniMoe video processing ValueError is unrelated to Gemma 4 mm_token_type_ids defaults." + "reason": "Different multimodal inputs: Qwen3Omni video processing vs Gemma4 mm_token_type_ids defaulting." }, { - "left": "issue:45823", + "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "Device-mismatch forward bug is unrelated to return_tensors being ignored." + "reason": "Different tokenizer behaviors: NER/batched padding crash vs return_tensors precedence bug." }, { - "left": "issue:44186", + "left": "issue:45823", "right": "issue:46192", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash is unrelated to return_tensors handling." + "reason": "Device-mapping mismatch is unrelated to return_tensors being ignored with text_kwargs." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Different processor regressions: tokenizer/image access vs wrong processor output size." + "reason": "Both are processor-related, but they concern different models and incompatible input-shape bugs." }, { - "left": "issue:43450", - "right": "issue:43575", + "left": "issue:43572", + "right": "issue:43957", "accept": false, - "reason": "Batched video shape bug is unrelated to tensor-parallel loading OOM." + "reason": "Unrelated config issues: missing pad_token_idx vs meta-device loading breakage." }, { - "left": "issue:43572", - "right": "issue:43957", + "left": "issue:43450", + "right": "issue:43575", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is unrelated to meta-device loading failures." + "reason": "Video processor batch-shape bug is unrelated to tensor-parallel OOM loading." }, { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "torch.compile SDPA failure and ZeRO-3 gradient explosion are different training/runtime bugs." + "reason": "Different training/inference problems: torch.compile SDPA failure vs ZeRO-3 gradient growth." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "Different import errors for different symbols; not the same underlying failure." + "reason": "Both are import errors, but they involve different symbols and likely different fixes." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "flash_attn lookup on Python 3.13 is unrelated to tensorflow_text backend mapping." + "reason": "Different backend-import problems with no shared code path." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Defaulting mm_token_type_ids and device_map auto forward mismatch are separate Gemma 4 issues." + "reason": "Same model family, but separate bugs: missing mm_token_type_ids default vs PLE device mismatch." }, { "left": "issue:29870", "right": "issue:29942", "accept": false, - "reason": "Bart generate UnboundLocalError is unrelated to FlashAttention 2 test failures." + "reason": "Unrelated issues: Bart generate() UnboundLocalError vs Flash Attention test failures." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "Trailing-slash import warning is unrelated to SigLIP2 implementation discrepancy." + "reason": "Local-path trailing slash warning is unrelated to SigLIP2 implementation discrepancy." }, { "left": "issue:43295", "right": "issue:43493", "accept": false, - "reason": "Custom processor.tokenizer regression is unrelated to SigLIP2 JAX/HF discrepancy." + "reason": "Processor/tokenizer regression and SigLIP2 implementation mismatch are different problems." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 documentation/example issues are unrelated to return_tensors handling." + "reason": "SigLIP2 documentation errors are unrelated to the return_tensors precedence bug." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "FineGrainedFP8Config failure is unrelated to a Qwen2.5-VL import error." - }, - { - "left": "issue:44521", - "right": "issue:45290", - "accept": false, - "reason": "Both involve apply_chat_template, but one is assistant_masks becoming zero and the other is a crash on tool-call messages." - }, - { - "left": "issue:43334", - "right": "issue:43525", - "accept": false, - "reason": "Both mention pad_token_id, but they are different model configs and not clearly the same bug." + "reason": "Qwen3Moe FP8 failure and Qwen2.5-VL import error are different model-specific issues." } ] }, @@ -9328,7 +9269,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28218", "issue:35141", @@ -9336,40 +9277,39 @@ "issue:41720", "issue:43262", "issue:43278", + "issue:43334", "issue:43335", "issue:43425", "issue:43450", "issue:43452", - "issue:43526", + "issue:43525", "issue:43550", "issue:43606", "issue:43653", "issue:43819", "issue:43824", - "issue:43827", "issue:43856", - "issue:43901", "issue:43950", "issue:43975", "issue:44112", "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", @@ -9380,21 +9320,23 @@ "issue:46932" ], "soft_pairs": [ + "issue:43334|issue:43525", "issue:43278|issue:43856", "issue:28218|issue:35141", - "issue:45381|issue:46209", - "issue:43824|issue:43975", + "issue:44521|issue:45290", "issue:44466|issue:45127", + "issue:43824|issue:43975", + "issue:45381|issue:46209", "issue:44737|issue:46894", "issue:43262|issue:43450", "issue:43975|issue:45520", "issue:43653|issue:44112", - "issue:43950|issue:46285", "issue:43452|issue:46512", - "issue:44442|issue:45198", + "issue:43950|issue:46285", "issue:44561|issue:46234", + "issue:44442|issue:45198", "issue:45200|issue:46899", - "issue:43526|issue:46192", + "issue:44479|issue:46192", "issue:43819|issue:44112", "issue:44898|issue:45412", "issue:43335|issue:43550", @@ -9404,214 +9346,198 @@ "issue:44610|issue:44857", "issue:43819|issue:45083", "issue:37428|issue:43425", - "issue:43606|issue:45083", - "issue:43827|issue:43901", - "issue:43335|issue:43653", - "issue:44792|issue:45563", - "issue:44857|issue:45593" + "issue:43606|issue:45083" ], - "prepared_review_unit_hash": "05784e981f70bf796c6f825158df220f9704cb935b544df53eeacfc2493fcb4d", + "prepared_review_unit_hash": "01ffcfd6f2f636907eb96d73781880afc2b4d92103f99de8201b569aaa901885", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36745, - "estimated_input_tokens": 9187, - "estimated_eval_tokens": 18630 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35639, + "estimated_input_tokens": 8910, + "estimated_eval_tokens": 18076 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This looks like a loose similarity bucket of mostly unrelated Transformers bug reports. I don\u2019t see any soft pair that is the same underlying bug/change, so all soft edges are rejected. The most central issue is 46894, but only as a representative anchor, not as a duplicate match.", - "confidence": 0.96, - "canonical_issue_reason": "issue:46894 is the best cluster anchor by discussion/inbound-reference strength and has a concrete, well-scoped bug report.", + "summary": "This cluster is very heterogeneous: the soft pairs mostly share a subsystem name or a vague regression theme, but they describe different code paths and different fixes. I would not merge any of the proposed soft edges.", + "confidence": 0.33, + "canonical_issue_reason": "Best representative is issue:44521 because it is open, has the most engagement/inbound refs, and targets a central user-facing API (`apply_chat_template`) with broad multimodal impact. That said, the cluster is not a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46894 is the strongest overall issue candidate in this set, but it is not a duplicate of the others.", + "best_issue_reason": "Issue:44521 is the strongest single issue to anchor the cluster: active, high-discussion, and on a shared API path that can attract related fixes. It is more representative than the narrower model-specific or regression-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both are missing `pad_token_id`, but they affect different configs/models (`Qwen3VLTextConfig` vs `Llama4Config`) and likely need separate fixes." + }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Different failures: dtype drift in embedding layers vs Qwen3 MoE training memory usage." + "reason": "Different bugs: dtype mismatch in evaluation vs memory inefficiency in Qwen3 MoE training." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer spacing after added tokens is unrelated to embedding resize/post_init reinitialization." + "reason": "Tokenizer spacing after added tokens is unrelated to embedding reinitialization during `post_init`." }, { - "left": "issue:45381", - "right": "issue:46209", + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both touch `apply_chat_template`, but one is about multimodal assistant masks and the other about tool-call messages with no content; different failure modes and fixes." + }, + { + "left": "issue:44466", + "right": "issue:45127", "accept": false, - "reason": "Qwen2.5-VL rope/position-id bug is unrelated to FP8 inference failure on B200." + "reason": "Serialization inconsistency of `lm_head.weight` is distinct from LoRA merge collapse with extended vocab." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "ImportError for a missing Qwen2.5-VL class is unrelated to DeepSeek detokenization." + "reason": "Import error for a Qwen VL class is unrelated to a detokenization regression in DeepSeek Coder." }, { - "left": "issue:44466", - "right": "issue:45127", + "left": "issue:45381", + "right": "issue:46209", "accept": false, - "reason": "Both touch tied weights, but one is serialization-device behavior and the other is LoRA merge collapse; not the same bug." + "reason": "Qwen2.5-VL video position IDs bug is unrelated to FP8 inference failing on B200." }, { "left": "issue:44737", "right": "issue:46894", "accept": false, - "reason": "XLNet CPU placement bug is unrelated to CohereASR training-loss shifting." + "reason": "XLNet CPU positional encoding bug and CohereASR double-shift training loss are different code paths." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate default is unrelated to batched video shape handling." + "reason": "Audio chat-template sampling-rate default and batched video shape handling are different processors and bugs." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Incorrect detokenization and a Python 3.13 flash_attn import KeyError are separate issues." + "reason": "Detokenization regression and Python 3.13 import-utils KeyError are unrelated." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "Tokenizer special-token registration bug is unrelated to a GraniteSpeech CI device-override test failure." - }, - { - "left": "issue:43950", - "right": "issue:46285", - "accept": false, - "reason": "Non-persistent buffer corruption on load is unrelated to CLIP text-model structure differences." + "reason": "Tokenizer special-token registration issue is unrelated to a stale CI test." }, { "left": "issue:43452", "right": "issue:46512", "accept": false, - "reason": "gguf_file loading failure is unrelated to stale config dtype after from_config()." + "reason": "`gguf_file` loading failure and `from_config(dtype=...)` stale config dtype are separate initialization problems." }, { - "left": "issue:44442", - "right": "issue:45198", + "left": "issue:43950", + "right": "issue:46285", "accept": false, - "reason": "FastSpeech2Conformer tokenizer loading and Wav2Vec2 save/tokenization failures are different model-specific bugs." + "reason": "Non-persistent buffer corruption on load is unrelated to CLIP text-model structural inconsistency." }, { "left": "issue:44561", "right": "issue:46234", "accept": false, - "reason": "Missing different import-utils symbols; these are separate API-regression reports." + "reason": "Missing utility after v5 removal and missing `is_falcon_mamba_ssm_available` are different import symbol regressions." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "FastSpeech2Conformer tokenizer loading and Wav2Vec2 save/tokenization failures are not the same bug." }, { "left": "issue:45200", "right": "issue:46899", "accept": false, - "reason": "Text-only mm_token_type_ids defaulting and quantized Gemma4 embedders dtype casting are distinct problems." + "reason": "Defaulting `mm_token_type_ids` to zeros is unrelated to unsafe dtype casts in quantized embedders." }, { - "left": "issue:43526", + "left": "issue:44479", "right": "issue:46192", "accept": false, - "reason": "BeitImageProcessorFast label reduction is unrelated to return_tensors being ignored with explicit text_kwargs." + "reason": "Video regression in multimodal processors and `return_tensors` being ignored are separate API issues." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "DAC latent reconstruction mismatch is unrelated to a GraniteSpeech CI test issue." + "reason": "DAC latent conversion mismatch and GraniteSpeech CI test staleness are unrelated." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Perceiver interpolation failure and RT-DETR memory release are separate model behaviors." + "reason": "Perceiver non-default resolution failure is unrelated to RT-DETR memory release behavior." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "SwitchTransformers config construction and Bamba torch.compile/SDPA failure are unrelated." + "reason": "SwitchTransformers sparse-layer construction and Bamba SDPA compile failure are distinct model issues." }, { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Default mm_token_type_ids and Qwen2.5-VL still-image position-id scaling are different bugs." + "reason": "`mm_token_type_ids` defaulting and Qwen2.5-VL rope-index scaling are different multimodal bugs." }, { "left": "issue:45200", "right": "issue:46932", "accept": false, - "reason": "Text-only Gemma4 defaults and quantized vision/audio dtype casting are not the same underlying issue." + "reason": "Default zero `mm_token_type_ids` is not the same as quantized embedder dtype casting." }, { "left": "issue:41720", "right": "issue:46209", "accept": false, - "reason": "Both are hardware-related failures, but they involve different models and different root causes." + "reason": "Different hardware/runtime failures in different models: A800 auto device mapping vs B200 FP8 inference." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "OmDet input-size mismatch and LwDetr AMP crash are unrelated detection bugs." + "reason": "Processor output size mismatch and AMP/CUDA loss crash are unrelated." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "DAC.from_latents mismatch is unrelated to Qwen3 Omni MoE feature-length computation." + "reason": "DAC latent mismatch and Qwen3 Omni helper length computation are different bugs." }, { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "A missing internal flash-attention import and Torch 2.10 incompatibility are not the same bug." + "reason": "Missing flash-attention import helper and Torch 2.10 incompatibility are not the same underlying issue." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU offload device mismatch in Bark is unrelated to qwen3_omni_moe length computation." - }, - { - "left": "issue:43827", - "right": "issue:43901", - "accept": false, - "reason": "Both are docs-related, but they concern different APIs and different v5 behavior changes." - }, - { - "left": "issue:43335", - "right": "issue:43653", - "accept": false, - "reason": "SwitchTransformers sparse-layer config and BigBirdTokenizer special-token registration are unrelated." - }, - { - "left": "issue:44792", - "right": "issue:45563", - "accept": false, - "reason": "Janus image-generation test failure is unrelated to a stale generate() warning." - }, - { - "left": "issue:44857", - "right": "issue:45593", - "accept": false, - "reason": "AMP crash in LwDetrImageLoss and missing auxiliary losses in D-FINE are separate defects." + "reason": "CPU offload device mismatch in Bark is unrelated to Qwen3 Omni feature-length calculations." } ] }, @@ -9633,7 +9559,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28282", "issue:29942", @@ -9643,7 +9569,6 @@ "issue:35141", "issue:37428", "issue:38175", - "issue:39692", "issue:42490", "issue:43335", "issue:43388", @@ -9652,30 +9577,31 @@ "issue:43550", "issue:43606", "issue:43611", - "issue:43819", + "issue:43653", + "issue:43827", "issue:43854", + "issue:43901", "issue:44112", "issue:44206", "issue:44351", "issue:44448", "issue:44466", - "issue:44493", - "issue:44514", - "issue:44521", "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:46170", "issue:46285", - "issue:46326", "issue:46424", "issue:46431", "issue:46512", @@ -9685,224 +9611,224 @@ "issue:46858" ], "soft_pairs": [ + "issue:44792|issue:45563", + "issue:43335|issue:43653", + "issue:43827|issue:43901", + "issue:44857|issue:45593", "issue:46512|issue:46821", "issue:44206|issue:45584", - "issue:28282|issue:33357", "issue:45759|issue:46285", + "issue:28282|issue:33357", "issue:38175|issue:45412", "issue:29942|issue:32090", "issue:43388|issue:43526", - "issue:43611|issue:44466", "issue:43335|issue:43504", + "issue:43611|issue:44466", "issue:44855|issue:45083", "issue:46424|issue:46858", "issue:43550|issue:44112", "issue:43335|issue:45083", "issue:43335|issue:44112", + "issue:43335|issue:43606", "issue:46431|issue:46566", "issue:45520|issue:46170", - "issue:45290|issue:46326", - "issue:43335|issue:43606", "issue:44448|issue:44743", "issue:37428|issue:44351", "issue:44617|issue:46697", "issue:42490|issue:43854", - "issue:45588|issue:45859", "issue:34689|issue:35141", - "issue:43819|issue:44493", - "issue:44514|issue:44521", - "issue:39692|issue:43526" + "issue:45588|issue:45859" ], - "prepared_review_unit_hash": "f4d264bfad04850203a880f88a8816241588364909b8cd40963cf8d191eeced2", + "prepared_review_unit_hash": "78e6b97ab87741163e91808a6762fe58cebaf94ce369a0d64d6ace57fcea4645", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36299, - "estimated_input_tokens": 9075, - "estimated_eval_tokens": 18406 + "serialized_chars": 36255, + "estimated_input_tokens": 9064, + "estimated_eval_tokens": 18384 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No proposed soft edge looks like a true duplicate pair; the set is broadly heterogeneous and most matches only share a subsystem keyword (chat templates, cache, imports, loading) rather than the same concrete bug.", - "confidence": 0.94, - "canonical_issue_reason": "issue:44521 is the strongest representative issue here: it is active, has inbound references, and describes a concrete, reproducible bug in `apply_chat_template` with clear user impact.", + "summary": "These items are mostly unrelated issue reports with superficial overlap in Transformers regressions, model loading, cache handling, and docs/tests. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected. The closest thematic hub is issue #43335, but it still does not absorb the others into one duplicate cluster.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is also the best standalone issue to keep as the representative bug report because it is specific, reproducible, and well-discussed.", + "best_issue_reason": "issue #43335 is the closest hub in this set because a few other reports are loosely about model/config initialization, but the overall group is too heterogeneous for a true canonical duplicate.", "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 warning in paged generate()." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Sparse-layer config bug vs BigBirdTokenizer mask-token registration/empty decode bug; different code paths." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs issues, but they target different API deprecations and different pages." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "AMP/CUDA crash in LwDetrImageLoss vs D-FINE not using auxiliary losses; unrelated behavior." + }, { "left": "issue:46512", "right": "issue:46821", "accept": false, - "reason": "Different failures: stale `config.dtype` vs `window_size` attribute error in TimesFM 2.5." + "reason": "Stale config.dtype after from_config vs window_size AttributeError; different APIs and failures." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Unrelated code paths: feature extractor `center` arg crash vs Whisper empty-transcription failure after token alignment." + "reason": "Unsupported center arg in LaserFeatureExtractor vs Whisper empty-transcription failure after align_special_tokens." }, { - "left": "issue:28282", - "right": "issue:33357", + "left": "issue:45759", + "right": "issue:46285", "accept": false, - "reason": "ImportError for missing PyTorch vs MacOS bus error in community CLIP loading; not the same bug." + "reason": "Different model families and bugs: Qwen multimodal from_config unwrap vs CLIP text-model structure inconsistency." }, { - "left": "issue:45759", - "right": "issue:46285", + "left": "issue:28282", + "right": "issue:33357", "accept": false, - "reason": "Different model families and failure modes: Qwen multimodal `text_config` unwrap vs CLIP text model structure change." + "reason": "Missing PyTorch ImportError vs MacOS bus error for community CLIP; different failure modes." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 zero probabilities and RT-DETR memory release are unrelated bugs." + "reason": "Unexpected zero probabilities in Siglip2 vs RT-DETR memory release bug; unrelated." }, { "left": "issue:29942", "right": "issue:32090", "accept": false, - "reason": "Flash Attention 2 test failures are not the same as a Trainer `_gpu_broadcast_one` `NoneType` error." + "reason": "Flash Attention 2 test failures vs NoneType broadcast error in Trainer; different subsystems." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Different layers: metric gathering/truncation bug vs Beit image-processor label reduction bug." + "reason": "Both involve labels, but one is gather_for_metrics dropping tuple elements and the other is BeitImageProcessorFast reduce_labels returning one label." }, { - "left": "issue:43611", - "right": "issue:44466", + "left": "issue:43335", + "right": "issue:43504", "accept": false, - "reason": "Model loading with `base_model_prefix` is unrelated to tied-weight serialization differing by device." + "reason": "Sparse-layer zero-count config bug vs BeitForSemanticSegmentation legacy-field loading issue." }, { - "left": "issue:43335", - "right": "issue:43504", + "left": "issue:43611", + "right": "issue:44466", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug and Beit legacy-field loading bug are not the same issue." + "reason": "base_model_prefix loading regression vs inconsistent lm_head serialization; separate code paths." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 import parsing error in DebertaV2 vs Qwen3 Omni MoE feature-length helper behavior." + "reason": "TorchScript parse failure from comment placement vs helper _get_feat_extract_output_lengths behavior; unrelated." }, { "left": "issue:46424", "right": "issue:46858", "accept": false, - "reason": "Both cache-related, but one is `max_cache_length` handling and the other is a device-tensor seq-length arithmetic bug." + "reason": "max_cache_length only honored for quantized cache vs device-tensor seq_length breaking arithmetic; both cache-related but different bugs." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to a stale CI device-override test in GraniteSpeech." + "reason": "torch.compile SDPA failure in Bamba vs stale GraniteSpeech CI test; not the same bug." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "Different models and code paths: SwitchTransformers config creation vs Qwen3 Omni MoE helper output lengths." + "reason": "Different functionality: sparse-layer config creation vs feature-extractor length helper behavior." }, { "left": "issue:43335", "right": "issue:44112", "accept": false, - "reason": "SwitchTransformers config bug and GraniteSpeech CI test failure are unrelated." + "reason": "Model config bug vs stale CI test; no shared underlying defect." + }, + { + "left": "issue:43335", + "right": "issue:43606", + "accept": false, + "reason": "Sparse-layer config bug vs Bark CPU offload device mismatch; different models and code paths." }, { "left": "issue:46431", "right": "issue:46566", "accept": false, - "reason": "Missing chat template error and DiffusionGemma `meta`-device loading bug are different problems." + "reason": "Missing chat template error vs meta-device weights under device_map='auto'; different loading issues." }, { "left": "issue:45520", "right": "issue:46170", "accept": false, - "reason": "Different platform/dependency issues: `flash_attn` import key error on Python 3.13 vs Windows ROCm `torch.distributed` dependency." - }, - { - "left": "issue:45290", - "right": "issue:46326", - "accept": false, - "reason": "Both involve chat templating, but one crashes on tool-call messages and the other concerns `continue_final_message` for mistral_common tokenizers." - }, - { - "left": "issue:43335", - "right": "issue:43606", - "accept": false, - "reason": "SwitchTransformers config creation and Bark CPU-offload device mismatch are unrelated." + "reason": "Python 3.13 flash_attn import key error vs Windows ROCm torch.distributed dependency issue." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Pegasus v4/v5 output mismatch is not the same as recurrent-state reset behavior in modular Qwen3.5." + "reason": "Pegasus v4/v5 output difference vs Qwen3 recurrent-state reset with cache; different model-specific bugs." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Different missing symbols in different import paths: `_flash_supports_window_size` vs `HybridCache`." + "reason": "Both are import errors, but for different missing symbols and different underlying causes." }, { "left": "issue:44617", "right": "issue:46697", "accept": false, - "reason": "CUDA OOM in Sam3Video is unrelated to Swin2SR image over-padding." + "reason": "CUDA OOM in Sam3Video vs Swin2SR over-padding images; unrelated." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Trailing-slash local-path import warning and GLM-4.7-Flash model loading failure are different issues." - }, - { - "left": "issue:45588", - "right": "issue:45859", - "accept": false, - "reason": "Flash attention `s_aux=None` crash and Qwen3_5Moe CPU-offload incompatibility are different code paths." + "reason": "Trailing-slash local model import warning vs GLM-4.7-Flash unit-test loading failure; different triggers." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Llama 3.2 Vision model-loading regression and token-embedding resize/reinit bug are unrelated." - }, - { - "left": "issue:43819", - "right": "issue:44493", - "accept": false, - "reason": "DAC forward/from_latents mismatch and unexpected `position_id` keys are separate model bugs." + "reason": "Llama 3.2 Vision loading regression vs post_init reinitializing resized embeddings; not the same defect." }, { - "left": "issue:44514", - "right": "issue:44521", - "accept": false, - "reason": "Both touch `apply_chat_template`, but one is batched padding behavior and the other is multimodal assistant-mask generation." - }, - { - "left": "issue:39692", - "right": "issue:43526", + "left": "issue:45588", + "right": "issue:45859", "accept": false, - "reason": "SigLIP2 docs/model mismatch and Beit image-processor label reduction are unrelated." + "reason": "flash_attention.py s_aux=None AttributeError vs Qwen3_5MoeTextRotaryEmbedding CPU offload incompatibility; different problems." } ] }, @@ -9924,7 +9850,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:10000", "issue:15354", @@ -9939,24 +9865,25 @@ "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:43819", "issue:43825", "issue:44112", "issue:44206", "issue:44360", "issue:44448", "issue:44493", + "issue:44514", + "issue:44521", "issue:44561", "issue:44779", "issue:44857", @@ -9966,242 +9893,249 @@ "issue:45245", "issue:45290", "issue:45310", - "issue:45381", "issue:45459", "issue:45478", "issue:45593", "issue:45823", "issue:46234", + "issue:46326", "issue:47000" ], "soft_pairs": [ + "issue:45290|issue:46326", + "issue:43819|issue:44493", + "issue:39692|issue:43526", "issue:43606|issue:44857", "issue:29127|issue:32090", + "issue:44514|issue:44521", "issue:44206|issue:45823", - "issue:39692|issue:44493", "issue:43278|issue:43825", + "issue:39692|issue:44493", "issue:28218|issue:32090", - "issue:10000|issue:15354", "issue:43720|issue:44112", + "issue:10000|issue:15354", "issue:43701|issue:45083", - "issue:44448|issue:44779", "issue:45310|issue:45478", + "issue:44448|issue:44779", "issue:26413|issue:31501", "issue:45127|issue:45245", "issue:44857|issue:45072", "issue:32090|issue:33357", - "issue:42490|issue:43575", "issue:43550|issue:45083", + "issue:42490|issue:43575", "issue:38175|issue:44493", - "issue:45459|issue:46234", "issue:44360|issue:44561", + "issue:45459|issue:46234", "issue:45593|issue:47000", "issue:27957|issue:34689", "issue:29942|issue:35141", - "issue:29870|issue:32090", - "issue:43278|issue:43329", - "issue:37428|issue:42175", - "issue:45290|issue:45381" + "issue:29870|issue:32090" ], - "prepared_review_unit_hash": "e368acb149a85ea06e49200d8521d7a7485fc86cae02ebe82822fe63752a6e41", + "prepared_review_unit_hash": "2f0a53fc8bccf759cf65ec3c21ab931b7d57e3807a54decfc2e99db1123070b4", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 27, - "serialized_chars": 35317, - "estimated_input_tokens": 8830, - "estimated_eval_tokens": 17916 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36086, + "estimated_input_tokens": 9022, + "estimated_eval_tokens": 18300 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The only clear duplicate pair is the Qwen3.5 MoE `from_pretrained` regression. All other soft pairs look like distinct bugs or unrelated code paths and should remain separate.", - "confidence": 0.89, - "canonical_issue_reason": "issue:45310 is the better canonical for the duplicate Qwen3.5 MoE reports because it is the earlier report and describes the same `transformers>=5.4.0` `from_pretrained` failure on the same model.", + "summary": "This cluster is mostly a collection of unrelated issue reports that only share broad subsystem vocabulary. I found one clear duplicate pair about the Qwen3.5 MoE from_pretrained regression; the rest should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45310 is the best canonical issue because issue:45478 repeats the same Qwen3.5 MoE from_pretrained failure with the same version qualifier, and 45310 appears to be the earlier, fuller report.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest issue candidate overall: it is a precise bug report with a clear version gate, exact affected model, and a near-identical duplicate.", + "best_issue_reason": "issue:45310 is the strongest representative for the only confirmed duplicate pair in this cluster, with a clear, specific title and earliest occurrence among the matching reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45290", + "right": "issue:46326", + "accept": false, + "reason": "Both involve chat-template/tokenizer behavior, but one is a crash with tool-call assistant messages and the other is a continue_final_message bug for mistral_common tokenizers; not the same underlying defect." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "Different problems: DAC latent reconstruction vs widespread unexpected position_id keys during model loading." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Docs/example errors for SigLIP2 are unrelated to BeitImageProcessorFast reduce_labels returning the wrong label shape." + }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "Different failures and subsystems: CPU offload device mismatch for Bark vs AMP/CUDA crash in LwDetrImageLoss." + "reason": "Both are runtime failures, but one is CPU offload device mismatch and the other is float16 AMP loss crash; different code paths and symptoms." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "Unrelated problems: LayoutLMv3 error-message clarity vs Trainer `_gpu_broadcast_one` NoneType type error." + "reason": "Clarifying a LayoutLMv3 error message is unrelated to a Trainer broadcast TypeError." }, { - "left": "issue:44206", - "right": "issue:45823", + "left": "issue:44514", + "right": "issue:44521", "accept": false, - "reason": "Different code paths: feature-extractor argument validation vs Gemma4 PLE forward device mismatch." + "reason": "Both touch apply_chat_template, but one is a batched-padding crash and the other is incorrect assistant_masks for multimodal inputs; too different to merge." }, { - "left": "issue:39692", - "right": "issue:44493", + "left": "issue:44206", + "right": "issue:45823", "accept": false, - "reason": "One is a SigLIP2 documentation/example issue; the other is a generic position-id warning across many models." + "reason": "Feature-extractor center-arg regression and Gemma4 PLE device mismatch are distinct failures." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "BF16/FP32 dtype drift in evaluation is unrelated to a pipeline error message about translation tasks." + "reason": "Embedding dtype drift in evaluate is unrelated to a pipeline error-message regression." }, { - "left": "issue:28218", - "right": "issue:32090", + "left": "issue:39692", + "right": "issue:44493", "accept": false, - "reason": "Tokenizer spacing bug and Trainer broadcast TypeError are different bugs." + "reason": "SigLIP2 docs/model mismatch and unexpected position-id keys are unrelated." }, { - "left": "issue:10000", - "right": "issue:15354", + "left": "issue:28218", + "right": "issue:32090", "accept": false, - "reason": "Different model/export issues: DistilBertModel error vs TorchScript `GeneratorExp` limitation during ViT export." + "reason": "Tokenizer spacing bug and Trainer broadcast TypeError are different subsystems and failures." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "Different CI failures with different causes: packed-weight loading vs stale device override test." + "reason": "Packed-weight loading in BitNet and a stale GraniteSpeech CI test are not the same bug." }, { - "left": "issue:43701", - "right": "issue:45083", + "left": "issue:10000", + "right": "issue:15354", "accept": false, - "reason": "Resume/checkpoint key mismatch is unrelated to a qwen3_omni_moe helper bug." + "reason": "German DistilBertModel issue and torch.jit.script GeneratorExp incompatibility are unrelated." }, { - "left": "issue:44448", - "right": "issue:44779", + "left": "issue:43701", + "right": "issue:45083", "accept": false, - "reason": "Different regressions: Pegasus output changes vs DeepSeek tokenizer correctness." + "reason": "Checkpoint resume key mismatch is unrelated to qwen3_omni_moe feature-extraction length behavior." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Same exact bug report: identical title, same affected model (`Qwen3.5 Moe`), and the same `transformers>=5.4.0` `from_pretrained` failure." + "reason": "Same concrete bug: transformers>=5.4.0 causing Qwen3.5 MoE from_pretrained errors; the titles match closely and 45478 is a later repeat." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Different v4-v5 behavior reports, but one is Pegasus generation output and the other is Deepseek tokenizer correctness." }, { "left": "issue:26413", "right": "issue:31501", "accept": false, - "reason": "Both mention training startup/resume pain, but the concrete failures are different and not the same bug." + "reason": "Resume-from-checkpoint epoch_iterator failure and FineWeb training not starting are too unspecific and not clearly the same underlying issue." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "LoRA embedding collapse is unrelated to the category-cardinality runtime error." + "reason": "LoRA/extended-vocab collapse is unrelated to a category-cardinality runtime error." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "Both are mixed-precision crashes, but they affect different models and different failure modes." + "reason": "Different dtype-related crashes in different models/code paths; not plausibly one merged fix." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "A Trainer broadcast NoneType error is not the same as a MacOS bus error with a community CLIP model." + "reason": "Trainer NoneType broadcast error and MacOS CLIP bus error are unrelated." }, { - "left": "issue:42490", - "right": "issue:43575", + "left": "issue:43550", + "right": "issue:45083", "accept": false, - "reason": "Import warning for local paths and tensor-parallel OOM are unrelated." + "reason": "torch.compile/SDPA failure in Bamba-9B-v2 is unrelated to qwen3_omni_moe helper behavior." }, { - "left": "issue:43550", - "right": "issue:45083", + "left": "issue:42490", + "right": "issue:43575", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to qwen3_omni_moe feature-length behavior." + "reason": "Local-path import warning and Qwen2-57B-A14B-Instruct OOM are unrelated load-time issues." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 zero probabilities and unexpected position-id keys are different model behaviors." + "reason": "Unexpected zero probabilities in SigLIP2 is a model-specific inference bug, not a position-id key regression." }, { - "left": "issue:45459", - "right": "issue:46234", + "left": "issue:44360", + "right": "issue:44561", "accept": false, - "reason": "Tokenizer/protobuf error handling and a missing import_utils availability helper are unrelated." + "reason": "A discussion about a missing ReLU in the DSA indexer is unrelated to removal of is_torch_fx_available breaking trust_remote_code models." }, { - "left": "issue:44360", - "right": "issue:44561", + "left": "issue:45459", + "right": "issue:46234", "accept": false, - "reason": "DSA indexer activation-function discussion is unrelated to `is_torch_fx_available` removal breaking remote-code models." + "reason": "Tokenizer protobuf error handling and missing import_utils helper are unrelated." }, { "left": "issue:45593", "right": "issue:47000", "accept": false, - "reason": "Different object-detection losses and different failure mechanisms; not the same underlying bug." + "reason": "Different object-detection loss issues with different failure modes; not the same concrete bug." }, { "left": "issue:27957", "right": "issue:34689", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta is unrelated to Llama 3.2 Vision model loading regression." + "reason": "Flash Attention 2 support for XLMRoberta and a Llama 3.2 Vision model-loading regression are unrelated." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "CI failures in Flash Attention tests do not match embedding reinitialization after resizing token embeddings." + "reason": "Flash Attention 2 CI failures and output embedding reinitialization on resize are unrelated." }, { "left": "issue:29870", "right": "issue:32090", "accept": false, - "reason": "Bart generation `decoder_attention_mask` bug is unrelated to a Trainer `_gpu_broadcast_one` NoneType error." - }, - { - "left": "issue:43278", - "right": "issue:43329", - "accept": false, - "reason": "BF16/FP32 dtype mismatch is unrelated to undefined variables in the multimodal video token-count path." - }, - { - "left": "issue:37428", - "right": "issue:42175", - "accept": false, - "reason": "Missing flash-attention import helper is unrelated to backend installation extras for TensorFlow." - }, - { - "left": "issue:45290", - "right": "issue:45381", - "accept": false, - "reason": "Chat-template tool-call crash and qwen2.5-vl video position-id bug are different prompt-processing issues." + "reason": "BART decoder_attention_mask initialization bug and Trainer broadcast NoneType error are different code paths." } ] }, "evaluator_result": { - "accept": false, - "feedback": "Rejecting because one key claim is not grounded in the packet: `issue:45310` does not have more discussion than `issue:45478` (discussion_activity is 2 vs 4). The duplicate-pair conclusion itself looks plausible, but the canonicalization rationale overstates the evidence, so the output is not conservative enough." + "accept": true, + "feedback": "The analyst output is grounded in the packet and stays conservative on the soft-edge verdicts. The Qwen3.5 MoE pair (issue:45310 / issue:45478) is a clear duplicate match with nearly identical titles and matching failure mode, and the other pairs are not overstated as duplicates." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -10217,7 +10151,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:25251", "issue:29127", @@ -10226,22 +10160,23 @@ "issue:33290", "issue:33357", "issue:34567", + "issue:37428", "issue:39692", "issue:40444", "issue:41720", + "issue:42175", "issue:42490", "issue:43257", "issue:43278", "issue:43299", + "issue:43329", "issue:43381", "issue:43526", "issue:43575", "issue:43701", "issue:43873", "issue:43927", - "issue:43976", "issue:44186", - "issue:44361", "issue:44448", "issue:44464", "issue:44466", @@ -10254,7 +10189,7 @@ "issue:45198", "issue:45229", "issue:45290", - "issue:45440", + "issue:45381", "issue:45588", "issue:45698", "issue:45823", @@ -10268,238 +10203,245 @@ "issue:47276" ], "soft_pairs": [ + "issue:43278|issue:43329", + "issue:37428|issue:42175", "issue:44464|issue:46424", + "issue:45290|issue:45381", "issue:43257|issue:43299", "issue:29870|issue:33290", + "issue:44855|issue:46399", "issue:44448|issue:44977", "issue:43278|issue:43381", - "issue:44855|issue:46399", - "issue:32090|issue:33290", "issue:40444|issue:43526", + "issue:32090|issue:33290", "issue:25251|issue:32090", - "issue:45290|issue:46655", "issue:43927|issue:45698", "issue:45920|issue:46192", "issue:44977|issue:45229", + "issue:45290|issue:46655", "issue:44530|issue:47276", "issue:44561|issue:44779", "issue:42490|issue:43873", "issue:43873|issue:45229", - "issue:29127|issue:34567", "issue:43381|issue:43701", - "issue:44871|issue:44977", + "issue:29127|issue:34567", + "issue:46421|issue:46858", "issue:44448|issue:44466", + "issue:44871|issue:44977", "issue:33357|issue:39692", - "issue:46421|issue:46858", "issue:44186|issue:45198", "issue:45588|issue:45823", "issue:43278|issue:43575", - "issue:44361|issue:45198", - "issue:41720|issue:45229", - "issue:43976|issue:45440" + "issue:41720|issue:45229" ], - "prepared_review_unit_hash": "56df003b2482ece3820e98c463c67996f0bc02adcfa939508e82c9d4db04a616", + "prepared_review_unit_hash": "2ae52087baa20118412a36bd9dcf48bcd04ccc4f7e80a08e5b1921ea2f38c7ce", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 36008, - "estimated_input_tokens": 9002, - "estimated_eval_tokens": 18260 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37062, + "estimated_input_tokens": 9266, + "estimated_eval_tokens": 18788 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated issues spanning tokenization, generation, caching, quantization, device placement, and model-loading regressions. I don\u2019t see any soft pair that clearly describes the same underlying bug/change, so all soft edges should stay rejected. Issue 43873 is the best triage anchor because it\u2019s the only open item and has the most inbound references/discussion.", - "confidence": 0.69, - "canonical_issue_reason": "issue:43873 has the highest engagement and inbound references, and it is the only open report, so it is the best representative anchor for the cluster.", + "summary": "These are mostly unrelated Transformers issue reports that only look superficially similar by subsystem (tokenizers, cache, generation, trainer, vision-language, quantization). I do not see any true duplicate pairs among the soft candidates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45920 is the broadest, highest-signal regression report in the set, with multiple affected model families and follow-up references, so it is the most representative issue if one must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the strongest single issue to surface because it is active, widely referenced, and broad enough to serve as the triage entry point.", + "best_issue_reason": "issue:45920 has the widest blast radius and strongest evidence of downstream impact; it is the best single issue to anchor the cluster, even though the cluster is not actually duplicate-heavy.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Different failures: training/eval dtype drift for embeddings vs an undefined helper in multimodal token counting." + }, + { + "left": "issue:37428", + "right": "issue:42175", + "accept": false, + "reason": "Completely unrelated: flash-attention import error vs missing TensorFlow in a torch-only install." + }, { "left": "issue:44464", "right": "issue:46424", "accept": false, - "reason": "Different bug classes: compiled-forward generation inconsistency vs cache max length being ignored for quantized cache." + "reason": "One is inconsistent generation under compiled forward; the other is cache length config being ignored for non-quantized caches." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Different code paths: chat template tool-call handling vs video vision_position_ids in Qwen2.5-VL." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Both involve Qwen3 MoE loading, but one is accelerate/deepspeed weight conversion and the other is a v5 loading regression for Qwen3VL MoE models." + "reason": "Both involve Qwen3 MOE, but one is accelerate/deepspeed weight conversion and the other is a v5 model-loading breakage; not the same bug." }, { "left": "issue:29870", "right": "issue:33290", "accept": false, - "reason": "UnboundLocalError in BART generate() is unrelated to DeepSpeed Adafactor OOM." + "reason": "UnboundLocalError in BART generation is unrelated to an OOM from Adafactor with DeepSpeed." + }, + { + "left": "issue:44855", + "right": "issue:46399", + "accept": false, + "reason": "Both mention registration/decorators, but the causes differ: Python 3.13 JIT parsing vs missing decorator on rotary embedding registration." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different models and failure modes: Pegasus v4/v5 output drift vs Qwen3.5 flash-attention generation failure." + "reason": "Pegasus v4/v5 output drift is a serialization/behavior regression; Qwen3.5 flash-attention failure is a separate generation bug." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "Embedding dtype mismatch in evaluate is not the same as gradient checkpointing being disallowed in eval mode." + "reason": "Embedding dtype changes in eval are unrelated to gradient checkpointing being disallowed in eval mode." }, { - "left": "issue:44855", - "right": "issue:46399", + "left": "issue:40444", + "right": "issue:43526", "accept": false, - "reason": "Python 3.13 import/parsing failure in DebertaV2 is unrelated to a missing decorator error in Qwen3.5 kernel registration." + "reason": "Different components and symptoms: multi-image IterableDataset finetuning failure vs BeitImageProcessorFast label reduction bug." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer _gpu_broadcast_one NoneType error is unrelated to Adafactor OOM under DeepSpeed." - }, - { - "left": "issue:40444", - "right": "issue:43526", - "accept": false, - "reason": "Different image-processor/training path bugs; multi-image IterableDataset failure is not the same as BeitImageProcessorFast label reduction." + "reason": "GPU broadcast NoneType error is not the same underlying problem as optimizer memory blow-up/OOM." }, { "left": "issue:25251", "right": "issue:32090", "accept": false, - "reason": "Pipeline top_k nesting regression has nothing to do with Trainer GPU broadcast of NoneType." - }, - { - "left": "issue:45290", - "right": "issue:46655", - "accept": false, - "reason": "Chat-template tool-call crash and video padding miscalculation are separate preprocessing bugs." + "reason": "Pipeline top_k nesting change and Trainer broadcast failure are different bugs in different execution paths." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Both touch save/load behavior, but custom token IDs being lost is a different serialization bug from loading the wrong custom module." + "reason": "Both are save/load-related, but one loses custom token IDs in config while the other reloads the wrong custom module." }, { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "Tokenizer wrong IDs across models is not the same as return_tensors being ignored when text_kwargs is passed." + "reason": "Wrong token IDs from AutoTokenizer is a real tokenizer regression; silently ignored return_tensors is an API-argument precedence issue." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Flash-attention generation failure for Qwen3.5 is unrelated to Gemma4 multi-GPU CUDA OOM." + "reason": "Flash-attention generation failure and multi-GPU OOM are unrelated symptoms with different root causes." + }, + { + "left": "issue:45290", + "right": "issue:46655", + "accept": false, + "reason": "Chat-template tool-call crash and video padding miscalculation are different multimodal preprocessing bugs." }, { "left": "issue:44530", "right": "issue:47276", "accept": false, - "reason": "PagedAttentionCache invalid group type is a concrete runtime crash, not the same as a misleading synced_gpus warning." + "reason": "Cache group-type crash on Qwen3.5 is unrelated to a misleading synced_gpus warning message." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Breaking trust_remote_code via removed helper is a different issue from DeepSeek tokenizer output regression." + "reason": "Trust_remote_code breakage from a removed utility is not the same as DeepSeek tokenizer output regression." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Local-path trailing slash warning is unrelated to quantization/offloading behavior." + "reason": "Local-path import warning and quantization offloading behavior are separate loader/offload issues." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Quantization offloading expectations and Gemma4 multi-GPU OOM are different problems." + "reason": "Quantization offloading not working is not the same as Gemma4 multi-GPU inference OOM." }, { - "left": "issue:29127", - "right": "issue:34567", + "left": "issue:43381", + "right": "issue:43701", "accept": false, - "reason": "LayoutLMv3 error-message clarity is unrelated to TrainerState token counting not updating." + "reason": "Eval-mode gradient checkpointing restriction and resume_from_checkpoint key mismatch are different trainer concerns." }, { - "left": "issue:43381", - "right": "issue:43701", + "left": "issue:29127", + "right": "issue:34567", "accept": false, - "reason": "Eval-mode gradient checkpointing restriction does not match resume_from_checkpoint key mismatch." + "reason": "LayoutLMv3 box-content error messaging is unrelated to TrainerState token accounting not updating." }, { - "left": "issue:44871", - "right": "issue:44977", + "left": "issue:46421", + "right": "issue:46858", "accept": false, - "reason": "Gemma-3 eos_token_id config inconsistency is unrelated to Qwen3.5 flash-attention generation failure." + "reason": "Both touch cache/compile, but one is spurious recompilation from lazy cache init and the other is device-tensor seq-length arithmetic breakage." }, { "left": "issue:44448", "right": "issue:44466", "accept": false, - "reason": "Both are v4/v5 regressions, but one is generation output drift and the other is serialization of lm_head.weight; not the same bug." + "reason": "Pegasus output differences across versions and lm_head serialization differing by device are distinct regressions." }, { - "left": "issue:33357", - "right": "issue:39692", + "left": "issue:44871", + "right": "issue:44977", "accept": false, - "reason": "MacOS bus error on a CLIP model is unrelated to SigLIP2 documentation example errors." + "reason": "Gemma eos_token_id config inconsistency is unrelated to Qwen3.5 flash-attention generation failures." }, { - "left": "issue:46421", - "right": "issue:46858", + "left": "issue:33357", + "right": "issue:39692", "accept": false, - "reason": "Spurious recompilation from lazy cache init is a different cache/compile issue than device-tensor seq length breaking slice arithmetic on non-CUDA backends." + "reason": "MacOS bus error on a community CLIP model and SigLIP2 docs/example mistakes are not the same underlying problem." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2 tokenizer NER/padding crash is unrelated to Wav2Vec2 save_pretrained/tokenization failure." + "reason": "LayoutLMv2 NER/padding crashes and Wav2Vec2 save_pretrained/tokenization failures affect different tokenizers and code paths." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "Flash-attention s_aux=None crash and Gemma4 PLE device mismatch are different code paths and symptoms." + "reason": "Flash-attention crashing on s_aux=None and Gemma4 PLE device mismatch are separate runtime bugs." }, { "left": "issue:43278", "right": "issue:43575", "accept": false, - "reason": "Embedding dtype drift in evaluate is unrelated to Qwen2-57B-A14B-Instruct TP OOM." - }, - { - "left": "issue:44361", - "right": "issue:45198", - "accept": false, - "reason": "MLukeTokenizer AttributeError is not the same underlying issue as Wav2Vec2 save/tokenization failure." + "reason": "Training/eval dtype drift is unrelated to tensor-parallel OOM when loading Qwen2-57B-A14B-Instruct." }, { "left": "issue:41720", "right": "issue:45229", "accept": false, - "reason": "Qwen3 auto device mapping cudaErrorAssert is distinct from Gemma4 multi-GPU OOM." - }, - { - "left": "issue:43976", - "right": "issue:45440", - "accept": false, - "reason": "Python version compatibility breakage is unrelated to DeepSeekV3MoE divergence from the remote implementation." + "reason": "Qwen3 auto device mapping cudaErrorAssert and Gemma4 multi-GPU OOM are different multi-GPU failures with different roots." } ] }, @@ -10521,7 +10463,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:29127", "issue:29870", @@ -10544,10 +10486,12 @@ "issue:43653", "issue:43701", "issue:43756", + "issue:43976", "issue:43994", "issue:44077", "issue:44206", "issue:44265", + "issue:44361", "issue:44387", "issue:44442", "issue:44479", @@ -10556,7 +10500,6 @@ "issue:44977", "issue:45198", "issue:45372", - "issue:45406", "issue:45412", "issue:45440", "issue:45459", @@ -10568,243 +10511,249 @@ "issue:45803", "issue:46121", "issue:46693", - "issue:46701", "issue:47217", "issue:47276" ], "soft_pairs": [ + "issue:43976|issue:45440", + "issue:44361|issue:45198", "issue:29942|issue:30064", "issue:43611|issue:46121", "issue:45198|issue:45803", "issue:44206|issue:44479", "issue:43412|issue:45588", - "issue:44568|issue:45491", "issue:29870|issue:34689", + "issue:44568|issue:45491", "issue:44077|issue:47276", "issue:43653|issue:44743", - "issue:42915|issue:42994", "issue:43756|issue:45440", + "issue:42915|issue:42994", "issue:44977|issue:45563", "issue:43653|issue:44442", "issue:44265|issue:45198", - "issue:43575|issue:43701", "issue:43065|issue:43994", + "issue:43575|issue:43701", "issue:29127|issue:33357", "issue:42490|issue:43381", - "issue:43065|issue:45412", "issue:45491|issue:46693", - "issue:43575|issue:44387", + "issue:43065|issue:45412", "issue:29127|issue:35141", + "issue:43575|issue:44387", "issue:43994|issue:45412", "issue:45372|issue:45588", "issue:43526|issue:47217", "issue:45440|issue:45593", - "issue:45459|issue:45712", - "issue:45406|issue:46701", - "issue:43334|issue:43531" + "issue:43334|issue:43531", + "issue:45459|issue:45712" ], - "prepared_review_unit_hash": "60ffa977217124cace515b2584c9a3cb623bb67f4d5c57f296db38dcd7f75d76", + "prepared_review_unit_hash": "ff5ebf186777ee67b8353f4740de7fbf1a28953ba8221dd8fc7c3b0ed3009357", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36608, - "estimated_input_tokens": 9152, - "estimated_eval_tokens": 18560 + "soft_pair_count": 30, + "serialized_chars": 36832, + "estimated_input_tokens": 9208, + "estimated_eval_tokens": 18672 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a duplicate set; it mixes many unrelated issues across different models, processors, and regressions. No soft edge looks like the same underlying bug or change.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "The items are a grab-bag of unrelated Transformer bugs across tokenizers, vision, quantization, export, attention, and loading paths. None of the soft pairs look like the same underlying issue, so no pair should be merged.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44387 is the strongest standalone representative: a concrete, version-scoped regression with clear impact (CUDA reserved memory/OOM) and a well-defined code path.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:44387 is the best issue to keep as the cluster representative because it is specific, actionable, and broadly impactful compared with the other unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python-version loading failure vs DeepSeek model implementation divergence; different models and code paths." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLuke tokenizer AttributeError on tasks vs Wav2Vec2 save/tokenization failure; unrelated subsystems." + }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Flash Attention test failures and void segmentation-map processing are unrelated subsystems and failure modes." + "reason": "Flash Attention 2 CI failures vs image processor void-segmentation handling; no shared bug." }, { "left": "issue:43611", "right": "issue:46121", "accept": false, - "reason": "One is a model-loading/base_model_prefix regression; the other is a rope-validation TypeError from JSON-loaded lists." + "reason": "Base-model-prefix loading regression vs rope-validation type error; different functionality." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure and Fuyu's bare except swallowing errors are different code paths and bugs." + "reason": "Wav2Vec2 save/tokenization failure vs Fuyu bare-except error masking; different root causes." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "Unsupported feature arg crash vs video-input regression for Qwen models; not the same defect." + "reason": "Unsupported center arg in a feature extractor vs Qwen video-input regression; unrelated." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "ExecuTorch export support issues and flash_attention sink-less-model AttributeError are unrelated." + "reason": "Executorch export/erfinv issue vs flash_attention s_aux=None crash; different code paths." }, { - "left": "issue:44568", - "right": "issue:45491", + "left": "issue:29870", + "right": "issue:34689", "accept": false, - "reason": "Tokenizer special-token behavior and Gemma3 NaN embeddings are different problems with different code paths." + "reason": "BART generate decoder mask bug vs Llama 3.2 Vision loading regression; unrelated." }, { - "left": "issue:29870", - "right": "issue:34689", + "left": "issue:44568", + "right": "issue:45491", "accept": false, - "reason": "BART generate() decoder_attention_mask bug and Llama 3.2 Vision loading breakage are unrelated." + "reason": "Tokenizer special-token addition bug vs Gemma3 NaN embeddings; different failure modes." }, { "left": "issue:44077", "right": "issue:47276", "accept": false, - "reason": "An optional post_init API change is not the same as a misleading synced_gpus warning." + "reason": "PatchTSMixer post_init behavior vs a misleading warning message; not the same bug." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "BigBirdTokenizer special-token registration and Qwen3.5 recurrent state reset are distinct bugs." + "reason": "BigBird tokenizer decode/special-token issue vs Qwen recurrent-state reset bug; unrelated." }, { - "left": "issue:42915", - "right": "issue:42994", + "left": "issue:43756", + "right": "issue:45440", "accept": false, - "reason": "FineGrainedFP8Config failure in Qwen3Moe and quantized model saving failure are separate issues." + "reason": "SmolLM3 RoPE-layer mismatch vs DeepSeekV3MoE divergence; different models and fixes." }, { - "left": "issue:43756", - "right": "issue:45440", + "left": "issue:42915", + "right": "issue:42994", "accept": false, - "reason": "Smollm3 RoPE-layer mismatch and DeepseekV3MoE divergence are different model-implementation discrepancies." + "reason": "Qwen3MoE FP8 config failure vs quantized-model saving failure; both quantization-related but not the same defect." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Flash-attention generation behavior and a stale num_return_sequences warning are not the same bug." + "reason": "Qwen3.5 flash-attention generation bug vs stale paged-generate warning; unrelated." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "BigBirdTokenizer decode output and FastSpeech2ConformerTokenizer loading failure are unrelated tokenizer issues." + "reason": "BigBirdTokenizer special-token bug vs AutoTokenizer loading failure for FastSpeech2ConformerTokenizer; different problems." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export failures with torch_compilable_check are not related to Wav2Vec2 save_pretrained/tokenization failures." + "reason": "torch.export failure on torch_compilable_check vs Wav2Vec2 save/tokenization issue; unrelated." }, { - "left": "issue:43575", - "right": "issue:43701", + "left": "issue:43065", + "right": "issue:43994", "accept": false, - "reason": "Tensor-parallel OOM on a specific model and resume_from_checkpoint key mismatch are different problems." + "reason": "Dummy Conv2d in Sam3PixelDecoder vs Siglip2 nonsense outputs; different models and symptoms." }, { - "left": "issue:43065", - "right": "issue:43994", + "left": "issue:43575", + "right": "issue:43701", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d and SigLIP2 nonsensical AutoModel/pipeline results are unrelated." + "reason": "Tensor-parallel load OOM vs resume_from_checkpoint key mismatch; unrelated loading/training bugs." }, { "left": "issue:29127", "right": "issue:33357", "accept": false, - "reason": "LayoutLMv3 box-information error messaging and a CLIP MacOS bus error are unrelated." + "reason": "LayoutLMv3 box-info messaging issue vs MacOS bus error loading a CLIP model; unrelated." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Trailing-slash import warning and gradient-checkpointing eval-mode restriction are different behaviors." - }, - { - "left": "issue:43065", - "right": "issue:45412", - "accept": false, - "reason": "Dummy pixel-decoder layer and RT-DETR memory not released are unrelated implementation issues." + "reason": "Trailing-slash import warning vs gradient-checkpointing eval-mode restriction; different behaviors." }, { "left": "issue:45491", "right": "issue:46693", "accept": false, - "reason": "Gemma3 NaN embeddings from mixed-length sliding-window batches and FA2 long-context performance regression are not the same bug." + "reason": "Gemma3 mixed-length NaNs vs long-context FA2 performance regression; both attention-related but not the same concrete bug." }, { - "left": "issue:43575", - "right": "issue:44387", + "left": "issue:43065", + "right": "issue:45412", "accept": false, - "reason": "OOM when loading a specific model with tp and increased reserved-memory regression under int4 quantization are different causes." + "reason": "Dummy decoder layer cleanup vs RT-DETR memory-release bug; unrelated." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "LayoutLMv3 error clarity and token-embedding reinitialization in post_init are unrelated." + "reason": "LayoutLMv3 error-message clarity vs token-embedding reinitialization on resize; different code paths." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "TP load OOM for a specific model vs int4 quantization reserved-memory regression; different causes and fixes." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 bad outputs and RT-DETR memory retention are unrelated model/runtime issues." + "reason": "Siglip2 bad outputs vs RT-DETR memory leak; unrelated." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Gemma 4 processor import failure due to ReasoningEffort and flash_attention sink-less crash are different dependency/code-path problems." + "reason": "Gemma 4 processor import error from mistral_common vs flash_attention crash on s_aux=None; unrelated." }, { "left": "issue:43526", "right": "issue:47217", "accept": false, - "reason": "BEiTImageProcessorFast reduce_labels bug and VL image-placeholder handling are different image-processing/model-input issues." - }, - { - "left": "issue:45440", - "right": "issue:45593", - "accept": false, - "reason": "DeepseekV3MoE divergence and D-FINE auxiliary-loss omission are unrelated model-specific bugs." + "reason": "BeitImageProcessorFast label reduction bug vs VL image-placeholder handling failure; different components." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Tokenizer protobuf error masking and leftover dummy classes in dummy_pt_objects.py are unrelated cleanup/bugfix tasks." + "reason": "Tokenizer protobuf exception handling vs dummy-class cleanup/check_repo issue; unrelated." }, { - "left": "issue:45406", - "right": "issue:46701", + "left": "issue:43334", + "right": "issue:43531", "accept": false, - "reason": "Gemma4Processor missing _tokenizer and DiffusionGemmaGenerationOutput pipeline failure are different breakages." + "reason": "Qwen3-VL pad_token_id AttributeError vs Qwen3-MoE sliding_window bug; different model/config defects." }, { - "left": "issue:43334", - "right": "issue:43531", + "left": "issue:45459", + "right": "issue:45712", "accept": false, - "reason": "Qwen3-VL pad_token_id loading failure and Qwen3-MoE sliding_window issue are separate configuration/runtime bugs." + "reason": "Tokenizer protobuf exception handling vs dummy-class cleanup/check_repo issue; unrelated." } ] }, @@ -10826,7 +10775,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:27957", "issue:29127", @@ -10835,6 +10784,7 @@ "issue:35141", "issue:42915", "issue:43066", + "issue:43262", "issue:43316", "issue:43381", "issue:43421", @@ -10846,13 +10796,12 @@ "issue:43931", "issue:43994", "issue:44077", + "issue:44206", "issue:44279", "issue:44442", "issue:44462", "issue:44514", "issue:44625", - "issue:44743", - "issue:44811", "issue:44857", "issue:45072", "issue:45245", @@ -10873,228 +10822,236 @@ "issue:46399", "issue:46640", "issue:46693", + "issue:46701", "issue:46821", "issue:47312" ], "soft_pairs": [ - "issue:44625|issue:45584", + "issue:45406|issue:46701", "issue:43421|issue:45491", "issue:44442|issue:45072", - "issue:45372|issue:45406", + "issue:44625|issue:45584", "issue:45250|issue:46640", + "issue:45372|issue:45406", "issue:45412|issue:46369", - "issue:43066|issue:44462", "issue:45245|issue:45563", "issue:43526|issue:45834", + "issue:43066|issue:44462", "issue:45372|issue:45459", "issue:27957|issue:33290", - "issue:33290|issue:34689", "issue:27957|issue:29127", + "issue:33290|issue:34689", "issue:33290|issue:35141", "issue:43526|issue:46693", "issue:43824|issue:44279", "issue:43994|issue:46369", "issue:43381|issue:43873", - "issue:44857|issue:45588", "issue:42915|issue:43931", + "issue:44857|issue:45588", "issue:45588|issue:46821", "issue:46121|issue:46399", "issue:43316|issue:47312", - "issue:44514|issue:45381", "issue:43646|issue:44077", - "issue:44743|issue:44811", - "issue:43761|issue:45859" + "issue:44514|issue:45381", + "issue:43761|issue:45859", + "issue:43262|issue:44206" ], - "prepared_review_unit_hash": "30ce66e272f9dde2172a8b2c4b1d51a925546fb71c77ecb20e9c3fdb9c03a18f", + "prepared_review_unit_hash": "f84be0d7e18c05671abb459d826a9b6ecddd9c16715974499a7ed02227cee295", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 27, - "serialized_chars": 35006, - "estimated_input_tokens": 8752, - "estimated_eval_tokens": 17760 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35913, + "estimated_input_tokens": 8979, + "estimated_eval_tokens": 18214 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the soft pairs span unrelated model bugs, tokenizer issues, config regressions, and flash-attention/performance topics. No soft edge looks like the same concrete bug/change, so I reject all proposed merges.", - "confidence": 0.95, - "canonical_issue_reason": "Use issue 45250 as the representative issue: it is the broadest and most generic Flash Attention-related ticket in an otherwise mixed cluster, while the other issues are mostly model-specific or orthogonal bugs.", + "summary": "This is a noisy, mostly heterogeneous issue cluster. The soft pairs are only superficially related, so I would not merge any of them. The strongest standalone issue in the set is the FA2 regression report with the clearest repro and discussion.", + "confidence": 0.31, + "canonical_issue_reason": "Issue 46693 is the most concrete and actionable report in the set: it describes a specific Flash Attention 2 long-context regression after 5.2.0, with substantial discussion and clear symptoms.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45250 is the best single issue to keep as a representative placeholder because it is the most general/top-level problem statement among these issues.", + "best_issue_reason": "Issue 46693 is the best issue to keep as the representative one because it is well-scoped, technically specific, and has the strongest evidence of a real underlying bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44625", - "right": "issue:45584", + "left": "issue:45406", + "right": "issue:46701", "accept": false, - "reason": "Different components and failures: Qwen3.5 num_labels propagation vs Whisper empty-transcription generation after align_special_tokens." + "reason": "Both are Gemma-related, but one is a processor attribute crash and the other is a pipeline/output issue; different code paths and root causes." }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "Tokenizer special-token post-processor updates are unrelated to Gemma3 NaN embeddings in mixed-length GPU batching." + "reason": "One is a tokenizer post-processor update bug, the other is a GPU NaN embedding issue; unrelated failures." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Tokenizer loading failure and a bfloat16 dtype mismatch are different bug paths, not one underlying change." + "reason": "AutoTokenizer loading a tokenizer class and dtype mismatches in inference are unrelated problems." }, { - "left": "issue:45372", - "right": "issue:45406", + "left": "issue:44625", + "right": "issue:45584", "accept": false, - "reason": "Both involve Gemma4 loading, but one is a mistral_common import problem and the other is a missing _tokenizer attribute in serve." + "reason": "Different models and different bugs: num_labels propagation vs Whisper empty-transcription generation failure." }, { "left": "issue:45250", "right": "issue:46640", "accept": false, - "reason": "Both mention Flash Attention 2, but the titles are too generic and do not show the same concrete code-path problem." + "reason": "Both mention Flash Attention 2, but they are too broad and do not clearly describe the same concrete bug." }, { - "left": "issue:45412", - "right": "issue:46369", + "left": "issue:45372", + "right": "issue:45406", "accept": false, - "reason": "RT-DETR memory leak and SAM3 litetext memory overflow are distinct model/runtime issues." + "reason": "Both touch Gemma 4 loading, but one is an import dependency issue and the other is a missing _tokenizer attribute crash." }, { - "left": "issue:43066", - "right": "issue:44462", + "left": "issue:45412", + "right": "issue:46369", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 is not the same as AutoTokenizer ignoring tokenizer.json." + "reason": "Both are memory-related, but they affect different models and different failure modes." }, { "left": "issue:45245", "right": "issue:45563", "accept": false, - "reason": "Category-count overflow and a stale generate warning are unrelated behaviors." + "reason": "Category-count validation and a stale generation warning are unrelated." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to Kosmos2.5 long OCR input indexing." + "reason": "Different vision processors and different symptoms; no shared code-path problem is evident." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer decoder type mismatch and ignoring tokenizer.json are distinct tokenizer-loading issues." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "ImportError from mistral_common and protobuf-related tokenizer error masking are different failure mechanisms." + "reason": "Both mention tokenizers, but one is a missing dependency import and the other is an exception-handling bug hiding real errors." }, { "left": "issue:27957", "right": "issue:33290", "accept": false, - "reason": "XLMRoberta Flash Attention 2 support and Adafactor+DeepSpeed OOM are different subsystems and symptoms." + "reason": "XLM-RoBERTa Flash Attention support and Adafactor OOM in DeepSpeed are unrelated." }, { - "left": "issue:33290", - "right": "issue:34689", + "left": "issue:27957", + "right": "issue:29127", "accept": false, - "reason": "Optimizer OOM in DeepSpeed is unrelated to a model-loading regression for Llama 3.2 Vision." + "reason": "Flash Attention support and an error-message clarity issue in LayoutLMv3 are unrelated." }, { - "left": "issue:27957", - "right": "issue:29127", + "left": "issue:33290", + "right": "issue:34689", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta is unrelated to a LayoutLMv3 error-message clarity issue." + "reason": "Optimizer OOM and model-loading breakage for a vision model are different bugs." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "OOM with Adafactor/DeepSpeed is not the same as embedding reinitialization after resize." + "reason": "Adafactor memory usage and embedding reinitialization after resize are unrelated." }, { "left": "issue:43526", "right": "issue:46693", "accept": false, - "reason": "Label-reduction behavior in an image processor is unrelated to FA2 long-context generation performance regression." + "reason": "Different models and different bug classes: label reduction vs Flash Attention performance regression." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "A specific missing Qwen2_5_VL import and a generic dependency issue are not the same underlying bug." + "reason": "A missing export/import and a vague dependency issue do not look like the same underlying problem." }, { "left": "issue:43994", "right": "issue:46369", "accept": false, - "reason": "Nonsensical AutoModel/pipeline outputs and memory overflow in sam3-litetext are different problems." + "reason": "Wrong model outputs and memory overflow are unrelated symptoms." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Gradient checkpointing in eval mode and offloading with quantization are unrelated code paths." + "reason": "Gradient checkpointing in eval mode and quantization offloading behavior are different concerns." }, { - "left": "issue:44857", - "right": "issue:45588", + "left": "issue:42915", + "right": "issue:43931", "accept": false, - "reason": "LwDetrImageLoss AMP crash and flash_attention.py s_aux=None AttributeError are separate failures." + "reason": "Both involve Qwen models, but the FP8 config failure and weight-shape mismatch are different code paths." }, { - "left": "issue:42915", - "right": "issue:43931", + "left": "issue:44857", + "right": "issue:45588", "accept": false, - "reason": "FineGrainedFP8Config failure in Qwen3Moe is unrelated to Qwen3-VL weight-shape mismatch on load." + "reason": "Float16 AMP loss crash and a Flash Attention sinkless-model AttributeError are unrelated." }, { "left": "issue:45588", "right": "issue:46821", "accept": false, - "reason": "A flash-attention AttributeError and a TimesFM window_size AttributeError are different bugs in different modules." + "reason": "Different subsystems: flash-attention integration vs TimesFM window_size handling." }, { "left": "issue:46121", "right": "issue:46399", "accept": false, - "reason": "JSON-loaded list handling in rope validation and a missing decorator for rotary kernel registration are not the same issue." + "reason": "Rope-validation TypeError and a kernel registration decorator error are unrelated." }, { "left": "issue:43316", "right": "issue:47312", "accept": false, - "reason": "Gemma3TextConfig API inconsistency is unrelated to kernel version incompatibility." - }, - { - "left": "issue:44514", - "right": "issue:45381", - "accept": false, - "reason": "Batched chat-template padding failure and wrong vision_position_ids in Qwen2.5-VL video input are different bugs." + "reason": "Configuration API discrepancy and kernel compatibility are different issues." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Custom model initialization breakage in Transformers v5 is broader and different from patchtsmixer's post_init restriction." + "reason": "Custom model initialization breakage is broader; the patchtsmixer post_init change is a specific model-level constraint, not the same bug." }, { - "left": "issue:44743", - "right": "issue:44811", + "left": "issue:44514", + "right": "issue:45381", "accept": false, - "reason": "Qwen3.5 recurrent-state reset with cache is unrelated to Whisper batch_decode skip_special_tokens handling." + "reason": "Both are Qwen2.5-VL, but one is batched chat-template padding and the other is video vision_position_ids; different failures." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility are different bugs." + "reason": "Hidden_states being None and CPU offload incompatibility are unrelated model behaviors." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and a feature extractor arg crash are different audio-preprocessing bugs." } ] }, @@ -11116,7 +11073,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:25251", "issue:27957", @@ -11125,7 +11082,6 @@ "issue:42175", "issue:42490", "issue:42617", - "issue:43262", "issue:43316", "issue:43493", "issue:43525", @@ -11136,17 +11092,19 @@ "issue:43720", "issue:43742", "issue:43749", + "issue:43761", "issue:43827", "issue:43873", "issue:43931", "issue:44062", "issue:44077", + "issue:44079", "issue:44190", - "issue:44206", "issue:44442", "issue:44479", "issue:44496", "issue:44743", + "issue:44811", "issue:44855", "issue:44871", "issue:44898", @@ -11167,8 +11125,8 @@ "issue:46697" ], "soft_pairs": [ - "issue:43262|issue:44206", "issue:44442|issue:44898", + "issue:44743|issue:44811", "issue:44977|issue:45245", "issue:45325|issue:46697", "issue:43720|issue:44442", @@ -11178,206 +11136,213 @@ "issue:27957|issue:29870", "issue:43873|issue:44871", "issue:43493|issue:43827", - "issue:44062|issue:45907", "issue:43643|issue:44913", "issue:43316|issue:45759", "issue:44964|issue:45440", + "issue:44062|issue:45907", "issue:43316|issue:43525", "issue:25251|issue:28282", "issue:43749|issue:43931", "issue:44987|issue:46178", - "issue:42617|issue:43575", "issue:44855|issue:45137", + "issue:42617|issue:43575", "issue:43550|issue:43701", "issue:43931|issue:45216", - "issue:43575|issue:43742", "issue:44190|issue:44496", - "issue:43701|issue:46506" + "issue:43701|issue:46506", + "issue:43575|issue:43742", + "issue:43761|issue:44079" ], - "prepared_review_unit_hash": "a5f72eea8086f833991b12e8f39bef33eb27186fa29a82b822d2a493d35dbdea", + "prepared_review_unit_hash": "ce4c1eac178814d55974bd777e98a99991b512847b015651618002d4d5b6b658", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 26, - "serialized_chars": 35010, - "estimated_input_tokens": 8753, - "estimated_eval_tokens": 17762 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35867, + "estimated_input_tokens": 8967, + "estimated_eval_tokens": 18190 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly superficial text-similarity matches across unrelated issues. None of the proposed soft pairs look like the same concrete bug/change path, so I reject all edges. If a single issue must anchor the set, issue:43749 is the strongest standalone representative by activity and inbound references.", - "confidence": 0.91, - "canonical_issue_reason": "issue:43749 is the strongest anchor in this set: it is active, has the most discussion and inbound references, and represents a substantial user-facing loading/regression report.", + "summary": "These are mostly unrelated issue reports spanning different models, APIs, and regressions. None of the soft pairs look like the same underlying bug or change, so all duplicate-style links should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43749 is the best overall issue to keep as a representative, but only as a standalone anchor; the rest of the items are not true duplicates of it.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43262", - "right": "issue:44206", + "left": "issue:44442", + "right": "issue:44898", "accept": false, - "reason": "Different bug surfaces: audio processor chat-template sampling-rate default vs image-classification feature-extractor center-arg crash." + "reason": "Different components and failure modes: tokenizer loading vs image classification interpolation; no shared code path." }, { - "left": "issue:44442", - "right": "issue:44898", + "left": "issue:44743", + "right": "issue:44811", "accept": false, - "reason": "Tokenizer loading failure and Perceiver image-padding/interpolation failure are unrelated code paths." + "reason": "Qwen recurrent-state cache bug vs Whisper processor decode option handling; unrelated models and APIs." }, { "left": "issue:44977", "right": "issue:45245", "accept": false, - "reason": "Flash-attention generation regression and category-cardinality limit error are distinct problems." + "reason": "Flash-attention generation bug for Qwen3.5 vs categorical cardinality runtime limit; clearly different issues." }, { "left": "issue:45325", "right": "issue:46697", "accept": false, - "reason": "Qwen2.5-VL rope/time-position scaling bug vs Swin2SR over-padding bug; different models and logic." + "reason": "Qwen2.5-VL rope position scaling bug vs Swin2SR over-padding in image processor; unrelated vision preprocessing paths." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "Packed-weight accelerate loading bug and AutoTokenizer failure are unrelated loading paths." + "reason": "BitNet packed-weight loading failure is unrelated to FastSpeech2ConformerTokenizer loading." }, { "left": "issue:42490", "right": "issue:44743", "accept": false, - "reason": "Local-path trailing-slash import warning is not the same as Qwen3.5 recurrent-state reset during cached generation." + "reason": "Local path trailing-slash warning vs Qwen recurrent-state reset; different subsystems and symptoms." }, { "left": "issue:42175", "right": "issue:46234", "accept": false, - "reason": "Torch-only install missing TensorFlow backend metadata vs missing import_utils availability helper are different issues." + "reason": "Packaging/backend dependency issue vs missing import_utils helper; not the same bug." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "Optional post_init validation/deprecation is unrelated to the Qwen video-input regression." + "reason": "PatchTSMixer post_init validation has nothing to do with Qwen video-input regression." }, { "left": "issue:27957", "right": "issue:29870", "accept": false, - "reason": "XLMRoberta Flash Attention 2 support and BART decoder_attention_mask initialization are separate code paths." + "reason": "Flash Attention 2 support for XLMRoberta is unrelated to BART decoder_attention_mask generate error." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization/offloading behavior and Gemma eos_token_id config mismatch are not the same bug." + "reason": "Quantization/offloading behavior is not the same as Gemma eos_token_id configuration mismatch." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 implementation mismatch vs stale docs referencing pipeline() after removals are unrelated." - }, - { - "left": "issue:44062", - "right": "issue:45907", - "accept": false, - "reason": "AddedToken keyword collision and list_repo_files revision forwarding are different APIs and failures." + "reason": "SigLIP2 HF/JAX numerical discrepancy is unrelated to docs still referencing removed pipeline()." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Missing fields after trust_remote_code config loading is unrelated to GPTNeoX rotary_pct resetting on reload." + "reason": "AutoConfig trust_remote_code missing fields vs GPTNeoX rotary_pct reload persistence; different config bugs." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Gemma3TextConfig API discrepancy and AutoModelForCausalLM text_config unwrapping are adjacent but not the same concrete defect." + "reason": "Gemma3TextConfig API discrepancy vs AutoModelForCausalLM text_config unwrapping for Qwen multimodal configs; related theme but not same bug." }, { "left": "issue:44964", "right": "issue:45440", "accept": false, - "reason": "Phi-4 multimodal load failure and DeepSeekV3MoE divergence are different models and problems." + "reason": "Phi-4 multimodal loading failure and DeepseekV3MoE divergence are distinct model-specific issues." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken special-arg duplication is unrelated to list_repo_files revision forwarding in CI." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Gemma3TextConfig discrepancy does not match Llama4Config missing pad_token_id." + "reason": "Gemma3TextConfig API mismatch is not the same as Llama4Config missing pad_token_id." }, { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Pipeline top_k nesting bug and AutoModel PyTorch-import error are entirely different concerns." + "reason": "Pipeline top_k output nesting bug is unrelated to AutoModel import requiring PyTorch." }, { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "FSDP CPU RAM efficient loading and Qwen3-VL shape mismatch are not the same underlying issue." + "reason": "FSDP CPU RAM efficient loading regression is unrelated to Qwen3-VL weight-shape mismatch." }, { "left": "issue:44987", "right": "issue:46178", "accept": false, - "reason": "Model-load failure for physical-intelligence/fast and missing tensorflow_text backend mapping are unrelated." + "reason": "Loading a specific repo with transformers>=5.1.0 is unrelated to tensorflow_text backend mapping." }, { - "left": "issue:42617", - "right": "issue:43575", + "left": "issue:44855", + "right": "issue:45137", "accept": false, - "reason": "3d_parallel.py runtime failure and tensor-parallel OOM on Qwen2-57B are different failure modes." + "reason": "Torch.jit indentation parsing issue is unrelated to DeepSpeed ZeRO3 deque pop error." }, { - "left": "issue:44855", - "right": "issue:45137", + "left": "issue:42617", + "right": "issue:43575", "accept": false, - "reason": "Python 3.13 import/torch.jit parsing error is unrelated to DeepSpeed ZeRO3 deque underflow." + "reason": "3d_parallel.py execution failure and Qwen2-57B-A14B-Instruct tp OOM are different problems." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "torch.compile SDPA failure in Bamba is not the same as resume_from_checkpoint key mismatch." + "reason": "Bamba torch.compile/SDPA bug is unrelated to resume_from_checkpoint key mismatch." }, { "left": "issue:43931", "right": "issue:45216", "accept": false, - "reason": "Weight-shape mismatch when loading Qwen3-VL is distinct from a save_pretrained regression in Qwen3.5." - }, - { - "left": "issue:43575", - "right": "issue:43742", - "accept": false, - "reason": "Tensor-parallel OOM and key error while loading MobileLLM are different loading failures." + "reason": "Model weight-shape mismatch when loading differs from Qwen3.5 save_pretrained checkpoint corruption." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in a training script is unrelated to an unrecognized-model config error." + "reason": "Local dataset loading in image classification script is unrelated to unrecognized model_type in config.json." }, { "left": "issue:43701", "right": "issue:46506", "accept": false, - "reason": "Checkpoint key mismatch and CPU/CUDA numerical-difference report do not share the same bug path." + "reason": "Checkpoint key mismatch is unrelated to CPU/CUDA output differences in PPDocLayoutV3." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "TP OOM on Qwen2-57B-A14B-Instruct is not the same as MobileLLM-125M key error." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment bug affect different layers and behaviors." } ] }, @@ -11399,7 +11364,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:16998", "issue:22355", @@ -11416,6 +11381,7 @@ "issue:43404", "issue:43441", "issue:43452", + "issue:43540", "issue:43606", "issue:43646", "issue:43701", @@ -11423,12 +11389,12 @@ "issue:43761", "issue:43825", "issue:44077", - "issue:44079", "issue:44162", "issue:44188", "issue:44442", "issue:44464", "issue:44496", + "issue:44514", "issue:44521", "issue:44534", "issue:44617", @@ -11450,7 +11416,6 @@ "issue:47276" ], "soft_pairs": [ - "issue:43761|issue:44079", "issue:43606|issue:43701", "issue:44442|issue:45803", "issue:27957|issue:32090", @@ -11459,9 +11424,9 @@ "issue:46829|issue:46932", "issue:43761|issue:44188", "issue:43295|issue:44871", + "issue:43316|issue:43716", "issue:43295|issue:43825", "issue:44871|issue:45584", - "issue:43316|issue:43716", "issue:42907|issue:45698", "issue:16998|issue:22355", "issue:44521|issue:47217", @@ -11475,192 +11440,193 @@ "issue:43352|issue:46640", "issue:44162|issue:44792", "issue:45803|issue:46365", - "issue:43065|issue:44617" + "issue:43065|issue:44617", + "issue:43540|issue:44514" ], - "prepared_review_unit_hash": "e1ed34e857af5c704417455c8785947e18fa7abcee28d653aebb7ba97c2bf539", + "prepared_review_unit_hash": "9c7284c371883b5f015ffc298c4935c41946f2f7362121863568d7d264b6425a", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 34753, - "estimated_input_tokens": 8689, - "estimated_eval_tokens": 17634 + "serialized_chars": 35322, + "estimated_input_tokens": 8831, + "estimated_eval_tokens": 17918 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab-bag of unrelated Transformers issues spanning multimodal processing, generation, config reload, FlashAttention, save/load, import, and trainer paths. None of the soft pairs look like the same concrete underlying bug/change, so there is no strong duplicate family here.", - "confidence": 0.95, - "canonical_issue_reason": "If a single representative is needed, issue #44521 is the most concrete and currently active user-facing bug, with a focused multimodal repro and the strongest external interest in the set.", + "summary": "This cluster is highly heterogeneous: the soft pairs mostly share broad Transformers keywords (multimodal, generation, save/load, FlashAttention) but point to different concrete failures and code paths. I would not merge any of the proposed pairs as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #44521 is the strongest representative only by cluster prominence: it has the most inbound references and broad multimodal chat-templating scope. That said, the overall cluster does not form a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #44521 is the best cluster anchor because it is well-scoped, open, and clearly described; the rest of the items are heterogeneous and don\u2019t form one duplicate group.", + "best_issue_reason": "#44521 is the most active and broadly referenced issue in the set, so it is the best single issue to anchor triage if one must be chosen.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43761", - "right": "issue:44079", - "accept": false, - "reason": "Different bugs: CLIPVisionModel hidden_states regression vs ModelOutput key reassignment when a previous value is None." - }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "Different code paths: CPU offload device mismatch vs resume_from_checkpoint key mismatch." + "reason": "Different failures: CPU offload device mismatch in Bark vs checkpoint key mismatch during resume_from_checkpoint." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "Tokenizer loading failure for FastSpeech2ConformerTokenizer is unrelated to FuyuBatchFeature's bare-except tensor conversion bug." + "reason": "Tokenizer loading failure for FastSpeech2Conformer is unrelated to Fuyu's bare-except error handling bug." }, { "left": "issue:27957", "right": "issue:32090", "accept": false, - "reason": "Different failures: Flash Attention 2 support for XLMRoberta vs Trainer _gpu_broadcast_one NoneType broadcast error." + "reason": "FlashAttention2 support for XLM-Roberta is unrelated to Trainer's NoneType broadcast error." }, { "left": "issue:42490", "right": "issue:47276", "accept": false, - "reason": "Import warning for a local path with a trailing slash is unrelated to the synced_gpus warning text in continuous batching." + "reason": "Both mention warnings, but one is about local-path import warnings and the other about synced_gpus messaging in continuous batching." }, { "left": "issue:43232", "right": "issue:43388", "accept": false, - "reason": "Generation kwargs update after sync_gpus and gather_for_metrics dropping tuple labels are separate utilities with different symptoms." + "reason": "Generation kwargs syncing and metric gathering are different code paths with different symptoms." }, { "left": "issue:46829", "right": "issue:46932", "accept": false, - "reason": "VideoPrism hidden_states/output bug and Gemma4 quantized embedder dtype casting are different model-specific issues." + "reason": "VideoPrism hidden-state output bug is distinct from Gemma4 quantized embedder dtype casting." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "Hidden_states=None in CLIPVisionModel is unrelated to attention kernel divergence under torch.compile." + "reason": "CLIPVisionModel hidden_states regression is unrelated to torch.compile attention-kernel divergence." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Custom processor/tokenizer regression and Gemma-3 eos_token_id config mismatch are distinct regressions." + "reason": "Custom processor/tokenizer access regression is different from Gemma-3 eos_token_id config inconsistency." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "Gemma3TextConfig API mismatch and Mistral-3 image-preprocessor dtype mismatch are separate issues." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "Processor.tokenizer/image-passing regression is not the same as the pipeline translation error-message regression." + "reason": "Processor/tokenizer regression is not the same bug as an inaccurate pipeline error message." }, { "left": "issue:44871", "right": "issue:45584", "accept": false, - "reason": "Gemma-3 eos_token_id config inconsistency is unrelated to Whisper empty-transcription failure after align_special_tokens." - }, - { - "left": "issue:43316", - "right": "issue:43716", - "accept": false, - "reason": "Gemma3TextConfig API discrepancy and Mistral-3 image-preprocessor dtype mismatch are different problems." + "reason": "Gemma-3 eos_token config mismatch is unrelated to Whisper empty-transcription generation after align_special_tokens." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Dequantized model save failures and wrong custom module loading after save_pretrained/from_pretrained are separate save/load bugs." + "reason": "Model save/load problems, but one is dequantized Ministral/Devstral saving and the other is wrong custom-module loading after save_pretrained." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "model_max_length question for DeBERTa-V3 is unrelated to the missing transformers.onnx import error." + "reason": "DeBERTa model_max_length question and missing transformers.onnx module are unrelated." }, { "left": "issue:44521", "right": "issue:47217", "accept": false, - "reason": "Both are multimodal-template related, but all-zero assistant masks and image-placeholder-in-user-text failures are not the same concrete bug." + "reason": "Both are multimodal chat-template related, but one is zero assistant masks and the other is failure when the image placeholder appears in user text; not the same concrete bug." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "patchtsmixer post_init validation and non-persistent buffer junk filling are different v5 issues with different code paths." + "reason": "PatchTSMixer post_init restriction and non-persistent buffer junk are different v5 issues." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Broad custom model initialization breakage does not match GPTNeoXConfig rotary_pct silently resetting on reload." + "reason": "Custom model initialization breakage is unrelated to GPTNeoX rotary_pct not persisting on reload." }, { "left": "issue:46234", "right": "issue:46291", "accept": false, - "reason": "Missing import_utils helper and kernels v0.15 pipeline import failure are unrelated import-time issues." + "reason": "Missing import_utils helper and kernels revision/version import failure are unrelated dependency/import bugs." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "gguf_file/from_pretrained breakage is a different root cause from unrecognized model_type in config.json." + "reason": "gguf_file loading regression is not the same as an unrecognized model due to missing model_type in config." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Gemma-3 eos_token_id mismatch is unrelated to sinq quantized model save/load corruption." + "reason": "Gemma-3 eos_token configuration mismatch is distinct from sinq quantized model save/load breakage." }, { "left": "issue:44464", "right": "issue:46421", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward and cache lazy-init recompilation are different performance/behavior bugs." + "reason": "Compiled-forward chunked generation inconsistency is different from cache lazy init causing spurious recompilation." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "lm_head weight tying in Mistral3ForConditionalGeneration is unrelated to Ministral-3 FlashAttention failure." + "reason": "Weight tying bug in Mistral3ForConditionalGeneration is unrelated to FlashAttention support failure in Ministral-3." }, { "left": "issue:43352", "right": "issue:46640", "accept": false, - "reason": "A specific unsupported-model FlashAttention 2 error is not the same as a broad open Flash Attention 2 issue with no concrete shared code path." + "reason": "A specific unsupported-model FlashAttention limitation is not enough to treat this as the same bug as a vague generic Flash Attention 2 issue." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 breakage for scientists workflows and a janus model generate-images test failure are unrelated." + "reason": "ESM2 workflow breakage and janus image-generation test failure are unrelated model-specific problems." }, { "left": "issue:45803", "right": "issue:46365", "accept": false, - "reason": "Bare except swallowing errors in FuyuBatchFeature is unrelated to Trainer import failure caused by datasets." + "reason": "Bare-except error swallowing in Fuyu preprocessing is unrelated to Trainer import failure caused by datasets." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Video CUDA OOM are different issues; the latter is a memory problem, not the former's architecture bug." + "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Video CUDA OOM are different problems; one is architecture cleanup, the other is resource usage." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Qwen3OmniMoe video-input value error and Qwen2_5_VLProcessor batched chat-template crash are different models and failure modes." } ] }, @@ -11682,7 +11648,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:25251", "issue:27957", @@ -11694,26 +11660,27 @@ "issue:43504", "issue:43525", "issue:43531", - "issue:43540", "issue:43572", "issue:43650", "issue:43673", "issue:43688", "issue:43931", "issue:44060", + "issue:44190", "issue:44242", "issue:44479", - "issue:44514", "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", @@ -11733,204 +11700,217 @@ "issue:47217" ], "soft_pairs": [ - "issue:43540|issue:44514", "issue:43572|issue:46178", - "issue:45381|issue:47217", "issue:44610|issue:46369", + "issue:45381|issue:47217", "issue:44242|issue:46894", "issue:43504|issue:45698", - "issue:45632|issue:46424", "issue:45542|issue:46234", - "issue:44479|issue:47217", + "issue:45632|issue:46424", "issue:44792|issue:44977", - "issue:44792|issue:45600", "issue:44534|issue:44913", + "issue:44792|issue:45600", + "issue:44479|issue:47217", "issue:43525|issue:43531", "issue:25251|issue:27957", "issue:45803|issue:46170", "issue:46082|issue:46697", - "issue:45127|issue:45200", "issue:43931|issue:44060", - "issue:43650|issue:46202", + "issue:45127|issue:45200", "issue:35141|issue:42490", "issue:46693|issue:47217", + "issue:43650|issue:46202", "issue:41950|issue:43475", "issue:43673|issue:44743", "issue:43388|issue:43688", - "issue:44855|issue:45588" + "issue:44855|issue:45588", + "issue:44190|issue:44964", + "issue:44792|issue:45405" ], - "prepared_review_unit_hash": "39229c094dd41f2056e514e340ff35ca2b6c389f897d6b47043be981371ee5f4", + "prepared_review_unit_hash": "27519010fae07f64358665c9a5ab63e5a7c8f6cf933ab0a3960557553a5952c0", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 25, - "serialized_chars": 34711, - "estimated_input_tokens": 8678, - "estimated_eval_tokens": 17612 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35586, + "estimated_input_tokens": 8897, + "estimated_eval_tokens": 18050 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft pairs are mostly superficial text matches across unrelated Transformers bugs; I do not see any true duplicate PR/issue pair here. The strongest anchor among the issues is the broad Qwen VL video regression report, but it still does not duplicate the other items.", - "confidence": 0.93, - "canonical_issue_reason": "Issue 44479 is the broadest, highest-signal report in this set, covering a multi-model video-input regression, so it is the best representative anchor even though the cluster is not a true duplicate group.", + "summary": "This cluster is highly heterogeneous: the items cover many unrelated bugs across configs, loading, pipelines, caching, and model-specific regressions. None of the soft pairs look like true duplicates or the same fixable code-path change.", + "confidence": 0.96, + "canonical_issue_reason": "issue:46693 is the most substantiated and broadly impactful report in the set, with the most discussion and multiple inbound references; it is the strongest representative if one issue must anchor the cluster, though the set is not a true duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "44479 is the most central issue by scope and impact: it describes a regression affecting several Qwen VL variants and has the clearest cross-cutting bug narrative among the candidates.", + "best_issue_reason": "issue:46693 is the best standalone issue candidate because it has the richest report quality and clear regression scope (FA2 long-context performance after 5.2.0).", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43540", - "right": "issue:44514", - "accept": false, - "reason": "Different failure points: video input processing for Qwen3OmniMoe vs batched chat-template handling in Qwen2.5_VLProcessor." - }, { "left": "issue:43572", "right": "issue:46178", "accept": false, - "reason": "Unrelated bugs: missing config field after v5 update vs backend mapping/import error for tensorflow_text." + "reason": "Different failures: missing config field in StableLmConfig vs an import_utils backend-mapping error for tensorflow_text." }, { - "left": "issue:45381", - "right": "issue:47217", + "left": "issue:44610", + "right": "issue:46369", "accept": false, - "reason": "Both are VL issues, but one is wrong vision_position_ids in video handling and the other is an image-placeholder-in-user-text parsing bug." + "reason": "Different model paths and symptoms: OmDet-Turbo processor size mismatch vs SAM 3 LiteText memory overflow." }, { - "left": "issue:44610", - "right": "issue:46369", + "left": "issue:45381", + "right": "issue:47217", "accept": false, - "reason": "Different subsystems and symptoms: processor output size mismatch vs memory overflow in sam3-litetext." + "reason": "Both are VL-related, but one is wrong vision_position_ids for video input while the other is image placeholder handling in user text." }, { "left": "issue:44242", "right": "issue:46894", "accept": false, - "reason": "Load-balancing loss gating is unrelated to CohereASR's training-loss double-shift bug." + "reason": "Both mention loss, but the concrete bug is different: missing load-balancing loss vs a training-loss double-shift issue in CohereASR/Moonshine pattern." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "One is legacy-field model loading for Beit; the other is custom-module resolution after save/load." - }, - { - "left": "issue:45632", - "right": "issue:46424", - "accept": false, - "reason": "Cache-path collision for local trust_remote_code models is unrelated to max_cache_length semantics." + "reason": "Both involve loading, but one is a legacy field in BEiT preset loading and the other is wrong custom module resolution after save_pretrained/from_pretrained." }, { "left": "issue:45542", "right": "issue:46234", "accept": false, - "reason": "TensorFlow backend detection failure is unrelated to a missing import_utils availability helper." + "reason": "Different import problems: tf backend installation handling vs a missing falcon-mamba availability helper in import_utils." }, { - "left": "issue:44479", - "right": "issue:47217", + "left": "issue:45632", + "right": "issue:46424", "accept": false, - "reason": "Both involve VL models, but the concrete bugs differ: a broad video regression versus user-text image-placeholder handling." + "reason": "Different cache subsystems: trust_remote_code cache-path collision vs cache max_cache_length semantics for quantized cache." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Unrelated: Janus image-generation test failure vs Qwen3.5 flash-attention generation failure." + "reason": "Different concrete bugs: a Janus image-generation test failure vs Qwen3.5 flash-attention generation breakage." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Unrelated issues: non-persistent buffer initialization junk vs GPTNeoXConfig rotary_pct not persisting on reload." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "A failed Janus test is not the same bug as stale auto_mappings references to removed Sam3LiteText configs." + "reason": "Different targets: Janus generate-images test failure vs stale auto_mappings references to removed Sam3LiteText configs." }, { - "left": "issue:44534", - "right": "issue:44913", + "left": "issue:44479", + "right": "issue:47217", "accept": false, - "reason": "Non-persistent buffer initialization junk is unrelated to GPTNeoX rotary_pct not persisting on reload." + "reason": "Both concern VL models, but one is a v5.3.0 video-input regression and the other is image placeholder placement in user text." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Missing pad_token_id in Llama4Config and Qwen3-MoE sliding_window behavior are different config bugs." + "reason": "Different model-specific bugs: missing pad_token_id on Llama4Config vs sliding_window handling in Qwen3-MoE." }, { "left": "issue:25251", "right": "issue:27957", "accept": false, - "reason": "Pipeline top_k nesting is unrelated to XLMRoberta Flash Attention 2 support." + "reason": "Completely different areas: pipeline top_k output nesting vs XLM-R with Flash Attention 2." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "Bare except swallowing interrupts in Fuyu tensor conversion is unrelated to the Windows ROCm distributed import issue." + "reason": "Different classes of bugs: bare except swallowing interrupts/errors vs Windows ROCm import/type dependency failure." }, { "left": "issue:46082", "right": "issue:46697", "accept": false, - "reason": "LlamaConfig head_dim validation and Swin2SR over-padding are different model/processor bugs." - }, - { - "left": "issue:45127", - "right": "issue:45200", - "accept": false, - "reason": "LoRA merge collapse with tied embeddings is unrelated to Gemma 4 mm_token_type_ids defaults." + "reason": "Different code paths: LlamaConfig validation for head_dim vs Swin2SR image over-padding in the processor." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Model weight-shape mismatch on load is not the same as an incorrect tied-weights warning." + "reason": "Both are model-loading/weights issues, but one is a shape mismatch on Qwen3-VL while the other is a tied-weights warning bug in Qwen3-Next." }, { - "left": "issue:43650", - "right": "issue:46202", + "left": "issue:45127", + "right": "issue:45200", "accept": false, - "reason": "The titles and bug reports are entirely unrelated; no shared code path is evident." + "reason": "Different bugs: LoRA merge with extended vocab causing collapse vs Gemma 4 mm_token_type_ids defaulting for text-only fine-tuning." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Embedding reinitialization after resize is unrelated to local-path slash import warnings." + "reason": "Different subjects entirely: embedding reinitialization after resizing vs trailing-slash import warning for local model paths." }, { "left": "issue:46693", "right": "issue:47217", "accept": false, - "reason": "FA2 long-context performance regression is a different issue from VL image-placeholder placement failure." + "reason": "Different regressions: FA2 long-context performance degradation vs VL image-placeholder failure." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "No shared bug pattern: a vague 'ADD THE DATA' request vs Gemma token roundtrip failure on \u2581." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Pipeline processor selection bug and missing Sam3VisionEncoderOutput attribute are unrelated." + "reason": "Different bugs in different subsystems: video-classification pipeline processor selection vs missing fpn_position_embeddings on Sam3VisionEncoderOutput." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both mention cache behavior, but one is missing GenerationMixin cache in chunked_prefill while the other resets recurrent states with seq_len>1." + "reason": "Both involve cache/state, but the concrete failures differ: missing GenerationMixin cache in chunked prefill vs recurrent states resetting for seq_len>1." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Metrics gathering drops labels in the last batch; auxiliary-loss normalization is a separate MoE training bug." + "reason": "Different training/eval bugs: gather_for_metrics dropping tuple labels vs auxiliary loss normalization in OLMoE/GPT Oss." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Completely different failure modes: Python 3.13 parsing/IndentationError on import vs AttributeError in flash_attention for sink-less models." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Different surfaces: local dataset loading in a training script vs loading a multimodal model with latest transformers." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Unrelated: Janus image-generation test failure vs dependency version bump for peft." } ] }, @@ -11952,7 +11932,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:25251", "issue:29870", @@ -11979,13 +11959,11 @@ "issue:44869", "issue:44913", "issue:44933", - "issue:44964", "issue:45072", "issue:45245", "issue:45276", "issue:45362", "issue:45381", - "issue:45405", "issue:45406", "issue:45563", "issue:45584", @@ -12001,236 +11979,249 @@ "issue:46897", "issue:46932", "issue:47013", - "issue:47312" + "issue:47312", + "issue:47321" ], "soft_pairs": [ - "issue:44190|issue:44964", - "issue:44792|issue:45405", "issue:29870|issue:35141", - "issue:43827|issue:43994", "issue:45072|issue:46932", + "issue:43827|issue:43994", "issue:45245|issue:46858", "issue:44734|issue:45406", "issue:43335|issue:44855", "issue:43644|issue:44913", + "issue:46399|issue:47312", "issue:44869|issue:46832", "issue:44617|issue:45584", - "issue:46399|issue:47312", "issue:45588|issue:47013", "issue:45632|issue:46617", "issue:41720|issue:45245", + "issue:43827|issue:47321", "issue:45381|issue:46369", "issue:43550|issue:44190", + "issue:39692|issue:47321", "issue:25251|issue:29870", - "issue:44242|issue:46897", "issue:45563|issue:45584", + "issue:44242|issue:46897", "issue:43854|issue:44792", "issue:43065|issue:44610", - "issue:45381|issue:46655", "issue:44933|issue:45362", - "issue:44479|issue:44734", "issue:43335|issue:44190", + "issue:44479|issue:44734", + "issue:45381|issue:46655", + "issue:39692|issue:44190", "issue:44617|issue:46369", "issue:43295|issue:45276", - "issue:39692|issue:44190", "issue:43526|issue:46697" ], - "prepared_review_unit_hash": "1fcdf58ba369334e13f4f353e57b0c21302c73907f558b232c851ff8838e1d1b", + "prepared_review_unit_hash": "33e9619f0be2850dafb51aca8e7ab38f19961bdf77ba9dca8397dd14901b40b0", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36945, - "estimated_input_tokens": 9237, - "estimated_eval_tokens": 18730 + "serialized_chars": 36321, + "estimated_input_tokens": 9081, + "estimated_eval_tokens": 18418 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All proposed soft links look like false positives. The items span unrelated bugs, models, and code paths, so there is no real duplicate cluster here.", + "summary": "All proposed soft links look like false positives: these issues span unrelated bugs across generation, processors, configs, docs, serving, and model-specific regressions.", "confidence": 0.97, - "canonical_issue_reason": "No canonical issue: the candidate links do not describe the same underlying bug or change.", + "canonical_issue_reason": "No single canonical issue fits this cluster; the soft pairs do not describe the same underlying bug, and the set is too heterogeneous to collapse around one issue.", "canonical_pr_reason": null, - "best_issue_reason": "No issue is a good global representative because the set is heterogeneous and the similarity is mostly superficial.", + "best_issue_reason": "If forced to pick a clean standalone report, issue 43827 is the most self-contained, but it is not a good cluster representative because the cluster is not a duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44190", - "right": "issue:44964", - "accept": false, - "reason": "Local dataset loading script failure vs model loading regression; different workflows and failure modes." - }, - { - "left": "issue:44792", - "right": "issue:45405", - "accept": false, - "reason": "Janus test failure vs PEFT version bump; unrelated concerns." - }, { "left": "issue:29870", "right": "issue:35141", "accept": false, - "reason": "Generate-time decoder_attention_mask bug vs embedding resize/post_init bug; different code paths." + "reason": "Different bugs: BART generate() decoder mask init vs token-embedding resize/post_init reinitialization." }, { - "left": "issue:43827", - "right": "issue:43994", + "left": "issue:45072", + "right": "issue:46932", "accept": false, - "reason": "Docs still referencing pipeline() vs incorrect SigLIP2 outputs; one is documentation, the other runtime behavior." + "reason": "Both involve dtype/quantization, but they affect different models and code paths; not the same concrete bug." }, { - "left": "issue:45072", - "right": "issue:46932", + "left": "issue:43827", + "right": "issue:43994", "accept": false, - "reason": "Different dtype/quantization issues in different models; not the same concrete bug." + "reason": "One is a docs reference to removed pipeline APIs; the other is a runtime SigLIP2 inference issue." }, { "left": "issue:45245", "right": "issue:46858", "accept": false, - "reason": "Category-count runtime error vs cache sequence-length tensor arithmetic; unrelated failures." + "reason": "Completely different failures: category-count limit vs static-cache seq-length tensor arithmetic." }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "KV-cache continuation indexing crash vs Gemma4Processor missing attribute; different serve bugs." + "reason": "Both hit transformers serve, but the crashes come from different code paths and different root causes." }, { "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug vs Python 3.13 import/torch.jit parsing error; unrelated." + "reason": "Unrelated issues: SwitchTransformers sparse-layer construction vs a Python 3.13 IndentationError in DebertaV2Model." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer initialization bug vs GPTNeoXConfig reload regression; different persistence issues." + "reason": "One is buffer initialization corruption in v5.0.0; the other is GPTNeoX config reload persistence." + }, + { + "left": "issue:46399", + "right": "issue:47312", + "accept": false, + "reason": "Kernel-related, but one is a missing decorator registration bug and the other is a broad incompatibility report." }, { "left": "issue:44869", "right": "issue:46832", "accept": false, - "reason": "Whisper token-stream decoding crash vs RT-DETR feature-level mismatch; unrelated subsystems." + "reason": "Whisper timestamp decoding crash and RT-DETR feature-level handling are unrelated." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "SAM3 video OOM vs Whisper empty-transcription failure; not the same bug." - }, - { - "left": "issue:46399", - "right": "issue:47312", - "accept": false, - "reason": "Missing decorator for a specific kernel registration vs broad kernels 0.16 incompatibility; too different in scope and cause." + "reason": "Different models and failures: Sam3Video OOM vs Whisper empty-transcription generation bug." }, { "left": "issue:45588", "right": "issue:47013", "accept": false, - "reason": "Flash-attention aux-state None handling vs xLSTM tuple-unpacking crash; different assumptions in different modules." + "reason": "Different code paths: flash-attention auxiliary-state handling vs xLSTM unpacking behavior in train mode." }, { "left": "issue:45632", "right": "issue:46617", "accept": false, - "reason": "Cache leaf-directory collision vs FileNotFoundError loading from cache; related area, but not the same underlying defect." + "reason": "Both touch trust_remote_code cache loading, but one is a leaf-directory collision and the other is a FileNotFoundError report." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Qwen3 auto device-map cuda assert vs category-limit runtime error; unrelated." + "reason": "Unrelated: Qwen3 auto device mapping cuda assert vs an embedding/category cardinality error." + }, + { + "left": "issue:43827", + "right": "issue:47321", + "accept": false, + "reason": "Docs pipeline-removal issue is unrelated to AMD quark class updates." }, { "left": "issue:45381", "right": "issue:46369", "accept": false, - "reason": "Qwen2.5-VL video position IDs bug vs SAM3 memory overflow; different models and causes." + "reason": "Different SAM/Qwen video-path regression vs SAM3 litetext memory overflow." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "torch.compile/SDPA model failure vs dataset-loading script issue; no overlap in root cause." + "reason": "torch.compile SDPA failure and local dataset loading script error are unrelated." }, { - "left": "issue:25251", - "right": "issue:29870", + "left": "issue:39692", + "right": "issue:47321", "accept": false, - "reason": "Pipeline output shape change vs BART generate UnboundLocalError; different behaviors and code paths." + "reason": "SigLIP2 docs/model mismatch is unrelated to AMD quark class update timing." }, { - "left": "issue:44242", - "right": "issue:46897", + "left": "issue:25251", + "right": "issue:29870", "accept": false, - "reason": "Router-logit load balancing loss omission vs Florence2 double-shift training loss bug; both loss-related but not the same bug." + "reason": "Pipeline top_k nesting behavior is unrelated to BART decoder_attention_mask initialization." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "Stale warning in paged generate vs Whisper empty transcription crash; different generation issues." + "reason": "Paged generate warning logic and Whisper empty-transcription failure are different bugs." + }, + { + "left": "issue:44242", + "right": "issue:46897", + "accept": false, + "reason": "Both concern training loss, but they are different model-specific loss bugs with different mechanics." }, { "left": "issue:43854", "right": "issue:44792", "accept": false, - "reason": "Model loading/unit-test failure for GLM-4.7-Flash vs Janus image generation test failure; unrelated." + "reason": "Different failures: GLM-4.7-Flash test loading vs Janus image generation test failure." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder vs OmDet-Turbo input-size mismatch; different model components." + "reason": "Sam3PixelDecoder dummy Conv2d and OmDet-Turbo input-size mismatch are unrelated." }, { - "left": "issue:45381", - "right": "issue:46655", + "left": "issue:44933", + "right": "issue:45362", "accept": false, - "reason": "Qwen2.5-VL vision_position_ids regression vs SmolVLM padding bug; both video-related but distinct root causes." + "reason": "Missing import from image_utils is unrelated to Qwen3.5 chat crashes." }, { - "left": "issue:44933", - "right": "issue:45362", + "left": "issue:43335", + "right": "issue:44190", "accept": false, - "reason": "Missing import from image_utils vs Qwen3.5 chat crash; unrelated." + "reason": "SwitchTransformers config bug and image-classification dataset loading are unrelated." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Video-input regression across Qwen models vs serve KV-cache indexing crash; different product surfaces." + "reason": "Both are regressions, but one is video-input handling and the other is serving KV-cache indexing." }, { - "left": "issue:43295", - "right": "issue:45276", + "left": "issue:45381", + "right": "issue:46655", "accept": false, - "reason": "Processor tokenizer API regression vs Gemma4 resize_token_embeddings behavior; different APIs and failure modes." + "reason": "Both involve video processors, but the bugs are different: vision_position_ids vs padding calculation." }, { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "SigLIP2 docs example/model-processor mismatch vs local dataset loading script failure; unrelated." + "reason": "SigLIP2 docs example problems do not match the local dataset loading failure." + }, + { + "left": "issue:44617", + "right": "issue:46369", + "accept": false, + "reason": "Both mention memory pressure, but they are different SAM3 models and different failure modes." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Processor.tokenizer regression is unrelated to Gemma4 embedding-resize propagation." }, { "left": "issue:43526", "right": "issue:46697", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug vs Swin2SR over-padding bug; different image-processor defects." + "reason": "Different image-processor bugs: label reduction vs over-padding on divisible sizes." } ] }, @@ -12252,24 +12243,22 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "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:43540", "issue:43643", + "issue:43644", "issue:43673", "issue:43854", + "issue:43994", "issue:44112", "issue:44246", "issue:44279", @@ -12290,7 +12279,6 @@ "issue:45381", "issue:45542", "issue:45584", - "issue:45593", "issue:45600", "issue:45632", "issue:45803", @@ -12300,248 +12288,250 @@ "issue:46489", "issue:46752", "issue:46858", + "issue:47145", "issue:47217", - "issue:47312" + "issue:47312", + "issue:47321" ], "soft_pairs": [ "issue:45127|issue:45357", "issue:43673|issue:46421", "issue:44779|issue:46489", "issue:45276|issue:45584", + "issue:47145|issue:47321", "issue:43540|issue:44734", "issue:44112|issue:45600", "issue:43854|issue:44964", "issue:43643|issue:44496", "issue:43317|issue:47312", - "issue:45290|issue:46752", "issue:44462|issue:45907", + "issue:43994|issue:47321", "issue:44521|issue:45381", + "issue:43644|issue:47321", + "issue:45290|issue:46752", "issue:27957|issue:35141", "issue:44246|issue:45003", "issue:43673|issue:46858", "issue:44737|issue:45803", "issue:44112|issue:44964", "issue:44964|issue:45600", - "issue:42175|issue:44279", "issue:44938|issue:45542", - "issue:42796|issue:44521", + "issue:42175|issue:44279", "issue:32090|issue:35141", "issue:43404|issue:45216", + "issue:42796|issue:44521", "issue:44496|issue:47217", "issue:45632|issue:46234", - "issue:43531|issue:45600", - "issue:43335|issue:45593", - "issue:36246|issue:41628", - "issue:36246|issue:38617" + "issue:43531|issue:45600" ], - "prepared_review_unit_hash": "6b5a8502f81717a58d3ec53ccaeef949bd6e20d49fb26c5d5d26dbbeec066451", + "prepared_review_unit_hash": "863ec8cb3dfc192419b857b61f5c64db56155b05600c0d53caf96e396351f4b6", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36638, - "estimated_input_tokens": 9160, - "estimated_eval_tokens": 18576 + "serialized_chars": 35896, + "estimated_input_tokens": 8974, + "estimated_eval_tokens": 18204 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. I only found one true duplicate-style pair: the DeepSeek tokenizer regression in v5+. The rest share broad subsystems (chat templates, cache, imports, multimodal, CI, loading) but describe different concrete failures and should stay separate.", - "confidence": 0.91, - "canonical_issue_reason": "issue:46489 is the clearest representative of the one real duplicate set: it names the affected model family, the v5+ regression, and explicitly frames the problem as a gap in the prior fix.", + "summary": "Mostly unrelated issues; only the DeepSeek tokenizer regression pair looks like the same underlying bug.", + "confidence": 0.84, + "canonical_issue_reason": "issue:46489 is the clearest representative of the duplicate DeepSeek tokenizer regression, and it explicitly ties the failure to the earlier PR gap.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46489 is the most actionable issue here because it gives the specific regression, affected model, and prior-fix context needed to route and fix the bug.", + "best_issue_reason": "issue:46489 has the most specific repro framing and the strongest linkage to the prior fix, so it is the best issue anchor here.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "Both mention Qwen3.5/save_pretrained, but the failures are different: LoRA+extended-vocab collapse vs incorrect visual encoder keys." + "reason": "Different regressions: LoRA/extended-vocab collapse vs incorrect visual-encoder keys on save_pretrained." }, { "left": "issue:43673", "right": "issue:46421", "accept": false, - "reason": "Both involve cache/chunked prefill, but one is a missing GenerationMixin cache and the other is spurious recompilation from lazy init; different code paths." + "reason": "Both involve cache/chunked prefill, but the concrete failures and code paths differ." }, { "left": "issue:44779", "right": "issue:46489", "accept": true, - "reason": "Same underlying DeepSeek tokenizer regression on v5+: both report incorrect tokenizer output, and 46489 explicitly calls out a gap in the earlier fix." + "reason": "Same DeepSeek tokenizer regression in v5+, with 46489 reading like a follow-up on the same bug/fix gap." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Resize-token-embeddings weight handling and Whisper empty-transcription generation are unrelated bugs." + "reason": "Unrelated: gemma4 resize_token_embeddings vs Whisper empty-transcription generation." + }, + { + "left": "issue:47145", + "right": "issue:47321", + "accept": false, + "reason": "Different areas entirely: FP8 checkpoint loading vs AMD quark class metadata." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-input validation and serve KV-cache continuation indexing are different failures." + "reason": "Different symptoms and paths: video input parsing vs serve/KV-cache continuation indexing." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "Both hit CI, but one is a stale GraniteSpeech test and the other is broken auto_mappings references; not the same bug." + "reason": "Both are CI-related, but they describe unrelated failures in different tests/code." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "Different model-loading failures for different models; no shared concrete code-path bug is evident." + "reason": "Both are model-loading failures, but for different models and different causes." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "One is missing fields from AutoConfig with trust_remote_code, the other is an unrecognized model/config model_type problem." + "reason": "Different config-loading bugs: missing fields with trust_remote_code vs missing model_type." }, { "left": "issue:43317", "right": "issue:47312", "accept": false, - "reason": "Dequantized model offload failure and kernel compatibility are separate issues." + "reason": "Different bugs: dequantized model offload loading vs kernel compatibility." }, { - "left": "issue:45290", - "right": "issue:46752", + "left": "issue:44462", + "right": "issue:45907", "accept": false, - "reason": "Both are apply_chat_template edge cases, but tool-call/no-content handling and empty-list IndexError are different concrete bugs." + "reason": "Different repo/file-resolution bugs: tokenizer.json selection vs list_repo_files revision forwarding." }, { - "left": "issue:44462", - "right": "issue:45907", + "left": "issue:43994", + "right": "issue:47321", "accept": false, - "reason": "Tokenizer.json lookup and list_repo_files revision forwarding are not the same failure mode." + "reason": "Unrelated: SigLIP2 inference correctness vs AMD quark class update." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "Assistant-mask generation and Qwen2.5-VL video vision_position_ids are distinct multimodal bugs." + "reason": "Both are multimodal, but assistant mask generation and vision_position_ids are different bugs." + }, + { + "left": "issue:43644", + "right": "issue:47321", + "accept": false, + "reason": "Unrelated: buffer initialization corruption vs AMD quark class update." + }, + { + "left": "issue:45290", + "right": "issue:46752", + "accept": false, + "reason": "Both crash in apply_chat_template, but on different inputs and different failure modes." }, { "left": "issue:27957", "right": "issue:35141", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta and embedding reinitialization after resize are unrelated." + "reason": "Different model/features: Flash Attention 2 support vs embedding reinitialization after resize." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import slowness and unsafe sys.modules access are related to imports only at a very broad level, not the same bug." + "reason": "Different issues: import slowness vs unsafe sys.modules access." }, { "left": "issue:43673", "right": "issue:46858", "accept": false, - "reason": "Both touch cache internals, but one is cache missing in chunked prefill and the other is device-tensor seq-length breaking static-cache arithmetic." + "reason": "Both are cache-related, but one is missing cache state and the other is device-tensor slice arithmetic." }, { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "XLNet CPU tensor allocation and a bare-except in Fuyu tensor conversion are unrelated." + "reason": "Different bugs: XLNet device placement vs Fuyu exception handling." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "A GraniteSpeech CI test and a Phi-4 multimodal loading failure are separate issues." + "reason": "Unrelated CI test failure vs model-loading regression." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "Model loading and stale auto_mappings references are different problems." - }, - { - "left": "issue:42175", - "right": "issue:44279", - "accept": false, - "reason": "Packaging extras missing TensorFlow backend vs a generic dependency issue are not the same concrete bug." + "reason": "Different areas: Phi-4 multimodal loading vs removed config references in auto_mappings." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 import failure and undefined tf backend detection are distinct import-time issues." + "reason": "Different environment/backend issues: Python 3.14 import failure vs missing TensorFlow backend." }, { - "left": "issue:42796", - "right": "issue:44521", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "Tokenizer BPE marker handling and multimodal assistant mask generation are unrelated." + "reason": "Different dependency problems with different packaging symptoms." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer broadcast type errors and embedding reinitialization are unrelated." + "reason": "Trainer broadcast NoneType error is unrelated to embedding resize/reinit." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Weight tying in Mistral3 and Qwen3.5 save_pretrained regression are different model-specific bugs." + "reason": "Different model-save/load bugs: lm_head tying vs incorrect checkpoint contents." + }, + { + "left": "issue:42796", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer BPE-marker handling is unrelated to multimodal assistant-mask generation." }, { "left": "issue:44496", "right": "issue:47217", "accept": false, - "reason": "Unrecognized config/model-type and VL image-placeholder handling are different failures." + "reason": "Different config/model-loading failures: missing model_type vs image-placeholder handling." }, { "left": "issue:45632", "right": "issue:46234", "accept": false, - "reason": "Cache-path collision with trust_remote_code and a missing import-utils helper are unrelated." + "reason": "Different trust_remote_code cache-path issue vs missing import_utils symbol." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding window behavior and removed Sam3LiteText config references are different issues." - }, - { - "left": "issue:43335", - "right": "issue:45593", - "accept": false, - "reason": "Sparse-layer creation in SwitchTransformersConfig and D-FINE auxiliary-loss behavior are unrelated." - }, - { - "left": "issue:36246", - "right": "issue:41628", - "accept": false, - "reason": "Missing export for Qwen2_5_VLImageProcessor and missing AutoImageProcessor import are different API surface issues." - }, - { - "left": "issue:36246", - "right": "issue:38617", - "accept": false, - "reason": "Different missing imports from different modules; no same underlying bug is shown." + "reason": "Different bugs: sliding_window behavior in Qwen3-MoE vs stale CI config references." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative about the soft edges. The DeepSeek v5+ pair is a plausible duplicate-style match, and the other pairwise dismissals are appropriately cautious given the issue titles." + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft-edge pair, issue:44779 vs issue:46489, is well supported by both titles referring to the same DeepSeek tokenizer regression across v5+. The remaining false verdicts are cautious and do not overstate linkage. No mergeability concerns apply here because this is an issue-pair evaluation, not a PR-pair merge decision." }, "evaluator_used": true, "retried": false, @@ -12560,12 +12550,16 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ + "issue:36246", + "issue:38617", + "issue:41628", "issue:41950", "issue:42907", "issue:43295", "issue:43299", + "issue:43335", "issue:43381", "issue:43441", "issue:43479", @@ -12585,7 +12579,6 @@ "issue:44336", "issue:44496", "issue:44683", - "issue:44779", "issue:44792", "issue:44933", "issue:44987", @@ -12594,7 +12587,6 @@ "issue:45216", "issue:45335", "issue:45375", - "issue:45405", "issue:45446", "issue:45584", "issue:45588", @@ -12605,260 +12597,264 @@ "issue:46424", "issue:46506", "issue:46515", - "issue:46614", "issue:46752", "issue:46932", - "issue:47085", "issue:47217" ], "soft_pairs": [ - "issue:43295|issue:45335", + "issue:43335|issue:45593", + "issue:36246|issue:41628", + "issue:36246|issue:38617", "issue:44496|issue:45042", + "issue:43295|issue:45335", "issue:45083|issue:46506", "issue:44295|issue:46506", "issue:41950|issue:44016", - "issue:43673|issue:46424", "issue:43976|issue:44987", + "issue:43673|issue:46424", "issue:43582|issue:46932", - "issue:44496|issue:45759", "issue:44683|issue:45446", + "issue:44496|issue:45759", "issue:43526|issue:43901", "issue:46082|issue:46121", "issue:43673|issue:44336", "issue:43381|issue:45593", "issue:44016|issue:44220", "issue:42907|issue:43441", - "issue:44327|issue:46752", - "issue:43650|issue:44220", "issue:44016|issue:44062", + "issue:43650|issue:44220", + "issue:44327|issue:46752", "issue:44792|issue:46424", - "issue:44279|issue:46515", "issue:43441|issue:45588", - "issue:45335|issue:45584", + "issue:44279|issue:46515", "issue:43976|issue:44279", "issue:43824|issue:44933", + "issue:45335|issue:45584", "issue:43479|issue:45375", - "issue:43901|issue:44496", "issue:43299|issue:45216", - "issue:43901|issue:47217", - "issue:46614|issue:47085", - "issue:43295|issue:44779", - "issue:44279|issue:45405" + "issue:43901|issue:44496", + "issue:43901|issue:47217" ], - "prepared_review_unit_hash": "c884598452b6c0dd8a58470996df2d7e65341e128dd09543bccc82b347144637", + "prepared_review_unit_hash": "43c09fa5e7621de44b3e60f07f5af79478b189f4a91127abacc07cc9da96970e", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37169, - "estimated_input_tokens": 9293, - "estimated_eval_tokens": 18842 + "serialized_chars": 37308, + "estimated_input_tokens": 9327, + "estimated_eval_tokens": 18910 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous and the soft edges are mostly superficial similarity matches, not true duplicates. A small subset shares a broad multimodal/loading theme, but each pair still points to a different concrete failure mode or code path.", - "confidence": 0.87, - "canonical_issue_reason": "Issue 43295 is the broadest and earliest multimodal regression report here, centered on processor/tokenizer/image handling. It is the best anchor for this cluster, even though most other items are distinct bugs.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bug reports. The soft pairs are only superficially similar (imports, loading regressions, cache issues, multimodal configs), but they do not look like the same underlying code-path bug, so I rejected all candidate merges.", + "confidence": 0.22, + "canonical_issue_reason": "issue:43824 is the strongest standalone report in the set: it is a concrete import failure for a public class, has discussion and inbound references, and is actionable without needing broader context.", "canonical_pr_reason": null, - "best_issue_reason": "43295 is the most representative issue for the nearby vision-language/processor regression subgroup and has the clearest general symptom framing.", + "best_issue_reason": "issue:43824 is the best overall issue to anchor this cluster because it has the clearest repro signal and strongest external interest among the listed bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43295", - "right": "issue:45335", + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "Different models and failure modes: sparse-layer config creation vs missing auxiliary losses." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and likely different export regressions." + }, + { + "left": "issue:36246", + "right": "issue:38617", "accept": false, - "reason": "Different code paths: processor/tokenizer image-handling regression vs resize_token_embeddings not updating decoder embeddings." + "reason": "Unrelated missing names from different modules; no shared code path is evident." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "One is model/config loading failure; the other is a PIL image-processor dependency regression on torchvision." + "reason": "Model-loading/config resolution vs image processor torchvision dependency; not the same bug." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Different subsystems: processor/tokenizer regression vs embedding resize behavior." }, { "left": "issue:45083", "right": "issue:46506", "accept": false, - "reason": "qwen3_omni_moe helper behavior is unrelated to PPDocLayoutV3 CPU/CUDA output divergence." + "reason": "Different bugs: qwen3_omni_moe feature-length helper vs CPU/CUDA divergence in PPDocLayoutV3." }, { "left": "issue:44295", "right": "issue:46506", "accept": false, - "reason": "Reading a registered buffer error is a different bug from CPU vs CUDA result mismatch." + "reason": "Position_ids buffer access issue is unrelated to cross-device output differences." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Video-classification image-processor lookup and a notebook syntax error are unrelated." + "reason": "Pipeline image-processor lookup bug and notebook syntax error are unrelated." }, { - "left": "issue:43673", - "right": "issue:46424", + "left": "issue:43976", + "right": "issue:44987", "accept": false, - "reason": "Both mention cache, but one is a missing GenerationMixin cache in chunked prefill and the other is cache_length handling for quantized cache only." + "reason": "A Python-version compatibility problem is not the same as a model-loading failure for one repo." }, { - "left": "issue:43976", - "right": "issue:44987", + "left": "issue:43673", + "right": "issue:46424", "accept": false, - "reason": "Python-version compatibility issue vs a concrete model-loading regression for physical-intelligence/fast." + "reason": "Both involve cache, but one is missing cache support in generation and the other is a max-length config limitation." }, { "left": "issue:43582", "right": "issue:46932", "accept": false, - "reason": "Apple Silicon TypeError in allocator warmup is unrelated to Gemma4 quantized dtype-casting." + "reason": "Apple Silicon allocator TypeError and Gemma4 dtype casting are different runtime issues." }, { - "left": "issue:44496", - "right": "issue:45759", + "left": "issue:44683", + "right": "issue:45446", "accept": false, - "reason": "Unrecognized model/config loading is different from AutoModelForCausalLM.from_config not unwrapping text_config." + "reason": "Compiled flex_attention failure on new torch versions is separate from a version-check/import bug." }, { - "left": "issue:44683", - "right": "issue:45446", + "left": "issue:44496", + "right": "issue:45759", "accept": false, - "reason": "Compiled flex_attention failure on torch>=2.9 is not the same as an incorrect version check for AuxRequest import." + "reason": "Both are config/model-loading related, but one is an unrecognized model and the other is unwrapping text_config for composite multimodal configs." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels bug is unrelated to TextClassificationPipeline docs mentioning return_all_scores." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to a text-classification docs/behavior mismatch." }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "LlamaConfig head_dim validation is a different issue from rope-validation ignore_keys list handling." + "reason": "LlamaConfig head_dim validation and rope-parameter JSON handling are different validation paths." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Cache-missing during chunked_prefill and ANSI codes in loading_report are unrelated." + "reason": "Generation cache behavior and ANSI output in loading_report do not share the same underlying problem." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient checkpointing in eval mode is unrelated to D-FINE auxiliary loss behavior when denoising is off." + "reason": "Gradient checkpointing eval-mode restriction is unrelated to missing auxiliary losses." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error and _torch_extract_fbank_features() are different failures." + "reason": "Notebook syntax error and fbank feature extraction are unrelated." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Saving dequantized Ministrals and FlashAttention failure are different Ministral bugs with different symptoms." + "reason": "Both mention Ministral, but saving dequantized models and FlashAttention loading failures are different code paths." }, { - "left": "issue:44327", - "right": "issue:46752", + "left": "issue:44016", + "right": "issue:44062", "accept": false, - "reason": "QA decode_spans top-k bounds error is unrelated to apply_chat_template on an empty list." + "reason": "Notebook syntax and AddedToken keyword conflict are unrelated." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "Placeholder issue title with no concrete bug is not the same as audio feature extraction failure." + "reason": "A placeholder issue title and an audio feature extraction bug are not the same change or bug." }, { - "left": "issue:44016", - "right": "issue:44062", + "left": "issue:44327", + "right": "issue:46752", "accept": false, - "reason": "Notebook syntax error is unrelated to AddedToken special kwarg duplication." + "reason": "QA span decoding top-k bounds error and empty chat-template input error are different edge cases." }, { "left": "issue:44792", "right": "issue:46424", "accept": false, - "reason": "janus generate_images test failure is not the same as cache max_cache_length semantics." + "reason": "Janus image-generation test failure is not the same as a cache-length configuration bug." }, { - "left": "issue:44279", - "right": "issue:46515", + "left": "issue:43441", + "right": "issue:45588", "accept": false, - "reason": "Generic dependency issue does not match a specific torch 2.4.1 incompatibility." + "reason": "FlashAttention failure for Ministral-3 and s_aux=None crash in flash_attention integration are distinct problems." }, { - "left": "issue:43441", - "right": "issue:45588", + "left": "issue:44279", + "right": "issue:46515", "accept": false, - "reason": "Both touch flash attention, but one is a Ministral model-specific failure and the other is an s_aux=None crash in integrations/flash_attention.py." + "reason": "General dependency issue and torch 2.4.1 incompatibility are not the same defect." }, { - "left": "issue:45335", - "right": "issue:45584", + "left": "issue:43976", + "right": "issue:44279", "accept": false, - "reason": "resize_token_embeddings behavior is unrelated to Whisper empty-transcription generation after align_special_tokens." + "reason": "Python support regression and dependency issue are unrelated." }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Both are import-related, but one is a missing exported VL model class and the other is a nonexistent image_utils import." + "reason": "Different missing imports from different modules; both are import errors but not the same symbol or path." }, { - "left": "issue:43479", - "right": "issue:45375", + "left": "issue:45335", + "right": "issue:45584", "accept": false, - "reason": "Default config initialization with None is different from strict dropping deepstack_visual_indexes." + "reason": "Embedding resize behavior and Whisper empty-transcription generation failure are different bugs." }, { - "left": "issue:43901", - "right": "issue:44496", + "left": "issue:43479", + "right": "issue:45375", "accept": false, - "reason": "Docs drift in TextClassificationPipeline is not the same as a model loading/config error." + "reason": "Default multimodal config initialization and missing deepstack_visual_indexes handling are different config issues." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Model loading regression and save_pretrained checkpoint regression are different code paths." + "reason": "Qwen3VL loading regression and Qwen3.5 save_pretrained regression are opposite directions and different code paths." }, { "left": "issue:43901", - "right": "issue:47217", - "accept": false, - "reason": "Pipeline docs mismatch is unrelated to VL models failing when the image placeholder appears in user text." - }, - { - "left": "issue:46614", - "right": "issue:47085", - "accept": false, - "reason": "Stale phonemizer backend state is unrelated to Rust tokenizer thread safety." - }, - { - "left": "issue:43295", - "right": "issue:44779", + "right": "issue:44496", "accept": false, - "reason": "Processor/tokenizer image handling regression is not the same as Deepseek tokenizer output changes." + "reason": "Docs/behavior mismatch in a pipeline is unrelated to an unrecognized model config error." }, { - "left": "issue:44279", - "right": "issue:45405", + "left": "issue:43901", + "right": "issue:47217", "accept": false, - "reason": "A generic dependency issue is not the same as bumping MIN_PEFT_VERSION beyond what is released on PyPI." + "reason": "Text-classification docs mismatch and VL image-placeholder failure are unrelated." } ] }, @@ -12880,11 +12876,12 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:39692", "issue:41950", "issue:42796", + "issue:43295", "issue:43299", "issue:43381", "issue:43577", @@ -12892,15 +12889,14 @@ "issue:43650", "issue:43653", "issue:43854", - "issue:43873", "issue:43901", "issue:43931", "issue:43976", "issue:44016", "issue:44075", "issue:44242", + "issue:44279", "issue:44291", - "issue:44336", "issue:44351", "issue:44368", "issue:44462", @@ -12920,243 +12916,243 @@ "issue:45593", "issue:45812", "issue:45823", - "issue:45907", "issue:46082", "issue:46123", "issue:46209", "issue:46369", "issue:46399", "issue:46512", - "issue:46851", + "issue:46614", "issue:47000", + "issue:47085", "issue:47145" ], "soft_pairs": [ - "issue:39692|issue:43901", "issue:41950|issue:45020", + "issue:43295|issue:44779", + "issue:44279|issue:45405", + "issue:39692|issue:43901", + "issue:46614|issue:47085", "issue:46123|issue:46369", "issue:42796|issue:43650", "issue:43577|issue:45216", "issue:44779|issue:45812", "issue:44075|issue:44368", - "issue:44466|issue:45276", "issue:43976|issue:45405", + "issue:44466|issue:45276", "issue:41950|issue:45823", "issue:43381|issue:47000", - "issue:43653|issue:45593", "issue:43299|issue:45357", + "issue:43653|issue:45593", "issue:46082|issue:46399", "issue:46082|issue:46512", "issue:44351|issue:45278", "issue:44462|issue:45081", "issue:43644|issue:47145", - "issue:43650|issue:46209", "issue:43299|issue:43854", - "issue:43931|issue:47145", "issue:44242|issue:45200", + "issue:43650|issue:46209", + "issue:43931|issue:47145", "issue:44016|issue:44291", "issue:44683|issue:45542", - "issue:45588|issue:47000", - "issue:45081|issue:45907", - "issue:43873|issue:47145", - "issue:44336|issue:46851" + "issue:45588|issue:47000" ], - "prepared_review_unit_hash": "bd0661070b2b0e0e614f3cd9c688a0347a8cadbc92f9404d2bcfcbb2ee2ea4c2", + "prepared_review_unit_hash": "861efadf39029a645eb776f1209d8156f2a8f9dadf945ba3ce31afee107ffc30", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36247, - "estimated_input_tokens": 9062, - "estimated_eval_tokens": 18380 + "serialized_chars": 36259, + "estimated_input_tokens": 9065, + "estimated_eval_tokens": 18386 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is semantically noisy: most items are different regressions across tokenizers, model loading/saving, pipelines, quantization, docs, and notebook/test issues. None of the soft pairs looks like the same concrete bug or change, so no dedupe merges are justified.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue cleanly represents this cluster. If forced, issue 45020 is the broadest regression umbrella, but it still does not subsume the rest.", + "summary": "This is a heterogeneous issue cluster with no strong duplicate family; the soft pairs mostly share only broad subsystem or version-regression themes, not the same concrete bug. No PRs are present. I\u2019d keep issue #45020 as the most representative standalone issue because it is a broad, high-signal regression report.", + "confidence": 0.72, + "canonical_issue_reason": "Issue #45020 is the best canonical-style anchor here: it describes a clear, broad regression affecting recent transformers versions and model loading, which is more representative than the narrower model-specific, docs-only, or dependency-only reports in the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45020 is the closest broad representative, but only as a loose umbrella; the set is too heterogeneous to treat it as a true duplicate target.", + "best_issue_reason": "Issue #45020 is the strongest single issue to keep: it is concrete, high-impact, and broadly useful as a reference for version-induced breakages, unlike the other items that are either unrelated or too specialized to serve as a cluster representative.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Different bugs: video-classification pipeline processor lookup vs remote_code model-loading regressions." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Different code paths: processor.tokenizer regression vs Deepseek tokenizer output corruption." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Dependency-related reports, but one is a vague dependency issue and the other is an unreleased PEFT version bump." + }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "Both are docs-related, but one is a SigLIP2 example error and the other is a TextClassificationPipeline API note; different problems." + "reason": "Both are docs-related, but they describe different example/API problems and not the same underlying bug." }, { - "left": "issue:41950", - "right": "issue:45020", + "left": "issue:46614", + "right": "issue:47085", "accept": false, - "reason": "Video-classification processor lookup and remote_code breakage are unrelated failure modes." + "reason": "Stale espeak backend state is unrelated to Rust tokenizer thread-safety regressions." }, { "left": "issue:46123", "right": "issue:46369", "accept": false, - "reason": "Final partial-batch dropping and sam3-litetext memory overflow are different code paths and symptoms." + "reason": "Final partial-batch handling and model memory overflow are distinct failures." }, { "left": "issue:42796", "right": "issue:43650", "accept": false, - "reason": "A tokenizer BPE-marker bug and an empty placeholder issue are not the same underlying bug." + "reason": "Tokenizer marker handling and a placeholder/non-issue report are not the same bug." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "One is dtype propagation on BLIP2 load; the other is incorrect Qwen3.5 save_pretrained output." + "reason": "Both involve model-state correctness, but one is dtype propagation in Blip2 and the other is Qwen3.5 save_pretrained serialization." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Both hit tokenizers, but Deepseek incorrect tokenization and Granite token IDs regressions are model-specific and likely separate fixes." + "reason": "Both are tokenizer regressions, but they affect different model families and likely different fixes." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Unused SGD args and a Qwen3.5 config warning are unrelated." + "reason": "Optimizer argument handling and a tie_word_embeddings warning are unrelated." }, { - "left": "issue:44466", - "right": "issue:45276", + "left": "issue:43976", + "right": "issue:45405", "accept": false, - "reason": "Serialization of tied lm_head weights and Gemma4 embedding-resize propagation are distinct issues." + "reason": "Python version compatibility and a PEFT version bump are separate dependency problems." }, { - "left": "issue:43976", - "right": "issue:45405", + "left": "issue:44466", + "right": "issue:45276", "accept": false, - "reason": "Python version compatibility and an unmet MIN_PEFT_VERSION bump are different dependency problems." + "reason": "lm_head serialization inconsistency is not the same as Gemma4 embedding resizing not propagating." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "Pipeline processor mismatch and Gemma4 PLE device placement are separate bugs." + "reason": "Different subsystems: pipeline processor lookup vs Gemma4 device placement during forward." }, { "left": "issue:43381", "right": "issue:47000", "accept": false, - "reason": "Gradient checkpointing in eval mode and RTDetrLoss infeasible cost matrix are unrelated." + "reason": "Gradient checkpointing in eval mode and RTDetr loss infeasibility are unrelated training/runtime bugs." }, { - "left": "issue:43653", - "right": "issue:45593", + "left": "issue:43299", + "right": "issue:45357", "accept": false, - "reason": "BigBirdTokenizer special-token registration and D-FINE auxiliary-loss behavior are different subsystems." + "reason": "Related model family, but the failures are different save/load paths and not mergeable as one fix." }, { - "left": "issue:43299", - "right": "issue:45357", + "left": "issue:43653", + "right": "issue:45593", "accept": false, - "reason": "Both involve Qwen3 family regressions, but one is model loading and the other is save_pretrained key serialization." + "reason": "Tokenizer special-token registration and D-FINE auxiliary loss logic are unrelated." }, { "left": "issue:46082", "right": "issue:46399", "accept": false, - "reason": "A config/head_dim validation issue is not the same as a Qwen3.5 rotary-position registration crash." + "reason": "LlamaConfig validation and Qwen3.5 kernel registration are different issues." }, { "left": "issue:46082", "right": "issue:46512", "accept": false, - "reason": "Explicit head_dim rejection and stale config.dtype after from_config are separate configuration bugs." + "reason": "One is config validation, the other is stale dtype metadata after from_config." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "A missing HybridCache import is too specific to merge with a broad import-error report." + "reason": "Both mention imports, but one is a missing symbol and the other is a broad post-upgrade import breakage." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json and Mistral regex patch crashes are different tokenizer-loading paths." + "reason": "AutoTokenizer ignoring tokenizer.json and a Mistral regex patch crash are different tokenizer-loading failures." }, { "left": "issue:43644", "right": "issue:47145", "accept": false, - "reason": "Non-persistent buffer junk initialization and missing FP8 weight_scale application are distinct corruption mechanisms." + "reason": "Both concern checkpoint/state loading, but one is junk non-persistent buffers and the other is missing FP8 weight_scale application." }, { - "left": "issue:43650", - "right": "issue:46209", + "left": "issue:43299", + "right": "issue:43854", "accept": false, - "reason": "A placeholder issue and an FP8 inference failure are unrelated." + "reason": "Different model-loading failures affecting different model families and code paths." }, { - "left": "issue:43299", - "right": "issue:43854", + "left": "issue:44242", + "right": "issue:45200", "accept": false, - "reason": "Different models and different failure modes; no shared concrete bug." + "reason": "Load-balancing loss omission and mm_token_type_ids defaulting are different model behaviors." }, { - "left": "issue:43931", - "right": "issue:47145", + "left": "issue:43650", + "right": "issue:46209", "accept": false, - "reason": "Shape-mismatch loading for Qwen3-VL and FP8 scale omission are separate loading bugs." + "reason": "A placeholder issue and an FP8 inference-on-B200 bug are unrelated." }, { - "left": "issue:44242", - "right": "issue:45200", + "left": "issue:43931", + "right": "issue:47145", "accept": false, - "reason": "Missing load-balancing loss and Gemma4 mm_token_type_ids defaults are unrelated training/model-input issues." + "reason": "Weight-shape mismatch on Qwen3-VL is unrelated to FP8 compressed-tensors weight_scale loading." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error and init_empty_weights argument handling are not the same bug." + "reason": "Notebook syntax error and init_empty_weights argument handling are completely different problems." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "flex_attention on torch>=2.9 and TensorFlow/tensorboard backend setup are unrelated." + "reason": "Compiled flex_attention on torch>=2.9 and a missing TensorFlow backend dependency are unrelated." }, { "left": "issue:45588", "right": "issue:47000", "accept": false, - "reason": "flash_attention s_aux handling and RTDetrLoss infeasible cost matrix are different crashes." - }, - { - "left": "issue:45081", - "right": "issue:45907", - "accept": false, - "reason": "Mistral tokenizer regex failure and list_repo_files revision forwarding are unrelated API issues." - }, - { - "left": "issue:43873", - "right": "issue:47145", - "accept": false, - "reason": "Both mention quantization/offloading, but one is runtime offloading behavior and the other is checkpoint loading without weight_scale; not the same bug." - }, - { - "left": "issue:44336", - "right": "issue:46851", - "accept": false, - "reason": "ANSI escape leakage in loading_report and Qwen3.5 variable-length position_ids are unrelated." + "reason": "flash_attention s_aux=None crash and RTDetrLoss infeasible cost matrix are distinct failures in different components." } ] }, @@ -13178,7 +13174,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:33290", "issue:42175", @@ -13193,7 +13189,7 @@ "issue:43673", "issue:43761", "issue:43784", - "issue:43824", + "issue:43873", "issue:43881", "issue:43901", "issue:43976", @@ -13204,12 +13200,12 @@ "issue:44261", "issue:44263", "issue:44336", - "issue:44373", "issue:44589", "issue:44623", "issue:44671", "issue:44749", "issue:44779", + "issue:45081", "issue:45103", "issue:45325", "issue:45362", @@ -13223,25 +13219,29 @@ "issue:45823", "issue:45834", "issue:45859", + "issue:45907", "issue:46851", "issue:47000", "issue:47145", "issue:47217" ], "soft_pairs": [ + "issue:45081|issue:45907", + "issue:44336|issue:46851", "issue:42994|issue:43317", + "issue:43873|issue:47145", "issue:43441|issue:45834", - "issue:45103|issue:45372", "issue:45859|issue:46851", + "issue:45103|issue:45372", "issue:43479|issue:43525", "issue:33290|issue:45656", - "issue:44246|issue:47217", "issue:44623|issue:44749", - "issue:44671|issue:44779", + "issue:44246|issue:47217", "issue:42175|issue:43976", + "issue:44671|issue:44779", "issue:44261|issue:44263", - "issue:44016|issue:45823", "issue:44623|issue:45397", + "issue:44016|issue:45823", "issue:44186|issue:47000", "issue:45542|issue:45800", "issue:43525|issue:43881", @@ -13254,206 +13254,210 @@ "issue:43575|issue:47145", "issue:42796|issue:43441", "issue:45325|issue:46851", - "issue:44589|issue:45362", - "issue:43824|issue:43976", - "issue:44336|issue:44373" + "issue:44589|issue:45362" ], - "prepared_review_unit_hash": "b5dfa6bdf70c610ca4c174295a5e83201e3350c8a0a62ef2808362b4a90ccd01", + "prepared_review_unit_hash": "f516b9719948c4fb3e8c00dfcf72a8f81c220d273b41b8b02e61def7901c3e46", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35332, - "estimated_input_tokens": 8833, - "estimated_eval_tokens": 17922 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36331, + "estimated_input_tokens": 9083, + "estimated_eval_tokens": 18422 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items do not form a true duplicate set; they span unrelated bugs, regressions, docs, packaging, and model-specific failures. I rejected all soft edges and used the most general/high-activity issue as the representative anchor.", - "confidence": 0.96, - "canonical_issue_reason": "Issue 44246 ('import transformers takes long sometimes') is the broadest and most discussion-heavy issue in the set, so it is the best representative anchor if one must be chosen.", + "summary": "This cluster is not a true duplicate set; it\u2019s a loose semantic bucket with many unrelated model-, tokenizer-, config-, and packaging-level issues. All soft-edge pairs should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "issue:43873 is the broadest and most active open issue in the list, so it is the least-specific representative if one must be chosen, though the overall cluster is heterogeneous.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44246 has the highest overall discussion/inbound activity and the broadest scope, making it the least arbitrary issue to keep as the cluster representative.", + "best_issue_reason": "issue:43873 is the best representative by scope and activity, but it still does not unify the cluster into a single bug.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral tokenizer regex crash vs repository file listing revision forwarding; unrelated code paths and symptoms." + }, + { + "left": "issue:44336", + "right": "issue:46851", + "accept": false, + "reason": "ANSI output formatting bug vs Qwen3.5 variable-length position_ids bug; no shared underlying defect." + }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Quantized model saving failure vs dequantized model loading/offload failure are different code paths and different bugs." + "reason": "Saving quantized models vs loading dequantized models with offload are related only at a high level; different failure modes." }, { - "left": "issue:43441", - "right": "issue:45834", + "left": "issue:43873", + "right": "issue:47145", "accept": false, - "reason": "FlashAttention failure in Ministral-3 is unrelated to Kosmos2.5 OCR long-input index errors." + "reason": "Both touch quantization/offload, but one is generic offloading behavior and the other is FP8 frozen checkpoint scaling; not the same bug." }, { - "left": "issue:45103", - "right": "issue:45372", + "left": "issue:43441", + "right": "issue:45834", "accept": false, - "reason": "Auto-docstring crashes from future annotations and Gemma 4 processor import failure are unrelated." + "reason": "Ministral FlashAttention failure and Kosmos2.5 OCR index error are unrelated model-specific crashes." }, { "left": "issue:45859", "right": "issue:46851", "accept": false, - "reason": "CPU-offload rotary embedding incompatibility is a different bug from variable-length position_ids handling." + "reason": "Both involve Qwen3.5 positional logic, but one is CPU offload incompatibility and the other is variable-length input handling; too different to merge." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring future-annotations crash vs mistral_common import error in Gemma 4 loading; unrelated subsystems." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Phi4MultimodalConfig default-init bug and Llama4Config missing pad_token_id are distinct config issues." + "reason": "Different config classes and fields (Phi4Multimodal vision/audio defaults vs Llama4 pad_token_id); not the same bug." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "AdaFactor OOM in DeepSpeed and double optimizer-step in DeepSpeed are different failures." - }, - { - "left": "issue:44246", - "right": "issue:47217", - "accept": false, - "reason": "Import-time slowness is unrelated to VL image-placeholder input handling." + "reason": "Deepspeed OOM on optimizer state vs optimizer step being called twice; distinct training-loop bugs." }, { "left": "issue:44623", "right": "issue:44749", "accept": false, - "reason": "Missing save_pretrained files and slow filtering after upgrade are different problems." + "reason": "processor.save_pretrained missing files vs a performance regression during filtering after upgrade; unrelated." }, { - "left": "issue:44671", - "right": "issue:44779", + "left": "issue:44246", + "right": "issue:47217", "accept": false, - "reason": "CamemBERT masked-LM prediction regression and DeepSeek tokenizer regression are different model/tokenizer bugs." + "reason": "Import latency complaint vs VL image-placeholder failure; different behavior and code paths." }, { "left": "issue:42175", "right": "issue:43976", "accept": false, - "reason": "TensorFlow backend packaging omission is unrelated to Python version compatibility." + "reason": "Packaging backend omission vs Python version compatibility regression; unrelated release/setup issues." }, { - "left": "issue:44261", - "right": "issue:44263", + "left": "issue:44671", + "right": "issue:44779", "accept": false, - "reason": "MLA RMS-eps precision issue and GlmMoeDsaIndexer split return-value issue are unrelated." + "reason": "CamemBERT masked LM regression vs Deepseek tokenizer regression; both v5 regressions but different components and fixes." }, { - "left": "issue:44016", - "right": "issue:45823", + "left": "issue:44261", + "right": "issue:44263", "accept": false, - "reason": "Notebook syntax error and Gemma4 device mismatch during forward are unrelated." + "reason": "LayerNorm epsilon precision issue vs torch.split return-value issue; no shared underlying bug." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "processor.save_pretrained missing files is not the same as Gemma-4 ZeRO3 from_pretrained failure." + "reason": "Missing processor save files vs gemma-4 zero3 from_pretrained; different failure modes." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error vs Gemma4 device mismatch under device_map=auto; unrelated." }, { "left": "issue:44186", "right": "issue:47000", "accept": false, - "reason": "LayoutLMv2Tokenizer padding/NER crash and RTDetrLoss infeasible cost matrix are different subsystems." + "reason": "LayoutLMv2 tokenizer padding/truncation crash vs RTDetrLoss infeasible cost matrix; different areas entirely." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "Missing TensorFlow backend from install extras is unrelated to torch/transformers incompatibility." + "reason": "TensorFlow backend packaging issue vs torch/transformers compatibility issue; not the same bug." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Llama4Config pad_token_id absence and glm-4v-9b load failure are different bugs." + "reason": "Llama4 missing pad_token_id vs glm-4v-9b loading failure; different model/config defects." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache missing in chunked prefill and CLIPVisionModel hidden_states regression are unrelated." + "reason": "GenerationMixin cache missing during chunked prefill vs CLIPVisionModel hidden_states regression; unrelated regression locations." }, { "left": "issue:43784", "right": "issue:45542", "accept": false, - "reason": "sentence-transformers import NameError and undefined tf backend are unrelated import/package issues." + "reason": "Import NameError in sentence-transformers integration vs undefined tf backend packaging; no common code path." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to intermittent import slowness." + "reason": "Docs still mentioning return_all_scores vs intermittent import latency; one is documentation, the other performance." }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "torch.compile attention-kernel divergence and ANSI code leakage are unrelated." + "reason": "torch.compile attention-kernel divergence vs ANSI code generation in loading_report; unrelated." }, { "left": "issue:43525", "right": "issue:43643", "accept": false, - "reason": "Missing pad_token_id and missing fields from trust_remote_code AutoConfig are similar in theme but not the same bug." + "reason": "Missing config field in Llama4 vs missing fields from AutoConfig trust_remote_code; both config-related but different root causes." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "auto_mappings CI breakage and leftover dummy classes are separate maintenance issues." + "reason": "auto_mappings stale config references vs dummy_pt_objects leftovers; both cleanup/CI issues but not the same artifact set." }, { "left": "issue:43575", "right": "issue:47145", "accept": false, - "reason": "TP load OOM for Qwen2-57B and FP8 compressed-tensors weight_scale omission are unrelated loading/quantization bugs." + "reason": "TP OOM on large model load vs FP8 weight_scale omission; both involve memory/quantization but different bugs." }, { "left": "issue:42796", "right": "issue:43441", "accept": false, - "reason": "Tokenizer BPE-marker handling and FlashAttention failure are different model-specific issues." + "reason": "Tokenizer BPE marker handling vs FlashAttention failure in a Ministral model; tokenizer vs attention backend, different fixes." }, { "left": "issue:45325", "right": "issue:46851", "accept": false, - "reason": "Qwen2.5-VL still-image rope scaling and Qwen3.5 variable-length position_ids are different implementations and bugs." + "reason": "Qwen2.5-VL temporal position scaling bug vs Qwen3.5 variable-length position_ids bug; similar theme, different model logic." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Float8 storage lookup error and Qwen3.5 chat crash are unrelated failures." - }, - { - "left": "issue:43824", - "right": "issue:43976", - "accept": false, - "reason": "Missing exported Qwen2.5-VL class and Python version support issue are unrelated." - }, - { - "left": "issue:44336", - "right": "issue:44373", - "accept": false, - "reason": "Terminal ANSI codes and docstring wording for position_ids are unrelated." + "reason": "Float8 storage lookup error vs Qwen3.5 chat crash; unrelated failure mechanisms." } ] }, @@ -13475,11 +13479,10 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:42491", "issue:42617", - "issue:42886", "issue:42915", "issue:43299", "issue:43526", @@ -13487,6 +13490,7 @@ "issue:43540", "issue:43550", "issue:43723", + "issue:43824", "issue:43976", "issue:44162", "issue:44242", @@ -13494,6 +13498,7 @@ "issue:44295", "issue:44297", "issue:44336", + "issue:44373", "issue:44485", "issue:44496", "issue:44871", @@ -13507,7 +13512,6 @@ "issue:45563", "issue:45588", "issue:45593", - "issue:46132", "issue:46291", "issue:46399", "issue:46506", @@ -13526,231 +13530,238 @@ "issue:47312" ], "soft_pairs": [ + "issue:43824|issue:43976", + "issue:44336|issue:44373", "issue:47176|issue:47177", "issue:43723|issue:44162", "issue:44246|issue:44496", - "issue:44871|issue:47145", "issue:45083|issue:46851", + "issue:44871|issue:47145", "issue:46291|issue:46399", "issue:44297|issue:45397", "issue:42617|issue:43976", - "issue:46655|issue:46726", "issue:46399|issue:46851", + "issue:46655|issue:46726", "issue:45003|issue:47217", "issue:42491|issue:43299", "issue:45072|issue:46829", - "issue:43540|issue:46752", "issue:43526|issue:45003", "issue:43531|issue:45362", "issue:45103|issue:46399", "issue:45561|issue:45563", - "issue:44246|issue:44871", + "issue:43540|issue:46752", "issue:45588|issue:46851", - "issue:45003|issue:46821", "issue:44295|issue:45397", + "issue:44246|issue:44871", + "issue:45003|issue:46821", "issue:46614|issue:46665", "issue:44336|issue:44485", "issue:43550|issue:46506", "issue:44242|issue:45593", - "issue:42915|issue:47312", - "issue:42886|issue:46132" + "issue:42915|issue:47312" ], - "prepared_review_unit_hash": "d34eca10430817ef92ac359509f05e69ee004618f2f29598dd45db28b43d9e2e", + "prepared_review_unit_hash": "36201e158f61c22f9daa3abb7ef7bbf96445801727c3f537d58b814874c5f52c", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35381, - "estimated_input_tokens": 8846, - "estimated_eval_tokens": 17948 + "soft_pair_count": 29, + "serialized_chars": 35663, + "estimated_input_tokens": 8916, + "estimated_eval_tokens": 18088 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the soft-similarity pairs share broad keywords or subsystems, but they do not look like the same underlying bug/change. I would not merge any of the candidate pairs as duplicates.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical issue stands out because the items are mostly unrelated. If a representative is needed, issue 42491 is the cleanest regression-style report with clear version context, but it does not subsume the rest.", + "summary": "The cluster is mostly heterogeneous issue reports; no soft pair looks like the same underlying bug/change closely enough to merge. The closest subgroup is around Qwen3-MoE/model-loading compatibility, but even those are distinct symptoms and code paths.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 42491 is the broadest and most central compatibility report in the set: a Qwen3-MoE LoRA model trained on hf4.x no longer works on hf5.x. It best represents the small Qwen3-MoE loading-regression theme, though the overall cluster is not truly cohesive.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 42491 is the best representative anchor only in the sense that it is concrete and version-specific; it is not a true umbrella for the cluster.", + "best_issue_reason": "Issue 42491 is the strongest representative issue because it describes a concrete, user-facing regression tied to the main branch / version transition, and it is broader than the other Qwen-specific variants. Still, it should not be treated as a duplicate of the unrelated issues here.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Different problems: missing model import vs Python version support regression." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "Unrelated: terminal ANSI formatting vs a docstring typo." + }, { "left": "issue:47176", "right": "issue:47177", "accept": false, - "reason": "Both are security-related, but one is a shell-operator allowlist bypass and the other is tar/zip slip arbitrary file write; different exploit paths and fixes." + "reason": "Both are security issues, but different attack surfaces and fixes: shell-command bypass vs archive path traversal." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer cache/offline loading failure and ESM2 model breakage are unrelated failures in different code paths." + "reason": "Tokenizer loading failure and ESM2 breakage are unrelated subsystems and symptoms." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import latency is a performance complaint; unrecognized model/config loading is a model-registration issue." + "reason": "Performance/import slowness is unrelated to unrecognized model config loading." }, { - "left": "issue:44871", - "right": "issue:47145", + "left": "issue:45083", + "right": "issue:46851", "accept": false, - "reason": "One is an eos_token_id config mismatch for Gemma-3; the other is missing weight_scale application for FP8 frozen checkpoints." + "reason": "Both involve Qwen-family models, but different code paths: feature-length helper vs position_ids handling." }, { - "left": "issue:45083", - "right": "issue:46851", + "left": "issue:44871", + "right": "issue:47145", "accept": false, - "reason": "Both involve Qwen models, but one is a helper-length calculation in qwen3_omni_moe and the other is variable-length position_ids handling in Qwen3.5." + "reason": "Different model/config issues: Gemma eos_token_id mismatch vs FP8 weight_scale loading." }, { "left": "issue:46291", "right": "issue:46399", "accept": false, - "reason": "Kernel version/revision validation for pipelines is a dependency/import issue; the Qwen3.5 crash is a registration/decorator problem in a specific model path." + "reason": "Both mention kernels, but they fail in different places for different reasons." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer save config mismatch and gemma-4 zero3 from_pretrained failure are distinct save/load issues." + "reason": "Tokenizer config save mismatch is unrelated to ZeRO3 from_pretrained loading." }, { "left": "issue:42617", "right": "issue:43976", "accept": false, - "reason": "Failure to run a parallel example script is unrelated to a Python-version compatibility problem." + "reason": "3d_parallel.py runtime failure is unrelated to Python-version packaging support." }, { - "left": "issue:46655", - "right": "issue:46726", + "left": "issue:46399", + "right": "issue:46851", "accept": false, - "reason": "Incorrect video padding in SmolVLM is unrelated to AutoProcessor missing-vision-backend error handling." + "reason": "Different Qwen3.5 failures: kernel decorator registration vs variable-length position_ids." }, { - "left": "issue:46399", - "right": "issue:46851", + "left": "issue:46655", + "right": "issue:46726", "accept": false, - "reason": "Both are Qwen3.5-related, but one fails when registering apply_rotary_pos_emb for kernels, while the other is about variable-length position_ids; different code paths." + "reason": "Both are vision-related, but one is padding math and the other is backend availability/error messaging." }, { "left": "issue:45003", "right": "issue:47217", "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils is unrelated to VL image-placeholder handling in user text." + "reason": "Sys.modules access bug is unrelated to VL image-placeholder handling." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both mention Qwen3 MoE loading across versions, but one is LoRA compatibility from hf4.x to hf5.x and the other is Qwen3VL MoE model loading in 5.0.0.dev0; not the same concrete bug." + "reason": "Same broad family, but different models and failure modes: LoRA hf4.x/hf5.x compatibility vs Qwen3VL MoE loading." }, { "left": "issue:45072", "right": "issue:46829", "accept": false, - "reason": "bfloat16 dtype mismatch in inference and incorrect hidden_states wiring in ImageClassifierOutput are different backend/output bugs." - }, - { - "left": "issue:43540", - "right": "issue:46752", - "accept": false, - "reason": "Video-input processing in Qwen3OmniMoe and empty-list chat template IndexError are unrelated." + "reason": "Different inference bugs in different model classes: dtype mismatch vs output field misassignment." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to modeling_utils sys.modules access." + "reason": "ImageProcessor label reduction bug is unrelated to modeling_utils sys.modules access." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Sliding_window behavior in Qwen3-MoE and a Qwen3.5 chat crash are different model/runtime issues." + "reason": "Sliding-window Qwen3-MoE bug and Qwen3.5 chat crash are not the same underlying defect." }, { "left": "issue:45103", "right": "issue:46399", "accept": false, - "reason": "Auto-docstring crashes with future annotations are unrelated to Qwen3.5 kernel registration failures." + "reason": "Auto-docstring annotation handling and kernel registration are unrelated." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "A pytest-xdist race on captured_info.txt is unrelated to a stale warning in paged generate()." + "reason": "Test xdist file-race bug is unrelated to a stale generate warning." }, { - "left": "issue:44246", - "right": "issue:44871", + "left": "issue:43540", + "right": "issue:46752", "accept": false, - "reason": "Import performance and Gemma eos_token_id inconsistency do not share the same failure mode." + "reason": "Video-input ValueError and empty-chat-list IndexError are different code paths." }, { "left": "issue:45588", "right": "issue:46851", "accept": false, - "reason": "Flash-attention s_aux handling and Qwen3.5 variable-length position_ids are different bugs." + "reason": "Flash-attention sinkless-model crash is unrelated to variable-length position_ids." }, { - "left": "issue:45003", - "right": "issue:46821", + "left": "issue:44295", + "right": "issue:45397", "accept": false, - "reason": "sys.modules access in modeling_utils is unrelated to TimesFM window_size AttributeError." + "reason": "Buffer read error with position_ids is unrelated to gemma-4 ZeRO3 loading." }, { - "left": "issue:44295", - "right": "issue:45397", + "left": "issue:44246", + "right": "issue:45593", + "accept": false, + "reason": "Import slowness is unrelated to D-FINE auxiliary-loss logic." + }, + { + "left": "issue:45003", + "right": "issue:46821", "accept": false, - "reason": "Reading a buffer-like position_ids after registration and gemma-4 zero3 from_pretrained are not the same issue." + "reason": "Safe sys.modules access and TimesFM window_size AttributeError are unrelated." }, { "left": "issue:46614", "right": "issue:46665", "accept": false, - "reason": "Stale phonemizer backend state and inconsistent from_pretrained behavior inside Python classes are only loosely related through statefulness, not the same bug." + "reason": "Phonemizer backend state bug is unrelated to from_pretrained behavior inside Python classes." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "ANSI code leakage in loading_report is unrelated to GLM-5 RoPE implementation concerns." + "reason": "ANSI code emission and GLM-5 RoPE implementation are different areas and bugs." }, { "left": "issue:43550", "right": "issue:46506", "accept": false, - "reason": "torch.compile+SDPA failure in Bamba-9B-v2 and CPU/CUDA result drift in PPDocLayoutV3 are different backend correctness problems." + "reason": "torch.compile SDPA failure and CPU/CUDA result mismatch are distinct issues." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both mention missing auxiliary losses under a flag, but they concern different models and different loss mechanisms; not the same code-path bug." + "reason": "Missing load-balancing loss and missing auxiliary losses are similar phrasing but different models and mechanisms." }, { "left": "issue:42915", "right": "issue:47312", "accept": false, - "reason": "Qwen3Moe FP8 config failure and kernels 0.16 incompatibility are different integration/version issues." - }, - { - "left": "issue:42886", - "right": "issue:46132", - "accept": false, - "reason": "Offline tokenizer cache loading failure and AttentionInterface.register behavior change are unrelated." + "reason": "Qwen3MoE FP8 config failure is unrelated to kernels 0.16 incompatibility." } ] }, @@ -13772,11 +13783,12 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36246", "issue:41950", "issue:42757", + "issue:42886", "issue:43257", "issue:43295", "issue:43352", @@ -13789,16 +13801,13 @@ "issue:44077", "issue:44162", "issue:44220", - "issue:44242", "issue:44295", "issue:44351", "issue:44451", "issue:44466", "issue:44479", - "issue:44623", "issue:44749", "issue:44779", - "issue:44857", "issue:44877", "issue:44987", "issue:44995", @@ -13814,247 +13823,249 @@ "issue:45600", "issue:45701", "issue:45759", + "issue:46132", "issue:46170", "issue:46369", "issue:46506", "issue:46655", "issue:46726", "issue:46752", - "issue:47217" + "issue:47217", + "issue:47321" ], "soft_pairs": [ + "issue:42886|issue:46132", "issue:44779|issue:45292", - "issue:43540|issue:46655", "issue:43901|issue:45003", + "issue:43540|issue:46655", "issue:43257|issue:45137", "issue:41950|issue:44220", "issue:46726|issue:47217", - "issue:44479|issue:46655", "issue:36246|issue:42757", "issue:36246|issue:44351", - "issue:45375|issue:45759", "issue:36246|issue:43596", + "issue:45375|issue:45759", + "issue:44479|issue:46655", "issue:45292|issue:45701", - "issue:43295|issue:45292", "issue:44451|issue:46752", - "issue:45137|issue:45600", + "issue:43295|issue:45292", "issue:44162|issue:45200", + "issue:45137|issue:45600", "issue:44295|issue:44749", - "issue:44466|issue:45335", "issue:43352|issue:45250", + "issue:44466|issue:45335", + "issue:46170|issue:47321", "issue:44060|issue:44077", "issue:45003|issue:45561", "issue:43475|issue:46369", "issue:44877|issue:45375", "issue:44995|issue:45200", - "issue:44987|issue:45459", "issue:44038|issue:46506", "issue:44479|issue:46170", - "issue:44242|issue:44857", - "issue:44623|issue:46726" + "issue:44987|issue:45459" ], - "prepared_review_unit_hash": "c2e1d8f1ddbfb5788bbc892abb39a8f531feff29d5a4d1c009c231887de5ccc0", + "prepared_review_unit_hash": "a01e317b0cabddff92db36d4ada940ee96bd6079afcac538f8eedfad24a723ce", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35713, - "estimated_input_tokens": 8929, - "estimated_eval_tokens": 18114 + "serialized_chars": 35703, + "estimated_input_tokens": 8926, + "estimated_eval_tokens": 18108 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are a mixed bag of unrelated bug reports (video/VL regressions, import errors, config issues, tokenizer/serialization problems). I don\u2019t see any true duplicate pairs among the soft edges, so all should be rejected. If forced to pick one issue as the cluster\u2019s representative, the broadest user-facing regression is the v5.3.0 video-input bug affecting multiple Qwen VL variants.", - "confidence": 0.68, - "canonical_issue_reason": "issue:44479 is the broadest concrete regression here: one bug report spanning multiple related Qwen VL/video models with a clear v5.3.0 breakage and actionable repro scope.", + "summary": "This set is a loose semantic cluster, not a duplicate set: it mixes unrelated import errors, config issues, tokenizer regressions, video-processor bugs, DeepSpeed failures, and model-specific regressions. I don\u2019t see any soft pair that looks like the same underlying bug or a PR-worthy consolidation target.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 44479 is the best anchor because it is a concrete, user-facing regression with a clear affected family (Qwen2.5-VL/Qwen3-VL/Qwen3.5/Qwen3.5-MoE) and a specific code-path class (video input handling). It is the most broadly scoped and representative bug report among the candidates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the strongest single representative because it is user-facing, version-specific, and covers multiple affected models rather than a one-off edge case.", + "best_issue_reason": "Issue 44479 is the strongest standalone triage target: it has the clearest regression framing, the broadest immediate impact within one bug family, and enough specificity to guide a fix without being overly generic.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44779", - "right": "issue:45292", + "left": "issue:42886", + "right": "issue:46132", "accept": false, - "reason": "Tokenizer output drift and resize_token_embeddings mismatch are different code paths and symptoms." + "reason": "Offline cache loading in tokenizers and AttentionInterface.register behavior changes are unrelated code paths and different failures." }, { - "left": "issue:43540", - "right": "issue:46655", + "left": "issue:44779", + "right": "issue:45292", "accept": false, - "reason": "Both are video-related, but one is a Qwen3OmniMoe ValueError and the other is a SmolVLM video-padding bug." + "reason": "DeepSeek tokenizer output corruption and resize_token_embeddings not updating output embeddings are different bugs in different subsystems." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to sys.modules access in modeling_utils." + "reason": "Docs mismatch for return_all_scores has nothing to do with unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:43540", + "right": "issue:46655", + "accept": false, + "reason": "Both involve video handling, but one is a Qwen3OmniMoe input validation failure and the other is a SmolVLM video padding calculation bug." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Both mention DeepSpeed, but one is Qwen3 MOE weight conversion and the other is a ZeRO3 deque IndexError." + "reason": "DeepSpeed-related, but one is Qwen3 MoE weight conversion under accelerate+deepspeed and the other is a ZeRO3 deque/indexing crash." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification image-processor lookup and fbank feature extraction are unrelated failures." + "reason": "Video-classification image-processor lookup and fbank feature extraction are different pipeline/audio code paths." }, { "left": "issue:46726", "right": "issue:47217", "accept": false, - "reason": "Both involve VL input handling, but one is missing vision backend/processor resolution and the other is empty-image-placeholder handling." + "reason": "Missing vision backend import handling is unrelated to VL placeholder parsing in user text." }, { - "left": "issue:44479", - "right": "issue:46655", + "left": "issue:36246", + "right": "issue:42757", "accept": false, - "reason": "Both are video bugs, but they affect different models and different failure points in the video pipeline." + "reason": "Different missing symbols from different packages: qwen2_5_vl image processor export vs huggingface_hub is_offline_mode." }, { "left": "issue:36246", - "right": "issue:42757", + "right": "issue:44351", "accept": false, - "reason": "Two separate ImportErrors for different symbols from different packages." + "reason": "Both are import errors, but they concern different exported APIs and different fixes." }, { "left": "issue:36246", - "right": "issue:44351", + "right": "issue:43596", "accept": false, - "reason": "Different missing imports: Qwen2_5_VLImageProcessor vs HybridCache." + "reason": "ImportError for a VL image processor and a DeepSpeed/BertModel IndexError are unrelated failures." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both are Qwen multimodal config issues, but one drops a config field and the other fails to unwrap text_config." + "reason": "One is strict config dropping a missing field; the other is AutoModelForCausalLM not unwrapping text_config for composite multimodal configs." }, { - "left": "issue:36246", - "right": "issue:43596", + "left": "issue:44479", + "right": "issue:46655", "accept": false, - "reason": "ImportError for a VL image processor is unrelated to a DeepSpeed/BertModel IndexError." + "reason": "Both are video-related, but the underlying bugs differ: Qwen VL regression vs SmolVLM padding logic." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "resize_token_embeddings behavior and tokenization version drift are different bugs." + "reason": "Embedding resize behavior and version-dependent tokenization are unrelated bugs." + }, + { + "left": "issue:44451", + "right": "issue:46752", + "accept": false, + "reason": "Model loading failure for ScandiBERT and apply_chat_template crashing on empty input are different user flows." }, { "left": "issue:43295", "right": "issue:45292", "accept": false, - "reason": "Processor/tokenizer regression with images is not the same as output_embeddings not updating after resize." + "reason": "Processor.tokenizer regression with images is unrelated to resize_token_embeddings not updating output embeddings." }, { - "left": "issue:44451", - "right": "issue:46752", + "left": "issue:44162", + "right": "issue:45200", "accept": false, - "reason": "Model loading failure for ScandiBERT and empty-list chat-template crash are unrelated." + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaults are different model-specific issues." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "DeepSpeed ZeRO3 deque error and stale auto_mappings references are separate problems." + "reason": "ZeRO3 initialization crash and stale auto_mappings references to removed configs are separate DeepSpeed/config problems." }, { - "left": "issue:44162", - "right": "issue:45200", + "left": "issue:44295", + "right": "issue:44749", "accept": false, - "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaulting are different model-specific bugs." + "reason": "Registering position_ids as a buffer causing read errors is unrelated to a slowdown after upgrading transformers." }, { - "left": "issue:44295", - "right": "issue:44749", + "left": "issue:43352", + "right": "issue:45250", "accept": false, - "reason": "Position_ids buffer access error is unrelated to the Chinese performance regression report." + "reason": "A model not supporting Flash Attention 2.0 and the generic Flash Attention 2.0 issue are not the same concrete bug." }, { "left": "issue:44466", "right": "issue:45335", "accept": false, - "reason": "Both involve embedding/serialization behavior, but the affected models and bug mechanics differ." + "reason": "Inconsistent lm_head.weight serialization by device is different from t5gemma resize_token_embeddings not affecting decoder.embed_tokens." }, { - "left": "issue:43352", - "right": "issue:45250", + "left": "issue:46170", + "right": "issue:47321", "accept": false, - "reason": "A model not supporting Flash Attention 2.0 is not the same as a generic Flash Attention issue." + "reason": "Windows/ROCm import problems from torch.distributed dependency are unrelated to an AMD quark class not being updated." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Incorrect tied-weights warning in Qwen3-Next is unrelated to PatchTSMixer post_init policy." + "reason": "Incorrect tied-weights warning in Qwen3-Next and patchtsmixer post_init policy are different model/config issues." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Unsafe sys.modules access and xdist race on captured_info.txt are different failure modes." + "reason": "Unsafe sys.modules access in modeling_utils and a pytest-xdist race on captured_info.txt are unrelated." }, { "left": "issue:43475", "right": "issue:46369", "accept": false, - "reason": "SAM 3 video attribute error and SAM 3 lite text memory overflow are distinct bugs." + "reason": "SAM 3 Video missing fpn_position_embeddings and sam3-litetext memory overflow are different failures in different model variants." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Strict config loading for granite_speech and missing Qwen3_5MoeVisionConfig field are different config issues." + "reason": "Strict config rejection for granite_speech and a missing field in Qwen3_5MoeVisionConfig are different schema issues." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "Stale indexer cache on second forward is unrelated to Gemma 4 token-type-id defaults." - }, - { - "left": "issue:44987", - "right": "issue:45459", - "accept": false, - "reason": "Model loading failure and tokenizer protobuf exception handling are unrelated." + "reason": "GlmMoeDsa stale indexer cache crash and Gemma 4 mm_token_type_ids defaulting are unrelated bugs." }, { "left": "issue:44038", "right": "issue:46506", "accept": false, - "reason": "Qwen3-VL-Moe bug and PPDocLayoutV3 CPU/CUDA mismatch are different models and behaviors." + "reason": "Qwen3-VL-Moe regression and CPU/CUDA result divergence in PPDocLayoutV3 are not the same code-path problem." }, { "left": "issue:44479", "right": "issue:46170", "accept": false, - "reason": "Video-input regression in Qwen VL models is unrelated to Windows ROCm/torch.distributed import failures." - }, - { - "left": "issue:44242", - "right": "issue:44857", - "accept": false, - "reason": "Missing load-balancing loss and AMP/CUDA image-loss crash are different issues." + "reason": "Video-input regression in Qwen VL models and Windows/ROCm distributed import failure are unrelated." }, { - "left": "issue:44623", - "right": "issue:46726", + "left": "issue:44987", + "right": "issue:45459", "accept": false, - "reason": "processor.save_pretrained missing files is unrelated to AutoProcessor vision-backend resolution." + "reason": "Loading failure for a specific model and tokenizer error masking when protobuf is absent are different failure modes." } ] }, @@ -14076,7 +14087,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:29870", "issue:30333", @@ -14093,14 +14104,16 @@ "issue:43931", "issue:44162", "issue:44188", + "issue:44242", "issue:44361", "issue:44368", "issue:44483", + "issue:44623", "issue:44749", + "issue:44857", "issue:44877", "issue:44977", "issue:44995", - "issue:45103", "issue:45229", "issue:45397", "issue:45412", @@ -14118,234 +14131,224 @@ "issue:46421", "issue:46431", "issue:46710", - "issue:46752", - "issue:46832", + "issue:46726", "issue:46851", - "issue:46858", - "issue:47020", "issue:47068", "issue:47145", "issue:47312" ], "soft_pairs": [ + "issue:44242|issue:44857", + "issue:44623|issue:46726", "issue:29870|issue:30333", "issue:44877|issue:46170", "issue:44995|issue:45588", "issue:44368|issue:44977", "issue:45397|issue:45656", - "issue:43653|issue:44162", "issue:42199|issue:46851", + "issue:43653|issue:44162", "issue:43475|issue:43540", "issue:42994|issue:47312", - "issue:45685|issue:46431", "issue:42886|issue:45229", - "issue:46170|issue:46832", + "issue:45685|issue:46431", "issue:45412|issue:46123", "issue:43723|issue:46710", "issue:43065|issue:43408", "issue:43931|issue:44368", "issue:44188|issue:46202", "issue:44368|issue:45563", - "issue:43906|issue:46202", "issue:45987|issue:46082", + "issue:43906|issue:46202", "issue:46132|issue:47145", - "issue:44483|issue:46202", "issue:44188|issue:46421", + "issue:44483|issue:46202", "issue:45593|issue:47068", - "issue:44361|issue:44749", - "issue:45103|issue:46858", - "issue:46752|issue:47020" + "issue:44361|issue:44749" ], - "prepared_review_unit_hash": "91715f06de3c77d1368a75ad1d816eb5e39e762df2d23b7dcb0090667533a6f1", + "prepared_review_unit_hash": "d367508cb85e5834681694e352c64da60d53e09e323ab0b312e3512e2598e0e9", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 35955, - "estimated_input_tokens": 8989, - "estimated_eval_tokens": 18234 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34913, + "estimated_input_tokens": 8729, + "estimated_eval_tokens": 17714 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster appears to be mostly false-positive similarity noise: the issues span unrelated subsystems (tokenizers, generation, vision, quantization, distributed training, etc.) with no clear shared underlying bug or change. None of the soft pairs look like true duplicates or mergeable PR-equivalents.", - "confidence": 0.94, - "canonical_issue_reason": "No strong canonical issue exists here because the cluster is heterogeneous; the items are not describing the same bug family. If forced to name a broad representative, issue 43723 is the closest generic tokenizer-loading report, but it still does not represent the cluster as a whole.", + "summary": "All candidate soft pairs are false positives: the items are heterogeneous and do not describe the same underlying bug/change. No PRs are present.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No issue in this set is a good global canonical representative. The broadest/most generic is issue 43723, but the cluster is too mixed for it to serve as a real dedup target.", + "best_issue_reason": "issue:44242 is the strongest standalone bug report here: it is concrete, reproducible, and has the clearest follow-up signal among the issues.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Both are detection/loss-related, but one is a missing MoE auxiliary loss and the other is an AMP/CUDA crash in LwDetrImageLoss; different failures and code paths." + }, + { + "left": "issue:44623", + "right": "issue:46726", + "accept": false, + "reason": "One is about save_pretrained missing files, the other about AutoProcessor failing when a vision backend is absent; related area, not the same bug." + }, { "left": "issue:29870", "right": "issue:30333", "accept": false, - "reason": "Unrelated failures: BART generate() UnboundLocalError vs MLflow job-status reporting for failed training jobs." + "reason": "Completely unrelated: BART generate() UnboundLocalError versus MLflow reporting a failed training job as still running." }, { "left": "issue:44877", "right": "issue:46170", "accept": false, - "reason": "Different problem areas: strict config loading for granite_speech vs Windows/ROCm import failure due to torch.distributed dependency." + "reason": "Different subsystems and symptoms: strict config loading for granite_speech versus Windows/ROCm import dependency issues." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "Not the same bug: stale indexer cache in GlmMoeDsa second forward pass vs flash_attention crash on sink-less models with s_aux=None." + "reason": "Both are runtime crashes, but one is a stale cache/indexer issue in GLM-MoE and the other is a flash-attention AttributeError on sink-less models." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "Different symptoms and code paths: tie_word_embeddings warning in Qwen3.5 fine-tuning vs flash-attention generation malfunction." + "reason": "One is a harmless warning about tie_word_embeddings during LoRA fine-tuning; the other is a generation correctness issue with flash-attention." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Different underlying issues: gemma-4 zero3 from_pretrained loading vs deepspeed optimizer step called twice." + "reason": "Both mention deepspeed, but one is gemma-4 from_pretrained under zero3 and the other is double optimizer stepping; different concrete bugs." }, { - "left": "issue:43653", - "right": "issue:44162", + "left": "issue:42199", + "right": "issue:46851", "accept": false, - "reason": "Tokenizer special-token decoding bug vs ESM2 model breakage; no concrete shared failure path." + "reason": "Different models and failures: DETR cardinality logic versus Qwen3.5 variable-length position_ids handling." }, { - "left": "issue:42199", - "right": "issue:46851", + "left": "issue:43653", + "right": "issue:44162", "accept": false, - "reason": "Both are model-specific, but the bugs differ: DETR cardinality/background-class error vs Qwen3.5 variable-length position_ids handling." + "reason": "Tokenizer special-token registration bug versus ESM2 being broken; no shared underlying code-path." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "Both involve multimodal/video models, but one is a missing Sam3VisionEncoderOutput attribute and the other is a Qwen3OmniMoe video-input ValueError." + "reason": "SAM3 video attribute missing versus Qwen3OmniMoe video-input validation error; unrelated model paths." }, { "left": "issue:42994", "right": "issue:47312", "accept": false, - "reason": "Too broad and unsupported: quantized model saving failure is not the same as generic kernels 0.16 incompatibility." - }, - { - "left": "issue:45685", - "right": "issue:46431", - "accept": false, - "reason": "Different subsystems and failures: MPS histogram on Int in moe code vs processor chat-template absence error." + "reason": "Quantized model saving failure is unrelated to kernel version incompatibility." }, { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Tokenizer cache/offline loading bug vs Gemma4 multi-GPU CUDA OOM; not the same bug." + "reason": "Offline tokenizer cache loading and multi-GPU Gemma inference OOM are different issues." }, { - "left": "issue:46170", - "right": "issue:46832", + "left": "issue:45685", + "right": "issue:46431", "accept": false, - "reason": "Import-time dependency issue vs RT-DETR forward TypeError from feature-level/backbone mismatch; unrelated paths." + "reason": "MPS histogram/int type error in MoE code versus chat template absence in a processor; different causes." }, { "left": "issue:45412", "right": "issue:46123", "accept": false, - "reason": "Memory not released on RT-DETR deletion vs MaskGenerationPipeline final-batch dropping results; different behavior and fix surface." + "reason": "Memory not released on deletion versus partial-batch is_last handling in a pipeline; unrelated bugs." }, { "left": "issue:43723", "right": "issue:46710", "accept": false, - "reason": "Tokenizer loading issue in v5 vs accuracy regression in DeepSeek-R1-Distill-Llama-8B after version upgrade; no same code path." + "reason": "Tokenizer loading issue versus accuracy regression after upgrade; same project area, not the same defect." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to the sam3_video/sam3_tracker model-type warning." + "reason": "A dummy Conv2d implementation and a model-type warning are not the same bug." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch is unrelated to Qwen3.5 tie_word_embeddings warning." + "reason": "Qwen3-VL weight-shape mismatch versus a Qwen3.5 warning about tied embeddings; different failure modes." }, { "left": "issue:44188", "right": "issue:46202", "accept": false, - "reason": "Torch.compile attention-kernel divergence is unrelated to Gemma tokenizer roundtrip failure on '\u2581'." + "reason": "Attention-kernel divergence under torch.compile is unrelated to Gemma pipeline roundtrip behavior on \u2581." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "Configuration warning about tie_word_embeddings is not the same as stale num_return_sequences warning in paged generate()." + "reason": "A config warning about tied embeddings is unrelated to a stale num_return_sequences warning in paged generate()." }, { - "left": "issue:43906", - "right": "issue:46202", + "left": "issue:45987", + "right": "issue:46082", "accept": false, - "reason": "Isolated reproduction of a prior issue vs Gemma pipeline tokenizer roundtrip failure; not the same bug." + "reason": "StaticCache return-type contract bug versus LlamaConfig head_dim validation; distinct code paths." }, { - "left": "issue:45987", - "right": "issue:46082", + "left": "issue:43906", + "right": "issue:46202", "accept": false, - "reason": "StaticCache.get_seq_length() returning a tensor is unrelated to LlamaConfig head_dim validation." + "reason": "An isolated reproduction of an external issue is not the same as a Gemma pipeline roundtrip failure." }, { "left": "issue:46132", "right": "issue:47145", "accept": false, - "reason": "AttentionInterface.register behavior change is unrelated to FP8 compressed-tensors checkpoints missing weight_scale on load." + "reason": "AttentionInterface.register behavior change is unrelated to FP8 compressed-tensors loading without weight_scale." }, { - "left": "issue:44483", - "right": "issue:46202", + "left": "issue:44188", + "right": "issue:46421", "accept": false, - "reason": "v5 chat/completions API rejection vs Gemma tokenizer roundtrip issue; different surfaces and failure modes." + "reason": "Different compile/cache problems: attention-kernel divergence versus spurious recompilation with chunked prefill." }, { - "left": "issue:44188", - "right": "issue:46421", + "left": "issue:44483", + "right": "issue:46202", "accept": false, - "reason": "Both touch performance/compile behavior, but one is attention-kernel divergence and the other is cache lazy-init recompilation with chunked prefill." + "reason": "API request rejection and Gemma tokenizer roundtrip failure are unrelated." }, { "left": "issue:45593", "right": "issue:47068", "accept": false, - "reason": "D-FINE auxiliary-loss bug is unrelated to deepspeed_sp_compute_loss host synchronization from CUDA-tensor conditional." + "reason": "Both involve loss computation, but one is missing auxiliary losses and the other is a potential host sync in a Python conditional." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer AttributeError on tasks is unrelated to the Chinese-described filtering slowdown regression after upgrading Transformers." - }, - { - "left": "issue:45103", - "right": "issue:46858", - "accept": false, - "reason": "auto_docstring annotations crash is unrelated to static-cache seq-length device tensor arithmetic in _prefill." - }, - { - "left": "issue:46752", - "right": "issue:47020", - "accept": false, - "reason": "Empty-list apply_chat_template IndexError is unrelated to XLMRobertaTokenizer passing a dict to Unigram." + "reason": "Tokenizer AttributeError on tasks versus a performance regression after upgrading Transformers; not the same defect." } ] }, @@ -14367,12 +14370,11 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28282", "issue:30990", "issue:39692", - "issue:41628", "issue:42994", "issue:43065", "issue:43232", @@ -14393,14 +14395,12 @@ "issue:44373", "issue:44623", "issue:44749", - "issue:44869", "issue:44871", - "issue:44933", "issue:44964", + "issue:45103", "issue:45137", "issue:45230", "issue:45310", - "issue:45356", "issue:45375", "issue:45397", "issue:45478", @@ -14412,261 +14412,258 @@ "issue:46399", "issue:46463", "issue:46726", + "issue:46752", "issue:46821", "issue:46851", + "issue:46858", "issue:46877", "issue:47000", + "issue:47020", "issue:47217" ], "soft_pairs": [ + "issue:45103|issue:46858", + "issue:46752|issue:47020", "issue:43994|issue:46726", - "issue:44297|issue:44623", "issue:43701|issue:44749", + "issue:44297|issue:44623", "issue:43065|issue:46123", "issue:44749|issue:45656", "issue:43646|issue:44162", "issue:44964|issue:45137", "issue:44623|issue:45656", - "issue:43475|issue:43643", "issue:43475|issue:46821", + "issue:43475|issue:43643", "issue:43784|issue:46399", "issue:42994|issue:43482", "issue:44373|issue:46851", "issue:39692|issue:46726", "issue:44871|issue:46132", + "issue:44188|issue:46851", "issue:46463|issue:46877", "issue:28282|issue:30990", "issue:45561|issue:47217", - "issue:44188|issue:46851", "issue:45375|issue:45478", "issue:46877|issue:47000", "issue:45230|issue:45397", - "issue:44869|issue:45356", - "issue:43723|issue:46326", - "issue:43232|issue:43673", - "issue:45310|issue:45375", "issue:44038|issue:44623", + "issue:45310|issue:45375", + "issue:43723|issue:46326", "issue:42994|issue:43525", - "issue:41628|issue:44933" + "issue:43232|issue:43673" ], - "prepared_review_unit_hash": "d44854b3acb9c2b35d49549c394eb9988dbbb08efbc19ff2e198dd333e762f97", + "prepared_review_unit_hash": "ad5d4eb5ca5bceb30ee33b635191253c08443d5021927ddab5deeb5fdc697cb6", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36066, - "estimated_input_tokens": 9017, - "estimated_eval_tokens": 18290 + "serialized_chars": 36160, + "estimated_input_tokens": 9040, + "estimated_eval_tokens": 18336 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly unrelated issues. The only clear duplicate-looking thread in the packet is the exact-title Qwen3.5 MoE from_pretrained error pair (45310/45478); issue 45375 looks like a separate, more specific config-field bug and should not be treated as the same thread from titles alone.", - "confidence": 0.58, - "canonical_issue_reason": "issue:45310 is the best canonical issue because it is part of the exact-title duplicate pair with issue:45478 and cleanly represents the Qwen3.5 MoE from_pretrained regression.", + "summary": "This cluster is mostly a semantic grab-bag of Transformer v5-era regressions and model-specific bugs; there is one clear duplicate issue pair, but most soft edges are not the same underlying problem.", + "confidence": 0.39, + "canonical_issue_reason": "issue:43646 is the broadest representative report in the set: it describes Transformers 5.0.0 breaking custom model initialization, which matches the general v5-regression pattern seen across many of the other issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest issue candidate: its title exactly matches issue:45478, while the other nearby Qwen3.5 item (45375) describes a different missing-field problem.", + "best_issue_reason": "issue:43646 is also the best single issue representative because it is broader than the model-specific reports and captures the common upgrade-breakage theme of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43994", - "right": "issue:46726", + "left": "issue:45103", + "right": "issue:46858", "accept": false, - "reason": "Both are vision-related, but one is about wrong outputs from AutoModel/pipeline and the other is a missing-vision-backend import/error path." + "reason": "Different subsystems: auto_docstring annotation parsing vs cache sequence-length arithmetic under torch.compile." }, { - "left": "issue:44297", - "right": "issue:44623", + "left": "issue:46752", + "right": "issue:47020", + "accept": false, + "reason": "Empty-list chat templating crash is unrelated to XLMRobertaTokenizer passing the wrong type to Unigram." + }, + { + "left": "issue:43994", + "right": "issue:46726", "accept": false, - "reason": "Both mention save_pretrained, but tokenizer config mismatch and missing processor files are different artifacts and code paths." + "reason": "Both involve vision models/processors, but one is a bad AutoModel/pipeline result and the other is an AutoProcessor backend-missing error." }, { "left": "issue:43701", "right": "issue:44749", "accept": false, - "reason": "Checkpoint key mismatch on resume is unrelated to a tokenizer/data-processing slowdown regression." + "reason": "resume_from_checkpoint key mismatch is unrelated to a training slowdown after upgrading Transformers." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both touch save_pretrained, but tokenizer_class mismatch and processor missing files are different serialization bugs." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "A dummy Conv2d in Sam3PixelDecoder and partial-batch dropping in MaskGenerationPipeline are different bugs in different components." + "reason": "Sam3PixelDecoder dummy Conv2d is unrelated to MaskGenerationPipeline dropping the final partial batch." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "A data filtering performance regression is not the same concrete bug as double optimizer stepping under deepspeed." + "reason": "Performance regression after upgrade is not the same as deepspeed causing optimizer.step to run twice." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Custom model initialization breakage and ESM2 breakage are both regressions, but not the same underlying failure." + "reason": "Generic custom model initialization breakage is too different from ESM2-specific model breakage." }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Phi-4 multimodal loading failure and a DeepSpeed ZeRO3 deque IndexError are unrelated." + "reason": "Phi-4 multimodal loading failure and DeepSpeed ZeRO3 deque underflow are unrelated code paths." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Missing processor files on save_pretrained and optimizer stepping twice are unrelated subsystems and symptoms." + "reason": "Processor save_pretrained file emission is unrelated to deepspeed double optimizer-step behavior." }, { "left": "issue:43475", - "right": "issue:43643", + "right": "issue:46821", "accept": false, - "reason": "Missing fpn_position_embeddings and missing fields from trust_remote_code config loading are different missing-field bugs." + "reason": "SAM 3 video missing field on vision output is unrelated to TimesFM window_size AttributeError." }, { "left": "issue:43475", - "right": "issue:46821", + "right": "issue:43643", "accept": false, - "reason": "These are too generic: different models and different missing-attribute problems." + "reason": "Missing fpn_position_embeddings and trust_remote_code AutoConfig field loss are different bugs." }, { "left": "issue:43784", "right": "issue:46399", "accept": false, - "reason": "An import-time NameError in sentence-transformers and a Qwen3.5 kernel registration crash are different failure paths." + "reason": "Import-time NameError in sentence-transformers is unrelated to Qwen3.5 rotary-position registration failure." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are separate serialization/loading problems." + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different failure modes." }, { "left": "issue:44373", "right": "issue:46851", "accept": false, - "reason": "A wrong docstring and a runtime position_ids bug are not the same issue." + "reason": "Wrong docstring for position_ids is documentation-only; variable-length position_ids is a runtime model bug." }, { "left": "issue:39692", "right": "issue:46726", "accept": false, - "reason": "SigLIP2 example errors and AutoProcessor's missing-backend error are different vision-stack bugs." + "reason": "SigLIP2 example errors are a documentation/example issue, not the same as missing vision-backend processor handling." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Gemma eos_token_id inconsistency and AttentionInterface.register behavior changes are unrelated." + "reason": "Gemma eos_token_id inconsistency and AttentionInterface.register behavior change are distinct configuration/API issues." + }, + { + "left": "issue:44188", + "right": "issue:46851", + "accept": false, + "reason": "torch.compile attention-kernel divergence is unrelated to Qwen3.5 variable-length position_ids handling." }, { "left": "issue:46463", "right": "issue:46877", "accept": false, - "reason": "Decoder input validation messaging and doc-lint stdout leakage are unrelated." + "reason": "T5 decoder-input validation messaging and auto_docstring stdout logging are unrelated." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "PyTorch import missing and Sentence Transformers loading stuck are too different to be duplicates." + "reason": "PyTorch-missing ImportError and Sentence Transformers loading hang are different root causes." }, { "left": "issue:45561", "right": "issue:47217", "accept": false, - "reason": "A pytest-xdist captured-info race is unrelated to VL image-placeholder handling." - }, - { - "left": "issue:44188", - "right": "issue:46851", - "accept": false, - "reason": "torch.compile attention-kernel divergence and variable-length position_ids are different code-path problems." + "reason": "pytest-xdist race in testing utils is unrelated to VL image placeholder failures." }, { "left": "issue:45375", "right": "issue:45478", - "accept": false, - "reason": "The titles point to different bugs: one is a missing deepstack_visual_indexes field, the other is a from_pretrained error message." + "accept": true, + "reason": "These are the same Qwen3.5 MoE from_pretrained error, with identical titles and likely the same underlying regression." }, { "left": "issue:46877", "right": "issue:47000", "accept": false, - "reason": "Import-time docstring diagnostics and RTDetrLoss infeasible cost matrix are unrelated." + "reason": "doc-lint diagnostics printing at import time is unrelated to RTDetrLoss infeasible cost matrix failure." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "A generic bug report is not enough to make it the same as gemma-4 zero3 from_pretrained." - }, - { - "left": "issue:44869", - "right": "issue:45356", - "accept": false, - "reason": "Whisper timestamp decode crash and Kimi tokenizer codec regression are both tokenizer-adjacent, but distinct bugs." - }, - { - "left": "issue:43723", - "right": "issue:46326", - "accept": false, - "reason": "AutoTokenizer loading failure and continue_final_message behavior for mistral_common tokenizers are different issues." + "reason": "A generic bug report is not the same underlying issue as gemma-4 zero3 from_pretrained failure." }, { - "left": "issue:43232", - "right": "issue:43673", + "left": "issue:44038", + "right": "issue:44623", "accept": false, - "reason": "Update-model-kwargs after sync_gpus and missing GenerationMixin cache during chunked_prefill are distinct generation regressions." + "reason": "Qwen3 VL bug and processor.save_pretrained missing files are not the same code-path problem." }, { "left": "issue:45310", "right": "issue:45375", "accept": false, - "reason": "Same broad Qwen3.5 area, but a generic from_pretrained error and a specific missing config field do not establish the same concrete bug." + "reason": "Both involve Qwen3.5 MoE, but one is a broad from_pretrained error and the other is a missing config field under @strict; not enough to treat as the same bug." }, { - "left": "issue:44038", - "right": "issue:44623", + "left": "issue:43723", + "right": "issue:46326", "accept": false, - "reason": "Qwen3-VL-Moe v5 bug and processor.save_pretrained missing files are different failures." + "reason": "AutoTokenizer loading regression and mistral_common continue_final_message behavior are different tokenizer issues." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized saving failure and missing pad_token_id on Llama4Config are unrelated." - }, - { - "left": "issue:41628", - "right": "issue:44933", - "accept": false, - "reason": "Both are import-related, but the missing symbols and failure modes differ." + "reason": "Quantized model saving failure and Llama4Config missing pad_token_id are unrelated." }, { - "left": "issue:45310", - "right": "issue:45478", + "left": "issue:43232", + "right": "issue:43673", "accept": false, - "reason": "This pair is not present in the soft-edge list, but the packet evidence indicates it is the only exact-title duplicate thread." + "reason": "sync_gpus generation state handling and chunked_prefill cache absence are different generation bugs." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster summary is a bit too confident for the evidence in the packet. The clearest duplicate pair in the titles is actually issue:45310 and issue:45478 (same Qwen3.5 MoE from_pretrained error wording), while issue:45375 has a different, more specific title about a missing deepstack_visual_indexes field. From the packet alone, that does not justify calling 45375/45478 a 'direct mirror' or 'the only obvious duplicate thread.' The soft-edge rejections are mostly conservative, but the one accepted pair looks stronger than the packet supports." + "accept": true, + "feedback": "The summary is grounded in the packet, and the lone accepted pair (issue:45375/issue:45478) is a clearly supported duplicate by title. The other soft-edge rejections stay conservative and do not overclaim shared root cause from 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 }, @@ -14681,7 +14678,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:30333", "issue:30990", @@ -14709,8 +14706,8 @@ "issue:44661", "issue:44743", "issue:44792", - "issue:44843", "issue:44857", + "issue:44869", "issue:44933", "issue:44945", "issue:45081", @@ -14718,6 +14715,7 @@ "issue:45103", "issue:45137", "issue:45310", + "issue:45356", "issue:45405", "issue:45406", "issue:45561", @@ -14728,12 +14726,13 @@ "issue:46701", "issue:46877", "issue:47059", - "issue:47068", - "issue:47085" + "issue:47068" ], "soft_pairs": [ - "issue:43873|issue:44945", + "issue:44869|issue:45356", + "issue:41628|issue:44933", "issue:42371|issue:45092", + "issue:43873|issue:44945", "issue:36010|issue:41628", "issue:44661|issue:47059", "issue:43278|issue:45137", @@ -14744,240 +14743,235 @@ "issue:45593|issue:46877", "issue:44038|issue:45310", "issue:41628|issue:44351", - "issue:44945|issue:45753", "issue:44222|issue:45310", + "issue:44945|issue:45753", "issue:43232|issue:44792", "issue:45561|issue:47059", "issue:42175|issue:44933", "issue:44743|issue:47068", "issue:45405|issue:46877", - "issue:30333|issue:33290", "issue:42831|issue:44222", + "issue:30333|issue:33290", "issue:42175|issue:45712", - "issue:42898|issue:44568", "issue:43784|issue:43824", - "issue:44496|issue:45561", "issue:42673|issue:43704", + "issue:42898|issue:44568", + "issue:44496|issue:45561", "issue:43976|issue:44933", - "issue:30990|issue:35141", - "issue:44843|issue:47085", - "issue:44568|issue:47085" + "issue:30990|issue:35141" ], - "prepared_review_unit_hash": "a070ce7bb489c1a286ffacf2a0c585095663efefcde9a9672985c022a22c4d53", + "prepared_review_unit_hash": "03de24922232bfe7d939356807d8e2ac912e495312196e8b8617241485acdc47", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36080, - "estimated_input_tokens": 9020, - "estimated_eval_tokens": 18296 + "serialized_chars": 36129, + "estimated_input_tokens": 9033, + "estimated_eval_tokens": 18322 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly noisy: the items share Transformers/tokenizer/model-adjacent vocabulary, but they describe different bugs, model families, or packaging regressions. I did not find any soft pair that looks like the same underlying fix.", - "confidence": 0.33, - "canonical_issue_reason": "issue:44843 is the clearest concrete core regression with a specific failure mode and downstream references; the rest are mostly unrelated point bugs rather than one duplicate thread.", + "summary": "The cluster is mostly heterogeneous issue noise. The clearest duplicate pair is the repeated Qwen3ForCausalLM VRAM leak report; the other pairs look related only at a broad subsystem level, not the same concrete defect.", + "confidence": 0.83, + "canonical_issue_reason": "issue:42673 is the best canonical issue candidate for the duplicate pair because it is the earlier, more discussed report with the exact same title and bug scenario as issue:43704.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44843 is the most actionable representative issue in the set: it has a precise trigger, a clear code path (_patch_mistral_regex / offline mode), and broader utility as the anchor if one item must be selected.", + "best_issue_reason": "issue:42673 is the strongest anchor for the one clear duplicate pair: it has the same title and underlying multi-dataloader VRAM leak problem as issue:43704, and it has more discussion context.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43873", - "right": "issue:44945", + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Both mention tokenizer/decoding behavior, but one is a Whisper timestamp crash and the other is a Kimi tokenizer regression; different failures and code paths." + }, + { + "left": "issue:41628", + "right": "issue:44933", "accept": false, - "reason": "Both are runtime/model-behavior complaints, but one is quantization/offloading and the other is pipeline-parallel output correctness; different bugs." + "reason": "Both are import-related, but they involve different missing symbols/modules and do not point to the same concrete bug." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 settings vs meta-init incompatibility for old InternVL2 are unrelated changes and failure modes." + "reason": "TF32 control settings and InternVL2 meta-init incompatibility are unrelated issues." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior and pipeline-parallel output corruption are different problems." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but for different symbols and different breakages; not the same underlying defect." + "reason": "Both are import errors, but for different APIs and with no clear shared root cause." }, { "left": "issue:44661", "right": "issue:47059", "accept": false, - "reason": "Model-mapping logic vs CLI completion compatibility with huggingface_hub are unrelated." + "reason": "`add-new-model-like` mapping logic and a CLI/huggingface_hub compatibility failure are unrelated." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Embedding dtype drift during evaluate and a ZeRO3 deque pop error are different training/runtime problems." + "reason": "Embedding dtype drift in eval and a DeepSpeed ZeRO3 deque error are distinct runtime issues." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Both crash with AttributeError, but in different helper paths with different root causes." + "reason": "Tokenizer regex patching and auto-docstring annotation handling fail in different subsystems." }, { "left": "issue:44857", "right": "issue:46877", "accept": false, - "reason": "AMP/CUDA loss crash and auto_docstring stdout diagnostics are unrelated subsystems." + "reason": "An AMP crash in LwDetrImageLoss and import-time docstring diagnostics are unrelated." }, { "left": "issue:45406", "right": "issue:46202", "accept": false, - "reason": "Gemma processor missing _tokenizer and Gemma pipeline roundtrip failure are different code paths." + "reason": "Both involve Gemma-related processing, but one is a processor attribute crash and the other is a tokenizer roundtrip failure." }, { "left": "issue:46202", "right": "issue:46701", "accept": false, - "reason": "Both involve Gemma pipelines, but one is a tokenizer roundtrip issue and the other is an image-text-to-text pipeline failure; too distinct to merge." + "reason": "Gemma pipeline roundtrip failure and image-text-to-text pipeline breakage are different bugs." }, { "left": "issue:45593", "right": "issue:46877", "accept": false, - "reason": "D-FINE auxiliary-loss behavior and doc-lint logging are unrelated." + "reason": "D-FINE auxiliary loss behavior is unrelated to auto_docstring logging." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen3 MoE variants, but they target different model families/versions and the exact failure details differ." + "reason": "Both mention Qwen3 MoE, but the versions, model variants, and failure modes differ." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Importing AutoImageProcessor vs importing HybridCache are separate API-export regressions." + "reason": "Different missing imports with no concrete shared fix." }, { - "left": "issue:44945", - "right": "issue:45753", + "left": "issue:44222", + "right": "issue:45310", "accept": false, - "reason": "Pipeline-parallel output correctness and Qwen3_5 image-induced looping are different bugs." + "reason": "FP8 save/load behavior and Qwen3.5 MoE from_pretrained errors are not the same bug." }, { - "left": "issue:44222", - "right": "issue:45310", + "left": "issue:44945", + "right": "issue:45753", "accept": false, - "reason": "FP8 save_pretrained and Qwen3.5 MoE from_pretrained are unrelated model/tooling issues." + "reason": "Pipeline-parallel output issues and a Qwen3_5 infinite loop on a specific image are unrelated." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs update after sync_gpus and a janus image-generation test failure do not share a concrete root cause." + "reason": "Generation kwargs handling after sync_gpus and a Janus image-generation test failure are different issues." }, { "left": "issue:45561", "right": "issue:47059", "accept": false, - "reason": "pytest-xdist file races and CLI completion breakage are separate test/runtime issues." + "reason": "pytest-xdist file races and CLI incompatibility with huggingface_hub are unrelated." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Missing tensorflow in the torch extra and a nonexistent image_utils import are different packaging/import problems." + "reason": "Backend dependency installation and an image_utils import failure are separate packaging/API problems." }, { "left": "issue:44743", "right": "issue:47068", "accept": false, - "reason": "Qwen3.5 recurrent-state reset and a host-sync concern in deepspeed_sp_compute_loss are unrelated model internals." + "reason": "Recurrent-state reset in modular_qwen3_5.py and a DeepSpeed host-sync concern are different code paths." }, { "left": "issue:45405", "right": "issue:46877", "accept": false, - "reason": "PEFT minimum version bump and auto_docstring import-time output are unrelated maintenance issues." - }, - { - "left": "issue:30333", - "right": "issue:33290", - "accept": false, - "reason": "MLFlow reporting on failed jobs and Adafactor OOM in DeepSpeed are entirely different topics." + "reason": "Minimum PEFT version bump and auto_docstring stdout noise are unrelated." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "FineGrainedFP8 accuracy and FP8 save_pretrained are both FP8-adjacent but not the same bug." + "reason": "Both reference FP8, but one is an accuracy issue and the other is save_pretrained behavior; not the same defect." }, { - "left": "issue:42175", - "right": "issue:45712", + "left": "issue:30333", + "right": "issue:33290", "accept": false, - "reason": "TensorFlow backend omission in a torch install and dummy pt object leakage without torch are different packaging/namespace issues." + "reason": "MLflow reporting on failed jobs and Adafactor OOM in DeepSpeed are unrelated." }, { - "left": "issue:42898", - "right": "issue:44568", + "left": "issue:42175", + "right": "issue:45712", "accept": false, - "reason": "Tokenizer cleanup behavior and add_special_tokens BOS/EOS insertion are separate tokenizer semantics regressions." + "reason": "Torch-extra backend packaging and leftover dummy PT objects are different maintenance issues." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Sentence-transformers import NameError and Qwen2_5_VL import error are unrelated import failures." - }, - { - "left": "issue:44496", - "right": "issue:45561", - "accept": false, - "reason": "Unrecognized model config and pytest-xdist capture file races have no shared underlying fix." + "reason": "Both are import errors on newer transformers, but they are missing different exported names and do not share a concrete bug." }, { "left": "issue:42673", "right": "issue:43704", - "accept": false, - "reason": "These are the same title, but the second is a later report of a similar VRAM leak; without code evidence they are still treated as separate reports rather than a proven duplicate pair." - }, - { - "left": "issue:43976", - "right": "issue:44933", - "accept": false, - "reason": "Python-version compatibility and an image_utils import problem are unrelated." + "accept": true, + "reason": "Identical title and the same VRAM leak scenario in multiple dataloader threads indicate a direct duplicate." }, { - "left": "issue:30990", - "right": "issue:35141", + "left": "issue:42898", + "right": "issue:44568", "accept": false, - "reason": "Sentence-Transformer loading hang and embedding reinitialization on resize are different model-loading issues." + "reason": "Both involve tokenizer behavior in v5, but cleanup spacing and special-token insertion are different bugs." }, { - "left": "issue:44843", - "right": "issue:47085", + "left": "issue:44496", + "right": "issue:45561", "accept": false, - "reason": "Offline model_info behavior in mistral regex patching and tokenizer thread-safety are distinct tokenizer regressions." + "reason": "Model config recognition failure and pytest-xdist file races are unrelated." }, { - "left": "issue:44568", - "right": "issue:47085", + "left": "issue:43976", + "right": "issue:44933", "accept": false, - "reason": "BOS/EOS insertion semantics and tokenizer thread-safety are not the same concrete bug." + "reason": "Python-version compatibility and an image_utils import error are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly grounded, but it overstates the evidence for the canonical issue. In the packet, issue:42673 has more discussion activity than issue:43704, but it does not have more inbound references (both show 0). The claim that this is the \u201conly true duplicate pair\u201d is also stronger than what the packet alone proves; the safe conclusion is that it is the clearest duplicate pair among the listed candidates, not that other duplicates are impossible." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -14992,7 +14986,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:22355", "issue:27957", @@ -15023,16 +15017,15 @@ "issue:44623", "issue:44734", "issue:44779", + "issue:44843", "issue:44869", "issue:45092", "issue:45103", "issue:45125", "issue:45200", "issue:45310", - "issue:45405", "issue:45600", "issue:45656", - "issue:45712", "issue:45812", "issue:46077", "issue:46123", @@ -15044,19 +15037,21 @@ "issue:47085" ], "soft_pairs": [ + "issue:44843|issue:47085", + "issue:44568|issue:47085", "issue:43479|issue:46829", "issue:42898|issue:45812", - "issue:44186|issue:47085", "issue:44623|issue:45310", + "issue:44186|issue:47085", "issue:42673|issue:46693", "issue:43723|issue:44568", "issue:39692|issue:46909", "issue:45600|issue:47059", - "issue:46077|issue:46123", - "issue:44016|issue:46231", "issue:43479|issue:44869", + "issue:44016|issue:46231", "issue:43479|issue:44079", "issue:42673|issue:47085", + "issue:46077|issue:46123", "issue:30333|issue:34689", "issue:27957|issue:30333", "issue:42898|issue:44779", @@ -15064,218 +15059,211 @@ "issue:42898|issue:43295", "issue:42898|issue:44448", "issue:30333|issue:36331", - "issue:22355|issue:28282", "issue:42222|issue:45200", + "issue:22355|issue:28282", "issue:44295|issue:45656", "issue:43874|issue:45092", "issue:44734|issue:45103", "issue:43116|issue:44190", "issue:44222|issue:44297", - "issue:45405|issue:45712", "issue:44075|issue:46231" ], - "prepared_review_unit_hash": "076879c2ba9a7f6b0ab0c4c865c93be819dd29a349eef1692a71b1dbca424840", + "prepared_review_unit_hash": "780f99d8f68988366e498b72a99bc62878f3e79d272ee3a84ee7777cf69987e2", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36593, - "estimated_input_tokens": 9149, - "estimated_eval_tokens": 18554 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36313, + "estimated_input_tokens": 9079, + "estimated_eval_tokens": 18414 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose grab-bag of unrelated issues; none of the soft pairs look like true duplicates. I\u2019d keep one strong, high-impact issue as the anchor rather than merging across topics.", - "confidence": 0.91, - "canonical_issue_reason": "`issue:45812` is a broad, high-impact silent regression (`AutoTokenizer` returning wrong IDs for all Granite models) with clear scope and multiple references, making it the best single issue to anchor the cluster.", + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, model-loading, training, docs, and serving issues. None of the soft pairs look like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "No single issue is a true duplicate anchor here; if forced to pick one, issue 47085 is the best current triage target because it is open and active, but the cluster itself is not coherent.", "canonical_pr_reason": null, - "best_issue_reason": "`issue:45812` is the most compelling standalone issue here: concrete regression, wide model impact, and likely easier to act on than the more niche or documentation-only reports.", + "best_issue_reason": "Issue 47085 is the most actionable current report in the set (open, recent, and discussed), though it is not truly representative of the unrelated cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44843", + "right": "issue:47085", + "accept": false, + "reason": "Offline AutoTokenizer/Mistral hub lookup bug vs tokenizer thread-safety regression; different code paths and symptoms." + }, + { + "left": "issue:44568", + "right": "issue:47085", + "accept": false, + "reason": "mDeBERTa add_special_tokens BOS/EOS regression is unrelated to Rust tokenizer thread safety." + }, { "left": "issue:43479", "right": "issue:46829", "accept": false, - "reason": "Different model families and different failure modes: config initialization defaults vs output field misassignment." + "reason": "Phi4MultimodalConfig default-init bug and VideoPrism output-field mismatch are unrelated model bugs." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Both involve tokenizer behavior changes, but one is about `clean_up_tokenization_spaces` and the other is a Granite ID regression; not the same bug." + "reason": "Both are v5 tokenizer regressions, but one is cleanup-space behavior and the other is Granite token-ID generation; not the same bug." }, { - "left": "issue:44186", - "right": "issue:47085", + "left": "issue:44623", + "right": "issue:45310", "accept": false, - "reason": "LayoutLMv2 NER/padding crash is unrelated to tokenizers losing thread safety." + "reason": "processor.save_pretrained file omission and Qwen3.5 MoE from_pretrained failure are unrelated save/load problems." }, { - "left": "issue:44623", - "right": "issue:45310", + "left": "issue:44186", + "right": "issue:47085", "accept": false, - "reason": "`processor.save_pretrained` missing files is unrelated to Qwen3.5 MoE `from_pretrained` loading errors." + "reason": "LayoutLMv2Tokenizer NER/padding crash is unrelated to thread-safety in fast tokenizers." }, { "left": "issue:42673", "right": "issue:46693", "accept": false, - "reason": "VRAM leak in `Qwen3ForCausalLM` with dataloader threads is not the same as a long-context FA2 performance regression." + "reason": "VRAM leak in Qwen3ForCausalLM multiple dataloader threads is not the same as FA2 long-context performance regression." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Generic `AutoTokenizer.from_pretrained` loading issue and a specific BOS/EOS regression for mDeBERTa are not the same underlying defect." + "reason": "Generic AutoTokenizer v5 loading issue vs a specific mDeBERTa add_special_tokens regression; different failures." }, { "left": "issue:39692", "right": "issue:46909", "accept": false, - "reason": "Both are broken docs examples, but they concern different examples and likely different mistakes/code paths." + "reason": "Both are broken doc examples, but for different models/examples and different error classes." }, { "left": "issue:45600", "right": "issue:47059", "accept": false, - "reason": "Broken auto-mapping config references in CI are unrelated to a Hugging Face hub CLI completion incompatibility." - }, - { - "left": "issue:46077", - "right": "issue:46123", - "accept": false, - "reason": "Gemma4 attention-mask bug and pipeline final-batch result dropping are different subsystems and behaviors." - }, - { - "left": "issue:44016", - "right": "issue:46231", - "accept": false, - "reason": "Notebook syntax error and documentation parameter-name mismatch are both docs-related but not the same problem." + "reason": "Removed-config auto_mappings CI breakage is unrelated to huggingface_hub CLI completion breakage." }, { "left": "issue:43479", "right": "issue:44869", "accept": false, - "reason": "Multimodal config default initialization is unrelated to Whisper word-timestamp decode crashing on trailing replacement characters." + "reason": "Multimodal config defaulting bug and Whisper timestamp decode crash are unrelated." }, { - "left": "issue:43479", - "right": "issue:44079", + "left": "issue:44016", + "right": "issue:46231", "accept": false, - "reason": "Config initialization bug and `ModelOutput` key assignment bug affect different code paths and objects." + "reason": "Notebook syntax error and training API doc parameter mismatch are different documentation issues." }, { - "left": "issue:42673", - "right": "issue:47085", + "left": "issue:46077", + "right": "issue:46123", "accept": false, - "reason": "A VRAM leak in Qwen3 CausalLM is not the same as fast tokenizers becoming non-thread-safe." + "reason": "Bidirectional-attention masking bug and final-partial-batch pipeline bookkeeping bug are different generation issues." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "MLflow reporting on failed training jobs is unrelated to Llama 3.2 Vision model loading regressions." + "reason": "MLflow reporting on failed training and Llama 3.2 Vision loading regression are unrelated." }, { "left": "issue:27957", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 behavior for XLMRoberta and MLflow job status reporting have no shared code path." + "reason": "Flash Attention 2 with XLMRoberta and MLflow failure reporting are unrelated subsystems." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions in v5, but one concerns tokenization-space cleanup and the other Deepseek token IDs; too different to merge." + "reason": "General v5 cleanup-space behavior change and Deepseek tokenizer incorrect IDs are not the same tokenizer bug." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Incorrect tied-weights warning and missing tensor-parallel plan are distinct Qwen3.5 MoE issues." + "reason": "Qwen3-Next tied-weights warning and missing _tp_plan for Qwen3.5 MoE are different model-architecture issues." }, { "left": "issue:42898", "right": "issue:43295", "accept": false, - "reason": "`clean_up_tokenization_spaces` behavior change and processor/tokenizer API regression are separate tokenizer-related problems." + "reason": "Token cleanup behavior change is unrelated to processor.tokenizer/image-passing regression." }, { "left": "issue:42898", "right": "issue:44448", "accept": false, - "reason": "Different regressions: tokenizer cleanup behavior versus generation output differences for Pegasus." + "reason": "Pegasus output differences between v4/v5 are a model-specific regression, not the clean_up_tokenization_spaces issue." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "MLflow failure reporting and `compute_loss(num_items_in_batch=...)` signature breakage are unrelated training issues." + "reason": "Failed-job MLflow reporting and CustomTrainer.compute_loss signature change are unrelated training issues." }, { - "left": "issue:22355", - "right": "issue:28282", + "left": "issue:42222", + "right": "issue:45200", "accept": false, - "reason": "Missing `transformers.onnx` module and missing PyTorch import are different import-time dependency problems." + "reason": "Broken vitpose models and Gemma 4 mm_token_type_ids defaults are unrelated multimodal model bugs." }, { - "left": "issue:42222", - "right": "issue:45200", + "left": "issue:22355", + "right": "issue:28282", "accept": false, - "reason": "VitPose models being broken and Gemma 4 needing default `mm_token_type_ids` are unrelated multimodal issues." + "reason": "Missing transformers.onnx module and missing PyTorch dependency are different import errors." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Buffer registration/reading issue is unrelated to optimizer steps being called twice under deepspeed." + "reason": "position_ids buffer access error and deepspeed double optimizer-step bug are unrelated." }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "Missing image-patch-count API and remote-code checkpoint meta-init incompatibility are different multimodal loading problems." + "reason": "Missing image-processor method in GLM46V and meta-init incompatibility in InternVL2 are different multimodal-loading failures." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "KV-cache continuation crash in `transformers serve` is unrelated to the auto-docstring annotations crash." + "reason": "serve KV-cache continuation indexing crash and auto_docstring future-annotations crash are unrelated." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Example-script multi-label output bug and local dataset loading failure in a different script are not the same issue." + "reason": "Classification example empty-results bug and local-dataset loading failure are different example-script issues." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "FP8 MoE save-pretrained behavior and tokenizer config class mismatch are separate save/serialization bugs." - }, - { - "left": "issue:45405", - "right": "issue:45712", - "accept": false, - "reason": "PEFT version pinning and leftover dummy classes in `dummy_pt_objects.py` are unrelated maintenance issues." + "reason": "FP8 save_pretrained MoE issue and tokenizer_class metadata mismatch are unrelated save-pretrained bugs." }, { "left": "issue:44075", "right": "issue:46231", "accept": false, - "reason": "SGD optimizer args being ignored and a docstring parameter-name mismatch are not the same underlying bug." + "reason": "SGD args ignored and documentation parameter-name mismatch are unrelated." } ] }, @@ -15297,7 +15285,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:30990", "issue:38175", @@ -15305,7 +15293,6 @@ "issue:42757", "issue:42796", "issue:42831", - "issue:42907", "issue:43122", "issue:43232", "issue:43475", @@ -15332,10 +15319,10 @@ "issue:45245", "issue:45310", "issue:45357", + "issue:45405", "issue:45656", "issue:45712", "issue:45834", - "issue:46050", "issue:46082", "issue:46132", "issue:46209", @@ -15349,23 +15336,24 @@ "issue:47276" ], "soft_pairs": [ - "issue:42796|issue:43723", + "issue:45405|issue:45712", "issue:44448|issue:46132", - "issue:43479|issue:43644", + "issue:42796|issue:43723", "issue:42831|issue:44623", "issue:44297|issue:45656", + "issue:43479|issue:43644", "issue:44933|issue:45712", "issue:45245|issue:46132", "issue:44222|issue:44623", "issue:46209|issue:46431", "issue:44373|issue:44485", "issue:44297|issue:45310", - "issue:43122|issue:46209", - "issue:43723|issue:44977", "issue:41628|issue:42757", + "issue:43723|issue:44977", + "issue:43122|issue:46209", "issue:43504|issue:47059", - "issue:43232|issue:47276", "issue:46614|issue:46932", + "issue:43232|issue:47276", "issue:45071|issue:46082", "issue:44360|issue:46762", "issue:44222|issue:44484", @@ -15373,228 +15361,227 @@ "issue:44190|issue:46082", "issue:30990|issue:38175", "issue:44263|issue:45092", - "issue:43232|issue:46424", - "issue:44327|issue:47059", "issue:44295|issue:44623", + "issue:44327|issue:47059", + "issue:43232|issue:46424", "issue:44263|issue:44373", "issue:43475|issue:46305", - "issue:42907|issue:46050", "issue:45092|issue:45357" ], - "prepared_review_unit_hash": "6bb8bffef150fcae6c461bed46e3ef54557f78a215d06ec2f21e573ebd90557d", + "prepared_review_unit_hash": "e4329a170ed0dec0b832f3a08aae2cecda6081ec8128cae695f6fa19762bef31", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36657, - "estimated_input_tokens": 9165, - "estimated_eval_tokens": 18586 + "serialized_chars": 36102, + "estimated_input_tokens": 9026, + "estimated_eval_tokens": 18308 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a collection of unrelated bug reports that only share broad themes like loading, tokenizers, save_pretrained, or v5 regressions. None of the soft pairs look like the same concrete underlying bug/change, so I would not merge any of them as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": null, + "summary": "These are mostly broad Transformers regressions that overlap only at the subsystem level; none of the soft pairs look like the same concrete bug/change, so all soft edges should be rejected.", + "confidence": 0.87, + "canonical_issue_reason": "issue:43723 is the broadest core-API bug here: a v5 tokenizer-loading failure with clear regression framing, making it the best representative among the tokenizer-related reports.", "canonical_pr_reason": null, - "best_issue_reason": "No true duplicate cluster is present. If one issue must serve as the broadest anchor, issue:43723 is the most generic concrete loading failure (AutoTokenizer.from_pretrained in v5), but it is still not a duplicate of the other items.", + "best_issue_reason": "issue:47059 has the strongest cross-cutting signal and most inbound references, with a concrete dependency-compatibility failure that is easy to reproduce and triage.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:42796", - "right": "issue:43723", - "accept": false, - "reason": "Both involve tokenizers, but one is a specific BPE-marker bug in Ministral-3-8B and the other is a general tokenizer-loading regression in v5; different underlying failures." + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "Unrelated problems: a PEFT version bump not yet on PyPI vs leftover dummy classes leaking into repo checks." }, { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "The first is a model output regression for Pegasus; the second is an AttentionInterface.register behavior change. Same broad area, not the same bug." + "reason": "Both touch model behavior, but one is Pegasus output drift across v4/v5 and the other is an AttentionInterface registration side effect." }, { - "left": "issue:43479", - "right": "issue:43644", + "left": "issue:42796", + "right": "issue:43723", "accept": false, - "reason": "One is a multimodal config initialization bug, the other is a non-persistent buffer serialization issue. No shared code-path bug." + "reason": "Tokenizer-specific BPE marker handling vs a general AutoTokenizer.from_pretrained loading failure; different bugs." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FP8 accuracy and processor.save_pretrained missing files are unrelated failures in different subsystems." + "reason": "FP8 accuracy regression and processor.save_pretrained missing files are separate code paths and failure modes." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save metadata mismatch is unrelated to deepspeed causing optimizer.step to run twice." + "reason": "Tokenizer metadata mismatch on save_pretrained is unrelated to optimizer stepping twice under deepspeed." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Config initialization bug for multimodal defaults vs a buffers-initialization/data corruption issue." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Both touch imports, but one is a missing import from image_utils and the other is leftover dummy classes leaking into namespace checks; not the same defect." + "reason": "One is a missing import from image_utils; the other is dummy classes leaking into namespace and checks." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category-count runtime error and AttentionInterface.register behavior change are unrelated." + "reason": "Category-count limit and AttentionInterface.register behavior are different functional issues." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both mention save_pretrained, but one is FP8 MoE model saving and the other is processor file persistence; different objects and symptoms." + "reason": "FP8 MOE save_pretrained failure vs processor save_pretrained omitting files; not the same bug." }, { "left": "issue:46209", "right": "issue:46431", "accept": false, - "reason": "FP8 inference on B200 and missing processor chat template are unrelated." + "reason": "B200 FP8 inference failure and missing chat template are unrelated runtime/configuration issues." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Docstring wording for position_ids is not the same issue as GLM-5 RoPE implementation correctness." + "reason": "A wrong docstring and a RoPE implementation discussion are not the same underlying defect." }, { "left": "issue:44297", "right": "issue:45310", "accept": false, - "reason": "Tokenizer_class mismatch during save_pretrained is not the same as Qwen3.5 MoE from_pretrained failure." + "reason": "Tokenizer save_pretrained class-name mismatch vs Qwen3.5 MoE from_pretrained error; different surfaces and causes." }, { - "left": "issue:43122", - "right": "issue:46209", + "left": "issue:41628", + "right": "issue:42757", "accept": false, - "reason": "Tokenizer behavior across versions and FP8 inference on B200 are unrelated." + "reason": "Both are import errors, but from different packages and unrelated symbols." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "General tokenizer-loading failure and Qwen3.5 flash-attention generation bug are different code paths." + "reason": "Tokenizer loading in v5 is unrelated to Qwen3.5 flash-attention generation behavior." }, { - "left": "issue:41628", - "right": "issue:42757", + "left": "issue:43122", + "right": "issue:46209", "accept": false, - "reason": "Both are import errors, but from different packages and symbols; no shared underlying bug." + "reason": "Tokenization drift across versions and FP8 inference on B200 do not share a concrete code path." }, { "left": "issue:43504", "right": "issue:47059", "accept": false, - "reason": "Beit pretrained loading with a legacy field is unrelated to a hub CLI compatibility failure." + "reason": "A legacy model-loading field issue is unrelated to a huggingface_hub CLI completion attr error." }, { - "left": "issue:43232", - "right": "issue:47276", + "left": "issue:46614", + "right": "issue:46932", "accept": false, - "reason": "One is a generation state update bug, the other is a misleading warning message; not the same concrete problem." + "reason": "Stale phonemizer backend state and Gemma4 quantized embedder dtype casting are different bugs." }, { - "left": "issue:46614", - "right": "issue:46932", + "left": "issue:43232", + "right": "issue:47276", "accept": false, - "reason": "Stale phonemizer language state and quantized vision/audio dtype casting are unrelated." + "reason": "Generation kwargs handling after sync_gpus and a misleading warning message are not the same issue." }, { "left": "issue:45071", "right": "issue:46082", "accept": false, - "reason": "PretrainedConfig type-checking regression is unrelated to LlamaConfig head_dim validation." + "reason": "PretrainedConfig type-checking regression vs LlamaConfig head_dim validation; related area, different defect." }, { "left": "issue:44360", "right": "issue:46762", "accept": false, - "reason": "DSA indexer ReLU discussion and Minimax attention numerical differences do not describe the same bug." + "reason": "DSA indexer missing ReLU and Minimax attention numerical differences are distinct implementation problems." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "FP8 MoE save_pretrained bug and the 50GB max_shard_size question are unrelated." + "reason": "A save_pretrained FP8 bug and a question about default shard size are not duplicates." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation keyword handling and Kosmos2.5 long OCR index error are different failures." + "reason": "Generation sync_gpus kwargs handling vs long-OCR-input index error in Kosmos2.5 are separate issues." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "Local dataset loading in a training script is unrelated to LlamaConfig head_dim validation." + "reason": "Local dataset loading in an example script and LlamaConfig head_dim validation are unrelated." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence Transformers loading hang and zero probabilities in siglip2 are unrelated model issues." + "reason": "Sentence Transformers loading hang and zero probabilities in SigLIP2 are different model-specific failures." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "torch.split return semantics in GlmMoeDsaIndexer are unrelated to remote-code checkpoint incompatibility in Transformers v5." + "reason": "torch.split return semantics in a DSA indexer and remote-code checkpoint incompatibility with meta init are unrelated." }, { - "left": "issue:43232", - "right": "issue:46424", + "left": "issue:44295", + "right": "issue:44623", "accept": false, - "reason": "Generation sync_gpus handling and cache max length behavior are not the same bug." + "reason": "Reading a buffer after registration and missing files in processor.save_pretrained are different failure modes." }, { "left": "issue:44327", "right": "issue:47059", "accept": false, - "reason": "QA decode_spans top_k bounds error is unrelated to the hub CLI _add_completion compatibility failure." + "reason": "QA decode_spans top_k bounds error and CLI completion attr mismatch are unrelated subsystems." }, { - "left": "issue:44295", - "right": "issue:44623", + "left": "issue:43232", + "right": "issue:46424", "accept": false, - "reason": "Reading a registered position_ids buffer and missing processor save files are different issues." + "reason": "Generation kwargs after sync_gpus and cache max length behavior are separate features/bugs." }, { - "left": "issue:43475", - "right": "issue:46305", + "left": "issue:44263", + "right": "issue:44373", "accept": false, - "reason": "SAM 3 Video missing attribute and MoonshineStreaming off-by-one mask logic are unrelated." + "reason": "A runtime torch.split/indexer bug and a docstring correction do not describe the same change." }, { - "left": "issue:42907", - "right": "issue:46050", + "left": "issue:43475", + "right": "issue:46305", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and sinq quantized model reload breakage are different save/load bugs." + "reason": "SAM 3 Video missing attribute and MoonshineStreaming off-by-one mask logic are different model code paths." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Both involve Qwen/InternVL model compatibility, but one is meta-init incompatibility and the other is incorrect visual encoder keys on save; not the same concrete defect." - }, - { - "left": "issue:46209", - "right": "issue:46431", - "accept": false, - "reason": "FP8 inference on B200 and missing chat template on a processor are unrelated." + "reason": "Meta-initialization incompatibility for remote-code checkpoints vs incorrect visual encoder keys on save_pretrained are different problems." } ] }, @@ -15616,19 +15603,18 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:22355", "issue:27957", "issue:28218", - "issue:39379", "issue:39401", "issue:41628", "issue:42796", "issue:42831", "issue:42886", + "issue:42907", "issue:42994", - "issue:43066", "issue:43122", "issue:43408", "issue:43576", @@ -15656,7 +15642,7 @@ "issue:45600", "issue:45753", "issue:46002", - "issue:46172", + "issue:46050", "issue:46305", "issue:46328", "issue:46369", @@ -15668,238 +15654,224 @@ "issue:47217" ], "soft_pairs": [ + "issue:42907|issue:46050", "issue:46369|issue:46909", "issue:46305|issue:46655", "issue:44279|issue:46328", "issue:22355|issue:27957", - "issue:43986|issue:46726", "issue:42831|issue:45310", + "issue:43986|issue:46726", "issue:44263|issue:44360", "issue:45381|issue:46305", - "issue:41628|issue:43824", "issue:43408|issue:46402", + "issue:41628|issue:43824", "issue:42831|issue:44038", "issue:44964|issue:47059", + "issue:44261|issue:45092", "issue:44483|issue:45406", "issue:44829|issue:46002", - "issue:44261|issue:45092", "issue:22355|issue:28218", "issue:43576|issue:43784", "issue:44188|issue:44373", "issue:43784|issue:45600", "issue:42831|issue:45753", "issue:43644|issue:46305", - "issue:39401|issue:44568", "issue:43756|issue:44060", + "issue:39401|issue:44568", "issue:43986|issue:47217", - "issue:42796|issue:43122", "issue:42886|issue:42994", - "issue:39379|issue:46172", - "issue:42886|issue:43066", - "issue:43408|issue:46369" + "issue:42796|issue:43122" ], - "prepared_review_unit_hash": "19bcfc355e575ae21da62d4022f2a82ee8e5455eaee517c160653cebf47b8fea", + "prepared_review_unit_hash": "26ce5c3a09a5f1d06678b1c02a0df124c93e6ea5f9ebce9961138efd72020b85", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36112, - "estimated_input_tokens": 9028, - "estimated_eval_tokens": 18312 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 34978, + "estimated_input_tokens": 8745, + "estimated_eval_tokens": 17746 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose thematic cluster of mostly unrelated Transformers issues (tokenizers, vision/model loading, CLI, quantization, and assorted regressions). The soft similarities are superficial; none of the candidate pairs look like true duplicates or mergeable PR-equivalents.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "This is not a true duplicate cluster: the items span many unrelated tokenizer, model-loading, quantization, CLI, and docs bugs. All soft edges look like superficial similarity rather than the same underlying defect.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue stands out because the set is heterogeneous and does not center on one shared bug or code path.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42886 is the clearest standalone bug report in the set: it has a concrete version-specific regression, explicit offline/cache trigger, and a focused expected-vs-actual description.", + "best_issue_reason": "No issue is a strong global representative; the pairings are mostly topic-level similarity, not duplicate reports of one defect.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both are about quantized model saving, but they target different quantization flows and failure modes (dequantized Ministral/Devstral vs sinq save/load)." + }, { "left": "issue:46369", "right": "issue:46909", "accept": false, - "reason": "Memory overflow in a model call vs a broken documentation example; unrelated failure modes." + "reason": "One is a sam3-litetext memory overflow bug; the other is a broken DIA documentation example. Different artifact types and code paths." }, { "left": "issue:46305", "right": "issue:46655", "accept": false, - "reason": "Different model components: a streaming mask off-by-one vs video padding calculation." + "reason": "Both involve video processing, but one is a MoonshineStreaming sliding-window off-by-one and the other is a SmolVLM padding-calculation bug." }, { "left": "issue:44279", "right": "issue:46328", "accept": false, - "reason": "Generic dependency issue vs a performance inefficiency in feature padding; not the same bug." + "reason": "A generic dependency issue and a pad() performance regression are unrelated problems." }, { "left": "issue:22355", "right": "issue:27957", "accept": false, - "reason": "Import error for transformers.onnx vs Flash Attention 2 support for XLM-Roberta; distinct problems." + "reason": "Import failure for transformers.onnx and XLMRoberta Flash Attention 2 are unrelated bugs." }, { - "left": "issue:43986", - "right": "issue:46726", + "left": "issue:42831", + "right": "issue:45310", "accept": false, - "reason": "AutoProcessor crash without torchvision vs missing vision backend unrecognized image processor; similar area but different triggers and fixes." + "reason": "FineGrainedFP8 accuracy drift and Qwen3.5 Moe from_pretrained errors are different issues in different subsystems." }, { - "left": "issue:42831", - "right": "issue:45310", + "left": "issue:43986", + "right": "issue:46726", "accept": false, - "reason": "FineGrainedFP8 accuracy issue vs Qwen3.5 MoE from_pretrained error; unrelated bugs." + "reason": "Both mention missing vision backends, but one is a video-model AutoProcessor crash without torchvision and the other is an unrecognized image processor path; not the same concrete failure." }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "Two different DSA indexer complaints (split returns vs missing ReLU); same subsystem, not same defect." + "reason": "GlmMoeDsaIndexer torch.split handling and a missing ReLU in the DSA indexer are different implementation bugs." }, { "left": "issue:45381", "right": "issue:46305", "accept": false, - "reason": "Wrong vision_position_ids for Qwen2.5-VL video input vs streaming window mask off-by-one; different code paths." + "reason": "Qwen2.5-VL video vision_position_ids and MoonshineStreaming mask off-by-one are unrelated code paths." }, { - "left": "issue:41628", - "right": "issue:43824", + "left": "issue:43408", + "right": "issue:46402", "accept": false, - "reason": "AutoImageProcessor import failure vs missing Qwen2_5_VLForConditionalGeneration import; both import issues but unrelated symbols." + "reason": "A sam3_video vs sam3_tracker config warning is not the same as Sam3Model text-encoder weights not loading." }, { - "left": "issue:43408", - "right": "issue:46402", + "left": "issue:41628", + "right": "issue:43824", "accept": false, - "reason": "Model-type warning for sam3_tracker vs missing Sam3Model text encoder weights; related family, but not the same bug." + "reason": "Different missing exports: AutoImageProcessor versus Qwen2_5_VLForConditionalGeneration." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "FineGrainedFP8 accuracy regression vs Qwen3-VL-Moe bug; no concrete overlap." + "reason": "FineGrainedFP8 accuracy issues and Qwen3-VL-Moe loading bugs are unrelated." }, { "left": "issue:44964", "right": "issue:47059", "accept": false, - "reason": "Model loading failure vs CLI test breakage from huggingface_hub API change; different layers entirely." + "reason": "Phi-4 multimodal loading and a huggingface_hub CLI compatibility failure are different problems." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Both are model-init related, but one is a missing rms_norm_eps precision issue and the other is a remote-code/meta-init incompatibility." }, { "left": "issue:44483", "right": "issue:45406", "accept": false, - "reason": "Chat/completions request rejection vs Gemma4Processor attribute error; unrelated service/model issues." + "reason": "An API request rejection and a Gemma4Processor missing _tokenizer crash are unrelated." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "Flash Attention 3 training degradation vs NaNs after checkout/init; both training symptoms but different causes." - }, - { - "left": "issue:44261", - "right": "issue:45092", - "accept": false, - "reason": "MLA RMS norm epsilon precision bug vs remote-code checkpoint/meta init incompatibility; distinct implementation issues." + "reason": "Flash-Attention-3 training degeneration and NaNs after checkout/init have different symptoms and likely different root causes." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "Missing transformers.onnx module vs tokenizer extra-space behavior; unrelated tokenizer/import issues." + "reason": "No module named transformers.onnx is unrelated to tokenizer spacing after added tokens." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "Broken env CLI command vs NameError importing sentence-transformers; different surface areas." + "reason": "A broken transformers env CLI command and a sentence-transformers import NameError are different regressions." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Attention-kernel divergence under torch.compile vs wrong docstring for position_ids; not the same underlying defect." + "reason": "torch.compile attention-kernel branching and a wrong docstring are not the same bug." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "Import NameError in sentence-transformers vs stale auto_mappings reference causing CI breakage; both import-related but separate." + "reason": "Import-time NameError in sentence-transformers and stale auto_mappings references are unrelated." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "FP8 accuracy issue vs Qwen3_5 infinite loop on a specific image; unrelated model behaviors." + "reason": "An FP8 accuracy issue and an infinite loop in Qwen3_5 on a specific image are different failures." }, { "left": "issue:43644", "right": "issue:46305", "accept": false, - "reason": "Non-persistent buffers junk in v5 vs sliding-window off-by-one; completely different code paths." - }, - { - "left": "issue:39401", - "right": "issue:44568", - "accept": false, - "reason": "Wrong offset_mapping for Qwen3 tokenizer vs missing BOS/EOS insertion for mdeberta-v3-base; both tokenizer bugs but distinct." + "reason": "Non-persistent buffer junk in Transformers 5.0 and a MoonshineStreaming mask off-by-one are unrelated." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Smollm3 RoPE-layer bug vs Qwen3-Next tied-weights warning; different models and mechanisms." + "reason": "Smollm3 RoPE-layer dropping and Qwen3-Next tied-weight warnings are both model-architecture issues, but not the same concrete defect." }, { - "left": "issue:43986", - "right": "issue:47217", + "left": "issue:39401", + "right": "issue:44568", "accept": false, - "reason": "Missing torchvision crash on AutoProcessor vs VL models failing when image placeholder is in user text; different repros and likely fixes." + "reason": "Both are tokenizer bugs, but one is wrong offset mapping in Qwen3 and the other is missing BOS/EOS insertion in mdeberta-v3." }, { - "left": "issue:42796", - "right": "issue:43122", + "left": "issue:43986", + "right": "issue:47217", "accept": false, - "reason": "BPE marker handling in Ministral tokenizer vs tokenization drift across versions; similar domain, not the same bug." + "reason": "AutoProcessor failing without torchvision and VL models failing on placeholders in user text are different code-path problems." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Tokenizer cache/offline load failure vs quantized model saving failure; unrelated problems." - }, - { - "left": "issue:39379", - "right": "issue:46172", - "accept": false, - "reason": "Incorrect full_text_row_masked_out_mask in mllama vs deepseek-v4 comb transpose issue; separate model internals." - }, - { - "left": "issue:42886", - "right": "issue:43066", - "accept": false, - "reason": "Offline cache load bug vs wrong tokenizer decoder type in v5; both tokenizer/config-related but not the same defect." + "reason": "Offline tokenizer cache loading and quantized model saving are unrelated." }, { - "left": "issue:43408", - "right": "issue:46369", + "left": "issue:42796", + "right": "issue:43122", "accept": false, - "reason": "sam3_tracker type warning vs memory overflow in sam3-litetext; different symptoms and likely causes." + "reason": "Both mention tokenization, but one is a Ministral BPE-marker bug and the other is a version-to-version tokenization change; not clearly the same defect." } ] }, @@ -15921,18 +15893,21 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:22355", "issue:27957", "issue:29159", "issue:30990", "issue:33357", + "issue:39379", "issue:39401", "issue:41628", "issue:42371", "issue:42757", + "issue:42886", "issue:42907", + "issue:43066", "issue:43122", "issue:43329", "issue:43408", @@ -15945,7 +15920,6 @@ "issue:43976", "issue:44038", "issue:44117", - "issue:44222", "issue:44263", "issue:44295", "issue:44297", @@ -15959,259 +15933,271 @@ "issue:44829", "issue:44945", "issue:44964", - "issue:45216", "issue:45278", "issue:45290", "issue:45381", - "issue:45478", "issue:46002", "issue:46172", "issue:46326", + "issue:46369", "issue:46762", "issue:46909", "issue:47059" ], "soft_pairs": [ + "issue:39379|issue:46172", + "issue:42886|issue:43066", + "issue:43408|issue:46369", "issue:42757|issue:45278", - "issue:45381|issue:46909", "issue:27957|issue:30990", "issue:43646|issue:46002", + "issue:45381|issue:46909", "issue:43540|issue:46909", "issue:30990|issue:33357", + "issue:44295|issue:44297", "issue:42907|issue:44964", "issue:43329|issue:47059", "issue:42907|issue:44945", - "issue:44295|issue:44297", - "issue:43122|issue:46326", "issue:44625|issue:44829", "issue:43976|issue:44945", - "issue:39401|issue:44488", - "issue:39401|issue:44521", + "issue:43122|issue:46326", "issue:41628|issue:45278", + "issue:39401|issue:44521", + "issue:39401|issue:44488", "issue:44117|issue:44704", "issue:42371|issue:44263", "issue:42907|issue:43937", - "issue:43122|issue:44521", "issue:44704|issue:45290", "issue:22355|issue:29159", - "issue:42907|issue:43452", + "issue:43122|issue:44521", "issue:43408|issue:43475", + "issue:42907|issue:43452", "issue:44360|issue:44485", "issue:44038|issue:44297", - "issue:46172|issue:46762", "issue:43824|issue:44351", - "issue:44038|issue:44295", - "issue:45216|issue:45478", - "issue:44038|issue:44222" + "issue:46172|issue:46762", + "issue:44038|issue:44295" ], - "prepared_review_unit_hash": "6dad0fe1676c328079b6bcd65eb3afd440fc90f2700fa335c7462cdd8a618b7e", + "prepared_review_unit_hash": "41ed60119f7d91753547b8b5a6caff202d1da7053e88d753344fd6903722ebc2", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 35919, - "estimated_input_tokens": 8980, - "estimated_eval_tokens": 18216 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36769, + "estimated_input_tokens": 9193, + "estimated_eval_tokens": 18642 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes import regressions, tokenizer behavior changes, multimodal/model-specific bugs, and a few unrelated framework issues. There are no PRs. The broadest umbrella issue is the generic import-regression report (#45278), but most soft pairs are not true duplicates.", - "confidence": 0.46, - "canonical_issue_reason": "Issue #45278 is the closest thing to an umbrella report: it captures broad import breakage after a version upgrade and can subsume several other import-related symptoms in the set. It is still only a loose canonical choice because the cluster overall is not a single bug family.", + "summary": "This cluster is largely heterogeneous: most items are separate bugs/features in different subsystems (imports, tokenizers, multimodal models, GLM/SAM, docs). The soft edges mostly reflect superficial overlap in keywords rather than the same underlying defect, so I reject all candidate duplicates.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45278 is the broadest and most general report here, centered on a regression causing multiple import failures after a version update; it best represents the small import-error subset of the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "#45278 is the best single issue to keep as the representative cluster item because it is the most general and most likely to explain other import-related reports; the rest are more specific one-off bugs or clearly different failure modes.", + "best_issue_reason": "issue:45278 is the best single issue to keep because it is broad, concrete, and likely to attract related import-regression reports, whereas the others are narrower and more clearly distinct.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:42757", - "right": "issue:45278", + "left": "issue:39379", + "right": "issue:46172", "accept": false, - "reason": "Both are import failures, but #42757 is a specific missing symbol from huggingface_hub while #45278 is a broad report of many import errors after an upgrade; not the same concrete bug." + "reason": "Different models and different bugs: mllama mask handling vs DeepSeek tensor transpose logic." }, { - "left": "issue:45381", - "right": "issue:46909", + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer-related, but one is offline cache loading and the other is decoder-type metadata in v5." + }, + { + "left": "issue:43408", + "right": "issue:46369", + "accept": false, + "reason": "Unrelated SAM3 issues: a model-type warning versus a memory overflow in a different model variant." + }, + { + "left": "issue:42757", + "right": "issue:45278", "accept": false, - "reason": "Different subsystems and symptoms: Qwen2.5-VL video position IDs vs a broken DIA documentation example." + "reason": "Both mention imports, but one is a specific huggingface_hub symbol regression and the other is a broad transformers upgrade breakage." }, { "left": "issue:27957", "right": "issue:30990", "accept": false, - "reason": "Flash Attention 2 on XLMRoberta and a Sentence-Transformers loading hang are unrelated failure modes." + "reason": "Different symptoms and code paths: Flash Attention 2 behavior in XLM-RoBERTa vs a loading hang in Sentence Transformers." }, { "left": "issue:43646", "right": "issue:46002", "accept": false, - "reason": "Custom model initialization breakage and NaNs in classification heads are different code paths and symptoms." + "reason": "Custom model init breakage and NaNs in classification heads are different failure modes." + }, + { + "left": "issue:45381", + "right": "issue:46909", + "accept": false, + "reason": "Qwen2.5-VL video position IDs and a DIA documentation example are unrelated." }, { "left": "issue:43540", "right": "issue:46909", "accept": false, - "reason": "Qwen3OmniMoe video-processing ValueError is unrelated to a DIA docs issue." + "reason": "Video-input processing bug in Qwen3OmniMoe is not the same as a broken docs example." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "A loading hang in Sentence Transformers is not the same as a MacOS bus error with a community CLIP model." + "reason": "Different models/platforms and distinct failures: loading stuck vs MacOS bus error." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "One is buffer access for position_ids; the other is tokenizer config metadata saved incorrectly." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and loading Phi-4 multimodal are different model and API paths." + "reason": "Dequantized save failure and multimodal model load failure are separate issues." }, { "left": "issue:43329", "right": "issue:47059", "accept": false, - "reason": "Undefined multimodal-token helper variables in video processing are unrelated to a CLI compatibility break with huggingface_hub." + "reason": "Video multimodal-token counting bug is unrelated to CLI completion breakage in huggingface_hub." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Checkpoint saving for dequantized models is unrelated to incorrect output under pipeline parallelism." - }, - { - "left": "issue:44295", - "right": "issue:44297", - "accept": false, - "reason": "Reading a buffer-backed position_ids tensor and mismatched tokenizer_class serialization are distinct save/load bugs." - }, - { - "left": "issue:43122", - "right": "issue:46326", - "accept": false, - "reason": "Tokenization changes across versions and continue_final_message handling for mistral_common are different tokenizer behaviors." + "reason": "Different areas and different failure modes: saving dequantized models vs pipeline-parallel output corruption." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "num_labels propagation in Qwen3.5 is unrelated to flash_attention_3 causing degenerate training." + "reason": "Qwen3.5 label propagation and Flash Attention 3 training degeneration are unrelated." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Python-version compatibility regression and pipeline-parallel output corruption do not share the same underlying bug." + "reason": "Python version compatibility and pipeline-parallel output correctness are not the same bug." }, { - "left": "issue:39401", - "right": "issue:44488", + "left": "issue:43122", + "right": "issue:46326", + "accept": false, + "reason": "Both involve tokenization, but one is version-to-version tokenization drift and the other is continue_final_message behavior with mistral_common." + }, + { + "left": "issue:41628", + "right": "issue:45278", "accept": false, - "reason": "Qwen3 tokenizer offset_mapping bug and failure to load cjvt/sleng-bert are separate model/tokenizer issues." + "reason": "Both are import errors, but they concern different missing symbols and different regression contexts." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Wrong offset mappings and zero assistant masks in multimodal chat templating are different tokenizer behaviors." + "reason": "Wrong offset mappings and zero assistant masks are distinct tokenizer/chat-template bugs." }, { - "left": "issue:41628", - "right": "issue:45278", + "left": "issue:39401", + "right": "issue:44488", "accept": false, - "reason": "Missing AutoImageProcessor import is a specific API export issue; #45278 is a broad umbrella for many import errors, not the same concrete failure." + "reason": "Tokenizer offset mapping bug and model-loading failure for cjvt/sleng-bert are unrelated." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES returning None and AutoProcessor not passing kwargs to cached_file are different call sites and bugs." + "reason": "Mapping-name None handling and AutoProcessor kwargs forwarding are different call sites and bugs." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 API guidance and a torch.split return-value issue in GlmMoeDsaIndexer are unrelated." + "reason": "TF32 API guidance and a torch.split indexer issue are unrelated." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Dequantized model saving and GLM-5 GenerationConfig validation are separate problems." - }, - { - "left": "issue:43122", - "right": "issue:44521", - "accept": false, - "reason": "Tokenizer output drift across versions is not the same as multimodal assistant_masks becoming all zeros." + "reason": "Dequantized save failure and invalid GenerationConfig for GLM-5 are separate defects." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Passing kwargs to cached_file and crashing on tool-call assistant messages are different apply_chat_template / processor bugs." + "reason": "AutoProcessor/cached_file kwargs forwarding is not the same as chat-template crashes on tool-call messages." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "A missing transformers.onnx module import and slow-vs-fast tokenizer inconsistency are unrelated." + "reason": "Missing transformers.onnx module and slow/fast tokenizer inconsistency are unrelated." }, { - "left": "issue:42907", - "right": "issue:43452", + "left": "issue:43122", + "right": "issue:44521", "accept": false, - "reason": "Saving dequantized checkpoints is not the same as gguf_file breaking from_pretrained loading." + "reason": "Tokenization drift across versions is not the same as multimodal assistant_mask generation." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "A model-type warning for sam3_video vs sam3_tracker and a missing fpn_position_embeddings attribute are different SAM3 issues." + "reason": "Both are SAM3-related, but one is a type warning while the other is a missing attribute in vision encoder output." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Dequantized save regression and gguf_file/from_pretrained loading breakage are different code paths." }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "A DSA indexer ReLU concern is unrelated to GLM-5 RoPE implementation." + "reason": "DSA indexer activation discussion and GLM-5 RoPE implementation are not the same issue." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe load failures and tokenizer_class mismatch on save are not the same bug." + "reason": "Qwen3-VL-Moe load bug and tokenizer save metadata mismatch are unrelated." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Missing Qwen2.5-VL and HybridCache imports are both import errors, but they are different symbols and likely different breakages." - }, - { - "left": "issue:44038", - "right": "issue:44295", - "accept": false, - "reason": "Qwen3-VL-Moe loading and buffer-backed position_ids errors are unrelated." + "reason": "Two import errors, but for different exported symbols and likely different regressions." }, { - "left": "issue:45216", - "right": "issue:45478", + "left": "issue:46172", + "right": "issue:46762", "accept": false, - "reason": "Both involve Qwen3.5 and version regressions, but one is save_pretrained checkpoint corruption while the other is a from_pretrained load error; not enough evidence they are the same concrete bug." + "reason": "Different numerical issues in different models: a tensor transpose bug versus attention output mismatch." }, { "left": "issue:44038", - "right": "issue:44222", + "right": "issue:44295", "accept": false, - "reason": "Qwen3-VL-Moe loading problems and FP8 save_pretrained issues are distinct model-path bugs." + "reason": "Qwen3-VL-Moe loading and position_ids buffer access are unrelated." } ] }, @@ -16233,7 +16219,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:22355", "issue:28348", @@ -16247,7 +16233,6 @@ "issue:42673", "issue:42757", "issue:42898", - "issue:43066", "issue:43116", "issue:43122", "issue:43232", @@ -16258,6 +16243,7 @@ "issue:43976", "issue:44038", "issue:44188", + "issue:44222", "issue:44351", "issue:44483", "issue:44485", @@ -16267,8 +16253,8 @@ "issue:44945", "issue:44995", "issue:45072", - "issue:45083", "issue:45200", + "issue:45216", "issue:45278", "issue:45357", "issue:45446", @@ -16285,6 +16271,8 @@ "issue:47312" ], "soft_pairs": [ + "issue:45216|issue:45478", + "issue:44038|issue:44222", "issue:43116|issue:45200", "issue:43122|issue:44779", "issue:22355|issue:29127", @@ -16299,217 +16287,201 @@ "issue:44188|issue:46328", "issue:46231|issue:46877", "issue:28348|issue:46640", - "issue:44483|issue:47217", "issue:42898|issue:46614", "issue:43611|issue:46002", - "issue:42673|issue:45072", + "issue:44483|issue:47217", "issue:43976|issue:45446", "issue:43937|issue:44485", + "issue:42673|issue:45072", "issue:45357|issue:45478", "issue:38617|issue:45278", "issue:44995|issue:46697", - "issue:44625|issue:46002", - "issue:43066|issue:44779", - "issue:45083|issue:45446", - "issue:43116|issue:43493", - "issue:44038|issue:45478" + "issue:44625|issue:46002" ], - "prepared_review_unit_hash": "1e2d0c05236ac7403e51885b71061d85f84110b4e1f4de5eb1280dab8d503d94", + "prepared_review_unit_hash": "c32e51b0a0e53f5ee0b561115c95988a1a6df58bd144e508d9627e5ff85e84e8", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 35546, - "estimated_input_tokens": 8887, - "estimated_eval_tokens": 18030 + "soft_pair_count": 26, + "serialized_chars": 34957, + "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 highly heterogeneous: most pairs are only loosely related by broad themes like tokenizer regressions, import errors, or model-family bugs, but they do not appear to be the same concrete underlying issue. I found no soft edge that should be merged.", - "confidence": 0.9, - "canonical_issue_reason": "issue:45278 is the broadest and most umbrella-like report in the set, covering multiple import failures after an upgrade, so it is the best representative issue if a single canonical node is needed.", + "summary": "No soft pairs look like true duplicates; they mostly share broad subsystems or release/version themes but describe different concrete bugs. The set is heterogeneous, so the best issue anchor is only a loose umbrella, not a real duplicate canonical.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45278 is the broadest regression-style report in the set and the closest umbrella for the import/version-breakage subset.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the strongest triage target because it captures a recurring, user-visible regression class rather than a narrow model-specific symptom.", + "best_issue_reason": "issue:45278 is the least specific and most representative of a cluster-wide regression theme, but it does not actually unify the rest of the issues.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 pretrained flows, but one is a save_pretrained regression and the other is a from_pretrained loading error; different code paths and symptoms." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Both mention MoE/transformers, but one is a Qwen3-VL-Moe bug and the other is an FP8 save_pretrained issue; not the same defect." + }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Different problems: example-script multi-label inference returning empty results vs Gemma 4 requiring mm_token_type_ids for text-only fine-tuning." + "reason": "Unrelated problems: example-script multi-label classification returns empty results versus Gemma 4 mm_token_type_ids defaulting behavior." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer-related, but they describe different models and different regressions; not the same bug." + "reason": "Both are tokenizer regressions, but for different models and different failure modes (generic tokenization change vs DeepSeek-specific incorrect output)." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "Import failure for transformers.onnx is unrelated to LayoutLMv3 error-message clarity." + "reason": "No meaningful overlap: an import error for transformers.onnx versus a LayoutLMv3 error-message clarity issue." }, { "left": "issue:39401", "right": "issue:46614", "accept": false, - "reason": "Wrong offset_mapping in Qwen3 and stale phonemizer backend state in Wav2Vec2PhonemeCTCTokenizer are distinct code paths." + "reason": "Different tokenizer implementations and bugs: Qwen3 offset_mapping versus Wav2Vec2PhonemeCTCTokenizer stale phonemizer language state." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "Embedding dtype drift in eval and incorrect output under pipeline parallelism are different symptoms with different causes." + "reason": "One is a dtype regression between train/eval, the other is incorrect output under pipeline parallelism; different failure mechanisms." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "One is a backend extra/installation packaging issue; the other is a Python 3.14 load failure." + "reason": "Both are import/startup-related, but one is a packaging/backend dependency issue and the other is a Python 3.14 compatibility failure." }, { "left": "issue:42222", "right": "issue:47312", "accept": false, - "reason": "VitPose breakage and kernel compatibility are unrelated." + "reason": "Completely different: broken vitpose model files versus kernel-version incompatibility." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor parallel training bug and empty multi-label results in the example script are not the same failure." + "reason": "Tensor parallel training bug and multi-label example-script behavior are unrelated." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different missing symbols in different packages; too broad to be the same bug." + "reason": "Both are ImportErrors, but for different missing symbols in different packages; likely separate version skew or deprecation issues." }, { "left": "issue:39379", "right": "issue:43232", "accept": false, - "reason": "mllama mask handling and generation kwargs after sync_gpus are separate generation issues." + "reason": "Both involve generation, but one is mllama mask handling and the other is sync_gpus/_update_model_kwargs_for_generation timing; not the same bug." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 implementation mismatch and Qwen3_5 infinite loop on one image are unrelated vision bugs." + "reason": "Different model families and failure types: SigLIP2 implementation mismatch versus Qwen3_5 infinite loop on a specific image." }, { "left": "issue:44188", "right": "issue:46328", "accept": false, - "reason": "Attention-kernel divergence under torch.compile and a padding performance issue are not the same underlying change." + "reason": "Attention-kernel divergence under torch.compile is unrelated to a SequenceFeatureExtractor.pad performance issue." }, { "left": "issue:46231", "right": "issue:46877", "accept": false, - "reason": "Docstring parameter mismatch and stdout diagnostics at import time are different documentation-tool problems." + "reason": "Documentation parameter-name mismatch and import-time docstring diagnostics are separate tooling/doc bugs." }, { "left": "issue:28348", "right": "issue:46640", "accept": false, - "reason": "Flash Attention 2 support request for GPT2LMHeadModel is not the same as a vague flash-attention issue report." - }, - { - "left": "issue:44483", - "right": "issue:47217", - "accept": false, - "reason": "Server request validation for /v1/chat/completions is distinct from multimodal placeholder handling in VL models." + "reason": "Both mention Flash Attention, but one is a GPT2LMHeadModel support request and the other is an unspecified issue; not the same concrete change." }, { "left": "issue:42898", "right": "issue:46614", "accept": false, - "reason": "Tokenizer-space cleanup behavior in v5 and stale phonemizer language state are unrelated." + "reason": "Tokenization-space cleanup behavior in v5 is unrelated to a phonemizer backend caching bug." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "Model loading breakage around base_model_prefix and NaNs in classification heads are different bugs." + "reason": "Model loading with base_model_prefix attribute versus NaNs in classification heads are different problems." }, { - "left": "issue:42673", - "right": "issue:45072", + "left": "issue:44483", + "right": "issue:47217", "accept": false, - "reason": "VRAM leak in multi-threaded Qwen3ForCausalLM use is unrelated to dtype mismatches in bf16 inference." + "reason": "Both touch chat/completions-style workflows, but one is request validation on the API and the other is multimodal placeholder placement in VL models." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python-version compatibility complaint and a PyTorch version guard bug in flex_attention are not the same issue." + "reason": "Both are version-check complaints, but they affect different compatibility gates and different code paths." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Invalid GenerationConfig for GLM-5 and RoPE implementation discussion are different concrete problems." + "reason": "GLM-5 generation config validation and RoPE implementation are distinct issues." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leakage in Qwen3ForCausalLM threads versus dtype mismatches in inference are unrelated." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "Incorrect save_pretrained visual-encoder keys and from_pretrained failures are separate Qwen3.5 MoE bugs." + "reason": "Both are Qwen3.5 save/load regressions, but one is incorrect visual encoder key serialization and the other is a from_pretrained loading error; too little evidence of a single root cause." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "A specific missing import and a broad umbrella of import errors are too general to merge as duplicates without the same concrete traceback." + "reason": "The latter is a broad import-regression report, but the former is a specific missing-symbol ImportError; not enough evidence they are the same defect." }, { "left": "issue:44995", "right": "issue:46697", "accept": false, - "reason": "Stale indexer cache crash and Swin2SR over-padding are unrelated." + "reason": "A stale cache crash on second forward pass is unrelated to image over-padding in an image processor." }, { "left": "issue:44625", "right": "issue:46002", "accept": false, - "reason": "num_labels propagation and NaNs after checkout/init affect different code paths." - }, - { - "left": "issue:43066", - "right": "issue:44779", - "accept": false, - "reason": "Different tokenizer regressions; no evidence they stem from the same fix or code path." - }, - { - "left": "issue:45083", - "right": "issue:45446", - "accept": false, - "reason": "Feature-extractor length calculation in qwen3_omni_moe is unrelated to AuxRequest import version checks." - }, - { - "left": "issue:43116", - "right": "issue:43493", - "accept": false, - "reason": "Multi-label classification example behavior and SigLIP2 implementation mismatch are unrelated." - }, - { - "left": "issue:44038", - "right": "issue:45478", - "accept": false, - "reason": "Qwen3-VL-Moe and Qwen3.5 MoE from_pretrained failures may be in the same family, but they are different concrete bugs and should not be merged." + "reason": "num_labels propagation in Qwen3.5 config and NaNs in classification heads are different configuration/initialization bugs." } ] }, @@ -16531,19 +16503,18 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:33453", "issue:39379", "issue:39401", - "issue:42222", "issue:42371", "issue:42673", "issue:42796", "issue:43064", "issue:43066", "issue:43116", - "issue:43441", + "issue:43493", "issue:43618", "issue:43673", "issue:43688", @@ -16553,24 +16524,25 @@ "issue:43874", "issue:43937", "issue:43992", + "issue:44038", "issue:44075", "issue:44117", - "issue:44263", "issue:44291", "issue:44355", - "issue:44380", "issue:44466", "issue:44610", "issue:44704", + "issue:44779", "issue:44936", "issue:45071", + "issue:45083", "issue:45092", "issue:45405", "issue:45446", + "issue:45478", "issue:45563", "issue:45588", "issue:46002", - "issue:46077", "issue:46231", "issue:46328", "issue:46710", @@ -16579,17 +16551,20 @@ "issue:46829", "issue:46909", "issue:46981", - "issue:47000", "issue:47116" ], "soft_pairs": [ - "issue:44610|issue:46909", + "issue:43066|issue:44779", + "issue:45083|issue:45446", + "issue:44038|issue:45478", + "issue:43116|issue:43493", "issue:44075|issue:44936", + "issue:44610|issue:46909", "issue:44355|issue:46231", "issue:43618|issue:46829", "issue:43066|issue:46710", - "issue:46726|issue:46981", "issue:42371|issue:43874", + "issue:46726|issue:46981", "issue:43116|issue:43825", "issue:42673|issue:42796", "issue:43992|issue:44704", @@ -16605,202 +16580,198 @@ "issue:43064|issue:43673", "issue:44117|issue:45588", "issue:43937|issue:45563", - "issue:44466|issue:46002", - "issue:44263|issue:46762", - "issue:42222|issue:43441", - "issue:44380|issue:46077", - "issue:42222|issue:47000" + "issue:44466|issue:46002" ], - "prepared_review_unit_hash": "e6ee710fde9a0a5983b77a1769933b1a81fad509be2066c499df66abe169076b", + "prepared_review_unit_hash": "219ba301a1720bd04e3b01820886436b92fd2b35a731beb0a109b4dd2085edb6", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35692, - "estimated_input_tokens": 8923, - "estimated_eval_tokens": 18102 + "serialized_chars": 35106, + "estimated_input_tokens": 8777, + "estimated_eval_tokens": 17810 }, "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/tokenizer/runtime/docs bugs, and none of the soft pairs look like the same underlying fix. No PRs are present.", - "confidence": 0.11, - "canonical_issue_reason": "issue:39401 is the clearest self-contained bug report here: a concrete tokenizer offset_mapping failure with a focused scope and repro potential. The cluster is otherwise too mixed to support a stronger canonical across all items.", + "summary": "This cluster is mostly heterogeneous: many items are separate v5 regressions or unrelated model/component bugs. The only loosely recurring theme is tokenizer-related regressions, but the soft pairs still don\u2019t look like true duplicates under a strict same-root-cause standard.", + "confidence": 0.66, + "canonical_issue_reason": "Issue 43066 is the clearest representative of the tokenizer-regression theme in this set and is specific enough to anchor the cluster, even though many neighboring issues are not actually duplicates of it.", "canonical_pr_reason": null, - "best_issue_reason": "issue:39401 is the most actionable standalone issue in the set because it names a specific failure mode and component, unlike the broader regression or documentation reports.", + "best_issue_reason": "43066 is the best single issue to represent the cluster because it is a concrete v5 tokenizer bug and sits closest to the small subset of tokenizer-related reports; the rest are too diverse to justify a different canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44610", - "right": "issue:46909", + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they describe different concrete failures and different models; not enough evidence they stem from the same code path." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Different subsystems and failure modes: a qwen3_omni_moe helper vs a flex_attention version check bug." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3-style MoE loading errors, but they target different model variants and distinct from_pretrained failures." + }, + { + "left": "issue:43116", + "right": "issue:43493", "accept": false, - "reason": "Different problems: processor image size mismatch vs a broken documentation example. No shared code path." + "reason": "A classification example returning empty results is unrelated to a SigLIP2 implementation discrepancy." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "Unrelated training optimizer bug vs trainer.evaluate() failure after training. Different execution paths and symptoms." + "reason": "Both are trainer/optimizer related, but one is about SGD args being ignored and the other is an evaluate-after-train failure." + }, + { + "left": "issue:44610", + "right": "issue:46909", + "accept": false, + "reason": "Processor input-size mismatch and a broken documentation example are unrelated." }, { "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": "Running compiled Python files and a docs parameter-name typo are not the same bug." }, { "left": "issue:43618", "right": "issue:46829", "accept": false, - "reason": "Both involve model outputs, but they affect different models and different fields; too broad to merge." + "reason": "Both concern output fields, but they affect different models and different returned attributes; not the same concrete defect." }, { "left": "issue:43066", "right": "issue:46710", "accept": false, - "reason": "Tokenizer decoder-type bug vs model accuracy regression after upgrade; not one concrete defect." + "reason": "A tokenizer-decoder regression and an accuracy regression in DeepSeek-R1 are too broad to treat as the same issue." }, { - "left": "issue:46726", - "right": "issue:46981", + "left": "issue:42371", + "right": "issue:43874", "accept": false, - "reason": "Processor backend-missing error vs repeated HTTP 429 retry behavior. Different failure modes." + "reason": "TF32 configuration guidance and a missing image-processor method are unrelated." }, { - "left": "issue:42371", - "right": "issue:43874", + "left": "issue:46726", + "right": "issue:46981", "accept": false, - "reason": "TF32 configuration guidance and a missing image-patch helper are unrelated." + "reason": "Both involve AutoProcessor, but one is about missing vision backend handling and the other is about repeated 429 retries." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Example script returning empty multi-label results is not the same as a pipeline error message regression." + "reason": "Both mention pipeline/example behavior, but one is a classification script bug and the other is a pipeline API/error-message mismatch." }, { "left": "issue:42673", "right": "issue:42796", "accept": false, - "reason": "VRAM leak in dataloader threads and tokenizer BPE-marker handling are unrelated." + "reason": "A VRAM leak in multi-threaded causal LM use is unrelated to a tokenizer BPE marker bug." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing encoder weights on load is separate from not forwarding kwargs to cached_file." + "reason": "One is a missing weight during UMT5 encoder loading; the other is AutoProcessor not forwarding kwargs to cached_file." }, { "left": "issue:46328", "right": "issue:47116", "accept": false, - "reason": "Performance issue in padding code vs streaming pipeline regression; no shared bug." + "reason": "A padding performance issue and a streaming pipeline regression are not the same underlying bug." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Docs still referencing removed pipelines and a PyTorch version check bug are unrelated." + "reason": "Docs still referencing pipeline() and a flex_attention version check are unrelated." }, { "left": "issue:39379", "right": "issue:43688", "accept": false, - "reason": "mllama mask handling and auxiliary-loss normalization are different model-specific errors." + "reason": "Incorrect mllama mask handling and auxiliary-loss normalization in OLMoE/GPT Oss are different model bugs." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Both are tokenizer-related, but one is a generic loading regression and the other is an offset_mapping bug; not the same defect." + "reason": "A generic tokenizer-loading regression and Qwen3 offset_mapping are only superficially related." }, { "left": "issue:43824", "right": "issue:46981", "accept": false, - "reason": "ImportError for a missing model class is unrelated to retry behavior on HTTP 429." + "reason": "An import error for a model class and AutoProcessor retry behavior on 429s are unrelated." }, { "left": "issue:43673", "right": "issue:46328", "accept": false, - "reason": "Generation cache regression and sequence feature padding performance are different code paths." + "reason": "Missing generation cache and SequenceFeatureExtractor padding inefficiency are unrelated." }, { "left": "issue:45092", "right": "issue:46762", "accept": false, - "reason": "Meta-initialization incompatibility for old checkpoints vs numerical attention discrepancies; not mergeable as one fix." + "reason": "Meta-init incompatibility for InternVL2 and numerical attention differences in Minimax are not the same bug." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "Version pin bump and a PyTorch version-check bug are separate release/compatibility issues." + "reason": "A released-version dependency bump and a PyTorch version check bug are unrelated." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "init_empty_weights argument mismatch and PretrainedConfig type-checking breakage are distinct v5 compatibility bugs." + "reason": "Both are v5 loading/configuration regressions, but they concern different failure points and different APIs." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "Distributed optimizer-state corruption and generation cache missing are unrelated trainer/generation problems." + "reason": "Wrong optimizer state handling in FSDP2/PEFT is unrelated to a missing GenerationMixin cache in chunked prefill." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping None-handling and flash_attention s_aux=None crash are different subsystems." + "reason": "A tokenizer mapping assumption and a flash_attention AttributeError are unrelated." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "Invalid GenerationConfig and a stale num_return_sequences warning are not the same bug." + "reason": "An invalid GenerationConfig and a stale warning in paged generate() are different issues." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "Serialization inconsistency and NaNs in classification heads are unrelated." - }, - { - "left": "issue:44263", - "right": "issue:46762", - "accept": false, - "reason": "torch.split return handling in one GLM component vs Minimax attention numerical differences are separate issues." - }, - { - "left": "issue:42222", - "right": "issue:43441", - "accept": false, - "reason": "VitPose breakage and Ministral FlashAttention failure are different model-specific regressions." - }, - { - "left": "issue:44380", - "right": "issue:46077", - "accept": false, - "reason": "Both are attention-related, but one is scaling config ignored in backends and the other is bidirectional mask construction; different concrete bugs." - }, - { - "left": "issue:42222", - "right": "issue:47000", - "accept": false, - "reason": "VitPose breakage and RTDetrLoss infeasible-cost-matrix crashes are unrelated." + "reason": "Serialization of tied weights depending on device is unrelated to NaNs in classification heads after checkout/init." } ] }, @@ -16822,7 +16793,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:38617", "issue:39379", @@ -16831,10 +16802,10 @@ "issue:42222", "issue:42371", "issue:42617", - "issue:42831", "issue:43122", "issue:43208", "issue:43408", + "issue:43441", "issue:43450", "issue:43742", "issue:43937", @@ -16842,23 +16813,23 @@ "issue:44117", "issue:44220", "issue:44261", + "issue:44263", + "issue:44380", "issue:44410", - "issue:44484", "issue:44496", "issue:44704", "issue:44779", "issue:44829", "issue:44877", - "issue:44933", "issue:44977", "issue:45071", "issue:45081", "issue:45216", - "issue:45278", "issue:45310", "issue:45341", "issue:45561", "issue:45632", + "issue:46077", "issue:46209", "issue:46231", "issue:46298", @@ -16869,13 +16840,18 @@ "issue:46894", "issue:46897", "issue:46981", + "issue:47000", "issue:47059", "issue:47110" ], "soft_pairs": [ + "issue:42222|issue:43441", + "issue:44263|issue:46762", + "issue:42222|issue:47000", + "issue:44380|issue:46077", "issue:39379|issue:46897", - "issue:44117|issue:45081", "issue:42222|issue:46752", + "issue:44117|issue:45081", "issue:43937|issue:44977", "issue:42617|issue:45341", "issue:45071|issue:46512", @@ -16883,221 +16859,224 @@ "issue:39401|issue:44779", "issue:43408|issue:43742", "issue:44410|issue:45310", - "issue:43208|issue:46894", "issue:44829|issue:45216", + "issue:43208|issue:46894", "issue:45632|issue:46981", "issue:45071|issue:47059", - "issue:45341|issue:45561", - "issue:44704|issue:47110", "issue:45341|issue:46328", + "issue:44704|issue:47110", + "issue:45341|issue:45561", "issue:44877|issue:46231", "issue:38617|issue:41628", "issue:39401|issue:46209", "issue:43122|issue:46298", "issue:43937|issue:44496", "issue:42371|issue:44261", - "issue:38617|issue:44220", "issue:44261|issue:46762", - "issue:44933|issue:45278", - "issue:44484|issue:45310", - "issue:42371|issue:42831" + "issue:38617|issue:44220" ], - "prepared_review_unit_hash": "8c5038d04cebdfe9fda542d43b563062559e4fb3649ff9012896d510e8fcfee0", + "prepared_review_unit_hash": "9f8ba4e98084a331476daf5d9bc1bbeb35e5c9ed8252536c639218b3ace3a485", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35261, - "estimated_input_tokens": 8816, - "estimated_eval_tokens": 17888 + "node_count": 48, + "item_count": 48, + "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": "No soft pair looks like the same underlying bug or change. The similarities are mostly broad-topic overlaps (imports, tokenizers, configs, testing utils, video models), but the concrete failures, code paths, and affected models differ enough to reject all edges.", - "confidence": 0.93, - "canonical_issue_reason": null, + "summary": "This cluster is mostly a lexical bucket of unrelated Transformers issues. None of the soft edges look like the same underlying bug or change, so all should stay separate.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 42222 is the most discussion-heavy concrete regression in the set and the closest thing to a hub, but the cluster is still too heterogeneous to treat it as a true canonical duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45278 is the broadest representative of the import-regression reports, but it still isn\u2019t a safe canonical because the cluster is not actually one duplicate family.", + "best_issue_reason": "Issue 42222 is the best representative issue here because it is concrete, impactful, and has the most activity among the candidates, even though it does not actually duplicate the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:39379", - "right": "issue:46897", + "left": "issue:42222", + "right": "issue:43441", "accept": false, - "reason": "Different bugs in different models: mllama mask handling vs Florence2 training-loss double-shift." + "reason": "VitPose model breakage and Ministral FlashAttention failure are unrelated bugs." }, { - "left": "issue:44117", - "right": "issue:45081", + "left": "issue:44263", + "right": "issue:46762", + "accept": false, + "reason": "torch.split indexing behavior and Minimax numerical mismatch are different code paths." + }, + { + "left": "issue:42222", + "right": "issue:47000", + "accept": false, + "reason": "VitPose loading breakage is unrelated to RTDetr loss infeasibility." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT2 attention scaling and Gemma4 bidirectional masks are distinct model-specific bugs." + }, + { + "left": "issue:39379", + "right": "issue:46897", "accept": false, - "reason": "Tokenizer mapping fallback bug is unrelated to the Mistral regex patch crash." + "reason": "mllama mask handling and Florence2 double-shift training loss are unrelated." }, { "left": "issue:42222", "right": "issue:46752", "accept": false, - "reason": "VitPose model breakage is unrelated to an empty-chat-template IndexError." + "reason": "VitPose model breakage does not match apply_chat_template empty-list crash." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer registry None handling and Mistral regex patch crash differ materially." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "Different model-specific generation issues: invalid GenerationConfig vs flash-attention behavior in Qwen3.5." + "reason": "Invalid GenerationConfig and flash-attention generation regression are different failures." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "3d_parallel.py runtime failure is not the same as a testing_utils.py documentation/code bug." + "reason": "3d_parallel.py runtime failure is unrelated to a testing_utils documentation bug." }, { "left": "issue:45071", "right": "issue:46512", "accept": false, - "reason": "Type-checking breakage in PreTrainedConfig is unrelated to stale model.config.dtype after from_config()." + "reason": "PretrainedConfig type checking and stale config.dtype are not the same bug." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Incorrect batched video processor shapes are a different issue from AutoProcessor failing without torchvision." + "reason": "Batched video processor shape and torchvision-less AutoProcessor crash differ." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer bugs, but for different tokenizers and different failure modes (offset mapping vs tokenization output)." + "reason": "Qwen3 offset_mapping corruption and Deepseek tokenization regression are separate issues." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "sam3 tracker/video model warning is unrelated to loading MobileLLM-125M." + "reason": "sam3_video/sam3_tracker warning is unrelated to the MobileLLM key error." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Missing Qwen3Next projections is a model architecture bug, not the Qwen3.5 MoE from_pretrained error." + "reason": "Missing QKV/gate projections and Qwen3.5 MoE from_pretrained failure are different bugs." }, { - "left": "issue:43208", - "right": "issue:46894", + "left": "issue:44829", + "right": "issue:45216", "accept": false, - "reason": "xLSTM training blockers are unrelated to the CohereASR double-shift training-loss bug." + "reason": "FlashAttention-3 training degeneration and save_pretrained checkpoint regression do not match." }, { - "left": "issue:44829", - "right": "issue:45216", + "left": "issue:43208", + "right": "issue:46894", "accept": false, - "reason": "Flash-attention training degeneration is not the same as an incorrect save_pretrained checkpoint for Qwen3.5." + "reason": "xLSTM small-model blockers and CohereASR double-shift loss bug are unrelated." }, { "left": "issue:45632", "right": "issue:46981", "accept": false, - "reason": "Cache-path collisions for local models are unrelated to repeated HTTP 429 retries in AutoProcessor." + "reason": "Cache-path collision and repeated 429 retries are different AutoProcessor problems." }, { "left": "issue:45071", "right": "issue:47059", "accept": false, - "reason": "PreTrainedConfig type-checking regression is unrelated to CLI completion breakage from huggingface_hub." + "reason": "Config type-checking regression and CLI completion breakage are unrelated." }, { "left": "issue:45341", - "right": "issue:45561", + "right": "issue:46328", "accept": false, - "reason": "A small testing_utils bug is not the same as the xdist captured_info.txt race condition." + "reason": "Testing utils bug and SequenceFeatureExtractor.pad performance issue are separate." }, { "left": "issue:44704", "right": "issue:47110", "accept": false, - "reason": "Missing kwargs to cached_file is unrelated to special_tokens being a list instead of a dict." + "reason": "Cached_file kwargs forwarding and special_tokens list handling are different code paths." }, { "left": "issue:45341", - "right": "issue:46328", + "right": "issue:45561", "accept": false, - "reason": "Both mention testing/utility code, but one is a bug and the other is a performance issue in SequenceFeatureExtractor.pad." + "reason": "Generic testing_utils note and xdist captured_info.txt race are not the same defect." }, { "left": "issue:44877", "right": "issue:46231", "accept": false, - "reason": "Strict config loading for granite_speech is unrelated to the set_training doc parameter mismatch." + "reason": "Strict config loading and a docs parameter-name mismatch are unrelated." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols and different modules." + "reason": "Missing exported symbol layer_type_validation and AutoImageProcessor import failure are distinct." }, { "left": "issue:39401", "right": "issue:46209", "accept": false, - "reason": "Tokenizer offset mapping regression is unrelated to FP8 inference failure on B200." + "reason": "Tokenizer offset mapping and FP8 inference on B200 are unrelated." }, { "left": "issue:43122", "right": "issue:46298", "accept": false, - "reason": "Tokenizer behavior change is unrelated to a noisy logging message." + "reason": "Tokenization regression across versions and noisy logging are not the same issue." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "Invalid GenerationConfig and unrecognized model config are separate loading failures." + "reason": "GenerationConfig invalidity and unrecognized model_type are different load-time failures." }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 API guidance and MLA rms_norm_eps precision error are different topics and code paths." - }, - { - "left": "issue:38617", - "right": "issue:44220", - "accept": false, - "reason": "ImportError for layer_type_validation is unrelated to the fbank feature extraction bug." + "reason": "TF32 API warning and rms_norm_eps precision bug are unrelated." }, { "left": "issue:44261", "right": "issue:46762", "accept": false, - "reason": "Similar numerical symptoms, but one is a missing config epsilon and the other is a separate attention implementation discrepancy." - }, - { - "left": "issue:44933", - "right": "issue:45278", - "accept": false, - "reason": "A missing import from image_utils is not the same as a broad wave of import errors after upgrading." - }, - { - "left": "issue:44484", - "right": "issue:45310", - "accept": false, - "reason": "save_pretrained shard-size behavior is unrelated to the Qwen3.5 MoE from_pretrained failure." + "reason": "LayerNorm epsilon precision issue and Minimax attention mismatch are different bugs." }, { - "left": "issue:42371", - "right": "issue:42831", + "left": "issue:38617", + "right": "issue:44220", "accept": false, - "reason": "TF32 settings guidance is unrelated to FineGrainedFP8 accuracy issues." + "reason": "Import error in configuration_utils is unrelated to fbank feature extraction." } ] }, @@ -17119,10 +17098,12 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:27957", "issue:41628", + "issue:42371", + "issue:42831", "issue:43012", "issue:43066", "issue:43097", @@ -17151,26 +17132,27 @@ "issue:44485", "issue:44493", "issue:44829", + "issue:44933", "issue:44987", "issue:45003", "issue:45071", + "issue:45278", + "issue:45310", "issue:45341", - "issue:45440", "issue:45464", "issue:45478", "issue:45588", "issue:45656", - "issue:45753", - "issue:46050", "issue:46172", - "issue:46369", "issue:46752", "issue:46897", "issue:46981", - "issue:47116", - "issue:47312" + "issue:47116" ], "soft_pairs": [ + "issue:44933|issue:45278", + "issue:44484|issue:45310", + "issue:42371|issue:42831", "issue:44075|issue:45656", "issue:44483|issue:45464", "issue:43097|issue:44493", @@ -17183,247 +17165,237 @@ "issue:46752|issue:47116", "issue:41628|issue:45071", "issue:43931|issue:44038", - "issue:43012|issue:43577", "issue:43874|issue:44261", + "issue:43012|issue:43577", "issue:43441|issue:44829", "issue:27957|issue:43352", - "issue:43097|issue:43452", "issue:43208|issue:46897", + "issue:43097|issue:43452", "issue:44484|issue:45478", "issue:44263|issue:46172", "issue:44360|issue:46172", "issue:43531|issue:45341", "issue:44117|issue:44987", - "issue:43986|issue:45341", "issue:44261|issue:44485", + "issue:43986|issue:45341", "issue:45003|issue:45341", "issue:43097|issue:45588", - "issue:44246|issue:46981", - "issue:45753|issue:46369", - "issue:45440|issue:46172", - "issue:43097|issue:46050", - "issue:43986|issue:47312" + "issue:44246|issue:46981" ], - "prepared_review_unit_hash": "5d54ffb0c253d5c7f71e93a8ae8dd1dfe5103fe2d5dfd0b1d12d5c3d379a261e", + "prepared_review_unit_hash": "74b60a1a0cc547a0872cd761da7389d387b8f7fdc14ad4059fd8ac8f9adcb1a7", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 36768, - "estimated_input_tokens": 9192, - "estimated_eval_tokens": 18640 + "soft_pair_count": 31, + "serialized_chars": 36419, + "estimated_input_tokens": 9105, + "estimated_eval_tokens": 18466 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is highly heterogeneous; these are mostly unrelated model-loading, tokenizer, FlashAttention, optimizer, and API regressions. I did not find any soft edge that looks like the same underlying bug/change, so all candidate merges should be rejected.", - "confidence": 0.84, - "canonical_issue_reason": "Issue #44246 is the broadest, most cross-cutting report in the set, but there is no strong duplicate core across the cluster; it is the least bad representative choice rather than a true canonical match.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers issues, with a few loose v5/tokenizer/import regressions and many model-specific bugs. I would not merge any of the soft pairs; they describe different underlying failures or code paths.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 45278 is the broadest representative of the cluster\u2019s dominant theme: post-upgrade Transformers v5 breakages causing import/load regressions. The cluster is heterogeneous, so this is only a loose canonical choice.", "canonical_pr_reason": null, - "best_issue_reason": "#44246 is the most generally framed issue and has the highest discussion activity, but the cluster does not form a real duplicate set.", + "best_issue_reason": "45278 is the most general and cluster-like issue among mostly narrow, model-specific reports. It best captures the shared upgrade-regression theme without being tied to one niche model.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "Both involve import failures, but 44933 is a specific nonexistent image_utils import while 45278 is a broad version-upgrade import regression. Not the same bug." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "Different problems: save_pretrained shard sizing vs Qwen3.5 MoE from_pretrained loading failure." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 configuration warnings/settings and FineGrainedFP8 accuracy issues are different precision-related topics, not one concrete defect." + }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "Different optimizer bugs: ignored SGD args vs optimizer step being called twice with deepspeed." + "reason": "SGD arguments being ignored is unrelated to optimizer.step being called twice under deepspeed." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Different chat/completions failures: request rejection vs streaming inference failure on a specific model." + "reason": "Both touch chat/completions, but one is a request acceptance/regression issue and the other is a streaming failure on a specific model. Different code paths." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Unrelated issues: removed tie_embeddings flag vs unexpected position-id key errors." + "reason": "Removal of tie_embeddings_and_encoder_decoder and unexpected position-id keys are unrelated breakages." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Different Qwen model-loading problems with different missing components and failure modes." + "reason": "Different Qwen-family model loading failures with different missing/mismatched components." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Both are tokenizer regressions, but they describe different concrete failures and likely different root causes." + "reason": "Both involve tokenizer changes in v5, but one is wrong decoder type and the other is changed tokenization output; not the same concrete bug." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "GLM DSA split semantics vs GLM-5 RoPE implementation; not the same bug." + "reason": "torch.split return-value handling in GlmMoeDsaIndexer is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:44077", "right": "issue:46981", "accept": false, - "reason": "Unrelated: patchtsmixer post_init policy vs AutoProcessor HTTP 429 retry behavior." + "reason": "patchtsmixer post_init policy and repeated 429 retries in AutoProcessor are unrelated." }, { "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Different Qwen families and different loading errors: shape mismatch vs missing projections." + "reason": "Model weight-shape mismatch and missing attn projections are different Qwen-related model bugs." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Generation cache bug and a testing_utils bug are unrelated." + "reason": "Generation cache missing in chunked_prefill and a testing_utils bug are unrelated." }, { "left": "issue:46752", "right": "issue:47116", "accept": false, - "reason": "Empty chat-template handling is unrelated to the streaming pipeline regression." + "reason": "Empty-list apply_chat_template crash and streaming pipeline regression are different API failures." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Import error for AutoImageProcessor is unrelated to PretrainedConfig type-checking breakage." + "reason": "AutoImageProcessor import failure and PreTrainedConfig type-checking breakage are unrelated." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "Both involve Qwen3-VL(-Moe), but the reported failures are different model-loading issues." + "reason": "Both are Qwen3-VL-related, but one is a weight-shape mismatch while the other is a from_pretrained error in a different model variant." }, { - "left": "issue:43012", - "right": "issue:43577", + "left": "issue:43874", + "right": "issue:44261", "accept": false, - "reason": "Both involve dtype propagation, but one is a PyTorch compile warning while the other is a BLIP2 loading bug." + "reason": "Missing image patch counting and an rms_norm_eps precision issue are not the same bug." }, { - "left": "issue:43874", - "right": "issue:44261", + "left": "issue:43012", + "right": "issue:43577", "accept": false, - "reason": "Missing image-patch method vs missing rms_norm_eps config are unrelated." + "reason": "A compile-time float32 warning and Blip2 dtype fields staying float32 are related only at a very high level, not as one fixable defect." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Both mention FlashAttention, but one is support/crash in inference and the other is degenerate training behavior." + "reason": "Both mention FlashAttention, but one is a model support/error path and the other is a training-quality regression with flash_attention_3." }, { "left": "issue:27957", "right": "issue:43352", "accept": false, - "reason": "Different model-specific FlashAttention 2 support gaps; same subsystem, not the same bug." + "reason": "One is enabling Flash Attention 2 for XLMRoberta, the other is a model not yet supporting Flash Attention 2.0." }, { - "left": "issue:43097", - "right": "issue:43452", + "left": "issue:43208", + "right": "issue:46897", "accept": false, - "reason": "Tie-embeddings config removal and gguf_file loading breakage are unrelated." + "reason": "xLSTM training bugs and Florence2 double-shift loss bug are unrelated model-specific issues." }, { - "left": "issue:43208", - "right": "issue:46897", + "left": "issue:43097", + "right": "issue:43452", "accept": false, - "reason": "xLSTM training blockers and Florence2 double-shift loss bug are different model-specific issues." + "reason": "A removed config attribute and gguf_file/from_pretrained failures are different loader/API problems." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "max_shard_size behavior is unrelated to Qwen3.5-MoE from_pretrained failures." + "reason": "max_shard_size behavior is unrelated to Qwen3.5 MoE loading errors." }, { "left": "issue:44263", "right": "issue:46172", "accept": false, - "reason": "Different GLM/DeepSeek internals and different failure mechanisms." + "reason": "GlmMoeDsaIndexer split handling and deepseek-v4 transpose/dtype bug are distinct implementation errors." }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "DSA indexer ReLU issue is unrelated to the DeepSeek comb dtype/transpose bug." + "reason": "Missing ReLU in DSA indexer and deepseek-v4 tensor transpose behavior are unrelated." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to a testing_utils bug." + "reason": "Qwen3-MoE sliding_window behavior and a testing_utils bug do not share the same underlying defect." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Tokenizer mapping None is too general to equate with the physical-intelligence/fast loading failure." + "reason": "Tokenizer mapping returning None and a physical-intelligence/fast load failure are only loosely related at the auto-tokenizer layer, not the same bug." }, { - "left": "issue:43986", - "right": "issue:45341", + "left": "issue:44261", + "right": "issue:44485", "accept": false, - "reason": "Video model AutoProcessor crash without torchvision is unrelated to testing_utils." + "reason": "rms_norm_eps precision issue and GLM-5 RoPE implementation are separate model internals." }, { - "left": "issue:44261", - "right": "issue:44485", + "left": "issue:43986", + "right": "issue:45341", "accept": false, - "reason": "Both are GLM-related, but one is a normalization epsilon issue and the other is a RoPE implementation issue." + "reason": "AutoProcessor/torchvision crash and testing_utils bug are unrelated." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "sys.modules access in modeling_utils and a testing_utils bug are unrelated." + "reason": "sys.modules access bug in modeling_utils and testing_utils bug are unrelated." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "Removed tie_embeddings flag and flash_attention.py s_aux=None crash are different code paths." + "reason": "Removed tie_embeddings_and_encoder_decoder and flash_attention s_aux=None crash are different failures." }, { "left": "issue:44246", "right": "issue:46981", "accept": false, - "reason": "Import slowness is not the same as repeated HTTP 429 retries during model loading." - }, - { - "left": "issue:45753", - "right": "issue:46369", - "accept": false, - "reason": "Qwen3_5 infinite-loop behavior and sam3-litetext memory overflow are different bugs." - }, - { - "left": "issue:45440", - "right": "issue:46172", - "accept": false, - "reason": "DeepSeekV3MoE divergence from remote implementation is unrelated to the Deepseek-v4 comb dtype/transpose bug." - }, - { - "left": "issue:43097", - "right": "issue:46050", - "accept": false, - "reason": "Tie-embeddings removal and sinq save/load quantization breakage are unrelated." - }, - { - "left": "issue:43986", - "right": "issue:47312", - "accept": false, - "reason": "A torchvision-dependent AutoProcessor crash is unrelated to a kernels 0.16 compatibility issue." + "reason": "Import latency and repeated 429 retries are different performance/network behaviors, not duplicates." } ] }, @@ -17445,7 +17417,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:38617", "issue:42757", @@ -17453,7 +17425,6 @@ "issue:42907", "issue:43012", "issue:43064", - "issue:43066", "issue:43097", "issue:43208", "issue:43502", @@ -17461,7 +17432,6 @@ "issue:43673", "issue:43873", "issue:43874", - "issue:43906", "issue:43986", "issue:43992", "issue:44038", @@ -17474,19 +17444,20 @@ "issue:44410", "issue:44451", "issue:44484", - "issue:44655", "issue:44704", "issue:44936", "issue:44938", "issue:45103", "issue:45375", - "issue:45712", + "issue:45440", "issue:45753", + "issue:46050", "issue:46123", "issue:46172", "issue:46178", "issue:46298", "issue:46365", + "issue:46369", "issue:46421", "issue:46655", "issue:46664", @@ -17494,25 +17465,30 @@ "issue:46710", "issue:46752", "issue:46821", - "issue:47116" + "issue:47116", + "issue:47312" ], "soft_pairs": [ + "issue:45753|issue:46369", + "issue:45440|issue:46172", + "issue:43097|issue:46050", + "issue:43986|issue:47312", "issue:44410|issue:45375", - "issue:43673|issue:46123", "issue:43874|issue:44263", + "issue:43673|issue:46123", "issue:43097|issue:43873", - "issue:44062|issue:44117", "issue:44038|issue:44410", "issue:43012|issue:46821", + "issue:44062|issue:44117", "issue:38617|issue:42757", "issue:44117|issue:46701", "issue:43064|issue:44315", "issue:46655|issue:46664", "issue:46298|issue:46752", - "issue:43986|issue:44484", "issue:46172|issue:46710", - "issue:42890|issue:46298", + "issue:43986|issue:44484", "issue:38617|issue:44351", + "issue:42890|issue:46298", "issue:44704|issue:45103", "issue:44936|issue:46365", "issue:43502|issue:43992", @@ -17521,226 +17497,210 @@ "issue:44380|issue:46421", "issue:44938|issue:46178", "issue:43064|issue:47116", - "issue:42907|issue:45753", - "issue:42907|issue:44655", - "issue:43012|issue:43550", - "issue:45712|issue:46298", - "issue:43066|issue:43906" + "issue:42907|issue:45753" ], - "prepared_review_unit_hash": "e94d41da64e628d06c74b85cde55590817e9ba891715f0776b20d503c19525ed", + "prepared_review_unit_hash": "a3d27c83dce933b3aa8193f04f08db83bf65e6ca16caeb41a45a18828cc0cfe9", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36089, - "estimated_input_tokens": 9023, - "estimated_eval_tokens": 18302 + "serialized_chars": 36020, + "estimated_input_tokens": 9005, + "estimated_eval_tokens": 18266 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most pairs only share a subsystem name or vague v5/regression language, not the same underlying bug. I reject all soft edges; there is no true duplicate-pr style grouping here.", - "confidence": 0.22, - "canonical_issue_reason": "Issue 43097 is the least-bad representative only because it is a concrete, user-facing v5 regression report with discussion and inbound references; however, the cluster is too mixed for a strong canonical duplicate.", + "summary": "This cluster is not a duplicate set: the items are a grab bag of unrelated Transformers issues (imports, model-specific bugs, training regressions, processor/API behavior, and a few versioning problems). The soft-similarity scores are driven by broad vocabulary overlap, not a shared underlying fix.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 43097 is the best single issue to anchor the cluster if one must be chosen, but it is not a genuine umbrella for the other items.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44410", - "right": "issue:45375", + "left": "issue:45753", + "right": "issue:46369", "accept": false, - "reason": "Different Qwen3 vision bugs: missing projections vs a missing config field." + "reason": "Both mention model behavior, but one is a Qwen3-specific infinite loop on an image and the other is a Trainer import issue tied to datasets; no shared code path." }, { - "left": "issue:43673", - "right": "issue:46123", + "left": "issue:45440", + "right": "issue:46172", "accept": false, - "reason": "Chunked prefill cache missing vs mask pipeline final-batch handling; unrelated code paths." + "reason": "Different models and failures: DeepseekV3 implementation divergence vs a deepseek-v4 tensor transpose/dtype bug." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "One is a v5 API removal complaint, the other is a quantization save/load regression; unrelated underlying problems." + }, + { + "left": "issue:43986", + "right": "issue:47312", + "accept": false, + "reason": "Video AutoProcessor crash without torchvision is unrelated to a kernels 0.16 compatibility issue." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen-family vision/model config issues, but one is missing projections in a layer and the other is a silently dropped config field; not the same bug." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Same GLM family, but image-processor API missing vs tensor split indexing bug are different failures." + "reason": "Both touch GLM-related code, but one is a missing image processor method and the other concerns torch.split return values in a different component." }, { - "left": "issue:43097", - "right": "issue:43873", + "left": "issue:43673", + "right": "issue:46123", "accept": false, - "reason": "v5 API removal/regression vs quantization offloading behavior; not the same bug." + "reason": "Both involve chunked prefill terminology, but one is missing GenerationMixin cache in v5 and the other is a pipeline batching bug that drops results." }, { - "left": "issue:44062", - "right": "issue:44117", + "left": "issue:43097", + "right": "issue:43873", "accept": false, - "reason": "Tokenizer AddedToken constructor error vs tokenizer mapping None; different root causes." + "reason": "Unrelated: tie_embeddings API removal vs offloading failing with quantization." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Both are Qwen-related, but one is a model/runtime bug and the other is missing layer projections." + "reason": "Different model families and failures; one is a generic transformers 5.0 issue with Qwen3-VL-Moe, the other is missing projections in qwen3next layer 0." }, { "left": "issue:43012", "right": "issue:46821", "accept": false, - "reason": "PyTorch bfloat16 warning vs TimesFM window_size AttributeError are unrelated." + "reason": "A PyTorch bfloat16 precision warning is not the same as a TimesFM window_size AttributeError." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizers AddedToken construction bug vs tokenizer mapping returning None during from_pretrained; similar area but different failure modes." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols in different modules and likely different causes." + "reason": "Both are import errors, but they refer to different symbols from different packages and likely different dependency/version breaks." }, { "left": "issue:44117", "right": "issue:46701", "accept": false, - "reason": "Tokenizer mapping fallback bug vs diffusion pipeline output failure; unrelated." + "reason": "TOKENIZER_MAPPING_NAMES/ from_pretrained logic is unrelated to the DiffusionGemmaGenerationOutput image-text pipeline failure." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "Different training issues: optimizer state corruption vs Liger kernel application with model_init." + "reason": "Different training-stack problems: optimizer state placement with FSDP2/PEFT vs Liger Kernel not applied with model_init." }, { "left": "issue:46655", "right": "issue:46664", "accept": false, - "reason": "Same SmolVLM video processor area, but padding calculation and num_frames semantics are distinct bugs." + "reason": "Both are SmolVLM video-processor issues, but one is padding calculation and the other is num_frames semantics; separate bugs." }, { "left": "issue:46298", "right": "issue:46752", "accept": false, - "reason": "Noisy logging vs empty-list chat template crash; unrelated." - }, - { - "left": "issue:43986", - "right": "issue:44484", - "accept": false, - "reason": "AutoProcessor/torchvision crash vs save_pretrained shard-size question; not the same issue." + "reason": "Noisy logging and apply_chat_template IndexError are unrelated." }, { "left": "issue:46172", "right": "issue:46710", "accept": false, - "reason": "DeepSeek tensor operation bug vs post-upgrade accuracy regression; too broad and different." + "reason": "Deepseek-v4 tensor transpose bug is not the same as an accuracy regression after upgrading Transformers." }, { - "left": "issue:42890", - "right": "issue:46298", + "left": "issue:43986", + "right": "issue:44484", "accept": false, - "reason": "Test flakiness from missing seeds vs logging noise; unrelated." + "reason": "AutoProcessor/torchvision crash is unrelated to max_shard_size default behavior in save_pretrained." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Different missing imports from different symbols/modules, not one duplicated export bug." + "reason": "Both are import-name failures, but one is missing layer_type_validation and the other is missing HybridCache; different symbols and likely different version mismatches." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "Seed-related test flakiness is not the same as a noisy logging message." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "cached_file kwarg forwarding vs auto_docstring annotations crash; different code paths." + "reason": "Passing kwargs to cached_file and auto_docstring crashing under future annotations are unrelated code paths." }, { "left": "issue:44936", "right": "issue:46365", "accept": false, - "reason": "Trainer lifecycle bug vs import conflict with datasets; unrelated." + "reason": "trainer.evaluate() after train() failing is unrelated to an import conflict caused by datasets when importing Trainer." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "Offline mode still making requests vs missing embed_tokens.weight on load; different problems." + "reason": "local_files_only leaking API requests is a networking/cache bug; UMT5Encoder missing embed_tokens.weight is a model loading/state-dict issue." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training blockers vs Bamba torch.compile SDPA failure; different models and symptoms." + "reason": "xLSTM training bugs for small models and Bamba torch.compile/SDPA failure are different model-specific issues." }, { "left": "issue:44451", "right": "issue:47116", "accept": false, - "reason": "ScandiBERT loading failure vs streaming pipeline regression; unrelated." + "reason": "Loading a specific model failing and a streaming pipeline regression in v5 are too broad and do not point to the same bug." }, { "left": "issue:44380", "right": "issue:46421", "accept": false, - "reason": "GPT2 attention scaling under SDPA/FlashAttention vs chunked-prefill recompilation; distinct issues." + "reason": "GPT2 attention scaling being ignored and cache lazy initialization causing recompilation are separate runtime behaviors." }, { "left": "issue:44938", "right": "issue:46178", "accept": false, - "reason": "Python 3.14 load failure vs unsupported backend mapping; not the same bug." + "reason": "Python 3.14 import failure and tensorflow_text backend registration error are different compatibility problems." }, { "left": "issue:43064", "right": "issue:47116", "accept": false, - "reason": "FSDP2/PEFT optimizer state corruption vs streaming pipeline regression; unrelated." + "reason": "FSDP2/PEFT optimizer-state corruption and streaming pipeline regression are unrelated." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Dequantized save failure vs Qwen3_5 infinite loop on a specific image; different models and code paths." - }, - { - "left": "issue:45712", - "right": "issue:46298", - "accept": false, - "reason": "Dummy class cleanup vs logging message; no shared underlying defect." - }, - { - "left": "issue:43012", - "right": "issue:43550", - "accept": false, - "reason": "Warning-only precision issue vs compile-time SDPA crash; different severity and cause." - }, - { - "left": "issue:43986", - "right": "issue:44484", - "accept": false, - "reason": "AutoProcessor crash vs shard size default discussion; unrelated." - }, - { - "left": "issue:44938", - "right": "issue:46178", - "accept": false, - "reason": "Python version compatibility vs backend-mapping validation; not the same failure." - }, - { - "left": "issue:43066", - "right": "issue:43906", - "accept": false, - "reason": "Wrong tokenizer decoder type vs isolated reproduction of another issue; no evidence they share a root cause." - }, - { - "left": "issue:43064", - "right": "issue:44315", - "accept": false, - "reason": "Trainer optimizer-state bug and Liger-kernel application bug are separate training paths." + "reason": "Saving dequantized Ministrals/Devstrals and Qwen3 infinite-loop-on-image are different model bugs with no shared fix." } ] }, @@ -17762,29 +17722,29 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:36010", "issue:36683", - "issue:38617", "issue:42175", "issue:42757", "issue:42831", "issue:42890", + "issue:42907", "issue:43012", "issue:43064", + "issue:43066", "issue:43097", "issue:43208", "issue:43352", - "issue:43502", "issue:43526", + "issue:43550", "issue:43582", - "issue:43618", "issue:43756", - "issue:43761", "issue:43824", "issue:43825", "issue:43827", + "issue:43906", "issue:43976", "issue:43992", "issue:44038", @@ -17804,6 +17764,7 @@ "issue:45478", "issue:45479", "issue:45584", + "issue:45712", "issue:46132", "issue:46172", "issue:46298", @@ -17813,18 +17774,22 @@ "issue:47276" ], "soft_pairs": [ + "issue:42907|issue:44655", + "issue:43012|issue:43550", + "issue:45712|issue:46298", + "issue:43066|issue:43906", "issue:43012|issue:44368", "issue:43825|issue:44655", "issue:43992|issue:44484", "issue:43827|issue:47116", - "issue:45584|issue:46365", "issue:43352|issue:43976", + "issue:45584|issue:46365", "issue:44038|issue:44230", "issue:44734|issue:47116", "issue:44162|issue:46172", "issue:44380|issue:46132", - "issue:43208|issue:43526", "issue:43352|issue:44380", + "issue:43208|issue:43526", "issue:43097|issue:46298", "issue:44077|issue:46365", "issue:44380|issue:45479", @@ -17837,207 +17802,217 @@ "issue:43064|issue:44380", "issue:43756|issue:44485", "issue:42831|issue:44261", - "issue:44483|issue:44514", "issue:36010|issue:42757", - "issue:43618|issue:43761", - "issue:38617|issue:43502" + "issue:44483|issue:44514" ], - "prepared_review_unit_hash": "78c73e960fc6c03e19a38813c24422faf59e81d86957f0bf3c97218ae239ef4a", + "prepared_review_unit_hash": "10ceace297638a28c73058fe3da40121a58c3685e84c2dd0c290590cb5433e7a", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35398, - "estimated_input_tokens": 8850, - "estimated_eval_tokens": 17956 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36554, + "estimated_input_tokens": 9139, + "estimated_eval_tokens": 18534 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly false-positive clustering: these issues span unrelated Transformers subsystems (imports, pipelines, model configs, training, FP8/RoPE, docs, CLI/server behavior). I would not merge any of the proposed soft edges; no PRs are present.", - "confidence": 0.86, - "canonical_issue_reason": "issue:47116 is the broadest representative of the only clear recurring subtheme here (v5 pipeline regression) and is more umbrella-like than the narrower single-model or single-error reports.", + "summary": "The cluster is very heterogeneous: the soft pairs mostly share only broad vocabulary (imports, save/load, warnings, pipeline, model families) but not the same concrete failure mode. I found no strong duplicate pair, and there are no PRs to triage here.", + "confidence": 0.19, + "canonical_issue_reason": "issue:36683 is the most actively discussed and clearly scoped runtime bug in the set; among unrelated items, it is a reasonable representative because it has the strongest standalone signal and no better duplicate anchor appears.", "canonical_pr_reason": null, - "best_issue_reason": "issue:47116 best captures a user-facing regression class with broader reuse as a reference point than the other, more specialized bug reports.", + "best_issue_reason": "issue:36683 is the best overall issue candidate because it is concrete, reproducible, and heavily discussed; it is also more representative than the many one-off environment/model-specific reports here.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both are save-related failures, but one is dequantized model saving and the other is Pipeline.save_pretrained; different code paths and likely different bugs." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "One is a PyTorch warning during bfloat16 compilation, the other is a torch.compile SDPA failure in Bamba; not the same concrete issue." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Left is a repo hygiene / dummy class export issue; right is a logging message complaint. No shared bug." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "The second is only an isolated reproduction reference; there is not enough evidence it is the same tokenizer-decoder bug." + }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Both mention warnings, but one is about bfloat16 compilation precision warnings and the other about tie_word_embeddings messaging; different bugs." + "reason": "A compilation precision warning is unrelated to a tie_word_embeddings warning in Qwen fine-tuning." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "v5 pipeline error-message regression vs inability to save Pipeline objects; different code paths and failure modes." + "reason": "Pipeline error-message behavior and Pipeline object save_pretrained support are different concerns." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder loading is unrelated to the max_shard_size question in save_pretrained()." + "reason": "Missing embed_tokens weight in UMT5 loading is not the same as save_pretrained max_shard_size behavior." }, { "left": "issue:43827", "right": "issue:47116", "accept": false, - "reason": "Docs still referencing pipeline() is a documentation cleanup issue, while 47116 is a runtime streaming pipeline regression." + "reason": "Docs still referencing pipeline() and a streaming pipeline regression are different problems." }, { - "left": "issue:45584", - "right": "issue:46365", + "left": "issue:43352", + "right": "issue:43976", "accept": false, - "reason": "Whisper empty-transcription failure after token alignment is unrelated to Trainer import failure caused by datasets interference." + "reason": "Flash Attention support for one model and Python 3.9/3.10 compatibility are unrelated." }, { - "left": "issue:43352", - "right": "issue:43976", + "left": "issue:45584", + "right": "issue:46365", "accept": false, - "reason": "Flash Attention 2 support for NemotronH is unrelated to Python 3.9/3.10 compatibility with Transformers 5.1.0." + "reason": "Whisper empty transcription failure and an import-trainer dependency issue do not match." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Both reference Qwen3-VL/MoE, but one is a transformers 5.0 bug report and the other is FP8 inference support; not the same change." + "reason": "Generic Qwen3-VL-Moe breakage in v5 and fp8 inference support are not the same concrete defect." }, { "left": "issue:44734", "right": "issue:47116", "accept": false, - "reason": "Serve /v1/responses KV-cache indexing bug is a server-side issue, not the streaming pipeline regression described in 47116." + "reason": "A /v1/responses KV-cache indexing crash is not the same as a streaming pipeline regression." }, { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "ESM2 breakage and DeepSeek-V4 tensor transpose/dtype handling are different model-specific bugs." + "reason": "ESM2 breakage and a DeepSeek tensor transpose bug are different model-specific failures." }, { "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "GPT2 attention scaling being ignored in SDPA/FlashAttention is unrelated to AttentionInterface.register changing registered function behavior." + "reason": "GPT2 attention scaling being ignored and AttentionInterface.register changing behavior are related to attention, but not the same bug." }, { - "left": "issue:43208", - "right": "issue:43526", + "left": "issue:43352", + "right": "issue:44380", "accept": false, - "reason": "xLSTM training blockers are unrelated to BeitImageProcessorFast reduce_labels returning only one label." + "reason": "FlashAttention unsupported for Nemotron and GPT2 attention scaling ignored are different code paths and symptoms." }, { - "left": "issue:43352", - "right": "issue:44380", + "left": "issue:43208", + "right": "issue:43526", "accept": false, - "reason": "Different bugs: unsupported FA2 for NemotronH versus GPT2 attention scaling being ignored in alternate backends." + "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are unrelated." }, { "left": "issue:43097", "right": "issue:46298", "accept": false, - "reason": "Removed tie_embeddings_and_encoder_decoder behavior is unrelated to a noisy logging message." + "reason": "A removed config option and a noisy logging message are not the same issue." }, { "left": "issue:44077", "right": "issue:46365", "accept": false, - "reason": "patchtsmixer post_init validation is unrelated to the Trainer import issue tied to datasets installation." + "reason": "patchtsmixer post_init validation and Trainer import breakage due to datasets are unrelated." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "GPT2 backend-specific attention scaling bug is not the same as zero-loss classification with num_labels=1." + "reason": "Attention scaling ignored under SDPA/FlashAttention is unrelated to degenerate zero-loss classification behavior." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Gemma3Config missing vocab_size and Apple Silicon caching_allocator_warmup TypeError are unrelated." + "reason": "Gemma3Config missing vocab_size and an Apple Silicon TypeError in caching allocator warmup are unrelated." }, { "left": "issue:43012", "right": "issue:47276", "accept": false, - "reason": "Both are warning-related, but one concerns bfloat16 compilation warnings and the other a misleading synced_gpus warning in continuous batching." + "reason": "A bfloat16 compile precision warning is not the same as a misleading synced_gpus warning in continuous batching." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Different import errors from different packages/models: huggingface_hub utility import vs missing Qwen2_5_VL class export." + "reason": "Both are import errors, but they concern different missing symbols from different packages and different fix paths." }, { "left": "issue:42175", "right": "issue:46365", "accept": false, - "reason": "TensorFlow backend packaging under pip install '.[torch]' is unrelated to the Trainer import problem caused by datasets." + "reason": "Backend selection under pip install extras and a Trainer import conflict with datasets are unrelated." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "FP8 save_pretrained behavior for MoE models is a different failure than Qwen3.5 MoE from_pretrained breakage in transformers>=5.4.0." + "reason": "FP8 save_pretrained failure and a Qwen3.5 MoE from_pretrained error are different directions and likely different bugs." }, { "left": "issue:42890", "right": "issue:47059", "accept": false, - "reason": "Missing set_seed in SAM-HQ tests is a test flake; the huggingface_hub 1.22.0 CLI breakage is a dependency/API incompatibility." + "reason": "A flaky integration test due to missing set_seed calls and a CLI incompatibility with huggingface_hub are unrelated." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Distributed Trainer/FSDP2 optimizer-state corruption is unrelated to GPT2 attention scaling being ignored in SDPA/FlashAttention." + "reason": "FSDP2/PEFT optimizer-state corruption and GPT2 attention scaling under SDPA are different failures." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Smollm3 RoPE-layer dropping and GLM-5 RoPE implementation are separate model-specific architecture bugs." + "reason": "Smollm3 RoPE layer dropping and GLM-5 RoPE implementation are both RoPE-related but not the same bug." }, { "left": "issue:42831", "right": "issue:44261", "accept": false, - "reason": "FineGrainedFP8 accuracy regression and MLA q_a_layernorm rms_norm_eps precision mismatch are different FP8-related issues." - }, - { - "left": "issue:44483", - "right": "issue:44514", - "accept": false, - "reason": "Chat/completions request rejection in v5.3 is unrelated to Qwen2_5_VLProcessor batched chat-template crashing with padding=False." + "reason": "FineGrainedFP8 accuracy drift and MLA q_a_layernorm epsilon handling are separate precision issues." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols and packages: GenerationMixin vs is_offline_mode." - }, - { - "left": "issue:43618", - "right": "issue:43761", - "accept": false, - "reason": "CLIPOutput attentions not assigned and CLIPVisionModel.forward returning hidden_states=None are related to CLIP but are different regression points." + "reason": "Both are import errors, but they involve different symbols and different source packages." }, { - "left": "issue:38617", - "right": "issue:43502", + "left": "issue:44483", + "right": "issue:44514", "accept": false, - "reason": "ImportError for layer_type_validation and unwanted API requests despite local_files_only=True are unrelated bug classes." + "reason": "A chat completions API acceptance failure and a batched chat-template crash are different code paths." } ] }, @@ -18059,29 +18034,30 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:29870", "issue:33453", "issue:36296", "issue:36683", - "issue:41669", + "issue:38617", "issue:42994", "issue:43010", "issue:43064", "issue:43299", "issue:43352", "issue:43475", + "issue:43502", "issue:43525", "issue:43575", "issue:43611", "issue:43618", + "issue:43761", "issue:43867", "issue:43976", "issue:43992", "issue:44075", "issue:44077", - "issue:44246", "issue:44355", "issue:44380", "issue:44393", @@ -18097,7 +18073,6 @@ "issue:45310", "issue:45375", "issue:45593", - "issue:46050", "issue:46077", "issue:46202", "issue:46365", @@ -18107,18 +18082,20 @@ "issue:47000", "issue:47059", "issue:47110", - "issue:47276", - "issue:47317" + "issue:47317", + "issue:47321" ], "soft_pairs": [ + "issue:43618|issue:43761", + "issue:38617|issue:43502", "issue:45276|issue:46077", - "issue:44380|issue:47276", "issue:43299|issue:43867", + "issue:43618|issue:44380", "issue:42994|issue:44655", + "issue:44655|issue:47321", "issue:44625|issue:46365", "issue:43611|issue:44655", "issue:44625|issue:46077", - "issue:44655|issue:46050", "issue:43352|issue:45593", "issue:36683|issue:43475", "issue:44355|issue:46877", @@ -18142,212 +18119,247 @@ "issue:44393|issue:47000", "issue:29870|issue:33453", "issue:36296|issue:43575", - "issue:43618|issue:44077", "issue:43867|issue:45375", - "issue:41669|issue:44246" + "issue:43618|issue:44077" ], - "prepared_review_unit_hash": "246e3b7e3577f3a4320b1b81168bb567165c33469636028a2acfb067446c9c04", + "prepared_review_unit_hash": "73cca387a9e84d8698db9f08233bf4023a5b11aa42e77399c0d4cacda15d0824", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 34, - "serialized_chars": 37430, - "estimated_input_tokens": 9358, - "estimated_eval_tokens": 18972 + "soft_pair_count": 35, + "serialized_chars": 37767, + "estimated_input_tokens": 9442, + "estimated_eval_tokens": 19140 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes unrelated loading/regression, config-attribute, save/load, import-time, tokenizer, training, and scheduler issues. The soft links are mostly superficial theme similarity, so none of the candidate pairs should be merged as duplicates.", + "summary": "This cluster is a loose grab-bag of Transformers v5 regression reports, mostly model-specific load/save/config issues plus a few unrelated runtime/import bugs. None of the soft pairs look like the same underlying bug or change.", "confidence": 0.94, - "canonical_issue_reason": "If one issue must anchor the cluster, issue 43611 is the broadest and most representative loading-regression report; however, the set is still too diverse for a true duplicate canonical.", + "canonical_issue_reason": "Issue 43611 is the broadest and most representative loading-regression report here: it describes a general v5 model instantiation failure, which is closer to the shared theme than the narrower model-specific attribute bugs.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43611 is the best single representative because it captures the broad model-loading regression theme that recurs most often across the cluster, even though it does not subsume the other bugs.", + "best_issue_reason": "43611 is the best single representative because it is a clear, general loading regression with a concrete failure mode, whereas the other candidates are mostly narrower, model-specific, or unrelated. ", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45276", - "right": "issue:46077", + "left": "issue:43618", + "right": "issue:43761", "accept": false, - "reason": "Both are Gemma4-related, but one is about resize_token_embeddings not updating embeddings, while the other is about bidirectional attention still building causal masks. Different code paths and fixes." + "reason": "Both are CLIP regressions, but one is missing attentions assignment and the other loses hidden_states; different outputs and different failure modes." }, { - "left": "issue:44380", - "right": "issue:47276", + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "ImportError for a missing symbol is unrelated to local_files_only still triggering network calls." + }, + { + "left": "issue:45276", + "right": "issue:46077", "accept": false, - "reason": "GPT2 attention scaling ignored under SDPA/FlashAttention is a model behavior bug; the other is a misleading warning in continuous batching. Same area, not the same bug." + "reason": "Same model family, but resize_token_embeddings not updating embeddings is distinct from bidirectional attention still building causal masks." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Qwen3VL MoE loading failure and sorted state_dict load failure are both loading problems, but they are different failure modes with different root causes." + "reason": "Qwen3VL MoE load breakage and sorted-state_dict load errors are different concrete loading problems." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "CLIP attentions missing is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Quantized model save failure is not the same as Pipeline.save_pretrained failing. Different objects and serialization paths." + "reason": "Quantized model saving failure is distinct from Pipeline save_pretrained failure." }, { - "left": "issue:44625", - "right": "issue:46365", + "left": "issue:44655", + "right": "issue:47321", + "accept": false, + "reason": "Pipeline serialization and AMD quark class updates are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:46365", "accept": false, - "reason": "num_labels propagation in Qwen3.5 is unrelated to Trainer import failing due to datasets being installed. Different subsystems." + "reason": "Qwen3.5 num_labels propagation has nothing to do with Trainer import failing because of datasets." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "base_model_prefix loading regression and Pipeline.save_pretrained failure are both save/load-related, but not the same concrete bug." + "reason": "Model loading via base_model_prefix is a different code path from saving Pipeline objects." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "Qwen3.5 num_labels propagation and Gemma4 bidirectional-attention masking are unrelated model bugs." - }, - { - "left": "issue:44655", - "right": "issue:46050", - "accept": false, - "reason": "Pipeline serialization failure and sinq quantized save/load break affect different APIs and likely different serialization logic." + "reason": "Different models and different bugs: num_labels propagation vs attention mode handling." }, { "left": "issue:43352", "right": "issue:45593", "accept": false, - "reason": "Flash Attention 2 unsupported for NemotronH is an unsupported-feature error; D-FINE auxiliary-loss handling is a training-loss logic bug." + "reason": "Flash Attention support warning for Nemotron is unrelated to D-FINE auxiliary losses when denoising is off." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Gemma3Config missing vocab_size and SAM3 vision output missing fpn_position_embeddings are different missing-attribute issues in different models." + "reason": "Missing vocab_size in Gemma3Config and missing fpn_position_embeddings in SAM3 output are different model-specific attribute regressions." }, { "left": "issue:44355", "right": "issue:46877", "accept": false, - "reason": "Compiled Python file import errors and auto_docstring stdout diagnostics are both import-time annoyances, but they are unrelated problems." + "reason": "Compiled-file import errors are not the same as auto_docstring printing doc-lint diagnostics at import time." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Missing vocab_size in Gemma3Config and patchtsmixer post_init restrictions are unrelated configuration/model-definition issues." + "reason": "Gemma3 config missing vocab_size and patchtsmixer post_init policy are unrelated config issues." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Python 3.9+/3.10+ compatibility regression and Python 3.14 load failure are version-compatibility issues, but not the same bug." + "reason": "Both are Python-version compatibility reports, but they target different versions and different breakages." }, { "left": "issue:46077", "right": "issue:46512", "accept": false, - "reason": "Gemma4 attention-mask construction and stale model.config.dtype after from_config are separate state/config propagation bugs." + "reason": "Attention-mask semantics in Gemma4 are unrelated to from_config(dtype=...) leaving config.dtype stale." }, { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "A state_dict sorting load error and an incorrect Qwen3.5 saved checkpoint are both save/load issues, but they have different symptoms and likely different causes." + "reason": "Sorting state_dict causing load errors is distinct from Qwen3.5 save_pretrained producing the wrong checkpoint." }, { "left": "issue:44938", "right": "issue:47059", "accept": false, - "reason": "Python 3.14 import failure and CLI completion breakage with huggingface_hub are unrelated compatibility regressions." + "reason": "Python 3.14 loading failure is unrelated to the CLI completion attribute error from huggingface_hub 1.22.0." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIP attentions no longer being assigned and Qwen3.5 checkpoint save corruption are different model/output vs serialization bugs." + "reason": "CLIP attentions not assigned and Qwen3.5 checkpoint saving are different subsystems and symptoms." }, { "left": "issue:36683", "right": "issue:47110", "accept": false, - "reason": "Gemma3Config missing vocab_size and special_tokens list lacking keys are unrelated attribute/type bugs." + "reason": "Gemma3 missing vocab_size and special_tokens being a list instead of a mapping are unrelated attribute-shape issues." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Gemma3Config missing vocab_size and Llama4Config missing pad_token_id are different config-attribute omissions in different model families." + "reason": "Missing vocab_size in Gemma3Config is not the same bug as Llama4Config missing pad_token_id." }, { "left": "issue:29870", "right": "issue:36296", "accept": false, - "reason": "Bart generate() UnboundLocalError and tensor parallel training bug are unrelated execution paths and failure modes." + "reason": "Bart generate UnboundLocalError and tensor-parallel training are different code paths and failure modes." }, { "left": "issue:44908", "right": "issue:46851", "accept": false, - "reason": "Scheduler kwargs being ignored and Qwen3.5 variable-length position_ids bugs are unrelated." + "reason": "Scheduler kwargs being ignored is unrelated to Qwen3.5 variable-length position_ids handling." }, { "left": "issue:43992", "right": "issue:47317", "accept": false, - "reason": "UMT5Encoder missing embed_tokens.weight and PPFormulaNet training-loss bug are unrelated model-specific issues." + "reason": "Missing embed_tokens.weight in UMT5Encoder is unrelated to PPFormulaNet training-loss behavior." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Compiled-file import errors and the Python 3.13 @torch.jit.script parsing/IndentationError issue stem from different parser/import edge cases." + "reason": "Both involve import-time failures, but compiled Python file errors and @torch.jit.script parsing with comments are different root causes." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "Cache/Layer update methods needing no_grad and broad import errors after upgrade are not the same bug." + "reason": "Cache update no_grad decoration and broad import errors after upgrade are not the same bug." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "FSDP2+PEFT optimizer-state corruption and a typo in cache strategies are unrelated." + "reason": "FSDP2/PEFT optimizer-state corruption is unrelated to a typo in cache strategy names." }, { "left": "issue:44075", "right": "issue:46365", "accept": false, - "reason": "SGD args ignored and inability to import Trainer without datasets are unrelated issues." + "reason": "SGD args being ignored is unrelated to Trainer import failure caused by datasets." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "UMT5Encoder missing embed_tokens.weight and Qwen3.5 Moe from_pretrained failure are distinct loading bugs in different model families." + "reason": "UMT5 missing embed_tokens.weight and Qwen3.5 MoE from_pretrained errors are different model-specific loading failures." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "Sorted-state_dict load errors and Gemma pipeline roundtrip failures are different serialization problems." + }, + { + "left": "issue:44393", + "right": "issue:47000", + "accept": false, + "reason": "Qwen3-VL 2D box hallucination/output errors are unrelated to RTDetrLoss cost-matrix infeasibility." + }, + { + "left": "issue:29870", + "right": "issue:33453", + "accept": false, + "reason": "Bart generation UnboundLocalError and tokenizer-loading regression are different bugs." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug and Qwen2-57B-A14B OOM on load are distinct issues." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "Sorted state_dict load error and missing deepstack_visual_indexes in Qwen3.5 vision config are different config/load issues." + "reason": "State_dict ordering load error and missing Qwen3.5 vision config field are not the same root cause." }, { - "left": "issue:41669", - "right": "issue:44246", + "left": "issue:43618", + "right": "issue:44077", "accept": false, - "reason": "Removing import * from models for import-time performance and import transformers being slow sometimes are related only at a very broad level, not the same change or bug." + "reason": "CLIP attentions regression and patchtsmixer post_init policy change are unrelated." } ] }, @@ -18369,20 +18381,20 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28282", "issue:30333", "issue:30990", "issue:31795", "issue:33453", + "issue:41669", "issue:42548", "issue:42890", "issue:42898", "issue:43010", "issue:43065", "issue:43232", - "issue:43299", "issue:43502", "issue:43525", "issue:43576", @@ -18395,6 +18407,7 @@ "issue:44062", "issue:44112", "issue:44222", + "issue:44246", "issue:44351", "issue:44355", "issue:44360", @@ -18416,11 +18429,11 @@ "issue:46285", "issue:46617", "issue:46664", - "issue:46726", "issue:47110", "issue:47317" ], "soft_pairs": [ + "issue:41669|issue:44246", "issue:42890|issue:45561", "issue:43010|issue:43976", "issue:43867|issue:45357", @@ -18450,243 +18463,228 @@ "issue:30333|issue:33453", "issue:44355|issue:44749", "issue:30990|issue:31795", - "issue:43010|issue:43232", - "issue:46664|issue:46726", - "issue:43502|issue:44222", - "issue:43299|issue:44556" + "issue:43010|issue:43232" ], - "prepared_review_unit_hash": "2f183c910098708e9edb18ca532ae55935c4cb1767723cdcc77e8e47d240ab7b", + "prepared_review_unit_hash": "1ce4a46ce9616d0e53a8a56028da4969eb9d9a6ba4be42fb3146b3b9cb7c6b6d", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 37251, - "estimated_input_tokens": 9313, - "estimated_eval_tokens": 18882 + "soft_pair_count": 31, + "serialized_chars": 36590, + "estimated_input_tokens": 9148, + "estimated_eval_tokens": 18552 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated single issues. The only plausible duplicate pair is the Qwen3.5 MoE loading regression: one report names the missing `_tp_plan`, the other reports a `from_pretrained` failure on the same model/version path.", - "confidence": 0.68, - "canonical_issue_reason": "issue:45125 is the most specific root-cause report; it directly identifies the missing `_tp_plan` needed for Qwen3_5MoeForConditionalGeneration tensor-parallel loading.", + "summary": "This cluster is mostly heterogeneous transformer issues. The only clear duplicate-style match is the import-time slowdown pair; the rest share broad subsystems but not the same concrete bug.", + "confidence": 0.77, + "canonical_issue_reason": "issue:41669 is the most concrete central bug in the cluster: it identifies a specific import-time performance regression and matches the closest soft duplicate (import takes long).", "canonical_pr_reason": null, - "best_issue_reason": "issue:45478 is the best representative symptom report because it clearly states the user-facing `from_pretrained` failure on transformers>=5.4.0 for Qwen3.5 Moe.", + "best_issue_reason": "issue:41669 is the best representative issue because it is specific, actionable, and anchors the only strong same-bug pair in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:41669", + "right": "issue:44246", + "accept": true, + "reason": "Both point to the same import-performance regression: transformers import is slow, with 41669 naming import * usage as the cause." + }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Both are test-related, but one is about missing `set_seed()` in SamHQ integration tests and the other is a pytest-xdist race on `captured_info.txt`; different bugs." + "reason": "Both are test/CI-related, but one is a missing set_seed flake and the other is an xdist file race; different bugs." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "Unrelated: cache/layer `update()` no_grad decoration vs Python version compatibility failure in Transformers 5.1.0." + "reason": "One is a no_grad decoration proposal for cache updates; the other is a Python version compatibility failure." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Both involve model loading/saving, but the concrete failures differ: sorted state_dict load error vs incorrect visual encoder keys on save_pretrained." + "reason": "Both involve model loading/saving, but the concrete failures differ: sorted state_dict loading vs incorrect visual encoder keys." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "One is a specific no_grad behavior request; the other is a broad complaint that 5.0 breaks many models." + "reason": "One is a specific generation/cache mutation issue; the other is a vague broad complaint about many models breaking." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "ImportError for missing PyTorch vs tokenizer loading regression; different code paths and root causes." + "reason": "Both are loading/import regressions, but one is a missing PyTorch dependency import error and the other is tokenizer loading regression." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 save_pretrained/moe serialization bug is distinct from Qwen3_5Moe tensor-parallel loading metadata." + "reason": "Both mention Qwen/MoE, but one is FP8 save_pretrained behavior and the other is missing tensor-parallel metadata." }, { "left": "issue:43618", "right": "issue:46285", "accept": false, - "reason": "Both mention CLIP, but one is about missing attentions assignment and the other about inconsistent module structure after v5.6; not the same bug." + "reason": "Both are CLIP-related structural issues, but they concern different classes and different API surface changes." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer ReLU discussion is unrelated to DeepseekV3MoE implementation divergence." + "reason": "Different code paths and symptoms: a DSA indexer ReLU discussion versus a DeepSeekV3MoE implementation divergence." }, { "left": "issue:42548", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different exported symbols (`PreTrainedModel` vs `HybridCache`) and likely different breakages." + "reason": "Both are import errors, but for different exported symbols and likely different broken exports." }, { "left": "issue:45125", "right": "issue:45478", - "accept": true, - "reason": "Same underlying Qwen3.5 MoE loading regression: one reports the missing `_tp_plan`, the other reports the resulting `from_pretrained` failure on the same model path." + "accept": false, + "reason": "Both are Qwen3.5-MoE related, but the reported failures are different concrete issues with different causes." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Both are Qwen3.5-related, but one is tensor-parallel loading metadata and the other is a config field being dropped by `@strict`." + "reason": "Related model family, but one is missing _tp_plan and the other is a config field being dropped; not the same bug." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "`local_files_only=True` still making requests is unrelated to the AddedToken duplicate-argument TypeError." + "reason": "Different failure modes: unwanted network access versus AddedToken argument conflict." }, { "left": "issue:43525", "right": "issue:47317", "accept": false, - "reason": "Llama4Config missing `pad_token_id` has no overlap with the PPFormulaNet training-loss issue." + "reason": "Unrelated domains: Llama4 config attribute error versus a FormulaNet training-loss bug." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Seed instability in tests is different from a stale device override CI failure in GraniteSpeech." + "reason": "Both are CI/test issues, but one is seed-related nondeterminism and the other is a stale device override regression." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Test seeding flakiness is unrelated to Phi-4-multimodal-instruct loading failure." + "reason": "Different failures: flaky integration tests versus inability to load a specific model." }, { "left": "issue:42898", "right": "issue:46664", "accept": false, - "reason": "Tokenizer-space cleanup behavior change and SmolVLM frame-count semantics are unrelated." + "reason": "Tokenization-space behavior change is unrelated to SmolVLM video frame semantics." }, { "left": "issue:44734", "right": "issue:46617", "accept": false, - "reason": "KV cache continuation crash in `transformers serve` is different from trust_remote_code cache FileNotFoundError." + "reason": "KV-cache continuation crash and trust_remote_code cache FileNotFoundError are different code paths." }, { "left": "issue:45071", "right": "issue:47317", "accept": false, - "reason": "PretrainedConfig type-checking breakage is unrelated to PPFormulaNet training-loss behavior." + "reason": "Type-checking regression versus a model-specific training-loss bug; no same underlying change." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Tokenizer loading issue and checkpoint reload incompatibility after upgrading versions are distinct problems." + "reason": "Both involve loading after v5 upgrades, but one is tokenizer loading and the other is checkpoint reload compatibility; too broad to merge." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "Int4 quantization memory regression is unrelated to checkpoint reload failure across versions." + "reason": "Memory/OOM regression under int4 quantization is unrelated to checkpoint reload failure." }, { "left": "issue:43010", "right": "issue:43065", "accept": false, - "reason": "No_grad on cache updates and dummy Conv2d in Sam3PixelDecoder are different implementation concerns." + "reason": "No shared bug: cache update no_grad decoration versus a dummy Conv2d in Sam3PixelDecoder." }, { "left": "issue:43824", "right": "issue:46617", "accept": false, - "reason": "Missing Qwen2_5_VL export is a different issue from cache-based trust_remote_code FileNotFoundError." + "reason": "Missing import for a specific Qwen class is not the same as cache-based trust_remote_code FileNotFoundError." }, { "left": "issue:43576", "right": "issue:46132", "accept": false, - "reason": "Broken `transformers env` CLI and AttentionInterface.register behavior change are unrelated." + "reason": "A broken CLI env command and AttentionInterface.register behavior change are different issues." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Offline download violation is unrelated to GLM-5 RoPE implementation discussion." + "reason": "local_files_only network leakage is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:43874", "right": "issue:46617", "accept": false, - "reason": "Missing image-patch helper in GLM46V is unrelated to trust_remote_code cache file lookup." + "reason": "Missing image-processor method and cache FileNotFoundError are unrelated." }, { "left": "issue:45850", "right": "issue:47110", "accept": false, - "reason": "Broad complaint that 5.0 breaks models is not the same as a `special_tokens` list-vs-dict error." + "reason": "A broad compatibility complaint is not the same as a specific special_tokens attribute error." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLflow run-state reporting on failed training jobs is unrelated to tokenizer loading regression." + "reason": "MLflow reporting on failed training is unrelated to tokenizer loading regression." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled Python file errors and Chinese report about slower filtering after upgrade do not share a concrete bug." + "reason": "Compiled-Python-file errors and upgrade-induced slowdown during filtering are different problems." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Sentence Transformers loading hang is different from documentation confusion about `input_ids` and `past_key_values`." + "reason": "Sentence Transformers loading hang and forward-argument documentation confusion are not the same bug." }, { "left": "issue:43010", "right": "issue:43232", "accept": false, - "reason": "Different generation-path bugs: no_grad decoration vs `_update_model_kwargs_for_generation` after `sync_gpus`." - }, - { - "left": "issue:46664", - "right": "issue:46726", - "accept": false, - "reason": "Video processor `num_frames` semantics and missing vision backend handling are different issues." - }, - { - "left": "issue:43502", - "right": "issue:44222", - "accept": false, - "reason": "Offline request leakage and FP8 moe save_pretrained are different code paths." - }, - { - "left": "issue:43299", - "right": "issue:44556", - "accept": false, - "reason": "Qwen3VL MoE loading breakage and checkpoint reload after version upgrade are different symptoms/root causes." + "reason": "Both involve generation internals, but one is cache update no_grad and the other is post-sync_gpus kwarg handling." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the duplicate call on issue:45125 vs issue:45478 is conservative and well-supported by the titles. The other soft-edge verdicts also stay appropriately cautious and do not overclaim evidence." + "feedback": "The analyst summary is grounded in the packet and stays conservative on the soft edges. The accepted 41669/44246 pair is a plausible same-bug match on import-time slowdown, and the rejected pairs are mostly supported by clear differences in failure mode or scope. No overstatement stands out." }, "evaluator_used": true, "retried": false, @@ -18705,7 +18703,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:29870", "issue:30333", @@ -18717,6 +18715,7 @@ "issue:41628", "issue:42548", "issue:42757", + "issue:43299", "issue:43352", "issue:43475", "issue:43502", @@ -18730,22 +18729,19 @@ "issue:44038", "issue:44222", "issue:44230", - "issue:44273", "issue:44315", "issue:44351", "issue:44355", "issue:44360", "issue:44485", "issue:44492", - "issue:44623", + "issue:44556", "issue:44961", "issue:45071", "issue:45125", - "issue:45230", "issue:45362", "issue:45375", "issue:45468", - "issue:45542", "issue:45850", "issue:46095", "issue:46143", @@ -18753,10 +18749,14 @@ "issue:46209", "issue:46617", "issue:46664", + "issue:46726", "issue:47059", "issue:47110" ], "soft_pairs": [ + "issue:46664|issue:46726", + "issue:43502|issue:44222", + "issue:43299|issue:44556", "issue:30333|issue:36296", "issue:29870|issue:31795", "issue:43827|issue:44961", @@ -18778,238 +18778,235 @@ "issue:44351|issue:46617", "issue:44230|issue:46095", "issue:45375|issue:47110", - "issue:45362|issue:45850", - "issue:44315|issue:46209", "issue:44485|issue:44961", + "issue:44315|issue:46209", + "issue:45362|issue:45850", "issue:44961|issue:45468", "issue:43525|issue:47110", "issue:43673|issue:44492", "issue:43352|issue:44961", - "issue:43867|issue:47110", - "issue:44273|issue:45542", - "issue:44485|issue:46664", - "issue:44623|issue:45230" + "issue:43867|issue:47110" ], - "prepared_review_unit_hash": "34d6d3f0f29162526c5b35b7f28e573436a4c7584e7bd1f45079f0821143cbfa", + "prepared_review_unit_hash": "2274bc1c8c7853ab3d2acef1199cd8ead0f20759ac640d45f537cf806c4ce248", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36515, - "estimated_input_tokens": 9129, - "estimated_eval_tokens": 18514 + "serialized_chars": 36090, + "estimated_input_tokens": 9023, + "estimated_eval_tokens": 18302 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: most pairs only share broad themes like v5 regressions, import failures, or model-loading problems, but they describe different concrete bugs. I found no true duplicate issue pairs; all soft edges should be rejected.", - "confidence": 0.84, - "canonical_issue_reason": "issue:43576 is the clearest, most concrete user-facing regression in the set and the best representative of the broader v5 breakage complaints, though it is not a duplicate of the others.", + "summary": "This cluster is mostly noisy and only loosely related by Transformers/v5 breakage themes. The strongest representative is the broad v5 regression report, but the soft pairs are not true duplicates and should stay separate.", + "confidence": 0.36, + "canonical_issue_reason": "issue:45850 is the broadest representative of the cluster\u2019s main theme: widespread v5 compatibility regressions affecting many models.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43576 has the strongest problem statement and the most directly actionable scope, so it is the best single issue to keep as the cluster representative.", + "best_issue_reason": "issue:45850 best captures the cluster\u2019s shared user pain around v5 breaking multiple model-loading paths, even though the cluster is not a tight duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:46664", + "right": "issue:46726", + "accept": false, + "reason": "Both involve processors, but one is video-frame semantics and the other is missing-vision-backend handling; different bugs." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "local_files_only network requests and FP8 save_pretrained are unrelated code paths." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both are v5 loading regressions, but they affect different models and likely different root causes." + }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLflow reporting on failed training jobs is unrelated to tensor-parallel training failures." + "reason": "MLFlow reporting on failed jobs is unrelated to tensor-parallel training behavior." }, { "left": "issue:29870", "right": "issue:31795", "accept": false, - "reason": "A generation-time UnboundLocalError in Bart is not the same as documentation confusion about forward arguments." + "reason": "A runtime generate() error is not the same as documentation confusion." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs still referencing pipeline() and an unrelated racoon issue are different problems." + "reason": "Docs referencing pipeline() and a one-word issue title are not the same underlying bug." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Both are loading-related, but one is a tokenizer regression and the other is a specific Llama 3.2 vision model-loading breakage." + "reason": "Different loading regressions affecting different models and likely different mechanisms." }, { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "Missing AutoImageProcessor and missing PreTrainedModel are separate import/export regressions." + "reason": "Import failures for different symbols do not indicate one shared duplicate." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "FP8 save_pretrained behavior and FP8 inference support for Qwen3-VL are different concrete issues." + "reason": "One is FP8 save_pretrained behavior; the other is FP8 inference support." }, { "left": "issue:46143", "right": "issue:46144", "accept": false, - "reason": "Passing through kwargs in RoFormer methods is not the same as using the attention interface." + "reason": "kwargs propagation and attention-interface compliance are related but distinct implementation bugs." }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "local_files_only still making API requests is unrelated to the racoon issue." + "reason": "local_files_only network leakage has nothing to do with the unrelated issue title." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled Python file errors and PretrainedConfig type-checking regressions do not describe the same bug." + "reason": "Compiled-file execution errors and PretrainedConfig type checking are separate concerns." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "A broken transformers env command and a config type-checking regression are different failure modes." + "reason": "CLI env command breakage is unrelated to config type-checking regressions." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "The v5 env command issue is unrelated to the Qwen2.5-VL import error." + "reason": "A broken env CLI command is not the same as an import error for a model class." }, { "left": "issue:43576", "right": "issue:47059", "accept": false, - "reason": "CLI breakage from the env command and a hub CLI completion attribute error are distinct issues." + "reason": "Both are CLI-related, but one is a transformers env command issue and the other is a huggingface_hub completion API mismatch." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Missing is_offline_mode in huggingface_hub is a dependency import problem, not the same as the transformers env command bug." + "reason": "Missing is_offline_mode import and env CLI breakage are unrelated failures." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "A general transformers 5.0/Qwen3-VL-Moe bug is not the same as a missing _tp_plan for tensor parallelism." + "reason": "General Qwen3-VL-Moe breakage is not the same as missing _tp_plan for tensor parallelism." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Documentation confusion about model.forward is unrelated to the layer_type_validation import error." + "reason": "Documentation confusion and an ImportError for layer_type_validation are different classes of issues." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor-parallel training failure and pipeline docs mentioning return_all_scores are unrelated." + "reason": "Tensor-parallel training bugs do not duplicate text-classification docs behavior." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger Kernel not applying with model_init and a missing ReLU in the DSA indexer are different bugs." + "reason": "Liger Kernel application and a DSA indexer ReLU discussion are unrelated." }, { "left": "issue:43475", "right": "issue:46617", "accept": false, - "reason": "SAM 3 Video attribute access and trust_remote_code cache loading are unrelated code paths." + "reason": "SAM 3 Video attribute missing and trust_remote_code cache FileNotFound are different failure modes." }, { "left": "issue:44351", "right": "issue:46617", "accept": false, - "reason": "HybridCache import failure and cached trust_remote_code FileNotFoundError are different problems." + "reason": "Missing HybridCache import is not the same as cache-file lookup failure." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "FP8 inference support for Qwen3-VL/FMoE is not the same as asking for a weight conversion script." + "reason": "FP8 inference support and a weight-conversion-script question are not duplicates." }, { "left": "issue:45375", "right": "issue:47110", "accept": false, - "reason": "Missing deepstack_visual_indexes in a config is unrelated to special_tokens being a list instead of a dict." + "reason": "A dropped config field and a special_tokens list error are unrelated." }, { - "left": "issue:45362", - "right": "issue:45850", + "left": "issue:44485", + "right": "issue:44961", "accept": false, - "reason": "A Qwen3.5-35B chat crash is not the same as the broad complaint that v5 breaks too many models." + "reason": "GLM-5 RoPE implementation and a title-only issue are not the same bug." }, { "left": "issue:44315", "right": "issue:46209", "accept": false, - "reason": "Liger Kernel application with model_init and FP8 inference on B200 are unrelated issues." + "reason": "model_init Liger Kernel application and FP8 inference on B200 are different problems." }, { - "left": "issue:44485", - "right": "issue:44961", + "left": "issue:45362", + "right": "issue:45850", "accept": false, - "reason": "GLM-5 RoPE implementation and an unrelated racoon issue do not match." + "reason": "A specific Qwen3.5-35B chat crash is not the same as a broad v5 breakage meta-issue." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "Racoon and Gemma-4 audio positional encoding are different model-specific bugs." + "reason": "The placeholder title does not match the Gemma-4 audio positional encoding bug." }, { "left": "issue:43525", "right": "issue:47110", "accept": false, - "reason": "Llama4Config missing pad_token_id is unrelated to special_tokens being a list." + "reason": "Llama4Config missing pad_token_id and a special_tokens keys error are unrelated." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "GenerationMixin cache missing during chunked_prefill and a typo in cache strategies are not the same underlying bug." + "reason": "Missing cache in chunked_prefill and a typo in cache strategies are not the same defect." }, { "left": "issue:43352", "right": "issue:44961", "accept": false, - "reason": "Nemotron flash-attention support and racoon are unrelated problems." + "reason": "Nemotron Flash Attention support and an unrelated title are not duplicates." }, { "left": "issue:43867", "right": "issue:47110", "accept": false, - "reason": "state_dict sorting load errors are unrelated to special_tokens structure errors." - }, - { - "left": "issue:44273", - "right": "issue:45542", - "accept": false, - "reason": "Lazy loading behavior and the TensorBoard/TensorFlow backend error are different issues." - }, - { - "left": "issue:44485", - "right": "issue:46664", - "accept": false, - "reason": "GLM-5 RoPE implementation and SmolVLM num_frames semantics are unrelated." - }, - { - "left": "issue:44623", - "right": "issue:45230", - "accept": false, - "reason": "Missing files from processor.save_pretrained is not the same as an uninformative bug report." + "reason": "State_dict sorting load errors and special_tokens handling are different bugs." } ] }, @@ -19031,7 +19028,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:31795", "issue:33453", @@ -19053,12 +19050,12 @@ "issue:44297", "issue:44315", "issue:44351", - "issue:44360", "issue:44393", "issue:44466", "issue:44485", "issue:44492", "issue:44556", + "issue:44623", "issue:44655", "issue:44704", "issue:44829", @@ -19070,7 +19067,7 @@ "issue:45230", "issue:45341", "issue:45381", - "issue:45468", + "issue:45542", "issue:45753", "issue:45850", "issue:46018", @@ -19083,6 +19080,9 @@ "issue:47176" ], "soft_pairs": [ + "issue:44273|issue:45542", + "issue:44485|issue:46664", + "issue:44623|issue:45230", "issue:44273|issue:45003", "issue:45381|issue:46664", "issue:41669|issue:43673", @@ -19095,8 +19095,8 @@ "issue:44556|issue:45092", "issue:42371|issue:44492", "issue:45850|issue:47059", - "issue:44246|issue:44273", "issue:44492|issue:45850", + "issue:44246|issue:44273", "issue:43408|issue:44315", "issue:41669|issue:46218", "issue:44492|issue:46328", @@ -19107,8 +19107,8 @@ "issue:44297|issue:45230", "issue:44704|issue:44908", "issue:43576|issue:44908", - "issue:44908|issue:45071", "issue:43576|issue:47176", + "issue:44908|issue:45071", "issue:44655|issue:44861", "issue:41669|issue:45341", "issue:41669|issue:44351", @@ -19118,259 +19118,264 @@ "issue:46018|issue:46172", "issue:43519|issue:46655", "issue:44315|issue:44829", - "issue:44485|issue:47176", - "issue:44360|issue:45468" + "issue:44485|issue:47176" ], - "prepared_review_unit_hash": "3be7e86154e986a69062a03fe21d5ea633819f2e7fca1bf1e108cc388345fee2", + "prepared_review_unit_hash": "daa0e7990b44d21c59096dad7e3e6467658c3b88e3fdb3f955e1c7c75ef98cec", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 37, - "serialized_chars": 38227, - "estimated_input_tokens": 9557, - "estimated_eval_tokens": 19370 + "soft_pair_count": 39, + "serialized_chars": 38873, + "estimated_input_tokens": 9719, + "estimated_eval_tokens": 19694 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly false-positive soft matches: the issues span unrelated bugs, performance regressions, model-specific compatibility problems, and CLI/config changes, with no pair sharing the same concrete failure mode.", - "confidence": 0.97, - "canonical_issue_reason": "issue:45850 is the broadest umbrella here, but the cluster is too heterogeneous for a strong duplicate anchor; it is only a loose summary of v5 regressions.", + "summary": "This set is not a true duplicate cluster: the issues span import/lazy-loading, tokenizer/processor save-load regressions, model-specific config errors, performance problems, and a security bypass. No soft pair looks like the same concrete bug/change closely enough to merge.", + "confidence": 0.67, + "canonical_issue_reason": "issue:44273 is the best representative if one issue must anchor the set: it is open, actively discussed, and the broadest import/lazy-loading regression among the candidates.", "canonical_pr_reason": null, - "best_issue_reason": "If forced to pick one representative, issue:45850 best captures the broad \u2018v5 breaks many models\u2019 theme, but it is too vague to be a precise dedupe target.", + "best_issue_reason": "issue:44273 has the strongest cluster-central signal and the most active discussion, so it is the most suitable issue to keep as the representative tracker.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Different problems: lazy loading vs a TensorFlow backend/tensorboard install error." + }, + { + "left": "issue:44485", + "right": "issue:46664", + "accept": false, + "reason": "Unrelated model bugs: GLM-5 RoPE implementation vs SmolVLM video frame semantics." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Too vague on the right side; no evidence both describe the same save_pretrained processor bug." + }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Both touch import/lazy-loading code, but one reports lazy loading failure and the other unsafe sys.modules access; different concrete bugs." + "reason": "Related import area, but one is lazy loading failure and the other is unsafe sys.modules access; not clearly the same bug." }, { "left": "issue:45381", "right": "issue:46664", "accept": false, - "reason": "Different models and different processor bugs: wrong vision_position_ids vs incorrect num_frames meaning/padding." + "reason": "Both are video-processing issues, but they concern different models and different calculations." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Import-time slowdown from wildcard imports is not the same as a missing GenerationMixin cache during chunked prefill." + "reason": "Import-performance issue vs generation cache missing during chunked prefill; different code paths and symptoms." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "A specific missing pad_token_id attribute is not the same as a vague \u2018v5 breaks many models\u2019 complaint." + "reason": "Specific missing config attribute vs a broad complaint that many models are broken after v5." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "Same model family, but different symptoms: 2D bbox output error versus an infinite loop on a specific image." + "reason": "Different Qwen3-VL failures: bounding-box output vs an infinite loop on one image." }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "Qwen3-VL output issues are unrelated to a PretrainedConfig type-checking regression." + "reason": "Model-specific vision bug vs a generic PretrainedConfig type-checking regression." }, { "left": "issue:44466", "right": "issue:44556", "accept": false, - "reason": "One is inconsistent lm_head.weight serialization; the other is checkpoint reload incompatibility after upgrade." + "reason": "Serialization inconsistency of tied weights vs inability to reload old checkpoints; related theme, but not the same concrete bug." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "Specific StableLmConfig pad token field regression is not the same as a broad version-wide breakage report." + "reason": "Specific StableLmConfig pad token regression vs a broad v5 compatibility complaint." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Optional post_init validation in patchtsmixer is unrelated to _get_tied_weight_keys crashing on a list." + "reason": "Different failures: optional post_init policy vs AttributeError in tied-weight-key handling." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both involve v5 loading problems, but one is generic checkpoint reload failure and the other is remote-code/meta-init incompatibility." + "reason": "Both involve v5 checkpoint compatibility, but one is general reload failure and the other is an InternVL2 meta-init incompatibility." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 behavior/API-settings guidance and a cache-strategy typo are unrelated." + "reason": "TF32 settings issue vs a typo in cache strategy text; unrelated." }, { "left": "issue:45850", "right": "issue:47059", "accept": false, - "reason": "A broad v5 model regression is unrelated to a CLI break caused by huggingface_hub 1.22.0." + "reason": "Broad v5 model breakage vs a CLI incompatibility with huggingface_hub 1.22.0." }, { - "left": "issue:44246", - "right": "issue:44273", + "left": "issue:44492", + "right": "issue:45850", "accept": false, - "reason": "Import slowness and lazy-loading malfunction are related only at a high level; they describe different problems." + "reason": "A cache-strategy typo does not match a broad model-breakage regression." }, { - "left": "issue:44492", - "right": "issue:45850", + "left": "issue:44246", + "right": "issue:44273", "accept": false, - "reason": "A typo in cache strategy docs/code is not the same as a general \u2018v5 breaks too many models\u2019 report." + "reason": "Import slowdown sometimes vs lazy loading malfunction; adjacent theme, but not same bug." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "Model-type mismatch warning and Liger Kernel not applied with model_init are different code paths." + "reason": "Model type warning vs Liger Kernel not applied; unrelated subsystems." }, { "left": "issue:41669", "right": "issue:46218", "accept": false, - "reason": "Import * performance regression is not the same as failing to import a specific class export." + "reason": "Import-star cleanup/performance issue vs missing exported class; not the same underlying defect." }, { "left": "issue:44492", "right": "issue:46328", "accept": false, - "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad inefficiency are unrelated." + "reason": "Typo in cache strategy naming vs inefficient padding conversion; unrelated." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "A flaky integration test due to missing set_seed is unrelated to scheduler kwargs being ignored." + "reason": "Test flakiness from missing set_seed calls vs scheduler kwargs being ignored." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python-version compatibility is not the same as a Qwen3-VL bounding-box output bug." + "reason": "Python version support issue vs a Qwen3-VL hallucination/bbox bug." }, { "left": "issue:44295", "right": "issue:45230", "accept": false, - "reason": "A buffer/position_ids access error is specific, while the other entry is an unspecific bug report." + "reason": "Concrete buffer/position_ids error vs an unspecified bug report." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash Attention 2 support warning and Qwen3-VL bbox errors are different model/runtime issues." + "reason": "Flash Attention support limitation vs a Qwen3-VL output bug." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer save metadata mismatch is unrelated to the generic bug report." + "reason": "Tokenizer save_pretrained mismatch vs an unspecified bug report." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "Both mention kwargs, but one is AutoProcessor forwarding and the other is scheduler configuration; different APIs and failures." + "reason": "Processor kwargs forwarding vs scheduler kwargs handling; different APIs." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "The CLI env command breakage and inverse_sqrt scheduler kwargs bug are unrelated." + "reason": "Broken env CLI command vs inverse_sqrt scheduler kwargs handling." }, { - "left": "issue:44908", - "right": "issue:45071", + "left": "issue:43576", + "right": "issue:47176", "accept": false, - "reason": "Scheduler kwargs handling is unrelated to PretrainedConfig type checking." + "reason": "CLI command breakage vs a shell-operator allowlist bypass; different security and functionality concerns." }, { - "left": "issue:43576", - "right": "issue:47176", + "left": "issue:44908", + "right": "issue:45071", "accept": false, - "reason": "A broken env CLI command and a shell-operator allowlist bypass are different security/CLI issues." + "reason": "Scheduler configuration bug vs PretrainedConfig type-checking regression." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Pipeline save_pretrained failure and tied-weight-key attribute crash are separate serialization/model-state bugs." + "reason": "Pipeline save_pretrained problem vs tied-weight-key AttributeError; unrelated save/load paths." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Import-time wildcard cleanup and a testing_utils bug do not share the same failure mode." + "reason": "Large import slowdown issue vs a small testing_utils bug." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import performance regression is unrelated to missing HybridCache export/import failure." + "reason": "Import-star/lazy-load performance regression vs missing HybridCache export; not the same bug." }, { "left": "issue:43976", "right": "issue:46218", "accept": false, - "reason": "Python 3.9 compatibility and inability to import Qwen3_5ForCausalLM are different issues." - }, - { - "left": "issue:31795", - "right": "issue:33453", - "accept": false, - "reason": "Documentation confusion about forward args is unrelated to tokenizer loading regression." + "reason": "Python compatibility issue vs missing Qwen3_5 import; unrelated." }, { "left": "issue:43976", "right": "issue:47176", "accept": false, - "reason": "Python-version support and shell command-substitution allowlist bypass are unrelated." + "reason": "Python version support vs shell command allowlist bypass; different domains." }, { "left": "issue:46018", "right": "issue:46172", "accept": false, - "reason": "Both are DeepSeek-V4 related, but one is a gating logic complaint and the other is a dtype/transpose bug." + "reason": "Related to deepseek-v4, but one is a gating logic complaint and the other is a dtype/transpose bug." }, { "left": "issue:43519", "right": "issue:46655", "accept": false, - "reason": "Qwen3VL timestamp calculation and SmolVLM video padding are different processor bugs." + "reason": "Different video bugs: timestamp calculation vs padding calculation." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel application with model_init is unrelated to flash_attention_3 degenerate training." + "reason": "Training-time Liger Kernel application issue vs flash_attention_3 causing degenerate training." }, { "left": "issue:44485", "right": "issue:47176", "accept": false, - "reason": "RoPE implementation discussion and a shell-operator allowlist bypass have no shared underlying bug." - }, - { - "left": "issue:44360", - "right": "issue:45468", - "accept": false, - "reason": "A DSA indexer ReLU discussion and Gemma4AudioRelPositionalEncoding are unrelated domains." + "reason": "RoPE implementation discussion vs a shell command substitution bypass; unrelated." } ] }, @@ -19392,7 +19397,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:28348", "issue:29870", @@ -19420,6 +19425,7 @@ "issue:44263", "issue:44273", "issue:44351", + "issue:44360", "issue:44485", "issue:44492", "issue:44623", @@ -19443,6 +19449,7 @@ "issue:47177" ], "soft_pairs": [ + "issue:44360|issue:45468", "issue:45468|issue:46894", "issue:43576|issue:44273", "issue:45230|issue:45310", @@ -19490,317 +19497,326 @@ "issue:43298|issue:44230", "issue:43717|issue:46144" ], - "prepared_review_unit_hash": "11b75bfdd519084f51c681474298ad9f0ca966ce750b1342783c2b7e17114e3b", + "prepared_review_unit_hash": "acf7b2061bd324d230419e2b71f5e818e92d4841cbd86a29120d446afb20d904", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 46, - "serialized_chars": 40068, - "estimated_input_tokens": 10017, - "estimated_eval_tokens": 20290 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 47, + "serialized_chars": 40900, + "estimated_input_tokens": 10225, + "estimated_eval_tokens": 20706 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly noisy lexical similarity: a few flash-attention and version-regression tickets are thematically close, but the majority are unrelated loading/tokenizer/utility bugs. I do not see any pair that should be merged as true duplicates under a strict duplicate-triage standard.", - "confidence": 0.34, - "canonical_issue_reason": "issue:45250 is the broadest umbrella title in the set for the flash-attention subgroup; it is the closest single issue to reuse as a canonical anchor, even though the overall cluster is heterogeneous.", + "summary": "Most items are unrelated issue reports; only the tokenizer v5 regression pair appears to be a plausible duplicate. No PRs are present.", + "confidence": 0.68, + "canonical_issue_reason": "issue:43066 is the most concrete/root-cause-oriented report in the only likely duplicate pair, so it works best as the canonical issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45250 is the least specific, most reusable issue title here and is the best available representative for the flash-attention-related subset; the rest are either narrower model-specific reports or unrelated regressions.", + "best_issue_reason": "issue:43066 names the specific tokenizer decoder-type regression in v5, making it the most actionable representative of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "Different subsystems: DSA indexer vs Gemma-4 audio positional encoding." + }, { "left": "issue:45468", "right": "issue:46894", "accept": false, - "reason": "Gemma audio RoPE vs CohereASR training-loss shift: different models and different bug classes." + "reason": "Unrelated model components and bugs; no shared code path." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken env CLI vs lazy-loading bug; unrelated symptoms and code paths." + "reason": "CLI env command breakage vs lazy-loading behavior are unrelated." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "Generic bug report vs Qwen3.5 MoE from_pretrained failure; too vague to be the same issue." + "reason": "Generic placeholder report vs a specific Qwen3.5 MoE loading error; not enough to treat as same bug." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching for audio models vs deepseekv4 weight conversion question; unrelated." + "reason": "Continuous batching audio support and DeepSeek weight conversion are unrelated feature requests." }, { "left": "issue:43976", "right": "issue:47177", "accept": false, - "reason": "Python version compatibility issue vs security RCE via shell operators; unrelated." + "reason": "Python version compatibility issue vs tar/zip-slip security bug." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Timestamp calculation bug in Qwen3VL vs GLM-5 RoPE implementation; different components." + "reason": "Timestamp calculation in Qwen3VL vs GLM-5 RoPE implementation." }, { "left": "issue:44485", "right": "issue:47177", "accept": false, - "reason": "RoPE math bug vs command-substitution security issue; no overlap." + "reason": "RoPE math bug vs archive extraction security issue." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading vs inverse_sqrt scheduler kwargs handling; unrelated." + "reason": "Lazy loading vs inverse_sqrt scheduler kwargs are unrelated." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Processor timestamp bug vs scheduler configuration bug; different subsystems." + "reason": "Different model/processor bug and scheduler bug." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Forward-doc confusion vs Llama 3.2 vision loading regression; not the same defect." + "reason": "Documentation confusion vs Llama 3.2 Vision loading regression." }, { "left": "issue:43576", "right": "issue:46218", "accept": false, - "reason": "CLI env command breakage vs import failure for a model class; different failures." + "reason": "Broken env command and missing Qwen3_5 module import are distinct." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import-star cleanup/perf issue vs VRAM leak in multi-threaded inference; unrelated." + "reason": "Import-star cleanup/performance vs VRAM leak in Qwen3ForCausalLM." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Different model-specific implementation bugs (GLM-5 RoPE vs Gemma audio positional encoding)." + "reason": "Different model internals and different failure modes." }, { "left": "issue:45850", "right": "issue:47177", "accept": false, - "reason": "Broad v5 breakage complaint vs shell-operator RCE; unrelated." + "reason": "Broad v5 breakage complaint vs shell/archive security vulnerabilities." }, { "left": "issue:45850", "right": "issue:47176", "accept": false, - "reason": "Broad v5 breakage complaint vs shell command allowlist bypass; unrelated." + "reason": "Broad model-regression complaint vs command-substitution allowlist bypass." }, { "left": "issue:44861", "right": "issue:47110", "accept": false, - "reason": "Tied-weights AttributeError vs special_tokens list/keys error; different APIs and failure modes." + "reason": "Tied-weight key crash vs special_tokens list handling bug." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak under multi-threading vs RoPE implementation issue; unrelated." + "reason": "VRAM leak in Qwen3ForCausalLM vs GLM-5 RoPE implementation." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 settings migration vs VRAM leak; different areas." + "reason": "TF32 configuration advice vs Qwen3 memory leak." }, { "left": "issue:43352", "right": "issue:47177", "accept": false, - "reason": "Flash-attention support request for Nemotron vs shell RCE security bug; unrelated." + "reason": "Flash Attention support gap vs command injection/file-write security issues." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "local_files_only network-leak bug vs timestamp calculation bug; unrelated." + "reason": "local_files_only network leak vs timestamp miscalculation." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch vs Gemma audio positional encoding; different code paths." + "reason": "CPU offload device mismatch vs Gemma-4 audio encoding issue." }, { "left": "issue:42913", "right": "issue:43066", - "accept": false, - "reason": "These are both v5 tokenizer regressions, but the titles point to different concrete symptoms and there is not enough evidence they are the same bug." + "accept": true, + "reason": "Both point to the same tokenizer v5 regression; the decoder-type bug plausibly explains the v4/v5 behavior change." }, { "left": "issue:45850", "right": "issue:46218", "accept": false, - "reason": "General v5 breakage complaint vs model import failure; too broad to duplicate." + "reason": "General v5 compatibility complaint vs missing Qwen3_5 import." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp bug vs missing processor.save_pretrained files; unrelated." + "reason": "Processor timestamp bug vs missing processor save_pretrained files." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return values in an indexer vs Gemma audio positional encoding; different implementation areas." + "reason": "GLM MoE split/indexing bug vs audio positional encoding bug." }, { "left": "issue:44861", "right": "issue:47059", "accept": false, - "reason": "Tied-weight key crash vs CLI test failure from huggingface_hub API change; unrelated." + "reason": "Tied-weight key AttributeError vs CLI completion breakage with huggingface_hub." }, { "left": "issue:43704", "right": "issue:45341", "accept": false, - "reason": "VRAM leak vs test utility bug; unrelated." + "reason": "Runtime VRAM leak vs a testing_utils bug." }, { "left": "issue:43352", "right": "issue:46218", "accept": false, - "reason": "Flash-attention support gap vs import failure for Qwen3_5ForCausalLM; different defects." + "reason": "Unsupported Flash Attention 2.0 model vs missing import/module for Qwen3_5." }, { "left": "issue:44351", "right": "issue:46218", "accept": false, - "reason": "HybridCache import failure vs Qwen3_5ForCausalLM import failure; related only by import surface, not the same missing symbol." + "reason": "HybridCache import failure vs Qwen3_5 module import failure." }, { "left": "issue:28348", "right": "issue:45250", "accept": false, - "reason": "GPT2-specific flash-attention support request and a generic flash-attention topic are related, but not clearly the same concrete fix." + "reason": "Specific GPT2LMHeadModel FA2 support request vs a generic Flash Attention 2.0 issue." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Tied-weight key AttributeError vs apply_chat_template crash on tool-call assistant messages; unrelated." + "reason": "Weight-key AttributeError vs chat-template/tool-call crash." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwarg bug vs unrelated placeholder issue title; no basis for duplication." + "reason": "Scheduler kwargs bug vs unrelated non-technical placeholder issue." }, { "left": "issue:44998", "right": "issue:47177", "accept": false, - "reason": "Placeholder issue title vs shell RCE security report; unrelated." + "reason": "Unemployment placeholder vs security vulnerability." }, { "left": "issue:44998", "right": "issue:47176", "accept": false, - "reason": "Placeholder issue title vs shell allowlist bypass; unrelated." + "reason": "Unemployment placeholder vs shell-operator allowlist bypass." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "Cache strategy typo vs placeholder issue title; unrelated." + "reason": "Cache-strategy typo vs unrelated placeholder issue." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "FP8 Qwen3-VL support vs missing processor save_pretrained files; unrelated." + "reason": "FP8 support request vs processor save_pretrained file omission." }, { "left": "issue:28348", "right": "issue:43352", "accept": false, - "reason": "Both mention Flash Attention 2.0, but one is GPT2LMHeadModel support and the other is Nemotron-specific support; separate model-specific requests." + "reason": "Both mention Flash Attention 2.0, but one is GPT2-specific support and the other is a different model's unsupported path." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression vs placeholder issue title; unrelated." + "reason": "CLIPOutput attentions regression vs unrelated placeholder issue." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow from_pretrained loading vs missing layer_type_validation import; different symptoms and root causes." + "reason": "Slow from_pretrained loading vs ImportError for layer_type_validation." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "FP8 inference support vs Qwen3.5 MoE from_pretrained error; related only by model family, not the same bug." + "reason": "FP8 inference support request vs Qwen3.5 MoE loading error." }, { "left": "issue:29870", "right": "issue:31515", "accept": false, - "reason": "BART generate() decoder_attention_mask bug vs slow checkpoint loading; unrelated." + "reason": "BART generate UnboundLocalError vs slow checkpoint loading." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLflow reporting bug on failed jobs vs slow checkpoint loading; unrelated." + "reason": "MLFlow job-finalization bug vs slow from_pretrained loading." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "from_pretrained performance issue vs tensor parallel training bug; unrelated." + "reason": "Loading performance issue vs tensor-parallel training bug." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Loading speed issue vs forward API documentation confusion; not the same problem." + "reason": "Slow checkpoint loading vs forward-API documentation confusion." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Audio continuous batching support vs FP8 Qwen3-VL inference support; different feature requests." + "reason": "Audio continuous batching limitation vs FP8 support request." }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Mamba init_weights usage vs RoFormer attention interface issue; unrelated." + "reason": "Mamba init_weights usage vs RoFormer attention-interface issue." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: only issues are present, and the sole accepted soft edge (42913/43066) is a conservative duplicate call based on closely related tokenizer-v5 descriptions. The rest of the soft-edge rejections are appropriately cautious and do not overstate evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -19815,7 +19831,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-523", + "cluster_id": "cluster-10000-524", "nodes": [ "issue:37934", "issue:42936", @@ -19839,14 +19855,14 @@ "issue:37934|issue:43668", "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "b2a34b414ba8f1d88dc28df1b6119d4b6a68c1db8f2f4aa80275340724356a69", + "prepared_review_unit_hash": "0707202ad3d81cead153ea2784e043490a923c4d03e91c897dcc68c2019b2048", "component_budget": { - "node_count": 523, - "item_count": 523, - "soft_pair_count": 1926, - "serialized_chars": 998732, - "estimated_input_tokens": 249683, - "estimated_eval_tokens": 499622 + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 }, "budget": { "node_count": 8, @@ -19861,78 +19877,78 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are a heterogeneous set of unrelated issue reports across different models/components. None of the soft pairs appear to describe the same underlying bug or change, so they should not be merged as duplicates.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "The cluster is highly heterogeneous: the issues cover unrelated models, tokenizer warnings, config typing, a typo, a loss-kwargs propagation bug, and an attention-interface mismatch. None of the soft pairs looks like the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "If a single issue must anchor the cluster, issue 42936 is the most concrete and discussion-heavy runtime bug, but it is not actually a duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 42936 is the clearest standalone bug report with concrete symptom description and the most discussion, so it is the best representative issue if one must be chosen; however, this cluster does not form a true duplicate group.", + "best_issue_reason": "Issue 42936 is the strongest representative only because it is a clear behavioral bug with the most discussion; the rest are mostly unrelated, one-off reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Different defects in different models: Mask2Former ignore_value handling vs DeepSeek-V4 shared expert gating." + "reason": "Different bugs in different models: Mask2Former ignore_value handling vs DeepSeek-V4 shared-expert gating." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "Unrelated problems: ModernBERTConfig type hint issue vs tokenizer regex warning regression." + "reason": "Config type-hint problem vs tokenizer regex warning after a version upgrade; unrelated failure modes." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "43240 is a generic kwargs-forwarding bug, while 42936 is a model-specific unused ignore_value issue; not the same code path." + "reason": "Mask2Former ignore_value usage is unrelated to kwargs not being forwarded to a loss function." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Generic loss-kwargs forwarding bug vs RoFormer attention interface usage; different subsystems and failure modes." + "reason": "Loss-kwargs plumbing bug and RoFormer attention-interface issue affect different code paths." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Mask2Former loss parameter handling and tokenizer regex warnings are unrelated." + "reason": "Different subsystems and symptoms: model ignore_value behavior vs tokenizer regex warning." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "Different models and code paths: Mask2Former loss config vs RoFormer attention implementation." + "reason": "No shared underlying bug; one is a loss-parameter issue, the other is an attention implementation concern." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Unused loss config value in Mask2Former is unrelated to a quantization attribute typo." + "reason": "Mask2Former loss handling and a typo in quantization code are not duplicates." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "A config type-hint correction is not the same as a quantization typo fix." + "reason": "Incorrect type hint in ModernBERTConfig is unrelated to a typo in a different conversion path." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "kwargs forwarding in loss computation is unrelated to a ModernBERT config typing mistake." + "reason": "Loss kwargs forwarding and config annotation accuracy are separate issues." }, { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Llama4TextL2Norm/RMS-norm question is unrelated to ModernBERTConfig norm_eps typing." + "reason": "A question about Llama4TextL2Norm semantics is unrelated to a ModernBERTConfig type hint bug." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Mask2Former ignore_value usage and ModernBERTConfig type hint are different bugs in different components." + "reason": "Different models and different defects: unused ignore_value vs incorrect norm_eps typing." } ] }, @@ -19965,52 +19981,52 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "314e07f67b13ae7323168aa47b10e780a688465711361ec8f2574a9cb5f8f59f", + "prepared_review_unit_hash": "8f85390143372ef7f6576cd3af9794c63b155bd07693e324001322ffb3eedc8e", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2876, - "estimated_input_tokens": 719, - "estimated_eval_tokens": 1694 + "serialized_chars": 2877, + "estimated_input_tokens": 720, + "estimated_eval_tokens": 1696 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2876, - "estimated_input_tokens": 719, - "estimated_eval_tokens": 1694 + "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 across different parts of the library, so the cluster should not be merged as duplicates.", + "summary": "These three issues are not duplicates of one another: one is about tokenizer truncation behavior, one is a feature request for ReduceLROnPlateau-like scheduler behavior, and one is about extracting word start/end timestamps from Wav2Vec2 outputs.", "confidence": 0.98, - "canonical_issue_reason": "No canonical issue: the items describe different problems/features \u2014 truncation-from-beginning behavior, ReduceLROnPlateau-like scheduling, and Wav2Vec2 word boundary timing.", + "canonical_issue_reason": "Issue #10519 is the most concrete and narrowly scoped of the set, so it is the best representative issue if one must be chosen, though it is not actually the same problem as the others.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good cluster representative because there is no common underlying bug or change request.", + "best_issue_reason": "#10519 is the best issue to anchor the cluster because it is specific and actionable; the other two are separate feature requests in different areas.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Different feature areas and goals: tokenizer truncation direction vs. extracting word start/end offsets from Wav2Vec2." + "reason": "Tokenizer truncation behavior and Wav2Vec2 word-boundary timing are unrelated problems in different code paths." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "Unrelated requests: truncation behavior in preprocessing versus learning-rate scheduler functionality." + "reason": "A truncation-option request for tokenizers is unrelated to a scheduler-feature request." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "No shared underlying bug/change: scheduler behavior request vs. Wav2Vec2 timing offsets." + "reason": "These are separate feature requests: learning-rate plateau scheduling versus speech word time offsets." } ] }, @@ -20042,7 +20058,7 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "c219a9ece4db3d89fe49efb0bb4d7a1ee1f17dc6da8eddea96a373e6ed807f12", + "prepared_review_unit_hash": "c2faf18d4344a4ae2c8299aac85704516576eafc8ab61e40cbe9b89f5faa3f1f", "component_budget": { "node_count": 3, "item_count": 3, @@ -20064,24 +20080,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three issues do not look like duplicates: one is about Tapas tokenization mismatching TensorFlow, one is about an auth-related error message for private models, and one is about failing to load bert-base-uncased. They share only broad model/hub context, not the same underlying bug.", - "confidence": 0.92, - "canonical_issue_reason": null, + "summary": "These three issues do not describe the same bug: one is about TAPAS tokenization mismatching TensorFlow, one is about a private-model download/auth error message, and one is a generic model-loading failure for bert-base-uncased. No duplicate cluster is supported here.", + "confidence": 0.89, + "canonical_issue_reason": "Issue #16618 is the broadest model-loading report and the closest thing to a representative issue in this set.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue cleanly represents the cluster because the reports concern different failure modes; 16618 is only the broadest of the three, but not a duplicate target for the others.", + "best_issue_reason": "#16618 is the most generally phrased and likely most reusable anchor, while #13244 and #15980 are clearly separate concerns.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "Tapas tokenization Different from Tensorflow Code", - "right": "Can't load the model for 'bert-base-uncased'.", + "left": "issue:13244", + "right": "issue:16618", "accept": false, - "reason": "Different problem domain: tokenization parity vs model loading failure." + "reason": "Tokenization parity for TAPAS is unrelated to a generic model-load failure for bert-base-uncased." }, { - "left": "Bad error message when downloading private model without being logged in.", - "right": "Can't load the model for 'bert-base-uncased'.", + "left": "issue:15980", + "right": "issue:16618", "accept": false, - "reason": "Auth/download error-message issue is not the same as a bert-base-uncased loading failure." + "reason": "Private-model auth/error-message behavior is a different problem from failing to load a public model." } ] }, @@ -20111,12 +20127,12 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "411a110435c44a2501e64f5bddd88ad84f49c2d8cce35612e97c4600021ec89f", + "prepared_review_unit_hash": "52cc011f3c96df745ad6ae9ab5630c850f8c3a2c1bbe33fbf93f63c45cb50f7d", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1590, + "serialized_chars": 1589, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -20124,7 +20140,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1590, + "serialized_chars": 1589, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -20133,18 +20149,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated community feature requests: one for Flash Attention 2 support across more architectures, the other for GGUF support across more architectures. They do not describe the same bug or change, so they should not be clustered as duplicates.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "These are two separate community contribution requests for different capabilities: Flash Attention 2 support across more architectures vs GGUF support across more architectures. They are not duplicate reports of the same underlying issue.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue: the issues target different features and code paths, so neither can represent the other.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 26350 is the more established discussion thread by activity, so it is the better cluster anchor if one is needed, but it is not a duplicate of 33260.", + "best_issue_reason": "Neither issue is a strong global representative for the pair because they are unrelated requests; if forced to pick by activity alone, issue 26350 has the longer discussion history, but it is not a duplicate of 33260.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different requested capabilities and code paths: Flash Attention 2 support vs GGUF support. No shared target or underlying issue." + "reason": "Different feature requests with different targets (Flash Attention 2 architecture support vs GGUF architecture support); not the same underlying bug/change." } ] }, @@ -20216,254 +20232,251 @@ "issue:46226|issue:47299", "issue:46931|issue:47299" ], - "prepared_review_unit_hash": "81a1aafaef4a880747f2f3cd5124a236db5574a6d79bc09fa71e5e35818be9c5", + "prepared_review_unit_hash": "6364a7874badb5b21681623ddacd740668e8d6acdc257c532cb05d8b33ad095e", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 35, - "serialized_chars": 18661, - "estimated_input_tokens": 4666, - "estimated_eval_tokens": 9588 + "serialized_chars": 18659, + "estimated_input_tokens": 4665, + "estimated_eval_tokens": 9586 }, "budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 35, - "serialized_chars": 18661, - "estimated_input_tokens": 4666, - "estimated_eval_tokens": 9588 + "serialized_chars": 18659, + "estimated_input_tokens": 4665, + "estimated_eval_tokens": 9586 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly about translation-docs request issues, but they are not the same concrete task: some are language-specific (Turkish, Spanish, Romanian), some are generic placeholder/template entries, and some look like unrelated spam/gibberish. Only the identical placeholder-template issues look like true duplicates.", - "confidence": 0.86, - "canonical_issue_reason": "issue:43834 is the best representative of the repeated generic i18n template pattern: it is an exact placeholder-style translation request and sits at the center of the similar template entries.", + "summary": "This cluster is mostly repeated i18n-doc template issues and some noisy/spam-like variants. They are textually similar, but they are not true duplicates because they refer to different languages or unrelated gibberish requests.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43834 is the cleanest generic template instance and best represents the repeated i18n-doc translation pattern, but it is still not a duplicate of the language-specific issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43834 is the clearest cluster representative because it matches the repeated template-like translation issue format and is the earliest of the near-identical generic entries.", + "best_issue_reason": "Issue 43834 is the best cluster representative: it is the most generic and centrally similar to the other template-like translation issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Turkish translation request vs a generic placeholder-template issue; different concrete tasks." + "reason": "Both are i18n-doc translation requests, but one is Turkish-specific and the other is a generic template; same category, not the same issue." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Turkish translation request vs a generic placeholder-template issue; different concrete tasks." + "reason": "Different language targets; the similarity comes from the shared template, not a shared underlying request." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Turkish-specific request vs a generic placeholder-template entry; not the same issue." + "reason": "Turkish-specific request vs generic placeholder template; not the same concrete issue." }, { "left": "issue:27088", "right": "issue:47299", "accept": false, - "reason": "Turkish translation request vs unrelated Thai/spam-like title; no shared concrete task." + "reason": "These are not the same request; one is a Turkish docs translation issue, the other is a noisy/spam-like generic translation title." }, { "left": "issue:27088", "right": "issue:38435", "accept": false, - "reason": "Turkish and Romanian translation requests are different languages, so they are distinct issues." + "reason": "Turkish and Romanian translation requests are separate issues, despite matching template language." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish-specific request vs generic placeholder-template issue; not the same task." + "reason": "Spanish translation request vs generic template issue; similar wording only." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Turkish vs Spanish translation requests are different concrete language tasks." + "reason": "Turkish and Spanish docs translations are distinct requests, not duplicates." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Spanish request vs generic placeholder-template issue; too broad to be duplicates." + "reason": "Different language targets; same template, different underlying issue." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish and Romanian translation requests are different language tasks." + "reason": "Spanish vs Romanian translation requests are separate issues." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Spanish request vs generic placeholder-template issue; not the same concrete issue." + "reason": "Same broad translation theme, but not the same language-specific issue." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Romanian request vs generic placeholder-template issue; not a duplicate." + "reason": "Romanian-specific request and a generic template issue are not duplicates." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Romanian request vs generic placeholder-template issue; not a duplicate." + "reason": "Romanian translation request vs another template instance; similarity is superficial." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "Romanian-specific request vs generic placeholder-template entry; different task." + "reason": "Different concrete request; one is language-specific, the other is generic." }, { "left": "issue:38435", "right": "issue:47299", "accept": false, - "reason": "Romanian translation request vs unrelated Thai/spam-like title; no concrete overlap." + "reason": "Romanian translation issue and spam-like generic translation title do not match the same underlying change." }, { "left": "issue:38435", "right": "issue:46226", "accept": false, - "reason": "Romanian request vs unrelated Thai/gibberish title; insufficient evidence of the same issue." + "reason": "These are not the same translation request; the titles are only loosely similar." }, { "left": "issue:43834", "right": "issue:44370", - "accept": true, - "reason": "These are the same placeholder i18n template issue: identical title pattern and near-identical body footprint, so likely duplicate submissions." + "accept": false, + "reason": "Both are placeholder-style translation issues, but they are separate instances rather than one duplicate." }, { "left": "issue:43834", "right": "issue:44365", - "accept": true, - "reason": "These are the same placeholder i18n template issue: identical title pattern and near-identical body footprint, so likely duplicate submissions." + "accept": false, + "reason": "Highly similar templates, but no evidence they describe the same concrete task." }, { "left": "issue:43834", "right": "issue:47299", "accept": false, - "reason": "Generic template translation issue vs unrelated Thai/spam-like title; not the same concrete request." + "reason": "Generic translation template vs spam-like variant; not the same issue." }, { "left": "issue:43834", "right": "issue:46226", "accept": false, - "reason": "Same broad translation theme, but the titles/content are different enough that they do not look like one concrete duplicate." + "reason": "These look like repeated template noise, not one mergeable duplicate issue." }, { "left": "issue:43834", "right": "issue:46931", "accept": false, - "reason": "Generic template translation issue vs unrelated Thai/non-English title; no clear duplicate relationship." + "reason": "Different language/request content; the overlap is only in template structure." }, { "left": "issue:44365", "right": "issue:44370", - "accept": true, - "reason": "These are the same placeholder i18n template issue: identical title pattern and near-identical body footprint, so likely duplicate submissions." + "accept": false, + "reason": "Both are generic translation-template issues, but they appear to be separate instances, not the same bug/change." }, { "left": "issue:44365", "right": "issue:47299", "accept": false, - "reason": "Generic template translation issue vs unrelated Thai/spam-like title; not the same issue." + "reason": "Template similarity is high, but the underlying request text is not the same." }, { "left": "issue:44365", "right": "issue:46226", "accept": false, - "reason": "Template-style issue vs unrelated Thai/gibberish title; not enough evidence for a duplicate." + "reason": "No clear shared concrete request beyond generic translation wording." }, { "left": "issue:44365", "right": "issue:46931", "accept": false, - "reason": "Template-style translation issue vs unrelated non-English title; different concrete request." + "reason": "Different content and no evidence of a shared underlying issue." }, { "left": "issue:44370", "right": "issue:47299", "accept": false, - "reason": "Generic placeholder translation issue vs unrelated Thai/spam-like title; not a duplicate." + "reason": "Generic translation template vs noisy variant; not a duplicate." }, { "left": "issue:44370", "right": "issue:46226", "accept": false, - "reason": "Template-style issue vs unrelated Thai/gibberish title; not the same concrete issue." + "reason": "These are separate template-like issues, not one specific task." }, { "left": "issue:44370", "right": "issue:46931", "accept": false, - "reason": "Template-style translation issue vs unrelated non-English title; different request." + "reason": "No indication they describe the same concrete translation request." }, { "left": "issue:46222", "right": "issue:47299", "accept": false, - "reason": "Both look spam-like, but the titles are different and there is no clear shared concrete issue." + "reason": "Both are noisy translation-like issues, but the titles/content do not match as the same request." }, { "left": "issue:44370", "right": "issue:46222", "accept": false, - "reason": "Generic template translation issue vs unrelated Thai title; not the same task." + "reason": "Template similarity only; one is generic English, the other is unrelated Thai-like text." }, { "left": "issue:44365", "right": "issue:46222", "accept": false, - "reason": "Generic template translation issue vs unrelated Thai title; not enough evidence of duplication." + "reason": "Surface similarity from repeated template wording, but not the same underlying issue." }, { "left": "issue:43834", "right": "issue:46222", "accept": false, - "reason": "Generic placeholder issue vs unrelated Thai title; similar theme only, not the same concrete issue." + "reason": "Generic translation template vs unrelated Thai text; too weak to merge." }, { "left": "issue:46222", "right": "issue:46226", "accept": false, - "reason": "Both are Thai/spam-like, but the titles differ and there is no clear evidence they are the same issue." + "reason": "Both are noisy/non-English variants, but there is no clear proof of the same concrete request." }, { "left": "issue:46226", "right": "issue:46931", "accept": false, - "reason": "Unrelated Thai/gibberish title vs another distinct non-English title; no duplicate signal." + "reason": "Different titles and no shared concrete task beyond superficial similarity." }, { "left": "issue:46226", "right": "issue:47299", "accept": false, - "reason": "Unrelated Thai/gibberish title vs another spam-like translation title; different concrete issue." + "reason": "Spam-like or template-like wording is similar, but the actual request is not the same." }, { "left": "issue:46931", "right": "issue:47299", "accept": false, - "reason": "Different non-English/spam-like titles with no evidence of the same underlying request." + "reason": "These are unrelated translation-like issues with only weak textual overlap." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The analysis is grounded in the packet and stays conservative. The distinction between language-specific translation requests, the repeated placeholder-template issues, and the Thai/gibberish-looking entries is supported by the titles/body footprints and the soft-similarity evidence. The duplicate call only for the near-identical placeholder-template issues is appropriately cautious." - }, - "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 }, @@ -20552,340 +20565,304 @@ "issue:42503|issue:46653", "issue:45027|issue:46653" ], - "prepared_review_unit_hash": "cd9d8ba3537e153361a80e1b4df7ece8c749b8bc42b60b6a0eac4b6ed1495e65", + "prepared_review_unit_hash": "b8d753ce1f0fb2d1cd2aabf936063e48bc3626cdc975c26214eb728496fa8ed1", "component_budget": { "node_count": 21, "item_count": 21, "soft_pair_count": 48, - "serialized_chars": 28618, - "estimated_input_tokens": 7155, - "estimated_eval_tokens": 14566 + "serialized_chars": 28616, + "estimated_input_tokens": 7154, + "estimated_eval_tokens": 14564 }, "budget": { "node_count": 21, "item_count": 21, "soft_pair_count": 48, - "serialized_chars": 28618, - "estimated_input_tokens": 7155, - "estimated_eval_tokens": 14566 + "serialized_chars": 28616, + "estimated_input_tokens": 7154, + "estimated_eval_tokens": 14564 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is a loose set of model-request issues with superficial title similarity, not true duplicates. The only clear hub is the RF-DETR request, but none of the pairings describe the same concrete change.", - "confidence": 0.88, - "canonical_issue_reason": "issue:36879 is the most central node by similarity links and is a clean, concrete model-add request, so it works best as an anchor if one must be chosen.", + "summary": "This is a loose cluster of model-request issues, not true duplicates. They share a generic 'add support/config/model' pattern, but the concrete models, tasks, and code paths are mostly unrelated, so every soft edge should be rejected.", + "confidence": 0.96, + "canonical_issue_reason": "If a single anchor is needed, issue 42503 ('Add ModernVBERT models') is the broadest and most representative model-addition request in the set, but it does not subsume the others as duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:36879 is the best representative issue for the cluster because it is specific, resolved, and has the broadest apparent overlap pattern, even though the overlaps are not duplicates.", + "best_issue_reason": "Issue 42503 is the most umbrella-like and central to the cluster\u2019s recurring theme of adding model support/configuration.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36718", "right": "issue:47004", "accept": false, - "reason": "RoMa keypoint matcher and LoMa keypoint detector/matcher are different model requests; similar naming is not enough to treat them as the same change." + "reason": "Related CV keypoint requests, but RoMa matcher and LoMa detector/matcher are different concrete model additions." }, { "left": "issue:36718", "right": "issue:41115", "accept": false, - "reason": "RoMa keypoint matching and MiniCPM3 architecture work are unrelated code paths." + "reason": "Different domains and changes: keypoint matching vs adding MiniCPM3 architecture support." }, { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "EuroBert config support and DEIMv2 support are distinct model additions." + "reason": "Both are model-addition requests, but EuroBert config and DEIMv2 are unrelated concrete models." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "A EuroBert config update is unrelated to adding Qwen3-TTS support." + "reason": "EuroBert config and Qwen3-TTS support are unrelated model families and code paths." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert and RF-DETR are different model families with different implementation needs." + "reason": "Different model requests: EuroBert config vs RF-DETR support." }, { "left": "issue:36817", "right": "issue:42954", "accept": false, - "reason": "EuroBert config support and XiaoMi MiMo support are unrelated model requests." + "reason": "EuroBert config and XiaoMi MiMo are distinct model additions, not the same change." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are both vision models, but they are not the same concrete model or code change." + "reason": "Both are vision-model requests, but RF-DETR and DEIMv2 are separate concrete integrations." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR model support and LFM2.5 Audio support are unrelated changes." + "reason": "RF-DETR detection support and LFM2.5 Audio support are unrelated." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR support and a weight-sparse transformer request do not share the same implementation path." + "reason": "RF-DETR model support and the weight-sparse transformer request address different code paths." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR detection support and Qwen3-ASR support are different model integrations." + "reason": "RF-DETR and Qwen3-ASR are unrelated model additions." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR and Qwen3-TTS are unrelated model requests." + "reason": "RF-DETR and Qwen3-TTS are distinct model support requests." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT support and Qwen3-ASR support are unrelated changes." + "reason": "NeoBERT support and Qwen3-ASR support are different model integrations." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT and a weight-sparse transformer request are different code paths." + "reason": "NeoBERT model addition and the circuit-sparsity transformer request are not the same underlying change." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT model support and LFM2.5 Audio support are not the same underlying change." + "reason": "NeoBERT and LFM2.5 Audio 1.5B are unrelated model requests." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are unrelated model additions." + "reason": "RF-DETR and NeoBERT are separate model-addition issues sharing only the generic template." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are different model families, so this is not a duplicate." + "reason": "NeoBERT and ModernVBERT are different concrete model additions, not duplicates." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture work and Qwen3-ASR support are unrelated." + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are unrelated." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 support and the weight-sparse transformer request are different changes." + "reason": "MiniCPM3 model architecture and circuitgpt support do not share the same code-path change." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are distinct model additions." + "reason": "MiniCPM3 and ModernVBERT are distinct model requests." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 architecture support is unrelated to Voxtral TTS support." + "reason": "MiniCPM3 support and Voxtral TTS support are unrelated model integrations." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are different model support requests." + "reason": "DEIMv2 support and Qwen3-TTS support are different model additions." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT models and Qwen3-ASR support are unrelated." + "reason": "ModernVBERT models and Qwen3-ASR support are unrelated concrete changes." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are different model families and code paths." + "reason": "ModernVBERT and Qwen3-TTS are separate model requests." }, { "left": "issue:42503", "right": "issue:42954", "accept": false, - "reason": "ModernVBERT models and XiaoMi MiMo support are unrelated." - }, - { - "left": "issue:42503", - "right": "issue:46653", - "accept": false, - "reason": "ModernVBERT model support and an issue titled with a personal name are clearly not the same change." + "reason": "ModernVBERT models and XiaoMi MiMo are distinct additions with no shared underlying bug." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like models with RoPE and Voxtral TTS support are unrelated." + "reason": "BERT-like RoPE support and Voxtral TTS support are unrelated." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT may be BERT-like, but this is still a specific model request versus a generic architectural feature request." + "reason": "Both are broad model-related requests, but one asks for ModernVBERT support and the other for RoPE handling in BERT-like models." }, { "left": "issue:42738", "right": "issue:42954", "accept": false, - "reason": "BERT-like RoPE support and XiaoMi MiMo support are not the same underlying implementation." + "reason": "BERT-like models with RoPE and XiaoMi MiMo are different model concerns." }, { "left": "issue:42738", "right": "issue:46653", "accept": false, - "reason": "BERT-like RoPE support and the Mirko Privitera issue are unrelated." + "reason": "BERT-like RoPE support and the Mirko Privitera issue are unrelated; the latter is not a model request." }, { "left": "issue:42954", - "right": "issue:43671", + "right": "issue:44205", "accept": false, - "reason": "XiaoMi MiMo and Qwen3-TTS are different model requests." + "reason": "XiaoMi MiMo and SAM3-LiteText are different model additions." }, { "left": "issue:42954", - "right": "issue:44205", + "right": "issue:43671", "accept": false, - "reason": "MiMo support and SAM3-LiteText support are unrelated." + "reason": "MiMo and Qwen3-TTS are unrelated model families." }, { "left": "issue:42954", "right": "issue:45177", "accept": false, - "reason": "MiMo support and DeepSeek-OCR2 support are different model integrations." + "reason": "MiMo and DeepSeek-OCR2 are distinct requests." }, { "left": "issue:42954", "right": "issue:46653", "accept": false, - "reason": "XiaoMi MiMo and Mirko Privitera are clearly unrelated." + "reason": "XiaoMi MiMo and the Mirko Privitera issue are unrelated." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are related by domain, but they are different tasks and likely different code paths." + "reason": "Qwen3-TTS and Qwen3-ASR are related by vendor and modality, but they are different concrete support changes and should not be merged." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS and Voxtral-4B-TTS are different TTS model requests, not the same concrete change." + "reason": "Both involve audio models, but Qwen3-TTS and Voxtral TTS are different model integrations." }, { "left": "issue:43671", - "right": "issue:45177", + "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS support and DeepSeek-OCR2 support are unrelated." + "reason": "Qwen3-TTS support and circuitgpt support are unrelated." }, { "left": "issue:43671", - "right": "issue:44121", + "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS and the weight-sparse transformer request do not describe the same fix." + "reason": "Qwen3-TTS and DeepSeek-OCR2 are different model requests." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio are both audio-related, but they are distinct model integrations." + "reason": "Both are audio-related, but Qwen3-ASR and LFM2.5 Audio 1.5B are separate concrete model supports." }, { "left": "issue:43837", - "right": "issue:44121", + "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR support and a weight-sparse transformer request are unrelated." + "reason": "Qwen3-ASR support and EUPE support are unrelated." }, { "left": "issue:43837", - "right": "issue:45027", + "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR and Voxtral TTS are different audio tasks and different model families." + "reason": "Qwen3-ASR support and the weight-sparse transformer request are different code paths." }, { "left": "issue:43837", - "right": "issue:45177", + "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR support and DeepSeek-OCR2 support are unrelated." + "reason": "Qwen3-ASR and Voxtral TTS are distinct model support requests." }, { - "left": "issue:43837", - "right": "issue:45322", + "left": "issue:37015", + "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and EUPE are unrelated model requests." + "reason": "NeoBERT support and Voxtral TTS support are unrelated model additions." }, { - "left": "issue:43909", - "right": "issue:44205", + "left": "issue:43837", + "right": "issue:45177", "accept": false, - "reason": "LFM2.5 Audio and SAM3-LiteText are different model additions." - }, - { - "left": "issue:43909", - "right": "issue:46645", - "accept": false, - "reason": "LFM2.5 Audio support and SWA+sink on Granite models are not the same change." - }, - { - "left": "issue:44121", - "right": "issue:45322", - "accept": false, - "reason": "Weight-sparse transformer support and EUPE support are unrelated requests." - }, - { - "left": "issue:43837", - "right": "issue:45027", - "accept": false, - "reason": "Qwen3-ASR and Voxtral-4B-TTS are distinct speech model requests." - }, - { - "left": "issue:37015", - "right": "issue:45027", - "accept": false, - "reason": "NeoBERT and Voxtral TTS are unrelated model support issues." - }, - { - "left": "issue:43837", - "right": "issue:45177", - "accept": false, - "reason": "Qwen3-ASR support and DeepSeek-OCR2 support are different changes." + "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model requests." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE and TIPSv2 are different model requests from different authorship and problem spaces." + "reason": "EUPE and TIPSv2 are both vision-language research models, but they are separate concrete additions." }, { "left": "issue:42503", "right": "issue:46653", "accept": false, - "reason": "ModernVBERT and Mirko Privitera are unrelated." + "reason": "ModernVBERT models and the Mirko Privitera issue are unrelated; the latter is not a model request." }, { "left": "issue:45027", "right": "issue:46653", "accept": false, - "reason": "Voxtral-4B-TTS support and Mirko Privitera are unrelated." + "reason": "Voxtral TTS support and the Mirko Privitera issue are unrelated." } ] }, @@ -20915,12 +20892,12 @@ "soft_pairs": [ "issue:38740|issue:46443" ], - "prepared_review_unit_hash": "0b37880632f18e5cb90d80bd0decb116dd3cbbe5a47f8f213b41231e4974df74", + "prepared_review_unit_hash": "2608ef635619629315066ddcfc213b58426237a38c057cfcda16267c956799ba", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1505, + "serialized_chars": 1506, "estimated_input_tokens": 377, "estimated_eval_tokens": 1010 }, @@ -20928,7 +20905,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1505, + "serialized_chars": 1506, "estimated_input_tokens": 377, "estimated_eval_tokens": 1010 }, @@ -20937,18 +20914,18 @@ "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, while the other is a Vision RoPE refactor. They should not be clustered as duplicates.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "These two issues are not duplicates: one is a docs request about adding Pruna, the other is a vision RoPE refactor. The soft match should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "No canonical issue: the issues describe unrelated work (documentation/feature listing vs a vision model refactor).", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good canonical representative for the other; they describe different areas and different requested work.", + "best_issue_reason": "issue:46443 is the more substantive technical issue; issue:38740 is a documentation request and not a good duplicate anchor for this cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:38740", "right": "issue:46443", "accept": false, - "reason": "Different subjects and scopes: documentation entry for Pruna vs a Vision RoPE code refactor. No shared code path or underlying bug/change." + "reason": "Different topics and scopes: a docs addition request versus a vision RoPE refactor; no shared bug or change." } ] }, @@ -20978,12 +20955,12 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "d0b0e633e0bdc8b24bed6afe804eda2c06410e009c27df1e10009e659a998079", + "prepared_review_unit_hash": "20efbfdefd64bdcca6557f20fa91741c46f901f55c5bb1df02dcae3bccd62fb5", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, + "serialized_chars": 1479, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -20991,7 +20968,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, + "serialized_chars": 1479, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -21000,18 +20977,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are not duplicates: one is about set block decoding, the other about adding DeepSeek v3.2 support. They concern different features and there is no evidence they target the same bug or change.", - "confidence": 0.97, - "canonical_issue_reason": "No clear canonical issue: the tickets describe unrelated work items with different user goals and no shared technical evidence.", + "summary": "These look like two different feature requests, not duplicates: one asks for a block decoding setting, the other asks for DeepSeek v3.2 support.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #43037 is the more specific, user-facing request and is the better representative of the cluster if one must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong cluster representative because the pair is not a duplicate cluster; if forced, 43037 is slightly broader and more feature-oriented.", + "best_issue_reason": "#43037 is the best issue candidate because it clearly states a concrete model-support request, whereas #41084 is more generic and ambiguous.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Titles and likely scopes differ substantially: decoding behavior vs model support. No shared files, targets, or concrete code-path evidence links them to the same underlying issue." + "reason": "The titles describe different underlying changes: block decoding configuration vs support for a specific model version. No shared target or code-path evidence suggests duplication." } ] }, @@ -21041,7 +21018,7 @@ "soft_pairs": [ "issue:41306|issue:41826" ], - "prepared_review_unit_hash": "92c3f719ae19f478c06d0adc9bd606757baa95e21c71cf16498aff1d2a974ec5", + "prepared_review_unit_hash": "6a550ff73ad490d306cc553c5a8b91812c5cfb360d7c7d536b050b1591a76fa5", "component_budget": { "node_count": 2, "item_count": 2, @@ -21063,18 +21040,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only by the word \"Tiled\" and both being optimization-focused issues, but they describe different feature requests in different code paths: one is about enabling TiledFusedLogitsLoss/FLCE in `from_pretrained()`, the other about adding TiledMLP for lower memory use.", - "confidence": 0.95, - "canonical_issue_reason": "No canonical duplicate issue: the requests target different components and changes (FLCE flag wiring vs. TiledMLP integration).", + "summary": "The two issues describe different feature requests in different areas: one is about enabling TiledFusedLogitsLoss/FLCE via `from_pretrained()`, while the other is about integrating TiledMLP for lower memory usage. They do not appear to be duplicates or the same underlying change.", + "confidence": 0.92, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:41306 is the more concrete and implementation-specific request, so it is the better representative issue if one must be chosen.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41306", "right": "issue:41826", "accept": false, - "reason": "Different underlying features and code paths; similar naming is not enough to treat them as the same bug or change." + "reason": "Different requested functionality and code paths: FLCE configuration in `from_pretrained()` vs. a TiledMLP memory-footprint integration. Shared ML/model-efficiency theme is too broad to treat as the same issue." } ] }, @@ -21104,7 +21081,7 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "577827bc3c25c8dfb8327bf86858826013fbe4adee524b16a806d34bccf0387c", + "prepared_review_unit_hash": "dff91ce0c664ed7506f4bfc626ee53cc8f794ef0c61efde14caa39ebc881d9c1", "component_budget": { "node_count": 2, "item_count": 2, @@ -21126,18 +21103,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two related-but-distinct Trainer issues: one about preserving customizability around `remove_unused_column`, the other about FLOPS estimation for custom trainers. They share the broad \u201ccustom trainer\u201d theme, but not the same underlying bug or change.", - "confidence": 0.94, - "canonical_issue_reason": null, + "summary": "The two issues are related only at a high level: both concern `Trainer` not handling custom trainers well, but they describe different problems and likely different fixes. I would not merge them as duplicates.", + "confidence": 0.87, + "canonical_issue_reason": "Issue #43598 is the broader and more central customizability complaint about `Trainer` behavior, so it works better as the representative issue in this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43598 is the more representative cluster item only in the sense that it concerns Trainer customizability, but it is not a duplicate of 43600.", + "best_issue_reason": "#43598 is the best cluster representative because it captures the more general Trainer extensibility concern, whereas #43600 is a narrower FLOPS-estimation case.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "They concern different Trainer behaviors: column pruning/customizability vs FLOPS estimation. Shared mention of custom trainers is too broad to treat as the same issue." + "reason": "They share the theme of custom trainers, but one is about `remove_unused_column` customizability and the other is about FLOPS estimation logic. Different code paths and different fixes, so not the same underlying bug." } ] }, @@ -21167,7 +21144,7 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "95dd0cbf1d734765d1e786553e91b490ab77ab069c187aae1189c1943122c83f", + "prepared_review_unit_hash": "358e64c354e7671df4cb5810651e7f2485ed375887bd0732fd148cbfcd038c8c", "component_budget": { "node_count": 2, "item_count": 2, @@ -21189,18 +21166,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are about different Qwen model support gaps: one requests AutoModel/AutoModelForConditionalGeneration registration for Qwen3-Omni, while the other requests AutoModelForSequenceClassification support for Qwen3.5. They share only a broad theme of model-registration work, not the same underlying bug or change.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "These are related only at a very high level (Qwen auto-class support) but they target different model families and different auto-class behaviors, so they should not be treated as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "issue:43684 is the broader, more central registration request and is a better representative of the Qwen auto-model support theme.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:43684 is the best single issue to keep as the representative because it covers a wider API registration gap rather than a narrower task-specific extension.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different model families and different auto-class targets (conditional generation registration vs sequence classification support). They are not the same underlying bug/change and would not plausibly be merged into one fix." + "reason": "Different underlying requests: Qwen3-Omni auto-registration for general/conditional generation vs Qwen3.5 sequence classification support. Same ecosystem, but not the same bug or change." } ] }, @@ -21237,12 +21214,12 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "e6204427c18f9d9c4a3124bcc042c4759adae9bd6a3c40397637aa83ce6839f4", + "prepared_review_unit_hash": "3ccb2674d358644f097cd1ce5c4cf43770fabf212b4a6b3a6b0e6d4768d4858f", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4264, + "serialized_chars": 4263, "estimated_input_tokens": 1066, "estimated_eval_tokens": 2388 }, @@ -21250,7 +21227,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4264, + "serialized_chars": 4263, "estimated_input_tokens": 1066, "estimated_eval_tokens": 2388 }, @@ -21259,58 +21236,58 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly low-information issue duplication: 43845, 43846, and 43859 share the same \"huggingface \" title, while 43835 looks unrelated. Based on the metadata available here, 43845 is the best canonical representative because it is the earliest of the near-identical huggingface reports.", - "confidence": 0.67, - "canonical_issue_reason": "issue:43845 is the earliest of the near-identical \"huggingface \" issues and the best representative of that group based on the available metadata.", + "summary": "The three \"huggingface \" issues look related by title and timing, but the packet does not provide enough concrete evidence to confirm duplicates; \"Fett\" appears unrelated. I would use issue 43845 as a provisional representative only because it is the earliest of the similar issues, not because duplicate status is proven.", + "confidence": 0.59, + "canonical_issue_reason": "issue:43845 is the earliest of the similar \"huggingface \" issues, so it is the best provisional canonical representative if they are later confirmed to be duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "Among the low-information \"huggingface \" reports, issue:43845 is the safest canonical issue because it is earliest and most representative from the metadata shown.", + "best_issue_reason": "issue:43845 is the strongest single issue candidate because it is earliest among the closely similar \"huggingface \" reports, though the evidence is only metadata-level similarity.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Different titles and no shared substantive details; insufficient evidence they describe the same underlying issue." + "reason": "The titles are different and the packet provides no body text, links, or shared targets showing the same bug." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Different titles and no content to tie them to the same bug." + "reason": "\"Fett\" and \"huggingface \" do not clearly point to the same underlying issue, and there is no concrete linking evidence." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Different titles and no shared issue-specific evidence." + "reason": "The titles are unrelated and the packet gives no explicit or structural evidence that they are duplicates." }, { "left": "issue:43845", "right": "issue:43846", - "accept": true, - "reason": "Same title, near-identical timestamps, and identical body length make these look like the same low-information report." + "accept": false, + "reason": "The titles are nearly identical, but the packet only shows soft similarity plus close timestamps; that is not enough to confirm a duplicate." }, { "left": "issue:43845", "right": "issue:43859", "accept": false, - "reason": "Same title is suggestive, but the body length and timing differ enough that the available metadata does not firmly show the same report." + "reason": "These two issues look similar by title and timing, but there is no body text or other concrete evidence proving the same report." }, { "left": "issue:43846", "right": "issue:43859", "accept": false, - "reason": "Same-title similarity alone is too weak here; the available metadata does not confirm the same underlying issue." + "reason": "Near-identical titles alone are insufficient here; the packet does not show explicit links, shared targets, or other proof of duplication." } ] }, "evaluator_result": { "accept": false, - "feedback": "The cluster-level summary is mostly grounded, but it overstates the evidence in the canonical-record rationale: the packet only shows titles, timestamps, and body lengths, so saying issue:43845 \u201cmatches the most similar duplicates exactly\u201d is too strong. The soft-edge verdicts are otherwise conservative, but I would reject on that unsupported precision. Safer wording would be that 43845 is the earliest of the near-identical \u2018huggingface \u2019 issues and the best representative of that group based on the available metadata." + "feedback": "The grounding is mostly okay, but the soft-edge verdicts are too strong for the evidence in the packet. The packet only shows near-identical titles, similar body lengths, and close creation times; it does not show body text, explicit links, or other concrete proof that these are duplicates. In particular, the claims about a \"title/body pattern\" and that the huggingface issues \"strongly suggest duplicate reports\" overstate the evidence. Since all three huggingface pairs are only supported by soft_similarity and the deterministic_accept flag is false, those edges should be treated more conservatively." }, "evaluator_used": true, "retried": true, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -21348,106 +21325,106 @@ "issue:44573|issue:47036", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "d7cd5a06c4387696c7616b61e9ca8ad56676b91f000efa97cc94dda38167d23b", + "prepared_review_unit_hash": "fd13b095b8bc320c77fad5f92e9d3816e76fa0c79a539ba7f31a7bb602be7c57", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 12, - "serialized_chars": 7352, - "estimated_input_tokens": 1838, - "estimated_eval_tokens": 3932 + "serialized_chars": 7361, + "estimated_input_tokens": 1841, + "estimated_eval_tokens": 3938 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 12, - "serialized_chars": 7352, - "estimated_input_tokens": 1838, - "estimated_eval_tokens": 3932 + "serialized_chars": 7361, + "estimated_input_tokens": 1841, + "estimated_eval_tokens": 3938 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is dominated by spammy or generic issues with no clear shared underlying bug or request. The soft similarities appear to be superficial, so none of the pairs should be merged as duplicates.", - "confidence": 0.93, - "canonical_issue_reason": "No clear canonical issue: the titles and available metadata do not indicate a single shared problem, and most items look like spam or generic low-signal submissions.", + "summary": "The cluster is dominated by spammy or extremely generic issues with no concrete shared bug report; the titles and minimal bodies do not support duplicate merging. The only somewhat substantive item is the generic documentation suggestion, but it still does not match the others.", + "confidence": 0.97, + "canonical_issue_reason": "issue:44418 is the only item with a non-spam, legible title and slightly more plausible issue content; the rest look like low-signal spam placeholders.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #47036 is the most specific and least noisy of the set, with a concrete topic and some discussion activity, but it still does not appear duplicate-related to the others.", + "best_issue_reason": "issue:44418 is the best representative issue in this set, but only as a generic placeholder\u2014not because it is a true duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are just spam-labeled, low-information issues; no evidence they describe the same underlying bug or request." + "reason": "Both look like spam placeholders; no concrete shared bug or change is described." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both are just spam-labeled, low-information issues; no evidence they describe the same underlying bug or request." + "reason": "Both look like spam placeholders with no substantive overlap beyond generic similarity." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam and the other is a generic documentation suggestion; they do not share a concrete problem statement." + "reason": "A spam placeholder and a generic documentation suggestion are not the same underlying issue." }, { "left": "issue:44371", "right": "issue:47036", "accept": false, - "reason": "Spam title versus a community benchmark tip; these are unrelated topics with no shared bug or change." + "reason": "No shared bug/change; one is spam-like and the other is a vague community tip." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "Spam title versus a specific mention of \"Google LLC\"; no indication of the same underlying issue." + "reason": "Spam placeholder vs. unrelated named-topic issue; no duplicate evidence." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "Generic documentation enhancement versus spam; too little substantive overlap to treat as the same issue." + "reason": "A generic docs suggestion and a spam placeholder are not the same underlying report." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "Generic documentation enhancement versus spam; too little substantive overlap to treat as the same issue." + "reason": "No concrete overlap; the titles point to unrelated low-signal issues." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are spam-like, but there is no concrete shared bug or request to justify duplicate merging." + "reason": "Both are spam placeholders, but there is no evidence they report the same specific problem." }, { "left": "issue:44572", "right": "issue:47036", "accept": false, - "reason": "Spam versus a benchmark tip; the titles suggest different subjects, not the same underlying problem." + "reason": "The items are unrelated in content and intent; similarity is only superficial." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "Spam versus a specific company-related title; no concrete duplication signal." + "reason": "No shared bug or code path is described, so they should not be merged." }, { "left": "issue:44573", "right": "issue:47036", "accept": false, - "reason": "Spam versus a benchmark tip; the titles suggest different subjects, not the same underlying problem." + "reason": "The issues do not describe the same underlying problem." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "Spam versus a specific company-related title; no concrete duplication signal." + "reason": "Only generic similarity is present; no duplicate-level evidence." } ] }, @@ -21477,7 +21454,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "b2137da56e8e635e3b604c2373f0659ec75aea8cab07f59df705b16cfde5290c", + "prepared_review_unit_hash": "c90f232a04742d55c9bcd9fc918a6a904c648986e70561012c140beacdd71d88", "component_budget": { "node_count": 2, "item_count": 2, @@ -21499,24 +21476,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster has two closed issues with the same title and very similar body length, so #45145 is the best canonical thread to keep. However, the evidence is only soft textual similarity; with no explicit links or shared targets, they should be treated as likely related reports rather than confidently merged as duplicates.", - "confidence": 0.83, - "canonical_issue_reason": "Issue #45145 is the better canonical issue because it was created first and has the only inbound reference, making it the more established thread.", + "summary": "The two closed issues are highly similar and #45145 is the best canonical candidate, but the available evidence only supports a soft similarity match rather than a confirmed duplicate.", + "confidence": 0.78, + "canonical_issue_reason": "#45145 is the better canonical target because it was created earlier and has an inbound reference; however, the packet does not prove the two reports are identical beyond strong title/body similarity.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #45145 is the best representative issue for the same reasons: it is earliest and already has an inbound reference, so it is the most suitable primary record.", + "best_issue_reason": "#45145 is the strongest single representative because it is earlier and already referenced, while #45151 appears closely related but not definitively the same report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", "accept": false, - "reason": "Same title and similar body size suggest a likely relation, but the evidence is only soft similarity (Jaccard ~0.63) with no explicit links or shared targets, so this is not strong enough to confidently mark them as duplicates of the same underlying issue." + "reason": "The issues have identical titles and similar lengths, but the packet only provides soft-similarity evidence. That is not enough to confirm they describe the exact same underlying issue or exclude scope differences." } ] }, "evaluator_result": { "accept": false, - "feedback": "The canonical issue choice is grounded, but the soft-edge verdict is too strong for the packet. The pair only has soft similarity evidence with a Jaccard of ~0.63 and no explicit links or shared targets; same title and similar body length suggest likely related reports, but not enough to confidently accept them as duplicates of the same underlying issue. The summary also overstates the evidence by calling them near-identical and effectively the same report." + "feedback": "The canonical choice (#45145) is grounded by the earlier timestamp and inbound reference, but the duplicate/merge verdict is too strong for the evidence given. The packet only shows a soft-similarity match with identical titles and similar body lengths; it does not directly establish identical content or the absence of scope differences. Please soften the language and avoid presenting the pair as confirmed duplicates." }, "evaluator_used": true, "retried": true, @@ -21597,7 +21574,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "15af1210204b3909785af8ae3ac2a5068d9da34b6b70a739e02f7de6b2d59498", + "prepared_review_unit_hash": "c5437194a67a708343378d870f8d5a9422e0be379376a67755d4a03170696374", "component_budget": { "node_count": 2, "item_count": 2, @@ -21619,18 +21596,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two unrelated, empty issues that look like spam or placeholder posts rather than duplicate bug reports. There is no evidence they describe the same underlying problem.", - "confidence": 0.98, + "summary": "Two closed issues with unrelated-looking personal-name titles and no body content; there is no evidence they describe the same underlying problem.", + "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Neither issue is a good canonical representative: both are bare, distinct-name entries with no description or shared technical context.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Reject: the titles are different person/name + date strings, both bodies are empty, and there is no shared issue content or bug description indicating the same underlying problem." + "reason": "The titles reference different people and dates, with no body, links, or technical details suggesting the same underlying bug or report." } ] }, @@ -21660,7 +21637,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "4ca5dbb043121a0391a678e3d9a425298222413c48a8453a07ab555d15b0fcc2", + "prepared_review_unit_hash": "72a0e84faac6331a6d79c72bfa71456961085c12cb6a05ff489ffe9bd0cbf1b4", "component_budget": { "node_count": 2, "item_count": 2, @@ -21682,18 +21659,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues only share a weak textual similarity in their titles; they name different people and different dates, with no bodies, links, targets, or other evidence of the same underlying report.", + "summary": "The two closed issues are just separate name/date entries with no shared subject matter, body, or linked code context. They do not look like duplicates or the same underlying bug/report.", "confidence": 0.98, - "canonical_issue_reason": "No canonical issue: these look like two separate items, not duplicates, so there is no clear single issue to represent the cluster.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong global representative because the cluster does not appear to be a duplicate set; both are equally sparse and unrelated.", + "best_issue_reason": "Neither issue is a good canonical candidate; both are unrelated standalone entries with no evidence of a shared problem or duplicate relationship.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "Different personal names and dates, with no shared description or references; the similarity is superficial only." + "reason": "Different names and different dates; no overlap in content, targets, or evidence of the same underlying issue." } ] }, @@ -21723,7 +21700,7 @@ "soft_pairs": [ "issue:45810|issue:46295" ], - "prepared_review_unit_hash": "485b908c56a66612c6018c502df5ca2914688f71d2ba8d5a96ab2b0dc22cb9fe", + "prepared_review_unit_hash": "65816f935ab5762d1027f0063d21928f9e38f712f6d1f76e94c7941d8680e219", "component_budget": { "node_count": 2, "item_count": 2, @@ -21745,18 +21722,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are not duplicates: one requests a new Qwen3_5 token-classification model head, while the other asks for FP8 training support. They target different functionality and code paths.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "The two issues are unrelated: one requests a new Qwen3_5 token-classification/value-model feature, while the other asks for FP8 training support. There is no evidence they describe the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "issue #46295 is the better cluster anchor because it is the broader, more generally scoped request; #45810 is a model-specific feature request that does not align with FP8 training support.", "canonical_pr_reason": null, - "best_issue_reason": "issue 46295 is the broader platform feature request and is more likely to be the primary tracking issue for FP8 work, but it is not a duplicate of the model-head request.", + "best_issue_reason": "issue #46295 is the most suitable issue to represent the cluster since it is more general and likely to attract broader discussion, whereas #45810 is narrowly about adding one model head.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45810", "right": "issue:46295", "accept": false, - "reason": "Different underlying requests: one is adding a token-classification/value-model head, the other is FP8 training support. No shared bug or change path." + "reason": "Different topics and code paths: adding Qwen3_5ForTokenClassification for a value model vs enabling FP8 training support. They do not look like the same underlying bug or change." } ] }, @@ -21796,92 +21773,92 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "7e2851d06c74e08e8e3cbcf17ede2257699b3cbff5adbf086d6a03bd98f69a5b", + "prepared_review_unit_hash": "351b2400ac403bf973088c758aeef969368a0da609df8c1693d0dafc4d3bc8df", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5718, - "estimated_input_tokens": 1430, - "estimated_eval_tokens": 3116 + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5718, - "estimated_input_tokens": 1430, - "estimated_eval_tokens": 3116 + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This is not one unified duplicate cluster. It contains two exact duplicate issue pairs (45870/45871 and 45872/45873) plus one unrelated dataset-promotional issue (45869).", - "confidence": 0.97, - "canonical_issue_reason": "No single issue cleanly represents the whole set, but 45872 is the best anchor because it is part of an exact duplicate pair and has the clearest, most specific request title.", + "summary": "This cluster contains one loosely related dataset post and two highly similar issue pairs, but the packet only provides soft_similarity evidence. That is not enough to assert deterministic duplicate pairs or a single canonical issue.", + "confidence": 0.86, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "45872 is the strongest issue to keep as the representative of its duplicate pair; the other items either duplicate it exactly (45873) or describe different requests/topics.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Dataset sample promotion is a different topic from the DataForge example issue." + "reason": "Same broad finance/data theme, but 45869 is a dataset/sample bundle post and 45870 is an example request; not the same concrete issue." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Dataset sample promotion is unrelated to the DataForge example issue." + "reason": "Related domain only; 45869 is about a dataset bundle, while 45871 is an example request." }, { "left": "issue:45870", "right": "issue:45871", - "accept": true, - "reason": "Exact duplicate: same title and matching body length, clearly the same example request." + "accept": false, + "reason": "Very similar titles and body size, but the packet only shows soft_similarity and no stronger linkage, so this is not safe to treat as a confirmed duplicate." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both mention financial time series, but one is an example and the other is a scraping-service/DataForge request; different underlying asks." + "reason": "Both mention DataForge and finance, but one is a prediction example and the other is a scraping-service request; different underlying change/request." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "The example issue is not the same request as the DataForge scraping-service issue." + "reason": "Same broad project, but prediction example vs scraping-service request are different concrete issues." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Same as above: example request versus scraping-service request, not the same underlying change." + "reason": "Different concrete request types: prediction example versus scraping-service request." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "The example issue is distinct from the DataForge scraping-service request." + "reason": "Different concrete request types: prediction example versus scraping-service request." }, { "left": "issue:45872", "right": "issue:45873", - "accept": true, - "reason": "Exact duplicate: same title and matching body length, clearly the same request." + "accept": false, + "reason": "Highly similar titles/body size, but only soft_similarity evidence is present; insufficient to confirm a duplicate pair." } ] }, "evaluator_result": { - "accept": true, - "feedback": "Grounded summary. It correctly identifies two tight duplicate pairs (45870/45871 and 45872/45873) and treats 45869 as a separate topic. The soft-edge decisions are conservative overall, and the canonical choice of 45872 is defensible as a representative anchor." + "accept": false, + "feedback": "The packet only supports soft similarity, not exact duplication. The analyst overstates the evidence by calling 45870/45871 and 45872/45873 \"exact duplicate\" pairs and by selecting a best issue on that basis. The conservative reading is that these are highly similar issue pairs, but the packet does not provide deterministic accept evidence or any stronger linkage. 45869 being only loosely related is fine, but the summary and verdicts are too strong overall." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -21904,7 +21881,7 @@ "soft_pairs": [ "issue:46439|issue:46441" ], - "prepared_review_unit_hash": "6842b84b40424b024cb163c619b55d3119f4b7e8bfefa82510fa0acdbf110e2d", + "prepared_review_unit_hash": "9c5a8219604400fa091d729cf9230a8fe8551cd1f6f1ee7ccd013493f8df95d7", "component_budget": { "node_count": 2, "item_count": 2, @@ -21926,24 +21903,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both issues concern static cache handling for linear-attention models, but they describe different failure modes: one is an `early_initialization` shape corruption leading to `RuntimeError`, the other is a `KeyError: 'linear_attention'` in Qwen3.5. Too little evidence that they are the same underlying bug, so treat them as separate issues.", - "confidence": 0.78, - "canonical_issue_reason": "Issue 46439 is the better canonical issue because it is more general and describes the broader static-cache/linear-attention corruption path, whereas 46441 is a model-specific crash report.", + "summary": "The two issues are thematically related, but the packet only supports title-level similarity. There isn\u2019t enough evidence to treat them as the same regression or the same failure mode, so the soft duplicate edge should be rejected.", + "confidence": 0.56, + "canonical_issue_reason": "Issue 46439 is the broader and more general report: static-cache early initialization corrupting linear-attention layers via shape corruption.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46439 is the best representative of the cluster because it names the more general defect and is not tied to a single model variant.", + "best_issue_reason": "Issue 46439 is the best canonical issue for the cluster because it describes the underlying static-cache/linear-attention problem more generally, while 46441 appears to be a more model-specific crash report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46439", "right": "issue:46441", "accept": false, - "reason": "Same broad area (static cache + linear attention) but different concrete symptoms and likely different root causes: shape corruption/`RuntimeError` vs missing registry entry/`KeyError`. Not enough to merge as duplicates." + "reason": "Reject: the packet only shows related static-cache/linear-attention wording. One issue reports wrong-shape `RuntimeError` during early initialization, the other reports a Qwen3.5 `KeyError: 'linear_attention'`; without shared targets, files, or body details, this is not strong enough to call them the same bug." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The pairing is not supported strongly enough by the packet. The titles suggest related static-cache / linear-attention crashes, but the analyst overstates the evidence by treating them as the same regression and same failure mode. With only title-level similarity and no body details or shared targets/files, the soft-edge verdict should be more conservative (e.g., undecided rather than accepted)." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -21967,7 +21947,7 @@ "soft_pairs": [ "issue:46551|issue:46552" ], - "prepared_review_unit_hash": "32d606b0b91b0898a242b46abe774c1c3aa649947333a8b4c99b88c4b7c46ca4", + "prepared_review_unit_hash": "1467a789812560af1fbbe88c2234b200aa5717c3b021c80eaef6ca27df159854", "component_budget": { "node_count": 2, "item_count": 2, @@ -21989,27 +21969,27 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues appear to be duplicate reports of the same Wav2Vec2CTCTokenizer v5 regression: they have the same title, equal body length, and strong soft-similarity. #46552 is the better canonical issue because it is open and already has inbound references.", - "confidence": 0.94, - "canonical_issue_reason": "#46552 is open and already has inbound references, so it is the better current tracking issue for this likely duplicate report.", + "summary": "These two issues are exact duplicates: same title, same underlying Wav2Vec2CTCTokenizer v5 special-token regression, and identical evidence. The open issue is the better canonical target.", + "confidence": 0.99, + "canonical_issue_reason": "issue #46552 is the same report but still open and already has inbound references, so it should serve as the canonical issue.", "canonical_pr_reason": null, - "best_issue_reason": "#46552 is the best issue to keep as the primary tracker because it is the active open report with inbound references and the same title.", + "best_issue_reason": "#46552 is the best issue to keep because it is open, identical in content, and already attracting references; #46551 can be treated as the duplicate/closed copy.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46551", "right": "issue:46552", "accept": true, - "reason": "Matching title and highly similar duplicate report about the same tokenizer regression; the available evidence strongly suggests the same underlying bug, though it does not prove the bodies are textually identical." + "reason": "Exact duplicate: identical title and reported bug, with matching text indicating the same tokenizer regression." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The conclusion is mostly grounded, but the writeup overstates the evidence in a few places. The packet shows identical titles, equal body length, and strong soft-similarity, but it does not prove the bodies are textually identical or that this is definitively the same report. Please soften claims like \"identical title/body\" or \"same exact title and description\" to something like \"matching title and highly similar duplicate report.\" The canonical choice of #46552 is otherwise reasonable because it is open and has inbound references." + "accept": true, + "feedback": "The summary is grounded in the packet: both issues have identical titles, equal body length, and perfect soft-similarity/Jaccard overlap, while #46552 is open and has inbound references. The duplicate/canonical judgment is conservative and supported by the evidence." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -22033,7 +22013,7 @@ "soft_pairs": [ "issue:46657|issue:46692" ], - "prepared_review_unit_hash": "6b765580b68e813e18e586ac8929234c2541e5a22e62fc50c488d4f5df82546c", + "prepared_review_unit_hash": "3de6dcc3830f10b8802b6d12e24742cc5eefc8006c6ccb91403c9138a1d0d13f", "component_budget": { "node_count": 2, "item_count": 2, @@ -22055,18 +22035,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are both about training failures in object detection, but they describe different underlying problems: a tensor size mismatch versus out-of-memory after several epochs in RT-DETR v2. They should not be treated as duplicates.", - "confidence": 0.93, + "summary": "These two issues do not look like duplicates: one reports a tensor-shape mismatch during object detection training, while the other reports out-of-memory after several epochs in RT-DETR v2. They point to different failure modes and likely different code paths.", + "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 46657 is the more concrete, self-contained bug report and reads like a specific runtime error rather than a broad training-stability symptom.", + "best_issue_reason": "Issue 46657 is the more concrete, narrowly described bug report, but it should not be treated as the same problem as 46692.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46657", "right": "issue:46692", "accept": false, - "reason": "Different failure modes and likely different code paths: one reports tensor shape mismatch during training, the other reports OOM after epochs in RT-DETR v2. Shared training context is too broad to merge." + "reason": "Reject: the titles and likely symptoms differ substantially\u2014tensor size mismatch vs OOM after epochs\u2014and there is no evidence they stem from the same underlying defect." } ] }, @@ -22096,12 +22076,12 @@ "soft_pairs": [ "issue:46822|issue:46823" ], - "prepared_review_unit_hash": "38fa17456ba78fe39776a272fcce6c089ad7cc4f8248a375560eb95fe9c76fd9", + "prepared_review_unit_hash": "25f7310998229ca1f2570f455c15705cfa12dfe1268f9abd0c5fb78bcc904108", "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 }, @@ -22109,7 +22089,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1674, + "serialized_chars": 1673, "estimated_input_tokens": 419, "estimated_eval_tokens": 1094 }, @@ -22118,18 +22098,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These issues describe different loading-time failures in different code paths: ZeRO-3 zero.Init partitioning vs 4-bit device_map/max_memory dispatch. They are not duplicates.", + "summary": "These two issues are about different failure modes in different loading paths: one is a ZeRO-3/zero.Init partitioning bug causing OOM on multi-GPU load, the other is a 4-bit device_map/max_memory dispatch error that incorrectly falls back to CPU. They are not duplicates and should not be merged.", "confidence": 0.96, - "canonical_issue_reason": "Issue 46822 is the better canonical anchor because it is open and describes a specific, high-signal ZeRO-3 OOM/loading failure with a concrete model path.", + "canonical_issue_reason": "Issue 46822 is the better representative of the cluster because it is an open, specific regression about ZeRO-3 zero.Init failing to partition a composite submodule and causing OOM during multi-GPU model loading.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46822 is the strongest representative issue here: it is active, clearly scoped, and tied to a concrete partitioning bug rather than a more general loading configuration mismatch.", + "best_issue_reason": "Issue 46822 is the best issue to keep as the cluster anchor: it describes a concrete code-path problem, is still open, and is more likely to need follow-up than the already-closed 46823.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46822", "right": "issue:46823", "accept": false, - "reason": "Different bugs and code paths: one is ZeRO-3 zero.Init failing to partition a composite submodule and causing OOM; the other is 4-bit loading with device_map/max_memory incorrectly dispatching to CPU and raising ValueError." + "reason": "Reject: the titles describe unrelated bugs in different subsystems and loading mechanisms. One is ZeRO-3 partitioning during zero.Init, the other is 4-bit dispatch with device_map/max_memory falling back to CPU despite fitting. They do not appear to share the same underlying defect." } ] }, @@ -22201,238 +22181,238 @@ "issue:46985|issue:47102", "issue:47049|issue:47102" ], - "prepared_review_unit_hash": "413b6b05229517c5cf3c320d427424b061f04118b645d9fe665bc635dc06a6f1", + "prepared_review_unit_hash": "83e1454d2d470d72fd10903330395786c364e339ed4198a720b770c92ef0869f", "component_budget": { "node_count": 11, "item_count": 11, "soft_pair_count": 34, - "serialized_chars": 18421, - "estimated_input_tokens": 4606, - "estimated_eval_tokens": 9468 + "serialized_chars": 18418, + "estimated_input_tokens": 4605, + "estimated_eval_tokens": 9466 }, "budget": { "node_count": 11, "item_count": 11, "soft_pair_count": 34, - "serialized_chars": 18421, - "estimated_input_tokens": 4606, - "estimated_eval_tokens": 9468 + "serialized_chars": 18418, + "estimated_input_tokens": 4605, + "estimated_eval_tokens": 9466 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates in the usual sense: they are a series of daily [serge] integration failure triage issues, each apparently tracking a different day\u2019s failures. They are highly similar in template and topic, but each is a separate dated triage report rather than one underlying bug.", - "confidence": 0.94, - "canonical_issue_reason": "Pick issue:47037 as the canonical issue only as the most representative/active triage entry in the set; it has the highest discussion activity and sits in the middle of the daily series. That said, these issues are date-scoped reports, so the cluster should not be treated as a single duplicate bug report.", + "summary": "This cluster is a series of similarly titled, date-stamped Serge integration triage issues. They read like repeated daily reports rather than the same underlying bug, so I rejected all soft duplicate edges. No PRs are present.", + "confidence": 0.93, + "canonical_issue_reason": "issue:46906 is the earliest dated entry in the series and has the longest body, making it the safest canonical representative.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:47037 is the best issue candidate because it has the strongest engagement and is a typical example of the recurring triage format, making it the most useful anchor for the cluster.", + "best_issue_reason": "issue:47037 has the highest discussion activity, so it is the most informative single issue to inspect in this cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46906", "right": "issue:46940", "accept": false, - "reason": "Same triage template, but different dates/day-specific reports; not the same underlying issue." + "reason": "Same triage template, but these are separate daily reports on different dates; no evidence they describe the same failure." }, { "left": "issue:46906", "right": "issue:46948", "accept": false, - "reason": "These are separate daily triage reports, not one duplicated bug report." + "reason": "Date-stamped triage entries from different days; title similarity alone is not enough to merge them." }, { "left": "issue:46906", "right": "issue:46969", "accept": false, - "reason": "Different dated triage entries; similarity is structural, not a shared concrete bug." + "reason": "These look like distinct daily integration triage reports, not the same underlying bug." }, { "left": "issue:46906", "right": "issue:46985", "accept": false, - "reason": "Same recurring triage series, but each issue corresponds to a different day." + "reason": "Reject: recurring triage item for a different date, with no concrete shared defect evidence." }, { "left": "issue:46906", "right": "issue:47037", "accept": false, - "reason": "Not duplicates: both are daily integration triage issues with different dates and likely different failure snapshots." + "reason": "Both are generic Serge integration triage reports, but they are separate dated instances rather than one bug." }, { "left": "issue:46940", "right": "issue:46948", "accept": false, - "reason": "Highly similar wording, but they are separate day-specific triage reports." + "reason": "Same repeating triage series, but each issue is tied to a different day and likely a different run." }, { "left": "issue:46940", "right": "issue:46969", "accept": false, - "reason": "Different dated integration triage instances; reject as duplicates." + "reason": "No concrete shared code-path failure; these are distinct dated triage entries." }, { "left": "issue:46940", "right": "issue:46985", "accept": false, - "reason": "Same template, different day and likely different failure set." + "reason": "Reject: same broad subsystem and title pattern, but not the same underlying issue." }, { "left": "issue:46940", "right": "issue:46998", "accept": false, - "reason": "These are recurring daily reports, not one underlying bug." + "reason": "These appear to be separate daily triage reports, not duplicates of one another." }, { "left": "issue:46948", "right": "issue:46969", "accept": false, - "reason": "Date-scoped triage issues; similarity alone is insufficient." + "reason": "Different date-stamped triage reports; nothing here indicates the same bug instance." }, { "left": "issue:46948", "right": "issue:46985", "accept": false, - "reason": "Same triage series, but each issue tracks a distinct day." + "reason": "Reject: recurring integration triage entries, but no shared concrete failure is shown." }, { "left": "issue:46948", "right": "issue:46998", "accept": false, - "reason": "Different daily integration-failure reports, not a duplicate bug." + "reason": "Same generic title, but they are separate daily reports rather than duplicate issues." }, { "left": "issue:46969", "right": "issue:46985", "accept": false, - "reason": "Separate dated triage reports, likely different snapshots of failures." + "reason": "These are sequential triage reports, not one underlying defect report." }, { "left": "issue:46969", "right": "issue:47037", "accept": false, - "reason": "Same topic and template, but not the same concrete issue instance." + "reason": "Different dated Serge integration triage items; title overlap is too generic to merge." }, { "left": "issue:46969", "right": "issue:47091", "accept": false, - "reason": "These appear to be daily triage logs rather than duplicate issue reports." + "reason": "Reject: same recurring triage template, but no evidence of identical failure context." }, { "left": "issue:46985", "right": "issue:46998", "accept": false, - "reason": "Different day, different triage run; reject duplicate merge." + "reason": "Different daily triage entries; not enough signal that they track the same bug." }, { "left": "issue:46985", "right": "issue:47037", "accept": false, - "reason": "Recurring triage format only; not a shared underlying bug." + "reason": "Same subsystem and template, but these look like separate daily failures or investigations." }, { "left": "issue:46985", "right": "issue:47049", "accept": false, - "reason": "These are separate daily integration triage issues." + "reason": "Reject: dated triage reports for different runs, not a single issue." }, { "left": "issue:46985", "right": "issue:47091", "accept": false, - "reason": "Similarity is due to templated triage text, not the same failure report." + "reason": "No concrete matching failure details; these are distinct triage artifacts." }, { "left": "issue:46985", "right": "issue:47102", "accept": false, - "reason": "Different dated triage entries; do not merge as duplicates." + "reason": "These are separate date-based triage reports, not the same underlying bug." }, { "left": "issue:46998", "right": "issue:47037", "accept": false, - "reason": "Distinct daily reports with shared theme but not the same issue." + "reason": "Generic title similarity only; the issues are from different dates and likely different incidents." }, { "left": "issue:46998", "right": "issue:47049", "accept": false, - "reason": "Separate day-specific triage issue, not a duplicate." + "reason": "Reject: recurring integration triage items, but no evidence they describe the same code-path problem." }, { "left": "issue:46998", "right": "issue:47053", "accept": false, - "reason": "Same series of reports, but each date is its own triage instance." + "reason": "Different daily triage reports; title match is too broad to call them duplicates." }, { "left": "issue:46998", "right": "issue:47091", "accept": false, - "reason": "These are not the same underlying bug; they are daily integration triage notes." + "reason": "These appear to be separate daily Serge triage issues, not one bug." }, { "left": "issue:47037", "right": "issue:47049", "accept": false, - "reason": "Both are daily triage reports; the date difference indicates distinct instances." + "reason": "Same series, but each is a dated triage entry; no shared concrete defect is shown." }, { "left": "issue:47037", "right": "issue:47053", "accept": false, - "reason": "Same recurring triage template, but separate day-scoped issues." + "reason": "Reject: similar titles, but they are distinct daily reports rather than the same failure." }, { "left": "issue:47037", "right": "issue:47091", "accept": false, - "reason": "Not duplicates: likely different failure snapshots from different days." + "reason": "No evidence of the same underlying bug; these are separate triage instances." }, { "left": "issue:47037", "right": "issue:47102", "accept": false, - "reason": "Different dated triage issue, so not the same underlying bug report." + "reason": "Different dated triage issues in a recurring series, not duplicates." }, { "left": "issue:47049", "right": "issue:47053", "accept": false, - "reason": "These are consecutive daily triage issues, not one merged bug." + "reason": "These are adjacent daily triage reports, but there is no concrete sign they refer to the same incident." }, { "left": "issue:47049", "right": "issue:47091", "accept": false, - "reason": "Same broad subsystem and template, but distinct daily reports." + "reason": "Reject: recurring Serge integration triage entries, not the same underlying bug report." }, { "left": "issue:47049", "right": "issue:47102", "accept": false, - "reason": "Separate date-specific triage entries; reject duplicate merge." + "reason": "Different dates and no shared failure details; treat as separate issues." }, { "left": "issue:47053", "right": "issue:47091", "accept": false, - "reason": "Different daily integration-failure triage reports." + "reason": "Same template and subsystem, but not enough evidence to merge them." }, { "left": "issue:47053", "right": "issue:47102", "accept": false, - "reason": "These are recurring triage logs, not one concrete duplicated issue." + "reason": "Reject: distinct daily triage reports with no demonstrated common root cause." }, { "left": "issue:47091", "right": "issue:47102", "accept": false, - "reason": "Same templated triage pattern, but different dates and likely different underlying failures." + "reason": "These look like separate day-by-day triage entries, not the same bug or change." } ] }, @@ -22462,7 +22442,7 @@ "soft_pairs": [ "issue:46983|issue:46984" ], - "prepared_review_unit_hash": "a4d287fc800fcd24690215376364663ec768cc6a13aad4273d5682ee6d19c4c1", + "prepared_review_unit_hash": "33356f13194da80e612623ed6b5c6abff5242172fa310bd1426c01462cf58b3f", "component_budget": { "node_count": 2, "item_count": 2, @@ -22484,24 +22464,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both items are closed issues with the same title, same apparent feature request, and very high text similarity, so they are duplicates about adding RWKV7 model support.", + "summary": "The two issues are clear duplicates: both request the same RWKV7 model support. Keep the more active report as the canonical issue.", "confidence": 0.98, - "canonical_issue_reason": "issue #46984 is the better canonical issue because it has the same request plus more discussion activity, suggesting it is the more active tracking thread.", + "canonical_issue_reason": "Issue #46984 is the better canonical choice because it has more discussion activity, suggesting it is the more developed or representative report.", "canonical_pr_reason": null, - "best_issue_reason": "issue #46984 is the strongest representative of the cluster due to its matching title and higher discussion activity.", + "best_issue_reason": "Issue #46984 is the best single issue to keep because it appears to have slightly more engagement while describing the same request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46983", "right": "issue:46984", "accept": true, - "reason": "Same title and highly similar issue text; both ask for RWKV7 model support, so they represent the same underlying request." + "reason": "Same title and same requested feature (RWKV7 model support); they describe the same underlying issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: both items are closed issues with the same title, the pair has strong soft-similarity evidence, and #46984 does have more discussion activity, making it a reasonable canonical choice. The soft-edge acceptance is conservative enough here." + "feedback": "The summary is grounded in the packet: both issues have the same title, and the only activity difference favors #46984 (discussion_activity 2 vs 0). The duplicate verdict is conservative enough given the strong similarity signal." }, "evaluator_used": true, "retried": false, @@ -22528,7 +22508,7 @@ "soft_pairs": [ "issue:47007|issue:47015" ], - "prepared_review_unit_hash": "88063b17880f543ac68058fbbd5682470551c946afd124001b350d33c5c53b4f", + "prepared_review_unit_hash": "66075f18a85e4141256cef6ee0d9f7a96b04712a6edbb528a8a7ae0d748c99fa", "component_budget": { "node_count": 2, "item_count": 2, @@ -22550,18 +22530,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are thematically similar but do not look like the same underlying request or bug; they read more like separate article/topic submissions about AI-driven vulnerability discovery.", - "confidence": 0.89, - "canonical_issue_reason": null, + "summary": "The two closed issues are thematically similar (both about AI/LLM-driven vulnerability discovery), but they do not read like the same concrete bug or actionable request. They appear to be separate discussion/proposal posts rather than duplicates of one underlying problem.", + "confidence": 0.86, + "canonical_issue_reason": "No single issue clearly subsumes the other; both are broad, overlapping topics but not the same specific report or request.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 47007 is the slightly better representative of the cluster because its title is more specific and descriptive, but it is not a true duplicate of 47015.", + "best_issue_reason": "Issue #47015 is the more directly framed concept/topic and is the better representative if one item must be picked for the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:47007", "right": "issue:47015", "accept": false, - "reason": "Similar broad cybersecurity/AI theme, but the titles frame different angles (LLM lessons vs transformer-based application) and there is no evidence they target the same underlying issue." + "reason": "Shared high-level subject matter only. The titles suggest different angles ('lessons from using advanced LLMs' vs 'novel application of transformers'), not the same underlying issue or change." } ] }, @@ -22615,12 +22595,12 @@ "issue:47265|issue:47309", "issue:47279|issue:47309" ], - "prepared_review_unit_hash": "8b0b5f50816cfbb3196f5d962a05c81b06e541978d4d70511349adb812d7d4f5", + "prepared_review_unit_hash": "273a5e2ba9963daf52e8be996b5b851c69f83b927130500d6e000265bbe42fd0", "component_budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 20, - "serialized_chars": 11159, + "serialized_chars": 11160, "estimated_input_tokens": 2790, "estimated_eval_tokens": 5836 }, @@ -22628,7 +22608,7 @@ "node_count": 7, "item_count": 7, "soft_pair_count": 20, - "serialized_chars": 11159, + "serialized_chars": 11160, "estimated_input_tokens": 2790, "estimated_eval_tokens": 5836 }, @@ -22637,132 +22617,132 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are a run of daily \u001b[1m[serge] integration failure triage\u001b[0m issues with highly similar templates, but they appear to be separate day-by-day triage artifacts rather than duplicates of one underlying bug.", - "confidence": 0.91, - "canonical_issue_reason": "Issue 47309 is the only open item and the latest triage entry, so it is the most useful representative for the cluster.", + "summary": "These are highly similar daily serge integration-triage issues, but the date-stamped titles indicate separate tracking tickets rather than duplicates of one underlying bug. No PRs are present.", + "confidence": 0.87, + "canonical_issue_reason": "issue:47279 is the strongest anchor in the series: it is closed, has the highest inbound references, and looks like the most central instance of the recurring triage thread. The date-stamped issues still appear to be separate daily tracker tickets, so this is only a loose representative choice.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 47309 is the best cluster representative because it is current, open, and likely the active place for ongoing triage discussion.", + "best_issue_reason": "issue:47279 is the best representative issue in the cluster because it is the most referenced and central of the closed daily triage tickets. It is not a true duplicate target, but it is the most suitable anchor if one must be chosen.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:47120", "right": "issue:47182", "accept": false, - "reason": "Same triage series, but different dates and no evidence of the same underlying failure." + "reason": "Same template and subsystem, but different dated triage tickets; these read like separate daily tracking issues rather than one bug." }, { "left": "issue:47120", "right": "issue:47309", "accept": false, - "reason": "Daily triage reports only; this is a later, distinct issue rather than the same bug." + "reason": "Date-stamped triage entries for different days; no evidence they describe the same underlying failure." }, { "left": "issue:47120", "right": "issue:47256", "accept": false, - "reason": "Looks like another day\u2019s integration-failure triage, not a duplicate of the same incident." + "reason": "Looks like another daily integration triage ticket, not a duplicate of the 2026-07-07 tracker." }, { "left": "issue:47120", "right": "issue:47279", "accept": false, - "reason": "Highly similar template, but it is a separate dated triage issue." + "reason": "Same recurring triage series, but distinct day-specific issue; not the same underlying bug report." }, { "left": "issue:47120", "right": "issue:47265", "accept": false, - "reason": "No shared concrete bug/change; only the same triage wording." + "reason": "Different daily tracker entry in the same series, with no concrete shared bug description shown." }, { "left": "issue:47182", "right": "issue:47309", "accept": false, - "reason": "Consecutive triage entries for different dates, so not the same underlying problem." + "reason": "Both are similarly titled triage tickets, but the date separation suggests independent daily reports." }, { "left": "issue:47182", "right": "issue:47256", "accept": false, - "reason": "Same series, but each issue tracks a different day\u2019s failures." + "reason": "Same broad integration-triage pattern, but not enough to treat as one duplicate issue." }, { "left": "issue:47182", "right": "issue:47265", "accept": false, - "reason": "Similar title/body, but no indication they refer to one bug." + "reason": "These are separate dated triage issues rather than one concrete bug ticket." }, { "left": "issue:47182", "right": "issue:47279", "accept": false, - "reason": "Different triage dates; similarity is templated, not substantive." + "reason": "Recurring tracker entry, but each date appears to be its own triage issue." }, { "left": "issue:47227", "right": "issue:47256", "accept": false, - "reason": "These are separate daily integration triage reports, not mergeable duplicates." + "reason": "Date-specific triage tickets; same title pattern does not imply the same underlying failure." }, { "left": "issue:47227", "right": "issue:47309", "accept": false, - "reason": "No shared code-path defect is evidenced beyond the recurring triage format." + "reason": "Different daily integration-failure triage reports, not a single duplicate issue." }, { "left": "issue:47227", "right": "issue:47279", "accept": false, - "reason": "Same broad tracker type, but distinct day-specific issue content." + "reason": "Same recurring series, but no shared bug details indicating a mergeable duplicate." }, { "left": "issue:47182", "right": "issue:47227", "accept": false, - "reason": "Adjacent dated reports, but not the same underlying failure." + "reason": "These appear to be successive daily triage logs, not the same issue." }, { "left": "issue:47227", "right": "issue:47265", "accept": false, - "reason": "Looks like a new daily triage instance rather than a duplicate." + "reason": "No concrete evidence of the same bug; just the same triage template on different dates." }, { "left": "issue:47256", "right": "issue:47309", "accept": false, - "reason": "Different dates and no concrete evidence of one shared bug." + "reason": "Different dated tracker issues in the same recurring integration-failure triage series." }, { "left": "issue:47256", "right": "issue:47265", "accept": false, - "reason": "Template match only; these are separate triage issues." + "reason": "Separate daily triage ticket, not the same underlying bug report." }, { "left": "issue:47256", "right": "issue:47279", "accept": false, - "reason": "Same integration-failure triage series, but not the same incident." + "reason": "Same pattern and subsystem, but different day-specific triage issue." }, { "left": "issue:47265", "right": "issue:47279", "accept": false, - "reason": "Different day\u2019s triage report, so they should stay separate." + "reason": "These are recurring triage logs; no sign they should be merged as duplicates." }, { "left": "issue:47265", "right": "issue:47309", "accept": false, - "reason": "No evidence they target the same concrete bug or fix." + "reason": "Different day-stamped integration triage issues, not one shared bug." }, { "left": "issue:47279", "right": "issue:47309", "accept": false, - "reason": "They look like sequential triage records, not duplicates of one issue." + "reason": "Same series, but the date and tracker context indicate separate issues rather than duplicates." } ] }, @@ -22792,12 +22772,12 @@ "soft_pairs": [ "issue:47138|issue:47139" ], - "prepared_review_unit_hash": "f813f3d87957c476f7b9e4b0ffabaf7def714e02b2f7693cc82c5f34bfe8b7ca", + "prepared_review_unit_hash": "949d2ba5cf49dfeb5510fd59366e439118e5e9f9e8db89f27c1a02c70c1df745", "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 }, @@ -22805,7 +22785,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1591, + "serialized_chars": 1590, "estimated_input_tokens": 398, "estimated_eval_tokens": 1052 }, @@ -22814,24 +22794,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two open issues are very likely duplicates based on their identical titles, identical body lengths, and near-simultaneous creation times. The packet does not include the issue bodies, so it cannot confirm that the text is identical, but the available metadata strongly points to the same bug report.", - "confidence": 0.86, - "canonical_issue_reason": "Issue #47138 is the better canonical issue because it is the earlier of two issues with the same title and matching body length, created only seconds apart.", + "summary": "#47138 and #47139 appear to be duplicate issue reports: they have identical titles and the similarity signal is maximal, but the issue bodies are not visible here, so the match should be treated as likely rather than certain. #47138 is the best canonical issue because it was created first and has slightly more inbound references.", + "confidence": 0.92, + "canonical_issue_reason": "#47138 is the earliest-created report and has more inbound references, so it is the best canonical issue to keep.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #47138 is the best global issue representative for the same reasons: it is the earliest of two near-identical issue reports and has no conflicting metadata.", + "best_issue_reason": "#47138 is the strongest single issue to track this bug because it appears to duplicate #47139 while having slightly more linkage/support.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:47138", "right": "issue:47139", "accept": true, - "reason": "Accept as a duplicate pair: the issues have identical titles, identical body lengths, and were created seconds apart, which is strong enough to indicate the same underlying report even though the body text itself is not provided." + "reason": "Identical titles and a 1.0 similarity signal make this look like a duplicate issue report, though the bodies are not visible here so this is best stated as an apparent duplicate." } ] }, "evaluator_result": { "accept": false, - "feedback": "The pair match is plausible, but the summary overstates what the packet proves. The packet supports identical titles, identical body lengths, near-simultaneous creation, and a soft-similarity score of 1.0; it does not show the bodies are the same or that they share inbound references. Please make the wording more conservative and grounded in the available fields." + "feedback": "The packet supports a strong similarity/likely-duplicate assessment, but the analyst states it too definitively. We only know the titles are identical and the similarity score is 1.0; we do not see the bodies, so phrases like \"same content,\" \"clear duplicate,\" and \"same underlying bug\" overstate the evidence. The canonical-choice rationale is grounded, but the soft-edge verdict and summary should be phrased more conservatively (e.g. \"appears to be a duplicate\")." }, "evaluator_used": true, "retried": true, @@ -22850,15 +22830,13 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-75", + "cluster_id": "cluster-40546-76", "nodes": [ "pull_request:43251", "pull_request:43254", - "pull_request:44517", "pull_request:44775", "pull_request:46196", "pull_request:46266", - "pull_request:46267", "pull_request:46376", "pull_request:46395", "pull_request:46836", @@ -22899,7 +22877,8 @@ "pull_request:47301", "pull_request:47302", "pull_request:47304", - "pull_request:47318" + "pull_request:47318", + "pull_request:47323" ], "soft_pairs": [ "pull_request:43251|pull_request:43254", @@ -22914,309 +22893,331 @@ "pull_request:47108|pull_request:47202", "pull_request:47204|pull_request:47243", "pull_request:47077|pull_request:47078", - "pull_request:47077|pull_request:47212", "pull_request:46266|pull_request:46957", + "pull_request:47077|pull_request:47212", "pull_request:47173|pull_request:47184", "pull_request:46376|pull_request:47202", "pull_request:47166|pull_request:47173", "pull_request:47077|pull_request:47275", - "pull_request:44517|pull_request:47082", "pull_request:44775|pull_request:47275", "pull_request:46990|pull_request:47253", + "pull_request:46995|pull_request:47323", "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:47163|pull_request:47284", - "pull_request:47003|pull_request:47204", "pull_request:46395|pull_request:47181", + "pull_request:47003|pull_request:47204", "pull_request:47275|pull_request:47302", "pull_request:47093|pull_request:47275", "pull_request:46196|pull_request:47210", "pull_request:46995|pull_request:47163", + "pull_request:46912|pull_request:47082", "pull_request:47163|pull_request:47173", "pull_request:44775|pull_request:47301", - "pull_request:46196|pull_request:46995", "pull_request:47204|pull_request:47295", + "pull_request:46196|pull_request:46995", "pull_request:47302|pull_request:47304", "pull_request:47077|pull_request:47301", "pull_request:44775|pull_request:47077", "pull_request:46376|pull_request:46995", - "pull_request:47069|pull_request:47318", - "pull_request:46995|pull_request:47181", - "pull_request:46267|pull_request:47181", - "pull_request:47093|pull_request:47181", - "pull_request:44517|pull_request:46376", - "pull_request:46836|pull_request:47275", - "pull_request:47166|pull_request:47292", - "pull_request:46266|pull_request:47301", - "pull_request:44775|pull_request:47304", - "pull_request:44775|pull_request:47181" + "pull_request:47069|pull_request:47318" ], - "prepared_review_unit_hash": "202c1330c9626068c64b6fd150c88f86ded1c3a3d826077784eaea22ccc1a9fc", + "prepared_review_unit_hash": "0d824e590206cc512162542d6364f2ad1d0a95da6beeec13951e0091ba1a9f44", "component_budget": { - "node_count": 75, - "item_count": 75, - "soft_pair_count": 217, - "serialized_chars": 263464, - "estimated_input_tokens": 65866, - "estimated_eval_tokens": 131988 + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 52, - "serialized_chars": 128339, - "estimated_input_tokens": 32085, - "estimated_eval_tokens": 64426 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 44, + "serialized_chars": 122076, + "estimated_input_tokens": 30519, + "estimated_eval_tokens": 61294 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous cluster of unrelated model/docs/test/infra PRs, with only a few true duplicate pairs: the fixed_cross_entropy kwargs fix and the CohereASR double-shift loss fixes. The rest are broad-subsystem or same-file-but-different-change overlaps and should stay separate.", - "confidence": 0.74, + "summary": "Most soft edges are false positives across unrelated docs, model additions, tests, or adjacent subsystems. The only plausible duplicate-like pair is the CohereASR loss double-shift follow-up, where 47090 is the best representative.", + "confidence": 0.84, "canonical_issue_reason": null, - "canonical_pr_reason": "47090 is the best canonical PR: it fixes the concrete CohereASR loss double-shift bug on the exact affected forward path, keeps the shared loss helper semantics, and is the most focused representative among the true duplicates.", + "canonical_pr_reason": "No issue artifacts here; among the PRs, 47090 is the cleanest representative of the CohereASR loss-double-shift fix because it is narrowly scoped and preserves the shared loss helper path.", "best_issue_reason": null, - "best_pr_reason": "47090 is the cleanest representative of the actual duplicate cluster and the easiest PR to treat as the canonical fix.", + "best_pr_reason": "47090 is the strongest standalone PR: it targets the exact CohereASR training-loss bug, keeps trainer-compatible loss handling, and is narrower than the broader predecessor PRs.", "soft_edge_verdicts": [ { "left": "pull_request:43251", "right": "pull_request:43254", - "accept": true, - "reason": "Same fixed_cross_entropy kwargs expansion in the same file; both add support for passing cross_entropy options through." + "accept": false, + "reason": "Same target issue, but the patches are just two overlapping implementations of the same helper change; not a clear duplicate PR pair to merge." }, { "left": "pull_request:46903", "right": "pull_request:47090", "accept": true, - "reason": "Both fix the same CohereASR training-loss double-shift in the forward loss path; 46903 is just broader because it also touches PPFormulaNet." + "reason": "Same CohereASR loss double-shift bug in the same forward path; 47090 is a narrower follow-up of the same fix and could be consolidated into one PR." }, { "left": "pull_request:46903", "right": "pull_request:46958", - "accept": true, - "reason": "Same underlying CohereASR loss double-shift bug and same affected code path, just implemented with different loss wiring." + "accept": false, + "reason": "Both touch CohereASR loss handling, but the fixes diverge materially and 46903 bundles an extra PPFormulaNet change, so this is not a clean duplicate pair." }, { "left": "pull_request:46958", "right": "pull_request:46974", "accept": false, - "reason": "Different problems: CohereASR training loss vs a MusicGen generate typo." + "reason": "Unrelated models and bugs: CohereASR loss logic versus a MusicGen generate typo." }, { "left": "pull_request:47268", "right": "pull_request:47284", "accept": false, - "reason": "Unrelated test fixes for different models and code paths." + "reason": "Different models and test failures; only superficial test-edit overlap." }, { "left": "pull_request:46954", "right": "pull_request:46958", "accept": false, - "reason": "Pytest fp32-module behavior change is unrelated to CohereASR loss shifting." + "reason": "Pytest-v9 module attribute fix versus CohereASR loss bug; unrelated code paths." }, { "left": "pull_request:47173", "right": "pull_request:47284", "accept": false, - "reason": "Different test suites and unrelated XPU/model fixes." + "reason": "Both are test-only changes, but they address different models and failures." }, { "left": "pull_request:47202", "right": "pull_request:47204", "accept": false, - "reason": "Both mention build/container tooling, but one removes a workflow and the other adds exporter deps to the dockerfile; not the same change." + "reason": "Both touch docker/build tooling, but one removes a CI image builder while the other adds exporter deps; not the same change." }, { "left": "pull_request:47284", "right": "pull_request:47292", "accept": false, - "reason": "Different model test fixes for different architectures." + "reason": "Different model test updates for different hardware expectations; not the same bug." }, { "left": "pull_request:47108", "right": "pull_request:47202", "accept": false, - "reason": "Doc-builder comment cleanup is unrelated to dropping the docker image builder workflow." + "reason": "Doc-builder pin comment cleanup versus docker-image builder removal; unrelated." }, { "left": "pull_request:47204", "right": "pull_request:47243", "accept": false, - "reason": "One changes exporter test/runtime dependencies, the other fixes Executorch exporter internals; distinct changes." + "reason": "Exporter dependency packaging versus Executorch exporter fixes; adjacent area but distinct changes." }, { "left": "pull_request:47077", "right": "pull_request:47078", "accept": false, - "reason": "Both edit Japanese docs, but one adds DistilBERT translation while the other only tweaks wording; not the same underlying change." + "reason": "Japanese docs additions/translation versus wording cleanup; same locale, but not the same underlying change." }, { - "left": "pull_request:47077", - "right": "pull_request:47212", + "left": "pull_request:46266", + "right": "pull_request:46957", "accept": false, - "reason": "Japanese docs translation vs kv-cache documentation update; unrelated." + "reason": "Model-conversion mappings for dinos versus finegrained FP8 decode/kernel optimization; unrelated." }, { - "left": "pull_request:46266", - "right": "pull_request:46957", + "left": "pull_request:47077", + "right": "pull_request:47212", "accept": false, - "reason": "Dino model conversion/modernization and finegrained FP8 decode changes are unrelated." + "reason": "Japanese docs translation versus KV-cache docs update; no shared bug or feature." }, { "left": "pull_request:47173", "right": "pull_request:47184", "accept": false, - "reason": "Different test adjustments for different model families and accelerator coverage." + "reason": "Separate test maintenance PRs for different models and accelerators." }, { "left": "pull_request:46376", "right": "pull_request:47202", "accept": false, - "reason": "Continuous batching multimodal support is unrelated to docker image builder removal." + "reason": "Continuous batching multimodal support versus docker builder removal; different subsystems and goals." }, { "left": "pull_request:47166", "right": "pull_request:47173", "accept": false, - "reason": "Quantization test updates and XPU expectation updates are separate test-only changes." + "reason": "Quantization test extensions versus XPU expectation fixes; not the same issue." }, { "left": "pull_request:47077", "right": "pull_request:47275", "accept": false, - "reason": "Japanese docs translation and adding EfficientViTSam are unrelated." - }, - { - "left": "pull_request:44517", - "right": "pull_request:47082", - "accept": false, - "reason": "Different model additions with only generic auto-mapping overlap." + "reason": "Japanese docs translation versus new model support; unrelated." }, { "left": "pull_request:44775", "right": "pull_request:47275", "accept": false, - "reason": "Docs on distributed training are unrelated to a new model architecture PR." + "reason": "Docs-only distributed training update versus new model support; not a duplicate change." }, { "left": "pull_request:46990", "right": "pull_request:47253", "accept": false, - "reason": "FSDP orchestration and tensor-parallel lm_head handling are different distributed-training fixes." + "reason": "Both are distributed-training related, but one is FSDP orchestration/loading and the other is tensor-parallel lm_head validation." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47323", + "accept": false, + "reason": "Different kernel availability and deepgemm multi-device fixes; shared accelerator context only." }, { "left": "pull_request:47275", "right": "pull_request:47301", "accept": false, - "reason": "New model support vs assisted-decoding docs; not the same change." + "reason": "New model support versus assisted-decoding docs for MTP; unrelated." }, { "left": "pull_request:47216", "right": "pull_request:47275", "accept": false, - "reason": "Compressed-tensors quantization changes are unrelated to adding EfficientViTSam." + "reason": "Compressed-tensors FP8 versus EfficientVit-SAM model support; different features." }, { "left": "pull_request:46912", "right": "pull_request:47179", "accept": false, - "reason": "RoPE test coverage and Granite model support are unrelated." + "reason": "Different model families and unrelated rotary/test changes." }, { "left": "pull_request:46196", "right": "pull_request:47202", "accept": false, - "reason": "SDPA dispatch support and docker image cleanup are unrelated." + "reason": "SDPA dispatch support versus CI docker builder removal; different layers of the project." }, { - "left": "pull_request:46836", - "right": "pull_request:47275", + "left": "pull_request:47163", + "right": "pull_request:47284", "accept": false, - "reason": "Unlimited OCR model support and EfficientViTSam model support are different model additions." + "reason": "Paged-attention XPU test coverage versus Mimo v2 flash model fixes; not the same underlying bug." }, { - "left": "pull_request:47163", + "left": "pull_request:46912", "right": "pull_request:47284", "accept": false, - "reason": "Paged-attention XPU tests and Mimo v2 flash changes are unrelated." + "reason": "RoPE tests versus Mimo model fixes; unrelated." }, { - "left": "pull_request:47003", - "right": "pull_request:47204", + "left": "pull_request:46836", + "right": "pull_request:47202", "accept": false, - "reason": "OpenVINO exporter support and dockerfile dependency changes are different layers of the export stack." + "reason": "Unlimited OCR model addition versus docker builder removal; no overlap in bug/change." }, { "left": "pull_request:46395", "right": "pull_request:47181", "accept": false, - "reason": "Param2MoE and Cosmos3 Edge are separate model additions with no shared bug/change." + "reason": "Two separate new-model PRs for different architectures; same broad pattern, but not a shared change." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47204", + "accept": false, + "reason": "OpenVINO exporter feature versus exporter dependencies in torch-light; not the same code-path problem." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47302", + "accept": false, + "reason": "Model-addition PR versus tokenizer doc formatting cleanup; unrelated." }, { "left": "pull_request:47093", "right": "pull_request:47275", "accept": false, - "reason": "OpenPangu-V2 and EfficientViTSam are distinct model integrations." + "reason": "OpenPangu-V2 model support versus EfficientVit-SAM model support; different model integrations." }, { "left": "pull_request:46196", "right": "pull_request:47210", "accept": false, - "reason": "Both are attention-related, but one is SDPA dispatch support and the other is packed-row varlen routing; different fixes." + "reason": "Both concern attention optimizations, but one is sdpa dispatch for models and the other is packed-row varlen routing; distinct bugs." }, { "left": "pull_request:46995", "right": "pull_request:47163", "accept": false, - "reason": "MLU availability detection and XPU paged-attention tests are unrelated." + "reason": "Flash-linear-attention availability on MLU versus paged-attention XPU tests; different concerns." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47082", + "accept": false, + "reason": "RoPE test improvements versus RoMa model support; unrelated." }, { "left": "pull_request:47163", "right": "pull_request:47173", "accept": false, - "reason": "Different accelerator-specific test fixes for different modules." + "reason": "Paged-attention XPU coverage versus XPU expectation updates for other model tests; not the same PR change." }, { "left": "pull_request:44775", "right": "pull_request:47301", "accept": false, - "reason": "Distributed-training docs and MTP assisted-decoding docs are unrelated." + "reason": "Docs for distributed training versus assisted-decoding docs for MTP; unrelated." }, { - "left": "pull_request:46196", - "right": "pull_request:46995", + "left": "pull_request:47204", + "right": "pull_request:47295", "accept": false, - "reason": "Different attention backend enablement changes with different code paths and hardware targets." + "reason": "Exporter deps in torch-light versus DiffusionGemma mask casting; no shared bug." }, { - "left": "pull_request:47204", - "right": "pull_request:47295", + "left": "pull_request:46196", + "right": "pull_request:46995", "accept": false, - "reason": "Executorch/ONNX dependency setup and DiffusionGemma mask casting are unrelated." + "reason": "Different attention/kernel features: sdpa dispatch versus flash-linear-attention availability." }, { "left": "pull_request:47302", "right": "pull_request:47304", "accept": false, - "reason": "Tokenizers fallback docs and MPS graph-cache docs are unrelated documentation changes." + "reason": "Tokenizer backend docs versus MPS graph-cache docs; unrelated documentation changes." }, { "left": "pull_request:47077", "right": "pull_request:47301", "accept": false, - "reason": "Japanese doc edits and MTP docs are unrelated." + "reason": "Japanese docs translation versus MTP docs; different documentation topics." }, { "left": "pull_request:44775", "right": "pull_request:47077", "accept": false, - "reason": "Distributed-training docs and Japanese DistilBERT docs are unrelated." + "reason": "Distributed-training docs versus Japanese DistilBERT translation; no overlap." + }, + { + "left": "pull_request:46376", + "right": "pull_request:46995", + "accept": false, + "reason": "Continuous batching multimodal support versus flash-linear-attention availability; different features." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47318", + "accept": false, + "reason": "Config property typing versus flash-attn fallback test gating; unrelated changes." } ] }, @@ -23238,15 +23239,16 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-75", + "cluster_id": "cluster-40546-76", "nodes": [ "pull_request:43451", "pull_request:44517", + "pull_request:44775", "pull_request:46196", "pull_request:46266", + "pull_request:46267", "pull_request:46376", "pull_request:46395", - "pull_request:46658", "pull_request:46836", "pull_request:46895", "pull_request:46912", @@ -23262,11 +23264,11 @@ "pull_request:47113", "pull_request:47143", "pull_request:47163", + "pull_request:47166", "pull_request:47173", "pull_request:47175", "pull_request:47179", "pull_request:47181", - "pull_request:47198", "pull_request:47204", "pull_request:47210", "pull_request:47216", @@ -23275,6 +23277,7 @@ "pull_request:47253", "pull_request:47275", "pull_request:47291", + "pull_request:47292", "pull_request:47294", "pull_request:47298", "pull_request:47301", @@ -23282,54 +23285,65 @@ "pull_request:47304", "pull_request:47306", "pull_request:47311", - "pull_request:47318" + "pull_request:47318", + "pull_request:47320", + "pull_request:47322" ], "soft_pairs": [ - "pull_request:46995|pull_request:47143", + "pull_request:47294|pull_request:47320", + "pull_request:46995|pull_request:47181", + "pull_request:46267|pull_request:47181", + "pull_request:47093|pull_request:47181", + "pull_request:44517|pull_request:46376", + "pull_request:47166|pull_request:47292", + "pull_request:46836|pull_request:47275", + "pull_request:46266|pull_request:47301", + "pull_request:44775|pull_request:47304", + "pull_request:44775|pull_request:47181", "pull_request:47093|pull_request:47143", - "pull_request:47082|pull_request:47108", + "pull_request:47082|pull_request:47210", "pull_request:47093|pull_request:47163", "pull_request:46266|pull_request:47216", "pull_request:47108|pull_request:47301", + "pull_request:47082|pull_request:47108", "pull_request:46376|pull_request:47291", + "pull_request:47082|pull_request:47275", "pull_request:47179|pull_request:47210", "pull_request:46196|pull_request:47179", - "pull_request:47078|pull_request:47275", "pull_request:46912|pull_request:46990", "pull_request:46990|pull_request:47181", + "pull_request:47078|pull_request:47275", "pull_request:46395|pull_request:47311", "pull_request:47311|pull_request:47318", "pull_request:47175|pull_request:47179", "pull_request:47210|pull_request:47302", - "pull_request:47198|pull_request:47318", "pull_request:47077|pull_request:47302", - "pull_request:47082|pull_request:47210", + "pull_request:47082|pull_request:47216", "pull_request:47163|pull_request:47179", "pull_request:46990|pull_request:47306", "pull_request:47108|pull_request:47302", - "pull_request:47082|pull_request:47301", "pull_request:46376|pull_request:47318", - "pull_request:47210|pull_request:47304", "pull_request:46836|pull_request:47218", + "pull_request:47210|pull_request:47304", "pull_request:46376|pull_request:47218", "pull_request:46957|pull_request:47216", "pull_request:47173|pull_request:47311", - "pull_request:46912|pull_request:47082", "pull_request:47108|pull_request:47216", "pull_request:46912|pull_request:47078", "pull_request:43451|pull_request:46990", "pull_request:47179|pull_request:47301", "pull_request:47113|pull_request:47204", "pull_request:47143|pull_request:47318", - "pull_request:47218|pull_request:47304", + "pull_request:47306|pull_request:47318", "pull_request:47218|pull_request:47311", - "pull_request:47163|pull_request:47218", + "pull_request:47218|pull_request:47304", + "pull_request:47166|pull_request:47218", "pull_request:47216|pull_request:47302", "pull_request:46266|pull_request:46376", "pull_request:47243|pull_request:47253", "pull_request:46912|pull_request:47294", + "pull_request:47143|pull_request:47320", "pull_request:47175|pull_request:47294", - "pull_request:47198|pull_request:47253", "pull_request:46895|pull_request:47306", "pull_request:46196|pull_request:46376", "pull_request:46196|pull_request:47301", @@ -23337,481 +23351,436 @@ "pull_request:47069|pull_request:47113", "pull_request:47275|pull_request:47304", "pull_request:44517|pull_request:47311", + "pull_request:47078|pull_request:47322", "pull_request:44517|pull_request:47301", "pull_request:44517|pull_request:47181", - "pull_request:46376|pull_request:47198", "pull_request:47078|pull_request:47243", - "pull_request:47069|pull_request:47082", "pull_request:47113|pull_request:47304", "pull_request:43451|pull_request:46266", + "pull_request:43451|pull_request:44517", "pull_request:46912|pull_request:47291", "pull_request:47069|pull_request:47302", "pull_request:47243|pull_request:47294", - "pull_request:43451|pull_request:47078", - "pull_request:47210|pull_request:47291", - "pull_request:46957|pull_request:47301", - "pull_request:46895|pull_request:47298", - "pull_request:47294|pull_request:47306", - "pull_request:47113|pull_request:47175", - "pull_request:47113|pull_request:47243", - "pull_request:46658|pull_request:47302", - "pull_request:47113|pull_request:47275", - "pull_request:46658|pull_request:47275" + "pull_request:46895|pull_request:47298" ], - "prepared_review_unit_hash": "0e0cbb5c84e7ab40b5cdef50a01b44cd014711a2d3281bec5ec3126478c6f150", + "prepared_review_unit_hash": "4499e173318c595d028d3f5a62260fe9b7e76e32092315b30b25681bb2ccff08", "component_budget": { - "node_count": 75, - "item_count": 75, - "soft_pair_count": 217, - "serialized_chars": 263464, - "estimated_input_tokens": 65866, - "estimated_eval_tokens": 131988 + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 }, "budget": { - "node_count": 43, - "item_count": 43, + "node_count": 47, + "item_count": 47, "soft_pair_count": 72, - "serialized_chars": 128828, - "estimated_input_tokens": 32207, - "estimated_eval_tokens": 64670 + "serialized_chars": 137274, + "estimated_input_tokens": 34319, + "estimated_eval_tokens": 68894 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No duplicate cluster emerges here: the PRs span unrelated model additions, doc-only changes, test expectation updates, generation/runtime fixes, and infrastructure tweaks. I would not merge any of the soft pairs as duplicates.", - "confidence": 0.97, + "summary": "This cluster is a loose bag of unrelated pull requests: model additions, documentation updates, test expectation tweaks, and infrastructure fixes. None of the soft pairs look like the same underlying change, so all should be rejected as duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "No single PR is a safe cluster representative because the items span unrelated code paths and topics; there is no coherent duplicate set here.", "soft_edge_verdicts": [ { - "left": "pull_request:46995", - "right": "pull_request:47143", + "left": "pull_request:47294", + "right": "pull_request:47320", "accept": false, - "reason": "Unrelated low-level import/availability tweak vs adding a new model; different code paths and goals." + "reason": "Causal-conv1d kernel helper work vs response-schema deprecation; different subsystems and goals." }, { - "left": "pull_request:47093", - "right": "pull_request:47143", + "left": "pull_request:46995", + "right": "pull_request:47181", "accept": false, - "reason": "Different model support additions; no shared bug or implementation path." + "reason": "Flash-linear-attention availability change vs new Cosmos3 Edge model support; unrelated." }, { - "left": "pull_request:47082", - "right": "pull_request:47108", + "left": "pull_request:46267", + "right": "pull_request:47181", "accept": false, - "reason": "RoMa model support and a doc-builder comment cleanup are unrelated changes." + "reason": "Serve chat parsing refactor vs model implementation add; no shared bug/change." }, { "left": "pull_request:47093", - "right": "pull_request:47163", + "right": "pull_request:47181", "accept": false, - "reason": "New model support vs paged-attention XPU test coverage; not the same underlying change." + "reason": "Two different model integrations (OpenPangu-V2 vs Cosmos3 Edge), not duplicates." }, { - "left": "pull_request:46266", - "right": "pull_request:47216", + "left": "pull_request:44517", + "right": "pull_request:46376", "accept": false, - "reason": "Different work: DINO modernization/conversion mapping vs compressed-tensors quantization docs/runtime." + "reason": "Qwen3 TTS model addition vs continuous-batching multimodal support; different change sets." }, { - "left": "pull_request:47108", - "right": "pull_request:47301", + "left": "pull_request:47166", + "right": "pull_request:47292", "accept": false, - "reason": "Doc-builder pin comment cleanup is unrelated to adding MTP documentation." + "reason": "Both are test updates, but for different models and hardware targets." }, { - "left": "pull_request:46376", - "right": "pull_request:47291", + "left": "pull_request:46836", + "right": "pull_request:47275", "accept": false, - "reason": "Both mention continuous batching, but one adds multimodal support while the other deprecates config plumbing; separate changes." + "reason": "Unlimited OCR model support vs EfficientViT-SAM model support; distinct artifacts." }, { - "left": "pull_request:47179", - "right": "pull_request:47210", + "left": "pull_request:46266", + "right": "pull_request:47301", "accept": false, - "reason": "Different features in different subsystems: model support vs packed SDPA attention routing." + "reason": "RoPE modernization/tests vs assisted-decoding docs; unrelated." }, { - "left": "pull_request:46196", - "right": "pull_request:47179", + "left": "pull_request:44775", + "right": "pull_request:47304", "accept": false, - "reason": "Different model families and fixes; no common code path." + "reason": "Parallelism docs restructuring vs MPS graph-cache docs; both docs-only but unrelated topics." }, { - "left": "pull_request:47078", - "right": "pull_request:47275", + "left": "pull_request:44775", + "right": "pull_request:47181", "accept": false, - "reason": "Japanese doc wording cleanup is unrelated to adding EfficientViT-SAM support." + "reason": "Docs navigation change vs model support PR; no same underlying change." }, { - "left": "pull_request:46912", - "right": "pull_request:46990", + "left": "pull_request:47093", + "right": "pull_request:47143", "accept": false, - "reason": "RoPE test updates and FSDP orchestration are separate concerns." + "reason": "OpenPangu-V2 model support vs NeuCodec model support; both model adds but not the same one." }, { - "left": "pull_request:46990", - "right": "pull_request:47181", + "left": "pull_request:47210", + "right": "pull_request:47302", "accept": false, - "reason": "FSDP orchestration changes are unrelated to Cosmos3 Edge model support." + "reason": "Packed-SDPA routing code vs tokenizer fallback docs; different layers entirely." }, { - "left": "pull_request:46395", - "right": "pull_request:47311", + "left": "pull_request:47093", + "right": "pull_request:47163", "accept": false, - "reason": "Param2MoE model support and gemma4 H100 test expectations are unrelated." + "reason": "Model integration PR vs XPU paged-attention test add; unrelated." }, { - "left": "pull_request:47311", - "right": "pull_request:47318", + "left": "pull_request:46266", + "right": "pull_request:47216", "accept": false, - "reason": "Test expectation tweaks for different areas; no shared underlying bug." + "reason": "RoPE tests/code vs compressed-tensors fp8 changes; not the same issue." }, { - "left": "pull_request:47175", - "right": "pull_request:47179", + "left": "pull_request:47108", + "right": "pull_request:47301", "accept": false, - "reason": "CI expectation updates for Olmo3 are unrelated to Granite model support." + "reason": "Doc-builder pin comment tweak vs assisted-decoding docs update; unrelated docs maintenance." }, { - "left": "pull_request:47210", - "right": "pull_request:47302", + "left": "pull_request:47082", + "right": "pull_request:47275", "accept": false, - "reason": "Packed SDPA routing and tokenizer fallback docs are unrelated changes." + "reason": "RoMa keypoint matcher vs EfficientViT-SAM model support; different models and code paths." }, { - "left": "pull_request:47198", - "right": "pull_request:47318", + "left": "pull_request:46376", + "right": "pull_request:47291", "accept": false, - "reason": "GPT backend fixes and flash-attn test helper changes touch different code paths." + "reason": "Continuous batching multimodal support vs deprecating CB config; related area, but not the same change." }, { - "left": "pull_request:47077", - "right": "pull_request:47302", + "left": "pull_request:47082", + "right": "pull_request:47210", "accept": false, - "reason": "Japanese DistilBERT docs and tokenizer fallback docs are unrelated." + "reason": "Vision model add vs attention-kernel routing; no concrete overlap." }, { - "left": "pull_request:47082", + "left": "pull_request:47179", "right": "pull_request:47210", "accept": false, - "reason": "RoMa model addition and packed SDPA attention dispatch are unrelated." + "reason": "Granite-SWA model support vs SDPA varlen packing; unrelated." }, { - "left": "pull_request:47163", + "left": "pull_request:46196", "right": "pull_request:47179", "accept": false, - "reason": "Paged-attention XPU tests and Granite model support are different changes." + "reason": "SDPA dispatch for wav2vec2-family audio models vs Granite-SWA model support; different scopes." + }, + { + "left": "pull_request:46912", + "right": "pull_request:46990", + "accept": false, + "reason": "RoPE test expansion vs FSDP orchestration; no shared bug or feature." }, { "left": "pull_request:46990", - "right": "pull_request:47306", + "right": "pull_request:47181", "accept": false, - "reason": "FSDP orchestration and Neuron experiment scripts are unrelated." + "reason": "FSDP orchestration changes vs Cosmos3 Edge model support; unrelated." }, { - "left": "pull_request:47108", - "right": "pull_request:47302", + "left": "pull_request:47078", + "right": "pull_request:47304", "accept": false, - "reason": "Doc-builder pin comment cleanup does not overlap with tokenizer fallback docs." + "reason": "Japanese doc wording cleanup vs MPS graph-cache docs; different documentation changes." }, { - "left": "pull_request:47082", - "right": "pull_request:47301", + "left": "pull_request:46395", + "right": "pull_request:47311", "accept": false, - "reason": "RoMa model support is unrelated to assisted-decoding docs for MTP." + "reason": "Param2MoE model implementation vs Gemma4 H100 expectation update; unrelated." }, { - "left": "pull_request:46376", + "left": "pull_request:47311", "right": "pull_request:47318", "accept": false, - "reason": "Continuous batching multimodal work and flash-attn test utility changes are different code paths." + "reason": "Gemma4 test expectations vs flash-attn availability helper; unrelated test infra changes." }, { - "left": "pull_request:47210", - "right": "pull_request:47304", + "left": "pull_request:47175", + "right": "pull_request:47179", "accept": false, - "reason": "Packed SDPA runtime support and an Apple Silicon doc page are unrelated." + "reason": "Olmo3 CI expectation tweak vs Granite-SWA model support; not the same artifact." }, { - "left": "pull_request:46836", - "right": "pull_request:47218", + "left": "pull_request:47210", + "right": "pull_request:47304", "accept": false, - "reason": "Unlimited OCR model support and test image-loading refactor are unrelated." + "reason": "Packed SDPA implementation vs docs-only MPS cache note; different layers." }, { "left": "pull_request:46376", "right": "pull_request:47218", "accept": false, - "reason": "Continuous batching features and image-loading test refactor do not address the same issue." + "reason": "Continuous-batching code vs image-loading refactor in tests; no duplicate bug." }, { - "left": "pull_request:46957", - "right": "pull_request:47216", - "accept": false, - "reason": "Finegrained FP8 kernel behavior and compressed-tensors quantization docs/runtime are different topics." - }, - { - "left": "pull_request:47173", - "right": "pull_request:47311", + "left": "pull_request:46836", + "right": "pull_request:47218", "accept": false, - "reason": "XPU expectation fixes and H100 expectation fixes are separate test-maintenance changes." + "reason": "Unlimited OCR model add vs test image-loading refactor; unrelated." }, { - "left": "pull_request:46912", - "right": "pull_request:47082", + "left": "pull_request:47243", + "right": "pull_request:47253", "accept": false, - "reason": "RoPE test adjustments and RoMa model support are unrelated." + "reason": "Exporter fixes vs tensor-parallel LM-head validation; different code paths." }, { - "left": "pull_request:47108", - "right": "pull_request:47216", + "left": "pull_request:47216", + "right": "pull_request:47291", "accept": false, - "reason": "Doc-builder pin cleanup and compressed-tensors quantization work are unrelated." + "reason": "Compressed-tensors FP8 work vs CB config deprecation; unrelated." }, { - "left": "pull_request:46912", - "right": "pull_request:47078", + "left": "pull_request:47173", + "right": "pull_request:47311", "accept": false, - "reason": "RoPE test updates and Japanese doc wording cleanup are unrelated." + "reason": "XPU Expectations test fixes vs Gemma4 H100 expectations; different test targets." }, { - "left": "pull_request:43451", - "right": "pull_request:46990", + "left": "pull_request:47108", + "right": "pull_request:47302", "accept": false, - "reason": "Molmo2 model addition and FSDP orchestration are unrelated." + "reason": "Doc-builder pin comment vs tokenizer-backend fallback docs; unrelated docs changes." }, { - "left": "pull_request:47179", + "left": "pull_request:47216", "right": "pull_request:47301", "accept": false, - "reason": "Granite model support and assisted-decoding MTP docs are unrelated." + "reason": "Quantization feature work vs assisted-decoding docs; not the same change." }, { "left": "pull_request:47113", - "right": "pull_request:47204", - "accept": false, - "reason": "Audio requirements fixes and exporter dependency changes are separate issues." - }, - { - "left": "pull_request:47218", "right": "pull_request:47304", "accept": false, - "reason": "Test image-loading refactor and Apple Silicon docs are unrelated." + "reason": "Audio requirements fix vs MPS docs update; unrelated." }, { - "left": "pull_request:47218", - "right": "pull_request:47311", + "left": "pull_request:43451", + "right": "pull_request:46266", "accept": false, - "reason": "Image-loading test refactor and H100 expectation tweaks are unrelated." + "reason": "Molmo2 model add vs DINO/DINOv2 modernization; distinct feature areas." }, { - "left": "pull_request:47163", - "right": "pull_request:47218", + "left": "pull_request:43451", + "right": "pull_request:44517", "accept": false, - "reason": "Paged-attention XPU tests and shared image-loading helpers are unrelated." + "reason": "Molmo2 model support vs Qwen3 TTS support; both model adds, but different models and code paths." }, { - "left": "pull_request:47216", - "right": "pull_request:47302", + "left": "pull_request:46912", + "right": "pull_request:47291", "accept": false, - "reason": "Compressed-tensors work and tokenizer fallback docs are unrelated." + "reason": "RoPE tests vs continuous-batching config deprecation; unrelated." }, { - "left": "pull_request:46266", - "right": "pull_request:46376", + "left": "pull_request:47069", + "right": "pull_request:47302", "accept": false, - "reason": "DINO modernization and continuous batching multimodal support are different features." + "reason": "`num_labels` typing fix vs tokenizer backend docs; unrelated." }, { "left": "pull_request:47243", - "right": "pull_request:47253", - "accept": false, - "reason": "Exporter fixes and tensor-parallel lm_head planning are separate runtime changes." - }, - { - "left": "pull_request:46912", "right": "pull_request:47294", "accept": false, - "reason": "RoPE tests and kernel-native causal_conv1d integration are unrelated." - }, - { - "left": "pull_request:47175", - "right": "pull_request:47294", - "accept": false, - "reason": "Olmo3 CI expectation changes and kernel integration work are unrelated." - }, - { - "left": "pull_request:47198", - "right": "pull_request:47253", - "accept": false, - "reason": "GPT backend fixes and tensor-parallel validation are different code paths." + "reason": "Exporter determinism fix vs kernel-function integration; different domains." }, { "left": "pull_request:46895", "right": "pull_request:47306", "accept": false, - "reason": "CohereASR training-loss fix and Neuron experiment scripts are unrelated." + "reason": "CohereASR training-loss fix vs Neuron experiment scripts; unrelated." }, { "left": "pull_request:46196", "right": "pull_request:46376", "accept": false, - "reason": "SDPA dispatch work and continuous batching multimodal support are unrelated." + "reason": "Audio SDPA support vs multimodal continuous batching; same broad area but different concrete changes." }, { "left": "pull_request:46196", "right": "pull_request:47301", "accept": false, - "reason": "wav2vec2 SDPA dispatch and MTP docs are unrelated." + "reason": "Audio attention dispatch vs assisted-decoding docs; no overlap." }, { "left": "pull_request:47253", "right": "pull_request:47294", "accept": false, - "reason": "Tensor-parallel lm_head validation and causal_conv1d kernel helpers are unrelated." + "reason": "Tensor-parallel validation vs causal_conv1d kernel functions; unrelated." }, { "left": "pull_request:47069", "right": "pull_request:47113", "accept": false, - "reason": "Configuration property typing and audio requirements fixes are unrelated." + "reason": "Config typing tweak vs audio requirements fix; separate issues." }, { "left": "pull_request:47275", - "right": "pull_request:47302", - "accept": false, - "reason": "EfficientViT-SAM model support and tokenizer fallback docs are unrelated." - }, - { - "left": "pull_request:44517", - "right": "pull_request:47311", + "right": "pull_request:47304", "accept": false, - "reason": "Qwen3-TTS model support and H100 test expectations are unrelated." + "reason": "EfficientViT-SAM model support vs MPS graph-cache docs; unrelated." }, { "left": "pull_request:44517", "right": "pull_request:47301", "accept": false, - "reason": "Qwen3-TTS model support and MTP docs are unrelated." + "reason": "Qwen3 TTS model add vs assisted-decoding docs; different topics." }, { "left": "pull_request:44517", "right": "pull_request:47181", "accept": false, - "reason": "Qwen3-TTS model support and Cosmos3 Edge model support are different changes." - }, - { - "left": "pull_request:46376", - "right": "pull_request:47198", - "accept": false, - "reason": "Continuous batching multimodal work and GPT backend fixes are unrelated." + "reason": "Qwen3 TTS vs Cosmos3 Edge model support; distinct model additions." }, { "left": "pull_request:47078", - "right": "pull_request:47243", + "right": "pull_request:47322", "accept": false, - "reason": "Japanese doc wording cleanup and exporter fixes are unrelated." - }, - { - "left": "pull_request:47069", - "right": "pull_request:47082", - "accept": false, - "reason": "num_labels typing change and RoMa model support are unrelated." + "reason": "Japanese docs wording vs AMD Quark config fix; unrelated docs vs code change." }, { "left": "pull_request:47113", "right": "pull_request:47304", "accept": false, - "reason": "Audio requirements fixes and Apple Silicon docs are unrelated." + "reason": "Audio requirements change vs MPS docs cleanup; not the same bug/change." }, { "left": "pull_request:43451", "right": "pull_request:46266", "accept": false, - "reason": "Two unrelated model additions; no shared bug or change request." + "reason": "Model addition vs Dino modernization/conversion mapping; unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:44517", + "accept": false, + "reason": "Different model families and different generated code; not duplicates." }, { "left": "pull_request:46912", "right": "pull_request:47291", "accept": false, - "reason": "RoPE test updates and continuous batching config deprecation are unrelated." + "reason": "RoPE test updates vs CB config deprecation; no common fix." }, { - "left": "pull_request:47069", - "right": "pull_request:47302", + "left": "pull_request:47179", + "right": "pull_request:47301", "accept": false, - "reason": "Config typing change and tokenizer fallback docs are unrelated." + "reason": "Granite-SWA support vs assisted-decoding docs; unrelated." }, { - "left": "pull_request:47243", - "right": "pull_request:47294", + "left": "pull_request:47113", + "right": "pull_request:47204", "accept": false, - "reason": "Exporter flattening fixes and causal_conv1d kernel integration are unrelated." + "reason": "Audio feature/proc requirements fix vs Dockerfile exporter dependency additions; unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:47078", + "left": "pull_request:47143", + "right": "pull_request:47318", "accept": false, - "reason": "Molmo2 model support and Japanese doc wording cleanup are unrelated." + "reason": "NeuCodec model add vs flash-attn availability helper; different functionality." }, { - "left": "pull_request:47210", - "right": "pull_request:47291", + "left": "pull_request:47306", + "right": "pull_request:47318", "accept": false, - "reason": "Packed SDPA attention routing and continuous batching config deprecation are different changes." + "reason": "Neuron experiment scripts vs flash-attn test helper; unrelated." }, { - "left": "pull_request:46957", - "right": "pull_request:47301", + "left": "pull_request:47218", + "right": "pull_request:47311", "accept": false, - "reason": "Finegrained FP8 runtime changes and MTP docs are unrelated." + "reason": "Image-loading helper refactor in tests vs Gemma4 expectation update; different test maintenance." }, { - "left": "pull_request:46895", - "right": "pull_request:47298", + "left": "pull_request:47218", + "right": "pull_request:47304", "accept": false, - "reason": "CohereASR loss fix and MI450 image-loading fallback changes are unrelated." + "reason": "Image-loading helper refactor vs MPS docs cleanup; unrelated." }, { - "left": "pull_request:47294", - "right": "pull_request:47306", + "left": "pull_request:47166", + "right": "pull_request:47218", "accept": false, - "reason": "Kernel integration work and Neuron experiment scripts are unrelated." + "reason": "Quantization test updates vs image-loading refactor; no shared issue." }, { - "left": "pull_request:47113", - "right": "pull_request:47175", + "left": "pull_request:47216", + "right": "pull_request:47302", "accept": false, - "reason": "Audio requirements fixes and Olmo3 CI expectation changes are unrelated." + "reason": "Compressed-tensors fp8 changes vs tokenizer backend docs; unrelated." }, { - "left": "pull_request:47113", - "right": "pull_request:47243", + "left": "pull_request:46266", + "right": "pull_request:46376", "accept": false, - "reason": "Audio requirements fixes and exporter fixes are unrelated." + "reason": "Dino modernization vs continuous-batching multimodal support; different areas." }, { - "left": "pull_request:46658", - "right": "pull_request:47302", + "left": "pull_request:47243", + "right": "pull_request:47294", "accept": false, - "reason": "Step-3.7 model tooling and tokenizer fallback docs are unrelated." + "reason": "Executorch exporter fixes vs causal_conv1d kernel functions; unrelated." }, { - "left": "pull_request:47113", - "right": "pull_request:47275", + "left": "pull_request:46912", + "right": "pull_request:47294", "accept": false, - "reason": "Audio requirements fixes and EfficientViT-SAM model support are unrelated." + "reason": "RoPE testing changes vs kernel helper PR; unrelated." }, { - "left": "pull_request:46658", - "right": "pull_request:47275", + "left": "pull_request:46895", + "right": "pull_request:47298", "accept": false, - "reason": "Step-3.7 Flash tooling and EfficientViT-SAM model support are unrelated." + "reason": "CohereASR loss fix vs MI450 image decoding fallback changes; unrelated." } ] }, @@ -23833,9 +23802,8 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-75", + "cluster_id": "cluster-40546-76", "nodes": [ - "pull_request:40546", "pull_request:41855", "pull_request:42668", "pull_request:43251", @@ -23860,7 +23828,7 @@ "pull_request:47069", "pull_request:47077", "pull_request:47090", - "pull_request:47096", + "pull_request:47113", "pull_request:47143", "pull_request:47163", "pull_request:47166", @@ -23868,24 +23836,35 @@ "pull_request:47175", "pull_request:47181", "pull_request:47184", - "pull_request:47198", "pull_request:47210", "pull_request:47216", "pull_request:47243", "pull_request:47253", + "pull_request:47275", "pull_request:47291", "pull_request:47292", + "pull_request:47294", "pull_request:47301", "pull_request:47302", "pull_request:47304", "pull_request:47306", "pull_request:47311", - "pull_request:47318" + "pull_request:47322", + "pull_request:47323" ], "soft_pairs": [ + "pull_request:46957|pull_request:47301", + "pull_request:47294|pull_request:47306", + "pull_request:47113|pull_request:47175", + "pull_request:47113|pull_request:47243", + "pull_request:46658|pull_request:47302", + "pull_request:46895|pull_request:47323", + "pull_request:47113|pull_request:47275", + "pull_request:46658|pull_request:47275", "pull_request:47069|pull_request:47175", - "pull_request:47306|pull_request:47318", - "pull_request:46658|pull_request:46895", + "pull_request:46912|pull_request:47253", + "pull_request:47302|pull_request:47322", + "pull_request:47069|pull_request:47323", "pull_request:46957|pull_request:47302", "pull_request:43451|pull_request:46957", "pull_request:41855|pull_request:43424", @@ -23922,8 +23901,9 @@ "pull_request:47163|pull_request:47166", "pull_request:47163|pull_request:47184", "pull_request:46509|pull_request:47181", - "pull_request:46376|pull_request:46509", "pull_request:46509|pull_request:47163", + "pull_request:46376|pull_request:46509", + "pull_request:47291|pull_request:47322", "pull_request:47163|pull_request:47292", "pull_request:46376|pull_request:47181", "pull_request:47216|pull_request:47301", @@ -23939,123 +23919,167 @@ "pull_request:47181|pull_request:47301", "pull_request:47163|pull_request:47311", "pull_request:46658|pull_request:47143", + "pull_request:46658|pull_request:47322", "pull_request:47143|pull_request:47163", - "pull_request:47198|pull_request:47301", - "pull_request:47198|pull_request:47311", - "pull_request:47181|pull_request:47198", "pull_request:47181|pull_request:47311", "pull_request:46376|pull_request:47311", - "pull_request:47143|pull_request:47301", - "pull_request:43451|pull_request:47243", - "pull_request:43451|pull_request:47291", - "pull_request:47096|pull_request:47291", - "pull_request:47253|pull_request:47306", - "pull_request:40546|pull_request:43451", - "pull_request:43451|pull_request:47175", - "pull_request:47096|pull_request:47175", - "pull_request:47166|pull_request:47311", - "pull_request:47184|pull_request:47311" + "pull_request:47143|pull_request:47322", + "pull_request:43451|pull_request:47243" ], - "prepared_review_unit_hash": "68a7ff2789a14ea556c71c855171ac4aebe0a620d9c3cd6373fd06f9521112b4", + "prepared_review_unit_hash": "e96ccc8e8ffb7338b5617c127147779d498fbc11454ee76986cd2c8258a83b2a", "component_budget": { - "node_count": 75, - "item_count": 75, - "soft_pair_count": 217, - "serialized_chars": 263464, - "estimated_input_tokens": 65866, - "estimated_eval_tokens": 131988 + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 }, "budget": { - "node_count": 46, - "item_count": 46, + "node_count": 47, + "item_count": 47, "soft_pair_count": 72, - "serialized_chars": 129073, - "estimated_input_tokens": 32269, - "estimated_eval_tokens": 64794 + "serialized_chars": 131684, + "estimated_input_tokens": 32921, + "estimated_eval_tokens": 66098 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated feature PRs or test/docs tweaks. The only clear duplicate set is the CohereASR training-loss double-shift fixes; 47090 is the best representative and canonical PR.", - "confidence": 0.97, + "summary": "Mostly a heterogeneous cluster: model additions, docs, tests, exporter/kernel fixes, and one real duplicate subgroup. The only clear duplicate set is the CohereASR training-loss double-shift trio (46895/46958/47090); the rest are unrelated and should not be merged as duplicates.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47090 is the cleanest and most explicit fix for the CohereASR double-shift bug, with the clearest explanation and the same target/files as the earlier variants.", + "canonical_pr_reason": "No single canonical for the whole cluster; the only true duplicate subgroup is the CohereASR training-loss double-shift trio, and 47090 is the clearest representative of that subgroup.", "best_issue_reason": null, - "best_pr_reason": "PR 47090 best represents the duplicate CohereASR loss bug: it directly addresses the right-shift issue, preserves loss reduction behavior, and supersedes the earlier variants.", + "best_pr_reason": "47090 is the best representative of the confirmed duplicate subgroup: it has the clearest title and explicitly describes the same CohereASR double-shift fix as 46895 and 46958.", "soft_edge_verdicts": [ { - "left": "pull_request:47069", + "left": "pull_request:46957", + "right": "pull_request:47301", + "accept": false, + "reason": "Different changes: finegrained FP8 decode/kernels vs MTP docs. Same repo area, not the same bug or mergeable fix." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47306", + "accept": false, + "reason": "Kernel-native causal_conv1d plumbing is unrelated to Neuron training experiments; no shared concrete code-path problem." + }, + { + "left": "pull_request:47113", "right": "pull_request:47175", "accept": false, - "reason": "Unrelated config typing change vs. OLMo3 XPU test expectation update." + "reason": "Audio feature/proc requirement fixes and OLMo3 CI expectation updates are unrelated test fixes." }, { - "left": "pull_request:47306", - "right": "pull_request:47318", + "left": "pull_request:47113", + "right": "pull_request:47243", + "accept": false, + "reason": "Audio feature/proc requirement fixes do not match ExecuTorch exporter flattening fixes." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47302", "accept": false, - "reason": "Completely different areas: Neuron experiment scripts vs flash-attn fallback test helper." + "reason": "Step3p7 model build/conversion work is unrelated to tokenizer-backend docs fallback wording." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47323", + "accept": false, + "reason": "CohereASR loss-shift fix and DeepGEMM multi-device support are different subsystems and different bugs." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47275", + "accept": false, + "reason": "Audio requirement fixes and new EfficientViTSam model support are unrelated." }, { "left": "pull_request:46658", - "right": "pull_request:46895", + "right": "pull_request:47275", + "accept": false, + "reason": "Step3p7 flash conversion and EfficientViTSam model addition are unrelated model work." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47175", + "accept": false, + "reason": "A configuration typing fix for num_labels is unrelated to OLMo3 test expectation changes." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47253", + "accept": false, + "reason": "RoPE test updates across models do not match tensor-parallel plan handling in lm_heads." + }, + { + "left": "pull_request:47302", + "right": "pull_request:47322", "accept": false, - "reason": "Step 3.7 model build/conversion work is unrelated to CohereASR loss handling." + "reason": "Docs wording for tokenizers fallback is unrelated to AMD Quark config class changes." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47323", + "accept": false, + "reason": "num_labels None handling and DeepGEMM multi-device fixes are unrelated." }, { "left": "pull_request:46957", "right": "pull_request:47302", "accept": false, - "reason": "Different changes: finegrained FP8 decode kernel selection vs tokenizer docs formatting." + "reason": "FP8 decode/kernel optimization and tokenizer docs fallback are unrelated." }, { "left": "pull_request:43451", "right": "pull_request:46957", "accept": false, - "reason": "Molmo2 model addition and FP8 kernel changes are unrelated subsystem work." + "reason": "Molmo2 model addition is unrelated to finegrained FP8 batching/runtime fixes." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Tokenizer method additions vs ExecuTorch dynamic-shape test are not the same bug." + "reason": "Mistral tokenizer API additions and ExecuTorch dynamic-shape tests are unrelated." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Mistral tokenizer API additions and MobileLLM model addition are unrelated." + "reason": "Mistral tokenizer missing methods and MobileLLM model support are unrelated changes." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor auto-loading robustness and MobileLLM model addition are different changes." + "reason": "Processor robustness work for AudioFlamingo3/auto processors is unrelated to adding MobileLLM." }, { "left": "pull_request:43424", "right": "pull_request:47301", "accept": false, - "reason": "ExecuTorch dynamic-shape testing and assisted decoding docs do not fix the same issue." + "reason": "ExecuTorch dynamic-shape testing and MTP docs are unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "New MobileLLM model support vs config annotation-order fix are unrelated." + "reason": "Adding MobileLLM and fixing config resolution order are unrelated code paths." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test/backend fixes are unrelated to the layrnorm rename cleanup." + "reason": "Mistral4 test fixes and layrnorm->layernorm renaming are different bugs and different code paths." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "ExecuTorch export tests and Mistral4 test/backend fixes are different problems." + "reason": "ExecuTorch dynamic-shape tests and Mistral4 test fixes are unrelated." }, { "left": "pull_request:43823", @@ -24067,103 +24091,97 @@ "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Cross-entropy kwargs forwarding and MobileLLM model addition are different changes." + "reason": "cross_entropy kwargs forwarding and MobileLLM model support are unrelated." }, { "left": "pull_request:44827", "right": "pull_request:46048", "accept": false, - "reason": "Mistral4 test fixes and TDT loss kernel integration are unrelated." + "reason": "Mistral4 tests and TDT loss kernel integration are unrelated." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Dynamic-shape ExecuTorch tests and MobileLLM model addition do not share the same code-path bug." + "reason": "ExecuTorch dynamic-shape testing and MobileLLM addition are unrelated." }, { - "left": "pull_request:43251", - "right": "pull_request:44827", + "left": "pull_request:43995", + "right": "pull_request:45213", "accept": false, - "reason": "Cross-entropy kwargs fix and Mistral4 fixes are unrelated." + "reason": "Falcon refactor and model-creation skill scaffolding are unrelated." }, { "left": "pull_request:44827", - "right": "pull_request:45655", - "accept": false, - "reason": "Mistral4 backend/test work is unrelated to cls.config resolution ordering." - }, - { - "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "Falcon output-interface refactor and model-creation skill PR are unrelated." + "reason": "Mistral4 test fixes and the model-creation skill are unrelated." }, { - "left": "pull_request:44827", - "right": "pull_request:45213", + "left": "pull_request:43995", + "right": "pull_request:44827", "accept": false, - "reason": "Mistral4 test fixes are unrelated to the model-creation skill draft." + "reason": "Falcon refactor and Mistral4 tests do not address the same concrete bug." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "MobileLLM model support and model-creation skill draft are not the same change." + "reason": "MobileLLM model addition and the model-creation skill are unrelated." }, { "left": "pull_request:46895", "right": "pull_request:47090", "accept": true, - "reason": "Both fix the same CohereASR double-shift training-loss bug in the same files and target the same issue." + "reason": "Same underlying CohereASR training-loss double-shift bug; both patch the same forward/loss path and could plausibly be merged into one fix." }, { "left": "pull_request:46958", "right": "pull_request:47090", "accept": true, - "reason": "Same CohereASR loss-shift bug, same code path, and same issue target; 47090 is the cleaner later variant." + "reason": "Same CohereASR double-shift bug and same loss computation path; 47090 is just a clearer variant of the same fix." }, { "left": "pull_request:46895", "right": "pull_request:46958", "accept": true, - "reason": "Both are variants of the same CohereASR double-shift loss fix and would plausibly merge into one PR." + "reason": "Same CohereASR loss-shift issue; the two PRs differ in implementation detail but target the same concrete bug." }, { "left": "pull_request:47173", "right": "pull_request:47292", "accept": false, - "reason": "Both are XPU-related test updates, but they touch different models and different underlying failures." + "reason": "Both are XPU test expectation updates, but they are model-specific failures on different code paths, not one shared bug." }, { "left": "pull_request:47166", "right": "pull_request:47184", "accept": false, - "reason": "Both adjust tests for accelerator-specific behavior, but they are different test suites and not the same bug." + "reason": "Quantization test expansions and Qwen3Next/Videomt accelerator test updates are unrelated." }, { "left": "pull_request:46912", "right": "pull_request:47175", "accept": false, - "reason": "Different model test expectation updates; no shared concrete bug." + "reason": "RoPE numerical test updates and OLMo3 CI expectation tweaks are different model/test changes." }, { "left": "pull_request:47301", "right": "pull_request:47302", "accept": false, - "reason": "Assisted-decoding docs vs tokenizer backend docs are unrelated documentation changes." + "reason": "MTP docs support and tokenizer fallback docs are unrelated documentation changes." }, { "left": "pull_request:47184", "right": "pull_request:47292", "accept": false, - "reason": "Different test files for different models; only broadly similar XPU maintenance." + "reason": "Qwen3Next/Videomt multi-accelerator test changes and Tipsv2-DPT XPU fixes are unrelated." }, { "left": "pull_request:47301", "right": "pull_request:47304", "accept": false, - "reason": "Different docs-only updates for unrelated features." + "reason": "MTP docs and MPS graph cache docs are unrelated." }, { "left": "pull_request:47077", @@ -24175,139 +24193,103 @@ "left": "pull_request:46658", "right": "pull_request:46836", "accept": false, - "reason": "Step 3.7 flash model work and Unlimited OCR model addition are separate model integrations." + "reason": "Step3p7 model build/conversion work and Unlimited OCR model addition are unrelated model-introduction PRs." }, { "left": "pull_request:47210", "right": "pull_request:47301", "accept": false, - "reason": "Packed SDPA implementation and MTP docs are unrelated." + "reason": "Packed SDPA varlen attention routing and MTP docs are unrelated." }, { "left": "pull_request:47003", "right": "pull_request:47243", "accept": false, - "reason": "OpenVINO exporter addition and ExecuTorch exporter fixes are different exporter backends and different bugs." + "reason": "OpenVINO exporter work and ExecuTorch exporter fixes are different exporter implementations and different bugs." }, { "left": "pull_request:46509", "right": "pull_request:47143", "accept": false, - "reason": "Nemotron Omni model support and NeuCodec model support are different additions." + "reason": "Nemotron Omni and NeuCodec are distinct model additions with different code and docs." }, { "left": "pull_request:46376", "right": "pull_request:46509", "accept": false, - "reason": "Continuous batching multimodal support and Nemotron Omni model support are different changes." + "reason": "Continuous batching multimodal support and Nemotron Omni model support are unrelated." }, { - "left": "pull_request:46509", - "right": "pull_request:47163", + "left": "pull_request:47291", + "right": "pull_request:47322", "accept": false, - "reason": "Model support PR vs paged-attention XPU test PR; no same underlying bug." + "reason": "Continuous-batching config deprecation and AMD Quark config changes are unrelated." }, { - "left": "pull_request:46376", - "right": "pull_request:47143", + "left": "pull_request:47163", + "right": "pull_request:47166", "accept": false, - "reason": "Continuous batching multimodal support and NeuCodec addition are unrelated." + "reason": "Paged-attention XPU tests and quantization test updates are unrelated." }, { - "left": "pull_request:46509", - "right": "pull_request:46836", + "left": "pull_request:47163", + "right": "pull_request:47184", "accept": false, - "reason": "Nemotron Omni addition and Unlimited OCR integration are different model PRs." + "reason": "Paged-attention XPU tests and Qwen3Next/Videomt accelerator test updates are unrelated." }, { "left": "pull_request:46509", - "right": "pull_request:47301", + "right": "pull_request:47181", "accept": false, - "reason": "Nemotron Omni model support and MTP docs support are unrelated." + "reason": "Nemotron Omni and Cosmos3 Edge are separate model additions, not the same underlying change." }, { - "left": "pull_request:46376", - "right": "pull_request:47301", + "left": "pull_request:46509", + "right": "pull_request:47163", "accept": false, - "reason": "Continuous batching multimodal support and MTP docs are unrelated." + "reason": "Model-addition PR and paged-attention XPU tests are unrelated." }, { - "left": "pull_request:46836", - "right": "pull_request:47143", + "left": "pull_request:46376", + "right": "pull_request:47163", "accept": false, - "reason": "Unlimited OCR and NeuCodec are different model additions." + "reason": "Continuous batching multimodal support and paged-attention XPU tests are unrelated." }, { - "left": "pull_request:47163", - "right": "pull_request:47181", + "left": "pull_request:47181", + "right": "pull_request:47301", "accept": false, - "reason": "Paged-attention XPU tests and Cosmos3 Edge model support are unrelated." + "reason": "Cosmos3 Edge model support and MTP docs support are unrelated." }, { "left": "pull_request:46376", "right": "pull_request:47311", "accept": false, - "reason": "Continuous batching multimodal support and Gemma4 H100 expectation tweaks are unrelated." + "reason": "Continuous batching multimodal support and Gemma4 H100 expectation updates are unrelated." }, { "left": "pull_request:47143", - "right": "pull_request:47301", - "accept": false, - "reason": "NeuCodec model support and MTP 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:43451", - "right": "pull_request:47291", - "accept": false, - "reason": "Molmo2 model addition and CB config deprecation are unrelated." - }, - { - "left": "pull_request:47096", - "right": "pull_request:47291", + "right": "pull_request:47163", "accept": false, - "reason": "Kimi exportability fixes and CB config deprecation are different areas." + "reason": "NeuCodec model support and paged-attention XPU tests are unrelated." }, { - "left": "pull_request:47253", - "right": "pull_request:47306", + "left": "pull_request:47181", + "right": "pull_request:47311", "accept": false, - "reason": "Tensor-parallel LM head support and Neuron experiment scripts are unrelated." + "reason": "Cosmos3 Edge model support and Gemma4 H100 test expectations are unrelated." }, { - "left": "pull_request:40546", - "right": "pull_request:43451", + "left": "pull_request:46376", + "right": "pull_request:47322", "accept": false, - "reason": "VibeVoice integration and Molmo2 model addition are separate new-model PRs." + "reason": "Continuous batching multimodal support and Quark config changes are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47175", - "accept": false, - "reason": "Molmo2 model addition and OLMo3 CI expectation updates are unrelated." - }, - { - "left": "pull_request:47096", - "right": "pull_request:47175", - "accept": false, - "reason": "Kimi exportability work and OLMo3 CI tweaks do not fix the same issue." - }, - { - "left": "pull_request:47166", - "right": "pull_request:47311", - "accept": false, - "reason": "Different XPU/test updates for different models; no shared bug." - }, - { - "left": "pull_request:47184", - "right": "pull_request:47311", + "right": "pull_request:47243", "accept": false, - "reason": "Different accelerator-specific test maintenance, not the same underlying issue." + "reason": "Molmo2 model addition and ExecuTorch exporter fixes are unrelated." } ] }, @@ -24329,198 +24311,246 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-75", + "cluster_id": "cluster-40546-76", "nodes": [ + "pull_request:40546", "pull_request:43451", "pull_request:46509", - "pull_request:46658", "pull_request:47003", "pull_request:47096", "pull_request:47143", "pull_request:47163", + "pull_request:47166", + "pull_request:47175", "pull_request:47181", + "pull_request:47184", + "pull_request:47253", + "pull_request:47291", "pull_request:47298", "pull_request:47306", "pull_request:47311", - "pull_request:47313" + "pull_request:47322", + "pull_request:47323" ], "soft_pairs": [ - "pull_request:43451|pull_request:46658", + "pull_request:43451|pull_request:47291", + "pull_request:47096|pull_request:47291", + "pull_request:47253|pull_request:47306", + "pull_request:40546|pull_request:43451", + "pull_request:43451|pull_request:47175", + "pull_request:47096|pull_request:47175", + "pull_request:47166|pull_request:47311", + "pull_request:47184|pull_request:47311", "pull_request:47181|pull_request:47306", "pull_request:43451|pull_request:47143", "pull_request:43451|pull_request:47163", + "pull_request:47298|pull_request:47323", "pull_request:43451|pull_request:46509", "pull_request:43451|pull_request:47311", "pull_request:43451|pull_request:47096", "pull_request:43451|pull_request:47003", "pull_request:47003|pull_request:47096", - "pull_request:43451|pull_request:47313", - "pull_request:47096|pull_request:47313", + "pull_request:43451|pull_request:47323", + "pull_request:47096|pull_request:47323", "pull_request:43451|pull_request:47298", + "pull_request:43451|pull_request:47322", "pull_request:47096|pull_request:47298", - "pull_request:47003|pull_request:47313", + "pull_request:47003|pull_request:47323", "pull_request:47003|pull_request:47298", - "pull_request:47298|pull_request:47313", "pull_request:47096|pull_request:47143", "pull_request:47096|pull_request:47163", - "pull_request:47143|pull_request:47313", - "pull_request:47143|pull_request:47298", - "pull_request:47163|pull_request:47313" + "pull_request:47143|pull_request:47298" ], - "prepared_review_unit_hash": "45ce1aa0a8ed1afc7b35b3c88997972251df2db3e6b54f101ac2629cda06ae54", + "prepared_review_unit_hash": "b8d96a1e1fa5a8e86a7e1d7e9a79ef80e5e2332137426b15870addfaad0a7bd8", "component_budget": { - "node_count": 75, - "item_count": 75, - "soft_pair_count": 217, - "serialized_chars": 263464, - "estimated_input_tokens": 65866, - "estimated_eval_tokens": 131988 + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 }, "budget": { - "node_count": 12, - "item_count": 12, - "soft_pair_count": 21, - "serialized_chars": 36841, - "estimated_input_tokens": 9211, - "estimated_eval_tokens": 18678 + "node_count": 18, + "item_count": 18, + "soft_pair_count": 27, + "serialized_chars": 50363, + "estimated_input_tokens": 12591, + "estimated_eval_tokens": 25438 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes unrelated model-addition PRs, exporter work, tests, and a DeepGEMM fix. None of the soft-edge pairs look like the same underlying change or a plausible single merged PR.", - "confidence": 0.96, + "summary": "This cluster is dominated by unrelated PRs that only look superficially similar because they are all model/support/test/infra changes. None of the soft edges look like true duplicates of the same underlying change.", + "confidence": 0.57, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43451 is the broadest representative of the repeated \u201cadd new model support\u201d pattern (docs, auto mappings, model package, tests), so it is the best anchor only as a cluster representative\u2014not because it duplicates the others.", + "canonical_pr_reason": "PR 43451 is the most central hub in the cluster and the broadest, most representative support PR; several other items are only loosely similar to it by pattern, not by concrete code-path.", "best_issue_reason": null, - "best_pr_reason": "PR 43451 is the clearest exemplar of the cluster\u2019s dominant shape (new model integration with docs/tests/registration). The rest are different subsystems or different models, so this is only a representative choice.", + "best_pr_reason": "PR 43451 is the best representative PR for the cluster because it is the hub with the strongest connectivity and the most complete implementation footprint among the items.", "soft_edge_verdicts": [ { "left": "pull_request:43451", - "right": "pull_request:46658", + "right": "pull_request:47291", + "accept": false, + "reason": "Both involve framework/config changes, but one adds Molmo2 support and the other deprecates continuous batching config; different code paths and goals." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47291", + "accept": false, + "reason": "Exporter/Kimi changes are unrelated to generation config deprecation; same broad area, but not the same bug or fix." + }, + { + "left": "pull_request:47253", + "right": "pull_request:47306", + "accept": false, + "reason": "Tensor-parallel validation and Neuron experiment scripts are unrelated changes; no shared concrete failure mode." + }, + { + "left": "pull_request:40546", + "right": "pull_request:43451", + "accept": false, + "reason": "Both are model-support PRs, but for different models (VibeVoice vs Molmo2) and different implementation details." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47175", "accept": false, - "reason": "Molmo2 model support vs Step-3.7 build/convert tooling; different codepaths and deliverables." + "reason": "Molmo2 support and Olmo3 CI expectation updates are different artifacts, with no shared underlying bug." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47175", + "accept": false, + "reason": "Kimi exportability and Olmo3 test expectation tweaks are unrelated; they do not fix the same code path." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47311", + "accept": false, + "reason": "Both adjust test expectations, but for different models and hardware targets; not a duplicate fix." + }, + { + "left": "pull_request:47184", + "right": "pull_request:47311", + "accept": false, + "reason": "Qwen3 Next test-annotation cleanup and Gemma4 H100 expectation updates are separate test-only changes." }, { "left": "pull_request:47181", "right": "pull_request:47306", "accept": false, - "reason": "Cosmos3 Edge model integration vs Neuron LoRA finetuning experiment script; unrelated functionality." + "reason": "Cosmos3 Edge support and Neuron experiment scripts are unrelated model/infrastructure additions." }, { "left": "pull_request:43451", "right": "pull_request:47143", "accept": false, - "reason": "Different model integrations (Molmo2 vs NeuCodec) with distinct files, tests, and APIs." + "reason": "Both add model support, but for different models (Molmo2 vs NeuCodec) with distinct files and runtime behavior." }, { "left": "pull_request:43451", "right": "pull_request:47163", "accept": false, - "reason": "Molmo2 support vs a small paged-attention XPU test update; not the same bug or feature." + "reason": "Molmo2 model support and XPU paged-attention test coverage are unrelated changes." + }, + { + "left": "pull_request:47298", + "right": "pull_request:47323", + "accept": false, + "reason": "Both mention hardware/runtime issues, but one patches image loading fallbacks for MI450 while the other changes DeepGEMM multi-device behavior; different concrete failures." }, { "left": "pull_request:43451", "right": "pull_request:46509", "accept": false, - "reason": "Molmo2 vs Nemotron Omni support; separate model additions." + "reason": "Molmo2 and Nemotron Omni are separate model integrations, not the same underlying feature." }, { "left": "pull_request:43451", "right": "pull_request:47311", "accept": false, - "reason": "Molmo2 model support vs a test expectation tweak for Gemma4 on H100; unrelated." + "reason": "Molmo2 support and Gemma4 H100 test expectations are unrelated model-specific updates." }, { "left": "pull_request:43451", "right": "pull_request:47096", "accept": false, - "reason": "Model addition vs exporter/Kimi exportability changes; different underlying change." + "reason": "Molmo2 support and Kimi exportability are different feature areas and code paths." }, { "left": "pull_request:43451", "right": "pull_request:47003", "accept": false, - "reason": "Molmo2 model support vs OpenVINO exporter infrastructure; different feature area." + "reason": "Molmo2 model support and OpenVINO exporter support are different subsystems and changes." }, { "left": "pull_request:47003", "right": "pull_request:47096", "accept": false, - "reason": "OpenVINO exporter implementation vs Kimi exporter support; same broad exporter subsystem but different concrete changes." + "reason": "OpenVINO exporter support and Kimi exportability are both exporter-related, but they target different models and runtime paths." }, { "left": "pull_request:43451", - "right": "pull_request:47313", + "right": "pull_request:47323", "accept": false, - "reason": "Molmo2 support vs DeepGEMM SM90 layout fix; unrelated." + "reason": "Model support for Molmo2 and DeepGEMM multi-device fixes are unrelated changes." }, { "left": "pull_request:47096", - "right": "pull_request:47313", + "right": "pull_request:47323", "accept": false, - "reason": "Kimi exportability changes vs DeepGEMM kernel layout fix; no shared bug/change." + "reason": "Kimi exportability and DeepGEMM multi-device fixes address different parts of the stack." }, { "left": "pull_request:43451", "right": "pull_request:47298", "accept": false, - "reason": "Molmo2 model support vs image decoding fallback for MI450 failures; unrelated." + "reason": "Molmo2 support and MI450 image-loading/model-failure fixes are unrelated code paths." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47322", + "accept": false, + "reason": "Molmo2 support and AMD Quark config class changes are distinct features with no shared fix." }, { "left": "pull_request:47096", "right": "pull_request:47298", "accept": false, - "reason": "Exporter/Kimi work vs image utility fallback for ROCm/MI450; different code paths." + "reason": "Exporter support for Kimi and MI450 runtime fallback fixes are unrelated." }, { "left": "pull_request:47003", - "right": "pull_request:47313", + "right": "pull_request:47323", "accept": false, - "reason": "OpenVINO exporter vs DeepGEMM padding behavior on SM90; separate domains." + "reason": "OpenVINO exporter support and DeepGEMM multi-device fixes are separate runtime changes." }, { "left": "pull_request:47003", "right": "pull_request:47298", "accept": false, - "reason": "OpenVINO exporter vs image decoding fallback; unrelated." - }, - { - "left": "pull_request:47298", - "right": "pull_request:47313", - "accept": false, - "reason": "Image decoding fallback vs DeepGEMM layout fix; different subsystems and failure modes." + "reason": "OpenVINO exporter work and MI450 image fallback fixes do not share a concrete bug or code path." }, { "left": "pull_request:47096", "right": "pull_request:47143", "accept": false, - "reason": "Kimi exporter enablement vs NeuCodec model addition; unrelated changes." + "reason": "Kimi exportability and NeuCodec support are different model integrations." }, { "left": "pull_request:47096", "right": "pull_request:47163", "accept": false, - "reason": "Kimi exporter changes vs paged-attention XPU test support; not the same underlying issue." - }, - { - "left": "pull_request:47143", - "right": "pull_request:47313", - "accept": false, - "reason": "NeuCodec model support vs DeepGEMM SM90 fix; no overlap beyond general library maintenance." + "reason": "Kimi exportability and XPU paged-attention tests are unrelated." }, { "left": "pull_request:47143", "right": "pull_request:47298", "accept": false, - "reason": "NeuCodec model support vs MI450 image-loading fallback; different features." - }, - { - "left": "pull_request:47163", - "right": "pull_request:47313", - "accept": false, - "reason": "Paged-attention XPU test update vs DeepGEMM padding/layout fix; unrelated." + "reason": "NeuCodec support and MI450 image-loading fixes are unrelated model/runtime changes." } ] }, @@ -24550,7 +24580,7 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "451d31f104d858f9beb7be5ef442e98458843bb2590f5e48b25649834e4ce8e5", + "prepared_review_unit_hash": "6fc7c337b5add6b7a211a4fdc8d2030ca261496352d4f5214f50a2462c22394d", "component_budget": { "node_count": 2, "item_count": 2, @@ -24572,18 +24602,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs are unrelated feature additions in different model families: TimesFM covariate forecasting vs. DINOv3 image classification support. They should not be clustered as duplicates.", - "confidence": 0.98, + "summary": "The two pull requests are unrelated feature additions in different model areas, so this cluster should be split rather than deduplicated.", + "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR for this cluster: the PRs target different code paths, different models, and different user-facing capabilities, so there is no single underlying change to deduplicate around.", + "canonical_pr_reason": "No canonical PR: PR 40755 adds TimesFM covariate forecasting, while PR 41224 adds DINOv3 image-classification support; they touch different models, code paths, and user-facing features.", "best_issue_reason": null, - "best_pr_reason": "Neither PR is a good duplicate target for the other. If a representative must be chosen, PR 40755 is slightly more substantial and self-contained, but it still does not serve as a canonical match for PR 41224.", + "best_pr_reason": "No single PR is a valid global representative for this cluster because the changes are unrelated. If forced, 41224 is the more narrowly scoped PR, but it is not a duplicate of 40755.", "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different model domains and changes: TimesFM covariate forecasting vs. DINOv3 image classification support. They do not fix the same bug or implement the same change, and would not be merged into one PR." + "reason": "Reject: these PRs implement different features for different architectures (TimesFM forecasting with covariates vs DINOv3 image classification) and do not fix the same underlying bug or change the same code path." } ] }, @@ -24676,7 +24706,7 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "9918c4b3497e98547ae7ad92bebf4cee12bc9b79eab633e03c9814322a7356f6", + "prepared_review_unit_hash": "a2255e5043b2da60e886c9cafaf826baf6f06af807e60e9dfaf1d7365f4a8408", "component_budget": { "node_count": 2, "item_count": 2, @@ -24698,18 +24728,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related GitHub Actions maintenance PRs, but they fix different action-version upgrades in different workflow groups, so they are not duplicates.", - "confidence": 0.92, + "summary": "These are two separate workflow-maintenance PRs, not duplicates: one upgrades several GitHub Actions for Node 24 compatibility, while the other bumps docker/build-push-action in multiple Docker image workflows. They overlap in workflow files but change different actions and address different upgrade tracks.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #42975 is the better canonical representative because it is the broader, clearly scoped Actions compatibility update across many workflows, centered on Node 24-compatible action versions.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR #42975 is the best fit for the cluster because it covers the widest set of workflow updates and has the clearest umbrella purpose, while #42976 is a narrower Docker action bump.", + "best_pr_reason": "PR 42975 is the broader and more descriptive maintenance change, covering multiple core GitHub Actions upgrades tied to Node 24 compatibility.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Reject: both are workflow dependency bumps, but they change different actions and different code paths\u2014#42975 updates checkout/cache/upload-artifact for Node 24 compatibility, while #42976 upgrades docker/build-push-action in Docker image workflows. They could not plausibly be merged into one single PR without becoming a mixed maintenance change." + "reason": "Different underlying changes: 42975 updates checkout/cache/upload-artifact and related workflow actions for Node 24 compatibility, while 42976 only upgrades docker/build-push-action. Shared workflow files alone are not enough to treat them as the same PR." } ] }, @@ -24744,12 +24774,12 @@ "pull_request:43448|pull_request:45490", "pull_request:45490|pull_request:46562" ], - "prepared_review_unit_hash": "af658cbae01f02e459f297d66580d6daae4bab303ee5ffdef410ff5ab583dbc6", + "prepared_review_unit_hash": "5fa43a3276a9dc573c55a56a33117bed1c16b5c1d944be8157812760b58a3dd4", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 13046, + "serialized_chars": 13045, "estimated_input_tokens": 3262, "estimated_eval_tokens": 6780 }, @@ -24757,7 +24787,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 13046, + "serialized_chars": 13045, "estimated_input_tokens": 3262, "estimated_eval_tokens": 6780 }, @@ -24766,36 +24796,36 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are four unrelated model-addition PRs that only overlap on standard integration plumbing (docs, auto-mapping, tests). They do not appear to fix the same bug or implement the same feature, so no soft-duplicate PR merges are justified.", - "confidence": 0.97, + "summary": "These are four separate model-addition PRs (ViT NEPA, Molmo, CTSM, Step3p7). They share only generic Transformers scaffolding such as docs TOC entries and auto-registration files; no pair looks like the same underlying change or bug fix.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No issue is present. Among the PRs, #45490 is the best cluster representative because it has the highest discussion/review activity and is a substantial model-addition change touching core model registration and related integrations; however, it is not a duplicate of the others.", + "canonical_pr_reason": "No true canonical PR: the PRs target different models and different code paths. If one must be used as a representative, #46562 is the most complete and review-heavy, but it is not a duplicate of the others.", "best_issue_reason": null, - "best_pr_reason": "#45490 is the strongest single representative of the set: it is the most actively discussed PR and includes broad model/auto-registration work, making it the most central artifact if one must be chosen.", + "best_pr_reason": "#46562 is the strongest standalone PR in this set: it has the broadest model-specific implementation surface and the most review activity. It is still not a duplicate of the other PRs.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo. Shared docs/auto-mapping filenames are generic integration churn, not the same underlying change or code-path fix." + "reason": "Different model additions (ViT NEPA vs Molmo). Shared files are only generic docs/auto-registry scaffolding, not the same underlying change." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different architectures and model packages (ViT NEPA vs CTSM). Overlap is limited to common registry/docs files, which is not enough to treat them as the same PR." + "reason": "Different model families and implementations. Overlap is limited to shared Transformers plumbing, which is not enough to treat them as one PR." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo and CTSM are distinct model implementations with separate model-specific files and behavior; shared auto/docs changes are only broad subsystem overlap." + "reason": "Molmo and CTSM are unrelated new-model PRs; shared filenames are broad integration points only, not a common code-path fix." }, { "left": "pull_request:45490", "right": "pull_request:46562", "accept": false, - "reason": "CTSM and Step3p7 are different model additions with different code paths. They both update auto-registration/docs, but do not plausibly merge into one concrete bug fix or feature." + "reason": "CTSM and Step3p7 are distinct model additions with different architectures and files; the overlap is generic auto/docs infrastructure, not the same change." } ] }, @@ -24844,7 +24874,7 @@ "pull_request:46539|pull_request:46922", "pull_request:46886|pull_request:46922" ], - "prepared_review_unit_hash": "b0b5bfb606bc2a2b17463263838d3b754623823cfc055ea8d4cad455ce37603e", + "prepared_review_unit_hash": "e8c6033e70abff46d5a751de5173c9c174356f44a095b749a8820d3e2f159b57", "component_budget": { "node_count": 7, "item_count": 7, @@ -24866,108 +24896,108 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated PRs that happen to be superficially similar (test tweaks, CI plumbing, generation validation, FP8 loading, etc.). The only strong duplicate-like pair is the two YOLOS PRs, which both target the same device-mismatch issue in the same file and issue thread.", - "confidence": 0.95, + "summary": "This cluster is mostly unrelated PR noise. The only true duplicate-like pair is the two YOLOS fixes, which target the same device-mismatch bug in the same file and share the same issue target; 46922 is the more complete representative.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:46922 is the best representative: it has the clearest concrete bug fix, the strongest issue linkage, and it subsumes the narrower YOLOS device-placement fix in pull_request:46886.", + "canonical_pr_reason": "PR 46922 is the best canonical PR: it directly fixes the YOLOS device-mismatch path in `forward()` and also includes the `_no_split_modules` adjustment, making it the more complete version of the same underlying fix than 46886.", "best_issue_reason": null, - "best_pr_reason": "pull_request:46922 is the most substantive and best-scoped PR in the cluster; the rest are either unrelated or too generic to serve as a stable canonical representative.", + "best_pr_reason": "PR 46922 is the strongest standalone candidate because it addresses the concrete YOLOS runtime bug at the actual failing code path and appears to subsume the narrower change in 46886.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Both are test-focused Whisper/CLIP/ViT cleanup PRs, but they touch different models and different test failures; no shared underlying bug." + "reason": "Both are test-focused, but they touch different models and different failures; the similarity is only superficial." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper tokenizer/test fix vs FP8 dequant/loading fix; completely different code paths and failure modes." + "reason": "Whisper test cleanup is unrelated to the FP8 dequant/loading fix; no shared code path or concrete bug." }, { "left": "pull_request:43291", "right": "pull_request:46380", "accept": false, - "reason": "Whisper test fix and generation watermarking validation fix are unrelated; only both are small correctness patches." + "reason": "Whisper tokenization/tests and generation watermark validation are different subsystems and different bugs." }, { "left": "pull_request:43291", "right": "pull_request:46922", "accept": false, - "reason": "Whisper decode/test changes are unrelated to YOLOS device-mismatch handling." + "reason": "Whisper decoding/tests do not overlap with the YOLOS device-mismatch fix." }, { "left": "pull_request:43665", "right": "pull_request:46539", "accept": false, - "reason": "Both are test/infra-oriented, but one is model unit-test edits and the other is CI environment shimming; not the same bug." + "reason": "These are unrelated test/workflow changes in different model areas and do not appear to fix the same issue." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "CLIP/ViT test edits do not match the FP8 loader/runtime fix; different subsystems and issues." + "reason": "One is ad hoc test changes; the other is a genuine FP8 dequantization/loading fix. Not the same bug." }, { "left": "pull_request:43665", "right": "pull_request:46380", "accept": false, - "reason": "Model test tweaks vs generation config validation; no shared concrete defect." + "reason": "Different models and different failure modes; only broad 'fix tests' similarity." }, { "left": "pull_request:43665", "right": "pull_request:46922", "accept": false, - "reason": "The CLIP/ViT test PR is unrelated to YOLOS runtime device placement." + "reason": "The CLIP/ViT test edits are unrelated to YOLOS inference/device placement behavior." }, { "left": "pull_request:44815", "right": "pull_request:46539", "accept": false, - "reason": "FP8 quantization/dequantization changes and CI memory/device-map plumbing address different problems." + "reason": "FP8 quantization/loading changes and Phimoe CI/device-memory test infrastructure are unrelated." }, { "left": "pull_request:44815", "right": "pull_request:46380", "accept": false, - "reason": "FP8 kernel/loading fix vs generation watermarking validation; unrelated code paths." + "reason": "They address different issues: FP8 dequantization vs generation config validation." }, { "left": "pull_request:44815", "right": "pull_request:46922", "accept": false, - "reason": "Both involve model internals, but the FP8 dequant issue is unrelated to YOLOS device mismatch." + "reason": "Different model families and different bug classes; no plausible single merged fix." }, { "left": "pull_request:46380", "right": "pull_request:46539", "accept": false, - "reason": "Generation config validation and CI model-job environment changes are different layers and not the same underlying bug." + "reason": "Generation config watermark validation and Phimoe integration-test CI setup are separate problems." }, { "left": "pull_request:46380", "right": "pull_request:46922", "accept": false, - "reason": "Generation watermarking validation does not overlap with YOLOS device placement or auto device_map behavior." + "reason": "These PRs touch different subsystems and fix different code paths." }, { "left": "pull_request:46539", "right": "pull_request:46922", "accept": false, - "reason": "CI runner CPU-memory limiting/patching is unrelated to YOLOS model-side device mismatch." + "reason": "CI memory limiting for Phimoe tests is unrelated to YOLOS device placement." }, { "left": "pull_request:46886", "right": "pull_request:46922", "accept": true, - "reason": "Both target the same YOLOS device-mismatch issue in the same file and under the same issue target; 46922 extends the fix with the same code-path concern." + "reason": "Both PRs target the same YOLOS device-mismatch bug in `src/transformers/models/yolos/modeling_yolos.py` and share the same issue target; 46922 effectively subsumes the narrower fix in 46886." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft-edge pair, 46886/46922, clearly looks like the same YOLOS bugfix family in the same file and issue thread, with 46922 subsuming the narrower fix. The other pair rejections are consistent with the diffs and do not overclaim overlap." + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft-edge pair, 46886/46922, is clearly mergeable: both target the same YOLOS device-mismatch issue in the same file and share the same issue target, with 46922 subsuming the narrower `_no_split_modules` change. The remaining rejections are reasonably supported by the diffs and titles as unrelated or at most superficially similar." }, "evaluator_used": true, "retried": false, @@ -24997,12 +25027,12 @@ "pull_request:43488|pull_request:43532", "pull_request:43532|pull_request:46747" ], - "prepared_review_unit_hash": "ff8058a9677d8976006f23f7be67ff721beb0b78e6a25b1fdd107bc5f38eb4bd", + "prepared_review_unit_hash": "bfb47689d01057127d3093d60b2eeaca6408ae9a09c0758c51281759fe6805bd", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 6347, + "serialized_chars": 6348, "estimated_input_tokens": 1587, "estimated_eval_tokens": 3430 }, @@ -25010,7 +25040,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 6347, + "serialized_chars": 6348, "estimated_input_tokens": 1587, "estimated_eval_tokens": 3430 }, @@ -25019,30 +25049,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three PRs are only superficially similar as '[don't merge]' repo/CI experiments; they do not appear to fix the same concrete bug or implement the same change.", - "confidence": 0.91, + "summary": "These are three separate '[don't merge]' test PRs with different concrete changes: packaging/dependency setup, assorted CI/comment workflow changes, and a PR security-gate workflow change. They are not duplicates of the same underlying bug or change.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No clear canonical PR: the three changes target different areas (packaging/test deps, self-comment CI/workflow plumbing, and PR security-gate workflow).", + "canonical_pr_reason": "PR 46747 is the cleanest single-purpose change: a small, focused workflow edit in one file. It is the best representative only in the sense of being the most self-contained artifact here, not because it duplicates the others.", "best_issue_reason": null, - "best_pr_reason": "PR 43532 is the broadest CI/workflow change and the closest central anchor for the superficial similarity, but it is still not a duplicate of the other PRs.", + "best_pr_reason": "PR 46747 is the most focused and least noisy of the three, with one workflow file and a single code-path change.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:46747", "accept": false, - "reason": "43488 changes setup/version/test dependencies, while 46747 only switches a security-gate workflow trigger; different code paths and goals." + "reason": "Different changes: 43488 edits setup/dependency metadata and a test file, while 46747 changes the PR security-gate workflow trigger. Same broad 'test' theme, but not the same underlying fix." }, { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "43488 is a packaging/bot-format test PR, whereas 43532 rewires self-comment CI and related workflow utilities; the shared Vit test file is incidental." + "reason": "43488 is about package/version/dependency setup plus a Vit test tweak; 43532 is about CI/comment workflows and related test/support files. The overlap is incidental, not a shared concrete bug or change." }, { "left": "pull_request:43532", "right": "pull_request:46747", "accept": false, - "reason": "Both touch workflow files, but 43532 is about self-comment CI behavior and 46747 is about PR security-gate triggering; not the same concrete fix." + "reason": "Both touch GitHub Actions, but they modify different workflows for different purposes: self-comment CI vs security gate. They are not plausibly the same mergeable change." } ] }, @@ -25072,12 +25102,12 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "aa5eedaa797a8586f2ebe52977fef8f3eb0929b20133ee7aafad3890a0d35e92", + "prepared_review_unit_hash": "6aefa12ef70689b4feec190b7038e3d305fde925897c7de55bd71ac6aa38f10e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6081, + "serialized_chars": 6082, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -25085,7 +25115,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6081, + "serialized_chars": 6082, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -25094,18 +25124,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate feature PRs for different pipelines: Promptable Concept Segmentation vs Promptable Visual Segmentation. They share scaffolding and documentation patterns, but they target different tasks, models, and test suites, so they should not be deduplicated.", - "confidence": 0.96, + "summary": "Two related but distinct PRs adding different segmentation pipelines. They share infrastructure/doc updates, but target different tasks, model integrations, and code paths.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 43613 is the better representative of the cluster because it is the broader, more complete promptable-visual-segmentation implementation and includes multiple model/test integrations; 43612 is a separate promptable-concept-segmentation pipeline.", "best_issue_reason": null, - "best_pr_reason": "No single canonical PR stands out because the two PRs implement different features. If forced to pick one representative, PR #43613 is slightly broader in model coverage, but it is not a duplicate of #43612.", + "best_pr_reason": "PR 43613 is the strongest standalone change: it adds a full visual segmentation pipeline, docs, auto-mapping, and tests across several relevant models. It is more substantial and coherent than 43612.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Different underlying changes: concept segmentation and visual segmentation are distinct pipelines with different model support and tests. Shared docs/registration files are boilerplate, not evidence of the same fix." + "reason": "Both are segmentation-related and share boilerplate docs/registry updates, but they implement different pipelines for different tasks and model sets. They are not the same underlying change and should not be merged as duplicates." } ] }, @@ -25135,7 +25165,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "efaa921e64cbdda1a5ba3bffc63f1cdad89d202a33d2ccf9123abf2055057453", + "prepared_review_unit_hash": "1184515d962f1fc5dff8050a36988813ec59a608197996093e76c996a431c557", "component_budget": { "node_count": 2, "item_count": 2, @@ -25157,18 +25187,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two distinct Trainer PRs touching the same file but fixing different extension points: one changes gradient-accumulation loss scaling, the other refactors signature-column filtering for custom trainers/PEFT models.", + "summary": "These PRs both touch `trainer.py` and add override hooks for custom trainers, but they address different code paths: gradient-accumulation loss scaling vs. dataloader column/signature filtering. They are not the same underlying change and should not be deduplicated.", "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No canonical PR: the two pull requests fix different trainer behaviors and are not mergeable into one coherent change.", "best_issue_reason": null, - "best_pr_reason": "No true duplicate here; if a single PR must anchor the cluster, 43651 is the more direct bugfix-style change, but it is not a substitute for 43663.", + "best_pr_reason": "No single best PR for a duplicate cluster here; each PR is a separate, focused improvement to a different trainer subsystem.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Different underlying code paths and behaviors: 43651 changes training loss scaling during gradient accumulation, while 43663 extracts signature-column computation for column filtering. Shared file alone is not enough to treat them as the same change." + "reason": "Different bugs/changes: 43651 adjusts loss scaling during training_step, while 43663 refactors signature-column filtering for dataset column pruning. Shared file alone is not enough." } ] }, @@ -25198,7 +25228,7 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "cb4985a67e17b0ae13dbbe8dd73c66d9a4ef331e603d41703b494411a2c4b806", + "prepared_review_unit_hash": "6702be00d97ef3b6aa42c9cee0dae55c1eb745916f771019ac1e88893fb3296d", "component_budget": { "node_count": 2, "item_count": 2, @@ -25220,28 +25250,25 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs appear to address the same import-time TypeAdapter NameError in src/transformers/cli/serve.py; 43842 is the narrower, more explicit fix, while 43656 includes broader typing cleanup.", - "confidence": 0.79, + "summary": "Two open PRs touch the same file and both address annotation-related NameErrors in the CLI, but they target different fixes and different issues: one is broader future-annotations/type-hint cleanup, the other is a narrow optional-pydantic TypeAdapter fix.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "43842 is the more focused fix for the serve.py import-time NameError when pydantic is absent, with less unrelated typing churn.", + "canonical_pr_reason": "PR #43842 is the cleaner canonical PR: it directly fixes the TypeAdapter NameError when pydantic is absent, with a focused change that matches its issue and title.", "best_issue_reason": null, - "best_pr_reason": "43842 is the best representative because it directly targets the runtime NameError and is the least invasive change.", + "best_pr_reason": "PR #43842 is the most representative of the cluster because it is a precise, self-contained fix for the concrete runtime failure in `src/transformers/cli/serve.py`.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", - "accept": true, - "reason": "Same concrete code path in src/transformers/cli/serve.py: both fix TypeAdapter annotation/import-time NameError behavior, just with different implementation styles." + "accept": false, + "reason": "Reject: although both are in `src/transformers/cli/serve.py` and concern annotation errors, #43656 is a broader `from __future__ import annotations` / type-hint cleanup tied to issue 43576, while #43842 specifically fixes a `TypeAdapter` NameError when pydantic is not installed for issue 43824. They are not the same concrete bug or change." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: both PRs touch src/transformers/cli/serve.py and the diffs support the claim that they are alternative fixes for the same TypeAdapter import-time NameError. The characterization of 43842 as narrower and 43656 as broader typing cleanup is conservative, and the two changes look mergeable into a single PR for the same bugfix." - }, - "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 }, @@ -25362,8 +25389,8 @@ "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:44071", "pull_request:44066|pull_request:44068", "pull_request:44018|pull_request:44066", "pull_request:44019|pull_request:44722", @@ -25371,8 +25398,8 @@ "pull_request:44019|pull_request:44085", "pull_request:44018|pull_request:44019", "pull_request:44019|pull_request:44071", - "pull_request:44068|pull_request:44722", "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", @@ -25427,340 +25454,388 @@ "pull_request:44044|pull_request:44154", "pull_request:44024|pull_request:44076" ], - "prepared_review_unit_hash": "997579d6316225db5b458bf68587cef720fa877cb71d1cbe8b095cc4491bf260", + "prepared_review_unit_hash": "8fda3c0773e46d177e288178f7bd4b14b788758cb56e1319c11765828f215589", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84513, - "estimated_input_tokens": 21129, - "estimated_eval_tokens": 42514 + "serialized_chars": 84505, + "estimated_input_tokens": 21127, + "estimated_eval_tokens": 42510 }, "budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 72, - "serialized_chars": 76945, - "estimated_input_tokens": 19237, - "estimated_eval_tokens": 38730 + "serialized_chars": 76938, + "estimated_input_tokens": 19235, + "estimated_eval_tokens": 38726 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of model-specific migrations to standardized output tracing (`capture_outputs` / `can_return_tuple`) under the same umbrella issue. Only same-model pairs in GPT-J and GPT-Neo look like true duplicates; the rest are parallel refactors in different code paths and should stay separate.", - "confidence": 0.88, + "summary": "This cluster is mostly a broad wave of model-specific output-tracing refactors, so most pairs are not true duplicates. The only clear duplicate-style overlaps are the GPT-J/CodeGen refactor set around 44066/44085/44722 and the GPT-Neo pair 44018/44068.", + "confidence": 0.82, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44722 is the strongest representative of the GPT-J/CodeGen duplicate family: it is a later, multi-file refactor of the same concrete output-tracing path and subsumes the near-duplicate GPT-J-only variants.", + "canonical_pr_reason": "44722 is the best representative of the duplicate subset: it targets the same GPT-J/CodeGen output-tracing code path as 44066/44085, touches the same files, and looks like the most complete version of that concrete change.", "best_issue_reason": null, - "best_pr_reason": "PR 44722 best represents one real duplicate set in the cluster because it covers the same GPT-J/CodeGen tracing migration as the other accepted duplicates and is the most complete version of that concrete change.", + "best_pr_reason": "Best PR for the cluster's main duplicate set. It most cleanly represents the shared GPT-J/CodeGen output-tracing refactor, while the rest of the cluster is mostly separate model-by-model work.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different models and code paths; same tracking issue alone is not enough to treat them as duplicates." + "reason": "Different model families and code paths; same tracking theme only." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Both are output-tracing refactors, but for unrelated model implementations, so they are not the same underlying change." + "reason": "CVT/FNet vs DeBERTa V2; not the same underlying change." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Both refactor the same GPT-J output-tracing code path, so they could plausibly be merged into one PR." + "reason": "Both are GPT-J output-tracing refactors on the same core code path; 44066/44085 overlap materially." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Different model families and implementations; this is the same migration pattern, not the same bug/change." + "reason": "ResNet/RegNet vs EfficientNet; separate model implementations." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "EfficientNet and GPT-J/CodeGen are different concrete code paths, so this is not a duplicate." + "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "Unrelated models; they share the umbrella refactor theme but not the same code-path change." + "reason": "Different models and implementation paths; only the refactor theme is shared." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa are separate implementations; not the same underlying fix." + "reason": "MobileNetV2 vs DeBERTa V2; no shared concrete bug or change." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J/CodeGen vs MGP-STR; same broad pattern, different concrete code path." - }, - { - "left": "pull_request:44066", - "right": "pull_request:44071", - "accept": false, - "reason": "These touch different model code paths, so they are not mergeable duplicates." + "reason": "GPT-J/CodeGen vs MGP-STR; same general refactor pattern but not the same code path." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Both are GPT-Neo output-tracing refactors for the same model code path and look mergeable as one PR." + "reason": "Both target GPT-Neo output tracing in the same model file/test area; plausible duplicate or direct revision of the same change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen and MPT are unrelated model implementations." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are different model implementations, despite similar refactor style." + "reason": "GPT-J/CodeGen vs GPT-Neo; similar refactor pattern, different code path." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "Different models and different code paths; same umbrella issue only." + "reason": "GPT-Neo vs GPT-J/CodeGen; not the same underlying model change." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet and GPT-J/CodeGen are unrelated concrete changes." + "reason": "ResNet and GPT-J/CodeGen are separate model families." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR and GPT-J/CodeGen are different model-specific changes." + "reason": "MGP-STR and GPT-J/CodeGen do not share a code path." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet and GPT-J are separate code paths; not duplicates." + "reason": "ResNet vs GPT-J; same theme only, not the same fix." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are different model implementations, so these should not be merged." + "reason": "GPT-Neo and ResNet are unrelated implementations." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "Different models and concrete paths; same output-tracing theme only." + "reason": "ResNet/RT-DETR backbone work vs MPT; no concrete overlap." }, { "left": "pull_request:44068", - "right": "pull_request:44722", + "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are different model families, so this is not the same change." + "reason": "GPT-Neo and Flaubert are different model code paths." }, { "left": "pull_request:44068", - "right": "pull_request:44116", + "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo vs Flaubert/XLM are unrelated model code paths." + "reason": "GPT-Neo vs GPT-J/CodeGen; related pattern, but not the same model change." }, { "left": "pull_request:44018", "right": "pull_request:44116", "accept": false, - "reason": "Different models and implementations; not a duplicate." + "reason": "GPT-Neo and Flaubert are unrelated." }, { "left": "pull_request:44071", "right": "pull_request:44722", "accept": false, - "reason": "MPT and GPT-J/CodeGen are separate concrete changes." + "reason": "MPT vs GPT-J/CodeGen are different implementations." }, { "left": "pull_request:44018", "right": "pull_request:44085", "accept": false, - "reason": "GPT-Neo and GPT-J are different code paths, even though both use output-tracing decorators." + "reason": "GPT-Neo vs GPT-J; same refactor theme, different code path." }, { "left": "pull_request:44018", "right": "pull_request:44086", "accept": false, - "reason": "Different model families; same refactor style is not enough." + "reason": "GPT-Neo and MGP-STR are unrelated model implementations." }, { - "left": "pull_request:44059", + "left": "pull_request:44068", "right": "pull_request:44086", "accept": false, - "reason": "GPT2 and MGP-STR are unrelated concrete implementations." + "reason": "GPT-Neo vs MGP-STR; not the same underlying change." }, { - "left": "pull_request:44098", - "right": "pull_request:44101", + "left": "pull_request:44071", + "right": "pull_request:44086", "accept": false, - "reason": "ViLT and Flaubert/XLM are different model code paths." + "reason": "MPT and MGP-STR touch different models and logic." }, { - "left": "pull_request:44017", - "right": "pull_request:44161", + "left": "pull_request:44085", + "right": "pull_request:44086", "accept": false, - "reason": "Segformer and LongT5 are unrelated changes." + "reason": "GPT-J and MGP-STR are unrelated code paths." }, { - "left": "pull_request:44017", - "right": "pull_request:44073", + "left": "pull_request:44059", + "right": "pull_request:44086", "accept": false, - "reason": "Different models and different code paths; not a duplicate." + "reason": "GPT-2 vs MGP-STR; shared refactor theme only." }, { - "left": "pull_request:44017", - "right": "pull_request:44722", + "left": "pull_request:44098", + "right": "pull_request:44101", "accept": false, - "reason": "Segformer and GPT-J/CodeGen are separate model-specific migrations." + "reason": "ViLT vs Flaubert/XLM; separate model families." }, { - "left": "pull_request:44068", - "right": "pull_request:44071", + "left": "pull_request:44059", + "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo and MPT are different model families." + "reason": "GPT-2 and GPT-J/CodeGen are different model implementations." }, { - "left": "pull_request:43997", - "right": "pull_request:44073", + "left": "pull_request:44017", + "right": "pull_request:44071", "accept": false, - "reason": "RegNet and VisualBert are unrelated concrete implementations." + "reason": "SegFormer and MPT are unrelated model paths." }, { "left": "pull_request:44074", "right": "pull_request:44101", "accept": false, - "reason": "TextNet and Flaubert/XLM are separate model code paths." + "reason": "TextNet and Flaubert/XLM are different code paths." }, { "left": "pull_request:44056", "right": "pull_request:44161", "accept": false, - "reason": "MPNet and LongT5 are different model implementations." + "reason": "MPNet and LongT5 are unrelated model implementations." }, { "left": "pull_request:43997", "right": "pull_request:44018", "accept": false, - "reason": "RegNet and GPT-Neo are different concrete changes." + "reason": "RegNet and GPT-Neo are unrelated." }, { "left": "pull_request:43997", "right": "pull_request:44161", "accept": false, - "reason": "RegNet and LongT5 are unrelated model-specific refactors." + "reason": "RegNet and LongT5 are separate model code paths." }, { "left": "pull_request:44074", "right": "pull_request:44161", "accept": false, - "reason": "TextNet and LongT5 are separate code paths." + "reason": "TextNet and LongT5 are unrelated models." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo vs GPT-2; similar refactor pattern, not the same change." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "Flaubert/XLM and LongT5 are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet vs MPNet; no concrete overlap." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are unrelated model code paths." }, { "left": "pull_request:44010", "right": "pull_request:44071", "accept": false, - "reason": "SqueezeBert and MPT are different model implementations." + "reason": "SqueezeBert and MPT are different implementations." }, { "left": "pull_request:44018", "right": "pull_request:44129", "accept": false, - "reason": "GPT-Neo and SpeechT5 are unrelated concrete paths." + "reason": "GPT-Neo and SpeechT5 are unrelated model paths." }, { "left": "pull_request:44010", "right": "pull_request:44073", "accept": false, - "reason": "SqueezeBert and VisualBert are separate model changes." + "reason": "SqueezeBert and VisualBert are different code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa V2 and TextNet are unrelated model implementations." }, { "left": "pull_request:44010", "right": "pull_request:44018", "accept": false, - "reason": "Different model families; not the same bug/change." + "reason": "SqueezeBert and GPT-Neo are not the same change." }, { - "left": "pull_request:44010", - "right": "pull_request:44044", + "left": "pull_request:44017", + "right": "pull_request:44161", "accept": false, - "reason": "SqueezeBert and DeBERTa are unrelated model-specific refactors." + "reason": "SegFormer and LongT5 are unrelated." }, { - "left": "pull_request:44044", - "right": "pull_request:44074", + "left": "pull_request:44073", + "right": "pull_request:44161", "accept": false, - "reason": "DeBERTa and TextNet are different concrete code paths." + "reason": "VisualBert and LongT5 are different model paths." }, { - "left": "pull_request:44024", - "right": "pull_request:44076", + "left": "pull_request:44073", + "right": "pull_request:44101", "accept": false, - "reason": "FocalNet and ImageGPT are separate models; similarity is only at the refactor pattern level." + "reason": "VisualBert vs Flaubert/XLM; only broad output-tracing theme matches." }, { - "left": "pull_request:44066", - "right": "pull_request:44722", - "accept": true, - "reason": "Both changes target the same GPT-J/CodeGen output-tracing code path and are plausible duplicates." + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are unrelated models." }, { - "left": "pull_request:44085", - "right": "pull_request:44722", - "accept": true, - "reason": "Both are GPT-J output-tracing refactors for the same underlying code path." + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert/XLM are different code paths." }, { - "left": "pull_request:44085", + "left": "pull_request:44059", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are separate refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are unrelated model implementations." + }, + { + "left": "pull_request:44071", "right": "pull_request:44129", "accept": false, - "reason": "GPT-J and SpeechT5 are different model implementations." + "reason": "MPT and SpeechT5 are different model code paths." }, { "left": "pull_request:43996", - "right": "pull_request:44007", + "right": "pull_request:44161", "accept": false, - "reason": "Different model families; the shared issue target is only a broad tracking umbrella." + "reason": "CVT/FNet and LongT5 are unrelated." }, { "left": "pull_request:43996", - "right": "pull_request:44129", + "right": "pull_request:44017", + "accept": false, + "reason": "CVT/FNet vs SegFormer; separate model code paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", "accept": false, - "reason": "CVT/FNet and SpeechT5 are unrelated concrete changes." + "reason": "SqueezeBert and DeBERTa V2 are unrelated implementations." }, { "left": "pull_request:44044", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTa and SpeechT5 are different code paths." + "reason": "DeBERTa V2 and SpeechT5 are different model paths." }, { "left": "pull_request:44044", - "right": "44154", + "right": "pull_request:44154", "accept": false, - "reason": "DeBERTa and ViTS are separate model-specific changes." + "reason": "DeBERTa V2 and VITS do not share a concrete fix." }, { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "FocalNet and ImageGPT are different model code paths, so this is not a true duplicate." + "reason": "FocalNet and ImageGPT are unrelated model code paths." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative: it correctly treats the cluster as mostly parallel model-specific output-tracing refactors, and it only elevates same-model GPT-J / GPT-Neo pairs as duplicates. The canonical choice of PR 44722 is supported by the packet as the later, more complete GPT-J/CodeGen variant. No soft-edge verdict looks overstated or non-mergeable for a single bugfix." + "feedback": "The cluster summary is grounded in the packet: it correctly centers the GPT-J/CodeGen duplicate set and treats the rest as mostly separate model-specific refactors. The accepted soft-edge pairs are plausible and mergeable as single bugfix PRs, and the rejected pairs are stated conservatively without overclaiming duplication." }, "evaluator_used": true, "retried": false, @@ -25795,20 +25870,20 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "ff6ff97b22d8b5cee37ad80e1014e7df4abf3188cf1209b018de736ba4200be7", + "prepared_review_unit_hash": "a46aabcb8bd3c72f7769b93be898d654c3e24bbff3ca4dcc7eb76e295d5426f0", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84513, - "estimated_input_tokens": 21129, - "estimated_eval_tokens": 42514 + "serialized_chars": 84505, + "estimated_input_tokens": 21127, + "estimated_eval_tokens": 42510 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 5, - "serialized_chars": 13068, + "serialized_chars": 13067, "estimated_input_tokens": 3267, "estimated_eval_tokens": 6790 }, @@ -25817,42 +25892,42 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a set of similar but model-specific PRs migrating different Transformers models to standardized output tracing/capture patterns. They share a theme, but each touches a different code path and should not be deduplicated as the same underlying change.", - "confidence": 0.9, + "summary": "These PRs are all model-specific migrations toward standardized output tracing, but they affect different architectures and code paths. They are related in theme, not duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the clearest canonical representative: it is explicitly issue-linked, self-contained, and shows the output-tracing migration in a concrete model implementation.", + "canonical_pr_reason": "PR #44073 is the best representative of the cluster: it shows the core standardized output-tracing refactor clearly, adds the model-specific output-capture mapping, and uses the new capture_outputs/can_return_tuple pattern in a canonical way.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the strongest standalone example of this refactor family because it has an explicit target issue and a clear standardized output-capture migration in SpeechT5.", + "best_pr_reason": "PR #44073 is the strongest single exemplar of the shared refactor theme, with explicit output-capture plumbing and a clean model migration. The others are model-specific variants rather than the same change.", "soft_edge_verdicts": [ { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "Same broad refactor theme, but VisualBert and TextNet are different model implementations with distinct forward paths; not the same concrete change." + "reason": "Same broad migration pattern, but different models and concrete code paths (VisualBert vs TextNet). This is not one fix that could be merged into a single PR." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "Both are output-tracing migrations, but MPNet vs VisualBert are separate model-specific refactors and could not be merged as one PR." + "reason": "Both are output-tracing refactors, but for different architectures (MPNet vs VisualBert) with different layer mappings and implementation details; not the same underlying change." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "Different models and code paths; the similarity is only at the pattern level, not the underlying bug/change." + "reason": "Shared refactor theme only. MPNet and TextNet have distinct model internals and output-capture wiring, so these are not duplicates." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model-specific migrations; they do not fix the same concrete issue." + "reason": "Both migrate to the new tracing helpers, but RegNet and SqueezeBert are unrelated model implementations and the diffs do not indicate the same concrete bug or patch." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 touch different output/capture implementations and shapes; similar intent, but not one mergeable change." + "reason": "SqueezeBert\u2019s decorator migration and SpeechT5\u2019s output/caching signature changes are different code-path changes in different models; they should not be merged as one PR." } ] }, @@ -25932,12 +26007,12 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "573dd81e8943423d0355bffa838b10d5fcb838ca4f8f2cbff07531cc8836586e", + "prepared_review_unit_hash": "45a6cdceef7fe949a867ae2aa714569765ee692e50e5a524d5549ebdb0a99b02", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40183, + "serialized_chars": 40182, "estimated_input_tokens": 10046, "estimated_eval_tokens": 20348 }, @@ -25945,7 +26020,7 @@ "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40183, + "serialized_chars": 40182, "estimated_input_tokens": 10046, "estimated_eval_tokens": 20348 }, @@ -25954,252 +26029,246 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is a set of independent model-specific output-tracing refactors. They share a common helper/decorator pattern, but they do not look like duplicates of the same underlying bug or change.", - "confidence": 0.97, + "summary": "This cluster is a set of model-specific pull requests that all refactor output tracing / tuple-return behavior, but they target different architectures and code paths. I found no true duplicate PR pairs among the soft candidates.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44003 is the best representative of the cluster: it is one of the larger output-tracing refactors and spans two closely related implementations, making it a strong exemplar of the shared refactor pattern.", + "canonical_pr_reason": "PR 44003 is the best representative of the cluster: it covers a broader, more central output-tracing refactor across the related Mamba/FalconMamba implementations, making it more representative than the single-model wrapper changes elsewhere.", "best_issue_reason": null, - "best_pr_reason": "PR 44003 best captures the cluster's core pattern, but the rest are separate model-specific applications rather than the same mergeable change.", + "best_pr_reason": "PR 44003 is the strongest single example of the common theme because it changes the core tracing pattern in closely related model code, rather than only adjusting one isolated model wrapper.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Same refactor theme, but timm_backbone and upernet are unrelated model code paths and this is not one concrete duplicate change." + "reason": "Different model families and files; both are output-tracing refactors, but not the same concrete change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Both add output-tracing plumbing, but they target different model families and cannot plausibly be one PR." + "reason": "Different architectures (`timm_backbone` vs `vision_text_dual_encoder`); same theme, not the same bug or patch." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Different architectures and forward implementations; shared helper usage is not enough to make these duplicates." + "reason": "`timm_backbone` and `univnet` are unrelated code paths; too broad to be duplicates." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "These are separate model implementations with different output-capturing changes, not the same code-path fix." + "reason": "Both are tracing refactors, but they affect different model families and distinct implementations." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Different models and distinct forward logic; only the broad refactor pattern is shared." + "reason": "Separate models and separate forward paths; no shared concrete fix." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNetV1 and UnivNet are unrelated model families, so these are not the same underlying change." + "reason": "`mobilenet_v1` and `univnet` are unrelated; similarity is only at the refactor theme level." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "They both refactor outputs, but in different model stacks and different files, so not duplicates." + "reason": "Different models and different output types; not mergeable as one PR." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "Unrelated model implementations; same tracing refactor pattern does not make this a mergeable duplicate." + "reason": "Different model families (`mobilenet_v1` vs `depth_anything`); same broad refactor pattern, not the same change." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "Different encoder-decoder wrappers and separate code paths; not one concrete bug or change." + "reason": "`mobilenet_v1` and `vision_encoder_decoder` touch unrelated code paths." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "MobileNetV1 vs speech_encoder_decoder are unrelated; these should not be deduped." + "reason": "Different model implementations; no shared bug or patch." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder and UnivNet are different model families and different forward paths." + "reason": "`vision_text_dual_encoder` and `univnet` are unrelated models, so this is only thematic similarity." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Same broad refactor theme only; the underlying code paths are unrelated." + "reason": "Different architectures and files; not the same concrete output-tracing change." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "These touch separate model implementations and are not the same concrete fix." + "reason": "`vision_text_dual_encoder` and `falcon_mamba/mamba` are distinct code paths; reject as duplicate." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different models and different tracing changes; not plausibly one PR." + "reason": "Different model families; both refactor outputs, but not the same fix." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are unrelated models; the overlap is only in helper style." + "reason": "`univnet` and `upernet` are unrelated implementations." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Different architectures and code paths; not a duplicate bug or change." + "reason": "Different model families and different forward signatures; not one underlying change." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "These are model-specific refactors in unrelated components, so they should remain separate." + "reason": "Separate model code paths; similarity is too coarse for duplicate triage." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "UperNet and RWKV are unrelated implementations; same output-tracing pattern is insufficient." + "reason": "`upernet` and `rwkv` are unrelated models with different output-capture logic." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "Different model families and separate forward logic; not mergeable as one change." + "reason": "Different architectures and modules; no plausible single merged PR." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Falcon/Mamba vs SuperPoint are unrelated code paths; this is only a thematic similarity." + "reason": "`falcon_mamba/mamba` and `superpoint` are unrelated models; only the refactor style matches." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Both are refactors around model outputs, but codegen and mamba are separate implementations and not the same fix." + "reason": "`mamba` family versus `codegen`; different code paths and different API changes." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Different model families and forward paths; shared helper imports do not indicate duplication." + "reason": "Different models and different refactor scopes; not the same underlying change." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet and CodeGen are unrelated models with distinct output semantics." + "reason": "`univnet` and `codegen` are unrelated; this is not a duplicate pair." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet and CodeGen are different subsystems; not the same underlying bug." - }, - { - "left": "pull_request:44000", - "right": "pull_request:44004", - "accept": false, - "reason": "These are separate model implementations; the similarity is only a generic refactor pattern." + "reason": "Different model code paths and different output contracts." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "Depth Anything and Vision-Encoder-Decoder are unrelated model code paths, so this is not a duplicate." + "reason": "`depth_anything` and `vision_encoder_decoder` are separate model families; same meta-pattern only." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "Different model families and different forward signatures; not the same change." + "reason": "`depth_anything` and `speech_encoder_decoder` are not the same concrete bug or change." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "Depth Anything and SuperPoint are unrelated implementations; do not merge as duplicates." + "reason": "Different models; one is depth estimation, the other keypoint detection." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Different model families and different output-capturing behavior; not the same concrete bug." + "reason": "Different model implementations; no shared code path." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "Depth Anything and DPR are separate model stacks; the overlap is only in refactor style." + "reason": "`depth_anything` and `dpr` are unrelated models." }, { "left": "pull_request:44026", - "right": "pull_request:44029", + "right": "pull_request:44027", "accept": false, - "reason": "Vision-Encoder-Decoder and RWKV are unrelated code paths, so this is not one mergeable fix." + "reason": "Two different encoder-decoder wrappers; similar refactor, but not the same PR." }, { "left": "pull_request:44026", - "right": "pull_request:44027", + "right": "pull_request:44028", "accept": false, - "reason": "Vision and speech encoder-decoder wrappers are different implementations; not a duplicate change." + "reason": "`vision_encoder_decoder` and `superpoint` address different subsystems." }, { "left": "pull_request:44026", - "right": "pull_request:44028", + "right": "pull_request:44029", "accept": false, - "reason": "Different model families and separate forward logic; only the high-level refactor theme matches." + "reason": "Different models and different output-tracing code paths." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Vision-Encoder-Decoder and DPR are distinct components; no shared underlying bug." + "reason": "`vision_encoder_decoder` and `dpr` are separate model families." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "Speech-Encoder-Decoder and SuperPoint are unrelated models with different output paths." + "reason": "`speech_encoder_decoder` and `superpoint` are unrelated changes." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Different model families and implementation details; not the same change." + "reason": "Different architectures; no concrete overlap beyond generic refactoring." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "Speech-Encoder-Decoder and DPR are separate code paths; same theme, not duplicate." + "reason": "`speech_encoder_decoder` and `dpr` are different code paths." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are unrelated model implementations, so these should not be merged." + "reason": "`superpoint` and `rwkv` are unrelated models." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "SuperPoint and DPR are different components with different forward semantics." + "reason": "Different model families and different tracing implementations." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are unrelated model families; the commonality is only the tracing refactor pattern." + "reason": "`rwkv` and `dpr` are unrelated; not mergeable as one PR." } ] }, @@ -26229,12 +26298,12 @@ "soft_pairs": [ "pull_request:44662|pull_request:45401" ], - "prepared_review_unit_hash": "80c2fcdfc06486ff9ccb473d47272b4cbc9c54dc4e434e1a7edf5d1c772248d0", + "prepared_review_unit_hash": "e2dceb60cb6e22461f95e1294acdf940a708e4822bccdd46f9ae2f8d4aeaaa4c", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6551, + "serialized_chars": 6552, "estimated_input_tokens": 1638, "estimated_eval_tokens": 3532 }, @@ -26242,7 +26311,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6551, + "serialized_chars": 6552, "estimated_input_tokens": 1638, "estimated_eval_tokens": 3532 }, @@ -26251,18 +26320,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate model-addition PRs that only overlap on common Transformers boilerplate; they target different models and code paths, so they are not duplicates.", - "confidence": 0.99, + "summary": "These are two separate model-addition PRs: one adds PenguinVL, the other adds Voxtral TTS. They share some boilerplate files for model registration/docs, but they do not fix the same code path or implement the same feature.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the artifacts implement unrelated models (PenguinVL vs Voxtral TTS), not the same underlying change.", + "canonical_pr_reason": "No duplicate PR pair here; the closest representative is #44662 because it is the broader, older model implementation PR, but it is not a duplicate of #45401.", "best_issue_reason": null, - "best_pr_reason": "45401 is the more tightly scoped PR and has an explicit issue target, but it still represents a different feature than 44662.", + "best_pr_reason": "#44662 is the better cluster representative because it is the more comprehensive and established implementation PR, while #45401 is a separate model integration.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different model implementations and domains: PenguinVL vision-language support vs Voxtral TTS text-to-speech support. Shared auto-registration/docs files are boilerplate, not a common change." + "reason": "Different models and different underlying implementations (PenguinVL vs Voxtral TTS). Shared auto/docs files are boilerplate, not evidence of the same bug/change." } ] }, @@ -26292,7 +26361,7 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "609616dee847a107cb451e611cedf03bcd5be74093b2df626de6633269165a26", + "prepared_review_unit_hash": "82b6e899cf7b709a4eacd4f442bcb77ec55baa6d7727c618df955805837a812d", "component_budget": { "node_count": 2, "item_count": 2, @@ -26314,18 +26383,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related around module/kernel fusion, but they target different implementations and code paths: one adds a generic Module Fusion API, while the other wires n-to-1 kernel fusion through KernelConfig and integration hooks. They do not look like the same PR.", - "confidence": 0.91, + "summary": "These are related by the general theme of model/kernel fusion, but they are not the same change: one introduces a standalone Module Fusion API, while the other implements n-to-1 kernel fusion through KernelConfig and integration hooks.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45363 is the more concrete, end-to-end feature implementation: it touches KernelConfig, model loading/integration code, and includes an executable example, making it the stronger representative of the kernel-fusion effort.", + "canonical_pr_reason": "No single canonical PR for a duplicate set: PR #44979 adds a new Module Fusion API surface, while PR #45363 adds a separate KernelConfig-based kernel fusion feature.", "best_issue_reason": null, - "best_pr_reason": "PR #45363 is the best fit for the cluster because it is broader and more directly user-facing, with explicit integration changes rather than only a new internal API scaffold.", + "best_pr_reason": "PR #45363 is the better representative of the cluster topic because it is the broader, more integrated user-facing fusion feature; PR #44979 is a separate foundational API addition.", "soft_edge_verdicts": [ { - "left": "pull_request:44979", - "right": "pull_request:45363", + "left": "pull_request:44979 (Module Fusion API)", + "right": "pull_request:45363 (n-to-1 kernel fusion via `KernelConfig`)", "accept": false, - "reason": "Different underlying changes: #44979 introduces a standalone Module Fusion API and test file, while #45363 implements n-to-1 kernel fusion via KernelConfig and integration plumbing. They share a theme but not the same concrete fix/change." + "reason": "They share a fusion theme, but the code changes implement different abstractions and use cases. PR #44979 adds a generic module-fusion API, while PR #45363 wires kernel fusion into model loading/integration paths; they would not plausibly be merged as one duplicate PR." } ] }, @@ -26355,12 +26424,12 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "238dc13dcdacb52f8459253feb81d37d1db53da2519b6b023bfd84e8c51e1998", + "prepared_review_unit_hash": "b4b574542e26fa1ed2da2bd59f2d7f82d5c94264f69d65fa3d35e5061b5c12ef", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4414, + "serialized_chars": 4413, "estimated_input_tokens": 1104, "estimated_eval_tokens": 2464 }, @@ -26368,7 +26437,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4414, + "serialized_chars": 4413, "estimated_input_tokens": 1104, "estimated_eval_tokens": 2464 }, @@ -26377,18 +26446,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests address the same Mistral regex AttributeError in `src/transformers/tokenization_utils_tokenizers.py` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; `45317` adds tests for the same fix.", + "summary": "Both PRs address the same Mistral tokenizer AttributeError in `_patch_mistral_regex` by switching from `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; the later PR adds tests and is the stronger candidate.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "`45317` is the stronger representative of the duplicate set: it fixes the same code path as `45086` and also adds coverage in `tests/models/auto/test_tokenization_auto.py`, making it the more complete PR for the shared bug.", + "canonical_pr_reason": "PR #45317 is the better canonical PR because it implements the same fix as #45086 but also includes test coverage, making it the more complete change set.", "best_issue_reason": null, - "best_pr_reason": "`45317` is the best single PR to keep because it includes the same code change as `45086` plus regression tests, so it most completely captures the underlying fix.", + "best_pr_reason": "PR #45317 is the best overall PR candidate: same concrete bug fix, clearer title, and added regression tests.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same underlying AttributeError in `_patch_mistral_regex`, same file, same fix; `45317` is essentially `45086` plus tests, so these are duplicate/overlapping PRs." + "reason": "Same underlying code-path bug in `_patch_mistral_regex` causing an AttributeError; the second PR is a test-augmented version of the same fix and could plausibly be merged as one change." } ] }, @@ -26418,7 +26487,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "e6868c6514b751b89f193b0f49851f399920067775d72393e5c2f83446d3db3e", + "prepared_review_unit_hash": "9826515c08658a68eb70c52442b47f3ec7eabe9084b25426fc6ea597ded8cc19", "component_budget": { "node_count": 2, "item_count": 2, @@ -26440,18 +26509,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related Nemotron-H architecture PRs, but they are not duplicates: one is a focused refactor around inheriting GraniteMoeHybrid and conversion-key remapping, while the other is a broader model split/dispatch update introducing NemotronHDense and NemotronHSparse plus docs and auto-mapping changes.", - "confidence": 0.9, + "summary": "These PRs are related NemotronH architecture cleanup work, but they are not the same change: one refactors NemotronH to inherit GraniteMoeHybrid and adds conversion mappings, while the other splits NemotronH into dense/sparse variants and updates docs/auto-mapping accordingly.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "No true duplicate pair here. If one PR must represent the cluster theme, #45569 is the better canonical candidate because it is the broader, user-facing architecture update and includes the new dense/sparse model split and documentation.", + "canonical_pr_reason": "PR #45569 is the broader, more representative NemotronH rework: it introduces the dense/sparse model split, updates docs, and touches the model auto-mapping path across the package.", "best_issue_reason": null, - "best_pr_reason": "#45569 is the most complete and recognizable change set for Nemotron-H support overall; #45115 is a narrower implementation refactor limited to conversion/compatibility details.", + "best_pr_reason": "PR #45569 is the best single representative of the cluster because it covers the more complete product-level change and subsumes the visible model-organization work, whereas #45115 is a narrower implementation refactor around inheritance/conversion mapping.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "Same model family, but different concrete changes: #45115 focuses on GraniteMoeHybrid-compatible refactoring and checkpoint conversion mappings, while #45569 introduces a broader NemotronH dispatch split into dense/sparse variants with docs, auto mappings, and tests. They are not mergeable as one duplicate PR." + "reason": "They share NemotronH files and a general architectural theme, but the concrete changes differ: #45115 is a refactor to inherit GraniteMoeHybrid and add conversion renames, while #45569 introduces separate dense/sparse variants plus docs and auto-mapping updates. They are related, not the same underlying PR." } ] }, @@ -26481,7 +26550,7 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "b2689cd0849aa0cd096a28c29689f1de6bae01efbfefed9141f3df4401378021", + "prepared_review_unit_hash": "199898ce010fc5fc892ac7a6f2b7d34dccc5a575effd432a9afc8050cc72a891", "component_budget": { "node_count": 2, "item_count": 2, @@ -26503,18 +26572,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests are typing-focused, but they address different changes: one is a broad repository-wide type-checking cleanup, while the other adds/adjusts specific typing protocols and related annotations around `modeling_utils` and helpers. They are related but not the same underlying fix.", - "confidence": 0.86, + "summary": "Both pull requests are typing-focused, but they are not the same change: #45415 is a broad type-checking sweep across many `src/transformers/*.py` files, while #45425 adds specific typing helpers/protocols and touches `modeling_utils`-related code. They overlap in a few filenames, but the concrete code-path changes differ enough that they should not be merged as duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45425 is the better canonical representative because it has a narrower, more coherent typing change centered on `_typing.py`/`modeling_utils` rather than a broad sweep across many files.", + "canonical_pr_reason": "PR #45425 is the more specific, self-contained typing change with a clearer scope and smaller blast radius. PR #45415 is a broad repository-wide type-checking sweep, so it is less suitable as a canonical representative for this cluster.", "best_issue_reason": null, - "best_pr_reason": "PR 45425 is the best standalone PR for this cluster because its scope is more specific and self-contained; PR 45415 is a large multi-file type-checking cleanup that is not the same concrete change.", + "best_pr_reason": "PR #45425 is the best representative because it targets a concrete typing refactor around `_typing.py`, `modeling_utils`, and related utilities, whereas #45415 is an expansive annotation pass that is too broad to serve as the best single exemplar.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Related by theme and a few shared files, but the PRs solve different typing tasks: 45415 is a broad type-checking pass across many modules, while 45425 adds specific typing protocols/aliases and adjusts a smaller set of typing-related call sites. They would not naturally merge into one single change." + "reason": "Do not treat as duplicates: both are typing-related, but #45415 is a wide type-checking sweep across many files, while #45425 adds specific typing protocols/helpers and related `modeling_utils` updates. They do not appear to fix the same concrete problem or be safely merged into one PR." } ] }, @@ -26544,7 +26613,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "1396a2b9ed1ea0b83d592cddc5826ad2952637d0eacb4b0030e13a88ac7395df", + "prepared_review_unit_hash": "2de982b868b57affdd9b7865289ccab1a05ce0eb16ccd2a9dd283e3a8da6dfdc", "component_budget": { "node_count": 2, "item_count": 2, @@ -26566,18 +26635,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two pull requests both touch expert parallelism, but they address different failures: one fixes EP + DeepSpeed ZeRO-3 loading under accelerate launch, the other fixes EP + FSDP2 parameter overwrites during rank-0 broadcast. They overlap only in broad subsystem and a shared file, not in the same concrete bug or mergeable change set.", - "confidence": 0.94, + "summary": "Both PRs concern expert parallelism in Transformers, but they fix different failure modes: one handles serialized distributed config / ZeRO-3 loading, the other fixes experts being overwritten during FSDP2-style sharded loading. They overlap in subsystem, not in the underlying bug.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR #45662 is the stronger representative because it targets a concrete loading corruption path for expert parameters and includes the most directly relevant code changes around sharded parameter assignment.", "best_issue_reason": null, - "best_pr_reason": "No clear duplicate canonical PR; if one had to be highlighted, PR 45662 is the broader EP correctness fix, but it is not the same bug as PR 45548.", + "best_pr_reason": "PR #45662 is the best overall PR here: it addresses a specific, user-visible expert-parallel loading bug with clear code-path coverage across loading and tensor-parallel integrations.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "Both concern expert parallelism, but 45548 fixes loading/serialization with DeepSpeed ZeRO-3 and accelerate launch, while 45662 fixes FSDP2 rank-0 broadcast overwriting experts. Different code paths and failure modes." + "reason": "Do not merge as duplicates: #45548 fixes EP + DeepSpeed ZeRO-3/accelerate launch serialization and adds has_ep support, while #45662 fixes experts being silently overwritten by rank-0 broadcast during sharded loading. They touch adjacent EP machinery but not the same bug or mergeable change set." } ] }, @@ -26607,7 +26676,7 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "d77b6054699bd62bce42cd745e8c38a3051e758e77dae31f130558a4b6a20cb6", + "prepared_review_unit_hash": "c53d48fec60e4187c3907553cfc9d7293087751881810550aebfaa59f51e931b", "component_budget": { "node_count": 2, "item_count": 2, @@ -26629,18 +26698,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same xdist-related captured_info logging problem and share the same issue target and files; one is a narrower code-only fix and the other adds the CI workflow change needed to preserve all debug logs.", - "confidence": 0.92, - "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45645 is the better representative because it includes the underlying xdist-safe logging fix plus the workflow update to collect all captured_info files in CI, making it the more complete version of the change.", - "best_issue_reason": null, - "best_pr_reason": "PR #45645 is the strongest standalone PR for this cluster: it addresses the same captured_info xdist collision and also preserves CI diagnostics, so it best captures the intended end-to-end fix.", + "summary": "Both PRs target the same xdist collision problem around patched testing debug artifacts, with PR 45645 effectively subsuming the core fix from 45639 and adding CI log collection support.", + "confidence": 0.83, + "canonical_issue_reason": "Issue 45561 is the shared tracker for the xdist collision in captured/patched testing debug output.", + "canonical_pr_reason": "PR 45645 is the stronger canonical PR because it includes the core xdist-safe artifact fix plus the CI workflow update to collect multiple captured_info files.", + "best_issue_reason": "Issue 45561 is the only explicit issue target and matches the underlying bug described by both PRs.", + "best_pr_reason": "PR 45645 is the best representative because it covers the same core code-path fix as 45639 and also addresses downstream CI log preservation.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying bug: xdist collisions in captured_info handling. The shared testing_utils changes and the workflow update in #45645 are directly complementary and could plausibly be merged into one PR." + "reason": "They share the same underlying xdist-safe captured-info fix in testing_utils; 45645 is a broader follow-on that appears to subsume 45639 rather than a separate change." } ] }, @@ -26736,7 +26805,7 @@ "soft_pairs": [ "pull_request:46126|pull_request:46269" ], - "prepared_review_unit_hash": "1b857edf20468798bd199a7ce43eab7eaf481bded18adb5f835e5faf2096da89", + "prepared_review_unit_hash": "c79cf511183156b9e58724aa55ef41be49ecfb68fcc1d77de2c4ca1fbdf3aa2c", "component_budget": { "node_count": 2, "item_count": 2, @@ -26758,18 +26827,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: PR #46126 is a targeted EP/Trainer integration change, while PR #46269 is a broad distributed-branch base touching many unrelated distributed and training areas. They share a few filenames, but not the same concrete code-path fix.", - "confidence": 0.93, + "summary": "These are related distributed-training PRs but not duplicates: one is a focused EP + Trainer integration, the other is a broad distributed branch base spanning FSDP, MoE, generation, and CI plumbing.", + "confidence": 0.9, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46126 is the more focused, self-contained change and best represents the specific EP + Trainer integration work in this cluster.", + "canonical_pr_reason": "PR #46269 is the broader umbrella change and the better representative of the distributed refactor/integration work, while #46126 is a narrower feature slice.", "best_issue_reason": null, - "best_pr_reason": "PR #46126 is the best representative because it targets one concrete feature/fix, whereas PR #46269 is a wide umbrella branch with many unrelated edits.", + "best_pr_reason": "PR #46269 is the strongest single PR to keep in a cluster because it covers the widest set of distributed changes and subsystems.", "soft_edge_verdicts": [ { "left": "pull_request:46126", "right": "pull_request:46269", "accept": false, - "reason": "Reject: overlapping files alone are not enough here. #46126 adds EP routing/trainer integration, but #46269 is a broad distributed branch base spanning FSDP, MoE, MXFP4, core loading, CI, and more. They do not look like the same underlying bug or mergeable change." + "reason": "Reject: they share some distributed files, but the concrete changes differ. #46126 is specifically EP + Trainer integration on top of DistributedConfig, while #46269 is a much larger branch-base PR adding FSDP CI and broader distributed/model integration. They do not look like the same underlying change or something that could plausibly be merged into one PR." } ] }, @@ -26837,220 +26906,220 @@ "pull_request:46504|pull_request:46716", "pull_request:46432|pull_request:46716" ], - "prepared_review_unit_hash": "9d75263a8b0534bedbed5d27b196fbf72f7d2a598df0358d1cfe69b89cf011d3", + "prepared_review_unit_hash": "1e3202457ce4f4cff1e6efa583b868ea52fe8e73bbcd5bfcb67cafb41a5deb34", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 31, - "serialized_chars": 26682, - "estimated_input_tokens": 6671, - "estimated_eval_tokens": 13598 + "serialized_chars": 26675, + "estimated_input_tokens": 6669, + "estimated_eval_tokens": 13594 }, "budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 31, - "serialized_chars": 26682, - "estimated_input_tokens": 6671, - "estimated_eval_tokens": 13598 + "serialized_chars": 26675, + "estimated_input_tokens": 6669, + "estimated_eval_tokens": 13594 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Mostly unrelated workflow-maintenance PRs. The only tight grouping is the pr-ci-security-gate edits, but those are separate hardening changes rather than duplicates of one concrete fix, so I would not merge them as the same PR.", - "confidence": 0.84, + "summary": "This cluster is mostly a bundle of unrelated CI/workflow maintenance PRs. The only tight subgroup is the pr-ci-security-gate.yml changes, but those PRs make different security/permission/reusable-workflow edits and are not duplicates of the same underlying fix.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46554 is the clearest substantive fix: it changes the security-gate workflow logic itself by checking out the PR head and actively blocking risky file changes.", + "canonical_pr_reason": "PR 46554 is the best single anchor if one must choose: it is the most substantive and central change within the only closely related subgroup (pr-ci-security-gate.yml), but it still represents a distinct workflow update rather than a duplicate.", "best_issue_reason": null, - "best_pr_reason": "PR 46554 is the best representative of the cluster because it makes the most concrete behavior change in the core workflow path, rather than a narrow permissions or formatting tweak.", + "best_pr_reason": "PR 46554 is the most representative PR in the cluster because it has the strongest connectivity and the clearest concrete workflow/security-gate logic change among the related items.", "soft_edge_verdicts": [ + { + "left": "pull_request:46138", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflows and different problems: self-comment CI argument handling vs build-ci-docker-images commit-message parsing." + }, { "left": "pull_request:46138", "right": "pull_request:46554", "accept": false, - "reason": "Different workflows and different behavior; only a superficial YAML/workflow similarity." + "reason": "Different workflows; one updates self-comment CI stdin handling, the other adds PR security-gate checks." }, { "left": "pull_request:46138", "right": "pull_request:46558", "accept": false, - "reason": "Unrelated workflows; one updates stdin handling, the other hardens PR CI security-gate permissions." + "reason": "Different workflows and unrelated changes; not the same bug or fix path." }, { "left": "pull_request:46138", "right": "pull_request:46563", "accept": false, - "reason": "Different files and different concerns; not the same underlying change." + "reason": "Different workflow update and different security-gate behavior; shared CI theme is too broad." }, { "left": "pull_request:46138", "right": "pull_request:46716", "accept": false, - "reason": "Separate workflow updates with no shared code-path bug." + "reason": "Different workflows and different code paths; not a duplicate change." }, { - "left": "pull_request:46138", - "right": "pull_request:46504", + "left": "pull_request:46432", + "right": "pull_request:46501", + "accept": false, + "reason": "SSH runner hardening is unrelated to anti-slop workflow edits." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46502", "accept": false, - "reason": "Both are small CI-script edits, but they fix different workflows and different command handling." + "reason": "SSH runner workflow changes are unrelated to assign-reviewers checkout behavior." }, { "left": "pull_request:46432", "right": "pull_request:46503", "accept": false, - "reason": "Different workflows and unrelated security/benchmark behavior." + "reason": "Different workflows and different runtime behavior; not the same underlying fix." }, { "left": "pull_request:46432", - "right": "pull_request:46501", + "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files and different permission/event changes." + "reason": "SSH runner hardening vs PR security-gate logic; related only at a very broad CI level." }, { "left": "pull_request:46432", - "right": "pull_request:46502", + "right": "pull_request:46558", "accept": false, - "reason": "Unrelated CI workflows; one is SSH runner hardening, the other reviewer assignment checkout behavior." + "reason": "Different workflows and distinct security concerns; not mergeable as one fix." }, { "left": "pull_request:46432", - "right": "pull_request:46554", + "right": "pull_request:46563", "accept": false, - "reason": "Both touch workflow security, but they address different problems and code paths." + "reason": "SSH runner changes are unrelated to the reusable PR CI security-gate update." }, { "left": "pull_request:46432", - "right": "pull_request:46558", + "right": "pull_request:46716", "accept": false, - "reason": "Different workflow security issues; not the same fix." + "reason": "Different workflows and different security-hardening contexts; not a duplicate." }, { "left": "pull_request:46501", - "right": "pull_request:46503", + "right": "pull_request:46502", "accept": false, - "reason": "Both are workflow config edits, but they change different jobs and triggers." + "reason": "Anti-slop workflow vs assign-reviewers workflow; different functionality and no shared bug." }, { "left": "pull_request:46501", - "right": "pull_request:46502", + "right": "pull_request:46503", "accept": false, - "reason": "Different workflow files and unrelated behavior changes." + "reason": "Distinct CI workflows with unrelated changes." }, { "left": "pull_request:46501", "right": "pull_request:46504", "accept": false, - "reason": "Separate workflow changes with no shared bug." + "reason": "Different workflows and different fixes; same broad CI area is not enough." }, { "left": "pull_request:46501", "right": "pull_request:46554", "accept": false, - "reason": "Only broad CI/workflow similarity; the actual changes are unrelated." + "reason": "Anti-slop workflow tweak is unrelated to PR security-gate logic changes." }, { "left": "pull_request:46502", "right": "pull_request:46503", "accept": false, - "reason": "Different workflows; one fixes checkout refing, the other adds a dispatch trigger." + "reason": "Assign-reviewers vs benchmark workflow; different code paths and purposes." }, { "left": "pull_request:46502", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow jobs and different command changes." + "reason": "Unrelated workflow files and distinct behavior changes." }, { "left": "pull_request:46502", "right": "pull_request:46554", "accept": false, - "reason": "Same general CI area, but different workflow logic and no shared concrete bug." + "reason": "Assign-reviewers checkout ref change is unrelated to PR security-gate hardening." }, { "left": "pull_request:46503", "right": "pull_request:46504", "accept": false, - "reason": "Unrelated workflow edits; not the same underlying problem." + "reason": "Benchmark gating change vs build-ci-docker-images commit-message parsing; different fixes." }, { "left": "pull_request:46503", "right": "pull_request:46554", "accept": false, - "reason": "Different workflows and different behavior changes." + "reason": "Different workflows and no shared concrete bug." }, { "left": "pull_request:46504", "right": "pull_request:46554", "accept": false, - "reason": "Both are workflow shell changes, but they fix different scripts and code paths." + "reason": "Build-ci-docker-images change is unrelated to PR security-gate workflow changes." }, { "left": "pull_request:46504", "right": "pull_request:46558", "accept": false, - "reason": "Different files and different hardening goals." + "reason": "Different workflow files and different operational concerns." }, { - "left": "pull_request:46554", - "right": "pull_request:46558", - "accept": false, - "reason": "Same workflow file, but one adds security checks and the other adjusts permissions/approval gating; not the same concrete fix." - }, - { - "left": "pull_request:46554", + "left": "pull_request:46504", "right": "pull_request:46563", "accept": false, - "reason": "The comment-only note is not the same change as the added security checks." + "reason": "Different workflow files; not the same underlying change." }, { - "left": "pull_request:46554", + "left": "pull_request:46504", "right": "pull_request:46716", "accept": false, - "reason": "Different security-gate hardening steps; not a duplicate code-path fix." + "reason": "Different workflow files and different fix scope." }, { - "left": "pull_request:46558", - "right": "pull_request:46563", + "left": "pull_request:46554", + "right": "pull_request:46558", "accept": false, - "reason": "Permissions/approval logic is different from adding explanatory comments." + "reason": "Same file, but different changes: one adds checkout/diff-based security checks, the other changes permissions and approval flow." }, { - "left": "pull_request:46558", - "right": "pull_request:46716", + "left": "pull_request:46554", + "right": "pull_request:46563", "accept": false, - "reason": "Related security hardening, but permissions and secrets changes are separate fixes." + "reason": "Same workflow file, but one switches to a reusable workflow and adds comments while the other adds concrete gate logic; not the same fix." }, { - "left": "pull_request:46563", + "left": "pull_request:46554", "right": "pull_request:46716", "accept": false, - "reason": "Comment-only documentation tweak versus a functional secrets change." - }, - { - "left": "pull_request:46504", - "right": "pull_request:46563", - "accept": false, - "reason": "Different workflows and unrelated edits." + "reason": "Same workflow file, but one changes security checks and the other changes secrets handling; distinct patches." }, { - "left": "pull_request:46432", + "left": "pull_request:46558", "right": "pull_request:46563", "accept": false, - "reason": "Unrelated workflow files and behavior." + "reason": "Both touch the same workflow file, but they address different aspects of the gate and are not the same code-path problem." }, { - "left": "pull_request:46504", + "left": "pull_request:46558", "right": "pull_request:46716", "accept": false, - "reason": "Different workflows and different shell/security tweaks." + "reason": "Different edits to the same workflow file: permissions/approval flow vs secrets inheritance; not duplicates." }, { - "left": "pull_request:46432", + "left": "pull_request:46563", "right": "pull_request:46716", "accept": false, - "reason": "Separate workflow hardening changes with no shared concrete bug." + "reason": "Reusable-workflow commentary/invocation change vs secrets configuration change; distinct fixes." } ] }, @@ -27143,7 +27212,7 @@ "soft_pairs": [ "pull_request:46310|pull_request:47162" ], - "prepared_review_unit_hash": "321a6206350374bd6d2f9696f151dbf34bae58b4e5540f37d1f7fbc2a3583833", + "prepared_review_unit_hash": "7807af6dc58bf033998dbc0e99c240c285965d8df0e3ac8c30956a16399d1b83", "component_budget": { "node_count": 2, "item_count": 2, @@ -27165,18 +27234,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs fix the same off-by-one bug in the MoonshineStreaming sliding-window right mask; PR 47162 is the broader superset because it includes the same MoonshineStreaming fix plus the analogous Gemma4 change and tests.", - "confidence": 0.95, + "summary": "Both PRs address an off-by-one in the right side of a sliding window mask, but PR 47162 also expands the fix to Gemma4 and adds tests, so it is broader than PR 46310 rather than a clean duplicate.", + "confidence": 0.87, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47162 is the better canonical PR because it contains the MoonshineStreaming fix from 46310, adds the same correction in Gemma4, and includes test coverage; 46310 is a narrower subset of the same fix.", + "canonical_pr_reason": "PR 47162 is the stronger canonical PR: it fixes the same MoonshineStreaming boundary bug as 46310, extends the correction to Gemma4, and includes test coverage.", "best_issue_reason": null, - "best_pr_reason": "PR 47162 is the best representative of the change set because it addresses the same off-by-one window bug with a broader, more complete patch and tests.", + "best_pr_reason": "PR 47162 is the best overall PR because it covers the broader code-path impact and includes tests, while 46310 is a narrower subset fix.", "soft_edge_verdicts": [ { "left": "pull_request:46310", "right": "pull_request:47162", - "accept": true, - "reason": "Same underlying off-by-one right-window bug in MoonshineStreaming; 47162 includes the same fix there and extends it to Gemma4, so these look like the same change family rather than distinct PRs." + "accept": false, + "reason": "Related off-by-one fixes, but not the same single change: 47162 also modifies Gemma4 and test coverage, so these are not a clean one-to-one duplicate pair." } ] }, @@ -27206,12 +27275,12 @@ "soft_pairs": [ "pull_request:46574|pull_request:46580" ], - "prepared_review_unit_hash": "a4f74146458885bb4727feaafa37bb8cd41abea29a3c0ce2a3b197c28507e768", + "prepared_review_unit_hash": "205e93df042b5bd7c3bd7d4127a5ef680af611b34fd08e1341901e06ffae286c", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4410, + "serialized_chars": 4409, "estimated_input_tokens": 1103, "estimated_eval_tokens": 2462 }, @@ -27219,7 +27288,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4410, + "serialized_chars": 4409, "estimated_input_tokens": 1103, "estimated_eval_tokens": 2462 }, @@ -27228,18 +27297,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs are not duplicates: one fixes CodeLlama decode behavior to preserve leading spaces, while the other fixes DeepSeek checkpoint-to-tokenizer-class resolution in auto tokenization.", + "summary": "The two PRs are unrelated: #46574 fixes CodeLlama decode stripping/leading-space handling, while #46580 fixes DeepSeek-Coder/DeepSeek-LLM tokenizer-class selection in auto tokenization.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No canonical PR: the diffs touch different code paths and solve different bugs, so neither PR is a duplicate of the other.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "#46574 is the more self-contained bugfix with a direct issue target, but it is not a representative match for #46580 because the underlying problems differ.", "soft_edge_verdicts": [ { "left": "pull_request:46574", "right": "pull_request:46580", "accept": false, - "reason": "Different bugs and code paths: 46574 changes CodeLlama tokenizer decoding/strip behavior, while 46580 changes AutoTokenizer selection for DeepSeek checkpoints. They would not be merged as one PR." + "reason": "Reject: one changes CodeLlama decoder initialization to preserve leading spaces; the other adds checkpoint-prefix overrides so DeepSeek models load the correct tokenizer class. Different bugs, different fix paths." } ] }, @@ -27269,12 +27338,12 @@ "soft_pairs": [ "pull_request:46696|pull_request:47215" ], - "prepared_review_unit_hash": "8ae8c67208d6a5db33c8f335964c95d052a829a96bd367fd9c062633bbfa0c60", + "prepared_review_unit_hash": "7990a36faed7c891ab9e64200610b8f66f1afa3faad1a7d69b252a84a3924b2c", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4580, + "serialized_chars": 4579, "estimated_input_tokens": 1145, "estimated_eval_tokens": 2546 }, @@ -27282,7 +27351,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4580, + "serialized_chars": 4579, "estimated_input_tokens": 1145, "estimated_eval_tokens": 2546 }, @@ -27291,18 +27360,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same FA2 long-context performance regression in `modeling_flash_attention_utils.py`; PR 47215 appears to be the more complete fix and likely subsumes PR 46696.", - "confidence": 0.77, + "summary": "Both PRs target the same FA2 long-context performance regression in `modeling_flash_attention_utils.py`, centered on eager-vs-compile handling of scalar lengths to avoid syncs. PR #47215 is the broader, later fix and adds related adjustments/tests, so it is the better canonical PR.", + "confidence": 0.84, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47215 is the better canonical PR because it addresses the regression more comprehensively in the same code path, adds related test/benchmark coverage, and looks like the fuller follow-up to the same issue.", + "canonical_pr_reason": "PR #47215 is the more complete fix for the shared FA2 long-context perf regression: it keeps scalar lengths as tensors only when compiling and otherwise converts eagerly, while also covering related helper paths and tests. PR #46696 is a narrower variant of the same underlying change.", "best_issue_reason": null, - "best_pr_reason": "PR 47215 is the strongest representative of the cluster: it fixes the long-context flash-attention slowdown more broadly and is more likely to be the final merged form of the change.", + "best_pr_reason": "PR #47215 is the best representative because it most fully addresses the regression and includes the supporting test/benchmark updates, making it the most useful single artifact for triage.", "soft_edge_verdicts": [ { "left": "pull_request:46696", "right": "pull_request:47215", "accept": true, - "reason": "Same underlying FA2 long-context decode/perf regression, same explicit issue target, and same helper file; 47215 looks like a broader version of the same fix rather than a different change." + "reason": "Same underlying FA2 long-context decode/perf regression in the flash-attention helper path; both fix eager-vs-compile scalar handling to avoid costly syncs and could plausibly be merged into one PR." } ] }, @@ -27332,40 +27401,40 @@ "soft_pairs": [ "pull_request:46773|pull_request:47087" ], - "prepared_review_unit_hash": "9edf2db6ca95fcce3bf3761f1068ce551abf39a055a062f714947c9bb8bf053c", + "prepared_review_unit_hash": "4330197a5f33cf9ad16398d4507f205088c119c96cb8a7aed55d5ad8cba9f856", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5433, - "estimated_input_tokens": 1359, - "estimated_eval_tokens": 2974 + "serialized_chars": 5432, + "estimated_input_tokens": 1358, + "estimated_eval_tokens": 2972 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5433, - "estimated_input_tokens": 1359, - "estimated_eval_tokens": 2974 + "serialized_chars": 5432, + "estimated_input_tokens": 1358, + "estimated_eval_tokens": 2972 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs in the masking/padding area, but they address different bugs and should not be deduplicated: one fixes batch-size-1 masking in linear-attention model code, the other fixes recurrent/cache-continuation masking in shared masking utilities.", - "confidence": 0.93, + "summary": "Two open PRs both touch padding-mask logic, but they address different bugs in different execution paths: one fixes batch-size-1 masking in linear-attention model forwards, the other fixes recurrent-layer masking being skipped on continued forwards/cache continuation.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:47087 is the better representative PR: it has an explicit issue target, touches the shared masking utility, and fixes a clearly scoped recurrent/cache-continuation bug.", + "canonical_pr_reason": "PR #47087 is the cleaner canonical representative because it has an explicit issue link, a narrower and well-scoped fix, and the diff targets a single shared masking utility rather than many model-specific forward paths.", "best_issue_reason": null, - "best_pr_reason": "pull_request:47087 is the strongest standalone PR in the cluster because it is more general within the masking stack and includes tests plus an explicit issue reference.", + "best_pr_reason": "PR #47087 is the strongest standalone candidate: it clearly states the bug, links to an issue, and fixes a concrete recurrent-mask code path with focused tests.", "soft_edge_verdicts": [ { "left": "pull_request:46773", "right": "pull_request:47087", "accept": false, - "reason": "Both concern padding masks, but the concrete failures differ: 46773 fixes an incorrect batch-size-1 check in linear-attention model implementations, while 47087 fixes masking being skipped on continued forwards/chunked prefill in recurrent-layer utilities. Different code paths and not mergeable as one PR." + "reason": "Reject: both concern padding masks, but #46773 fixes batch-size-1 masking being skipped in linear-attention model forwards, while #47087 fixes recurrent masking being skipped on continued forwards/chunked prefill/cache continuation. Different underlying bugs and not plausibly one merged PR." } ] }, @@ -27397,12 +27466,12 @@ "pull_request:47074|pull_request:47076", "pull_request:47074|pull_request:47195" ], - "prepared_review_unit_hash": "b083b362cc1c9dea6f244248caad73fb1b2d5ca1211a21b86b179c79499c4db5", + "prepared_review_unit_hash": "3c1a3355a7069df85c403c705e3de90f4ed93d39bc487abb2f069c5ce5d5686f", "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 }, @@ -27410,7 +27479,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 6434, + "serialized_chars": 6433, "estimated_input_tokens": 1609, "estimated_eval_tokens": 3474 }, @@ -27419,24 +27488,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three pull requests are not duplicates: they target different code paths in different subsystems (AnyToAny pipeline key normalization, Gemma4 audio placeholder validation, and generation prompt lookup length capping).", - "confidence": 0.97, + "summary": "These are three unrelated pull requests: one fixes a key-name bug in AnyToAnyPipeline, one changes Gemma4 processor placeholder validation, and one adjusts candidate-generation length capping. None appear to be duplicates or mergeable into a single PR.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:47074 is the best cluster representative because it is a focused, self-contained bug fix with a clear API-level correction and tests.", + "canonical_pr_reason": "PR 47074 is the clearest and most self-contained change, with the strongest discussion activity, but it is not actually a duplicate of the other two PRs.", "best_issue_reason": null, - "best_pr_reason": "pull_request:47074 is the strongest overall PR in the set: it fixes one concrete multimodal input mapping bug and has the most directly verifiable scope; the other two are unrelated changes.", + "best_pr_reason": "PR 47074 is the best representative only in the sense that it is the most focused and well-scoped of the three; it should not be treated as the same change as 47076 or 47195.", "soft_edge_verdicts": [ { "left": "pull_request:47074", "right": "pull_request:47076", "accept": false, - "reason": "Different underlying bugs: one normalizes AnyToAny pipeline input keys, the other validates Gemma4 audio placeholder batching in processors." + "reason": "Different code paths and bugs: 47074 normalizes AnyToAnyPipeline input keys, while 47076 validates Gemma4 audio/image placeholder batching. They touch different processors and fix different behaviors." }, { "left": "pull_request:47074", "right": "pull_request:47195", "accept": false, - "reason": "Different code paths and fixes: multimodal pipeline input key mapping vs. generation candidate length capping." + "reason": "Unrelated functionality: 47074 fixes multimodal pipeline input keys, while 47195 changes generation candidate length limiting. No shared concrete bug or mergeable combined patch is evident." } ] }, @@ -27466,7 +27535,7 @@ "soft_pairs": [ "pull_request:47105|pull_request:47129" ], - "prepared_review_unit_hash": "dbbb541607153bc021302f8b1bb90d97ef257bfe7adc99fa315f30528a7290ee", + "prepared_review_unit_hash": "12ae9a914892d4ccb1dce9188cf03e8b5d4c7227140baae64df08b97acd996ee", "component_budget": { "node_count": 2, "item_count": 2, @@ -27488,18 +27557,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs are unrelated: one re-enables a VITS model forward test, the other fixes an RT-DETR v2 weight converter for current model keys. They do not target the same bug or code path, so the soft edge should be rejected.", - "confidence": 0.99, + "summary": "These PRs are not duplicates: one re-enables a VITS model-forward test, while the other fixes RT-DETR v2 weight-conversion key mappings. They touch different code paths and should remain separate.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 47129 is the more substantive code fix, but there is no shared underlying change that would make these PRs a duplicate cluster.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "PR 47129 is the better representative of the cluster because it changes production conversion logic, whereas PR 47105 is only a test re-enable.", "soft_edge_verdicts": [ { "left": "pull_request:47105", "right": "pull_request:47129", "accept": false, - "reason": "Different models and different concerns: test coverage for VITS vs. converter/key-mapping logic for RT-DETR v2. No shared failing code path or mergeable single fix." + "reason": "Different subsystems and different defects: VITS test expectations versus RT-DETR v2 converter key mapping. They do not fix the same concrete bug and would not plausibly merge into one PR." } ] }, @@ -27563,230 +27632,203 @@ "pull_request:47282|pull_request:47310", "pull_request:47280|pull_request:47282" ], - "prepared_review_unit_hash": "7e77cda27de002eee9acd53936bbb5180e9cda9d188dcd8dd6ad2b7d674a3249", + "prepared_review_unit_hash": "d90ba3baa839bc0afa94f4ef4823f7aedf4effc3b9f4ecaa2c4a24c4047c34b6", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 27, - "serialized_chars": 29369, - "estimated_input_tokens": 7343, - "estimated_eval_tokens": 14942 + "serialized_chars": 29375, + "estimated_input_tokens": 7344, + "estimated_eval_tokens": 14944 }, "budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 27, - "serialized_chars": 29369, - "estimated_input_tokens": 7343, - "estimated_eval_tokens": 14942 + "serialized_chars": 29375, + "estimated_input_tokens": 7344, + "estimated_eval_tokens": 14944 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a mix of unrelated model-specific integration-test fixes. The only clear duplicate/near-duplicate relationships are the MiniMax-M3-VL load_error pair and the Edgetam/Whisper fixes that are also included in the umbrella PR 47224.", + "summary": "This cluster is mostly heterogeneous: one real duplicate pair stands out for MiniMax-M3-VL load_error/offload_folder handling, while the rest are unrelated model-specific fixes. The broad umbrella PRs are not duplicates of the focused ones because they bundle additional unrelated changes.", "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "47224 is the broad umbrella PR that directly contains the Edgetam and Whisper code fixes and best represents the shared serge integration-failure patch set in this cluster.", + "canonical_pr_reason": "PR 47224 is the broad umbrella fix with the widest scope and strongest centrality in the cluster, but it is not a clean duplicate of the focused model-specific PRs; it best represents the cluster\u2019s mixed serge-failure batch.", "best_issue_reason": null, - "best_pr_reason": "47224 is the most representative PR because it bundles the concrete code-path fixes for multiple failing models, rather than only a test-only or single-model patch.", + "best_pr_reason": "PR 47224 is the best representative because it is the largest umbrella change and links to multiple files/code paths seen in the cluster, even though the cluster does not form a single true duplicate set.", "soft_edge_verdicts": [ { "left": "pull_request:47122", "right": "pull_request:47281", "accept": false, - "reason": "Different models and different failure types; Edgetam import_or_config is unrelated to GLM46V model changes." + "reason": "Different models and code paths: Edgetam import_or_config/backbone kwargs bug versus GLM46V integration changes." }, { "left": "pull_request:47122", "right": "pull_request:47123", "accept": false, - "reason": "Both are integration-test-failure PRs, but for different models and different bugs (Edgetam import_or_config vs MiniMax load_error)." + "reason": "Both are test-failure fixes, but for different models and different failure causes; not the same underlying bug." }, { "left": "pull_request:47122", "right": "pull_request:47280", "accept": false, - "reason": "Different models and different fixes; 47280 is MiniMax-M3-VL, not the Edgetam code-path bug." + "reason": "Same symptom class, but different models and different artifacts; this is not the same concrete change." }, { "left": "pull_request:47122", "right": "pull_request:47224", - "accept": true, - "reason": "Same Edgetam failure; 47224 includes the same original_sizes/kwargs handling fix in the same Edgetam files." + "accept": false, + "reason": "47224 bundles the Edgetam fix with several unrelated fixes, so the PRs are not duplicates as a whole." }, { "left": "pull_request:47122", "right": "pull_request:47310", "accept": false, - "reason": "Unrelated models and code paths; Gemma3 cache behavior is not the Edgetam import_or_config issue." + "reason": "Edgetam backbone kwarg cleanup versus Gemma3 cache_implementation handling are unrelated." }, { "left": "pull_request:47123", "right": "pull_request:47280", "accept": true, - "reason": "Same MiniMax-M3-VL load_error fix on the same test file; 47280 is the later patch of the same offload_folder change." + "reason": "Same MiniMax-M3-VL load_error in the same test file; both add offload-folder handling, and 47280 is a fuller version of the same fix." }, { "left": "pull_request:47123", "right": "pull_request:47266", "accept": false, - "reason": "Different models and unrelated failure modes; load_error for MiniMax-M3-VL is not DeepSeekV32." + "reason": "Same load_error label, but different models and loading scenarios; not the same underlying bug." }, { "left": "pull_request:47123", "right": "pull_request:47281", "accept": false, - "reason": "Different models and different underlying bug; MiniMax load_error does not match GLM46V model work." + "reason": "MiniMax-M3-VL test loading versus GLM46V model changes are unrelated." }, { "left": "pull_request:47123", "right": "pull_request:47124", "accept": false, - "reason": "Different models and different runtime behavior; MiniMax load_error is unrelated to Falcon-Mamba output mismatch." + "reason": "Different model and failure mode: MiniMax load_error versus Falcon-Mamba output mismatch." }, { "left": "pull_request:47124", "right": "pull_request:47310", "accept": false, - "reason": "Different models and different fixes; Falcon-Mamba output mismatch is unrelated to Gemma3 cache handling." + "reason": "Falcon-Mamba output mismatch and Gemma3 cache_implementation behavior are distinct code paths." }, { "left": "pull_request:47124", "right": "pull_request:47150", "accept": false, - "reason": "Both are output-related test fixes, but for different models and different code paths (Falcon-Mamba vs Whisper generation length)." + "reason": "Falcon-Mamba generation output mismatch versus Whisper max_length adjustment are unrelated bugs." }, { "left": "pull_request:47124", "right": "pull_request:47282", "accept": false, - "reason": "Unrelated models and failures; Falcon-Mamba output mismatch is not the Mamba2 OOM/generation issue." + "reason": "Falcon-Mamba output mismatch versus Mamba2 OOM is not the same issue." }, { "left": "pull_request:47124", "right": "pull_request:47281", "accept": false, - "reason": "Different models and different bug classes; Falcon-Mamba output mismatch is unrelated to GLM46V integration work." + "reason": "Falcon-Mamba generation fix versus GLM46V changes are unrelated." }, { "left": "pull_request:47150", "right": "pull_request:47310", "accept": false, - "reason": "Different models and distinct generation issues; Whisper max_length handling is unrelated to Gemma3 cache_implementation." + "reason": "Both involve generation behavior, but Whisper max_length setting and Gemma3 cache_implementation defaulting are different concrete problems." }, { "left": "pull_request:47150", "right": "pull_request:47280", "accept": false, - "reason": "Different models and failure modes; Whisper generation length bug is unrelated to MiniMax-M3-VL load_error." + "reason": "Whisper generation fix versus MiniMax-M3-VL load_error are unrelated." }, { "left": "pull_request:47150", "right": "pull_request:47282", "accept": false, - "reason": "Different models and different symptoms; Whisper generation length is unrelated to Mamba2 OOM." + "reason": "Whisper generation logic and Mamba2 OOM test adjustments are unrelated." }, { "left": "pull_request:47150", "right": "pull_request:47281", "accept": false, - "reason": "Different models and unrelated changes; Whisper generation fix does not match GLM46V model work." + "reason": "Whisper generation behavior versus GLM46V model changes are unrelated." }, { "left": "pull_request:47150", "right": "pull_request:47224", - "accept": true, - "reason": "Same Whisper bug; 47224 includes the same generation_whisper max_length update for the same code path." + "accept": false, + "reason": "47224 includes the Whisper fix but also unrelated tokenizer/Edgetam changes, so it is not the same standalone PR/change." }, { "left": "pull_request:47266", "right": "pull_request:47280", "accept": false, - "reason": "Different models and different failure classes; DeepSeekV32 load_error is unrelated to MiniMax-M3-VL load_error." + "reason": "Different models and different load_error causes; not duplicates." }, { "left": "pull_request:47266", "right": "pull_request:47281", "accept": false, - "reason": "Different models and unrelated fixes; DeepSeekV32 load_error is not GLM46V integration work." - }, - { - "left": "pull_request:47266", - "right": "pull_request:47310", - "accept": false, - "reason": "Different models and code paths; DeepSeekV32 load_error is unrelated to Gemma3 generation cache behavior." + "reason": "DeepSeekV3.2 loading/offload handling versus GLM46V changes are unrelated." }, { "left": "pull_request:47124", "right": "pull_request:47266", "accept": false, - "reason": "Different models and different failure types; Falcon-Mamba output mismatch is unrelated to DeepSeekV32 load_error." - }, - { - "left": "pull_request:47266", - "right": "pull_request:47280", - "accept": false, - "reason": "Different models and failure classes; DeepSeekV32 load_error is unrelated to MiniMax-M3-VL load_error." + "reason": "Falcon-Mamba output mismatch versus DeepSeekV3.2 load_error are unrelated." }, { "left": "pull_request:47266", "right": "pull_request:47310", "accept": false, - "reason": "Different models and unrelated fixes; DeepSeekV32 load_error is not Gemma3 cache handling." + "reason": "DeepSeek load_error and Gemma3 generation cache behavior are different bugs." }, { "left": "pull_request:47280", "right": "pull_request:47281", "accept": false, - "reason": "Different models and bug types; MiniMax-M3-VL load_error does not match GLM46V model changes." - }, - { - "left": "pull_request:47280", - "right": "pull_request:47282", - "accept": false, - "reason": "Different models and failures; MiniMax-M3-VL load_error is unrelated to Mamba2 OOM." + "reason": "MiniMax-M3-VL load_error fix versus GLM46V changes are unrelated." }, { "left": "pull_request:47280", "right": "pull_request:47310", "accept": false, - "reason": "Different models and code paths; MiniMax-M3-VL load_error is unrelated to Gemma3 cache behavior." + "reason": "MiniMax-M3-VL load_error fix versus Gemma3 generation cache behavior are unrelated." }, { "left": "pull_request:47281", "right": "pull_request:47282", "accept": false, - "reason": "Different models and unrelated bugs; GLM46V model work is not the Mamba2 OOM/generation issue." + "reason": "GLM46V changes versus Mamba2 OOM test fix are unrelated." }, { "left": "pull_request:47282", "right": "pull_request:47310", "accept": false, - "reason": "Different models and fixes; Mamba2 OOM/generation tuning is unrelated to Gemma3 cache_implementation." - }, - { - "left": "pull_request:47280", - "right": "pull_request:47281", - "accept": false, - "reason": "Different models and unrelated failures; MiniMax-M3-VL load_error is not GLM46V model work." + "reason": "Mamba2 OOM-related test adjustment versus Gemma3 cache_implementation fix are unrelated." }, { "left": "pull_request:47280", "right": "pull_request:47282", "accept": false, - "reason": "Different models and different bug classes; MiniMax-M3-VL load_error is unrelated to Mamba2 OOM." + "reason": "MiniMax-M3-VL load_error fix versus Mamba2 OOM fix are unrelated." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet: the MiniMax-M3-VL pair (47123/47280) is a plausible same-bug duplicate, and 47224 does visibly bundle the Edgetam and Whisper fixes. The other pairwise rejections are conservative and stay within the model- and failure-specific evidence shown here." - }, - "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 }, @@ -27809,47 +27851,50 @@ "soft_pairs": [ "pull_request:47140|pull_request:47303" ], - "prepared_review_unit_hash": "e74f1a92251156c8cb11971a117d7688c9f55a34d7c23122f1dda2a6aeb46ae6", + "prepared_review_unit_hash": "db90b54083490ae5cda54a85df6d7ddde35819d1f8f4c2ee23267ee88f64b8fd", "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": "Two open PRs address related chat-CLI management URL handling in `src/transformers/cli/chat.py`, but they are not the same concrete fix: one normalizes the base URL in the interface, the other rewrites only the `load_model` request to the service root.", - "confidence": 0.82, + "summary": "Both PRs address the chat CLI management URL bug when the server is mounted under `/v1` versus the service root; 47303 is the broader root-routing variant, while 47140 is a narrower normalization approach.", + "confidence": 0.74, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47140 is the broader, more central fix for the `/v1` base-URL handling in the chat CLI and better represents the shared bug theme.", + "canonical_pr_reason": "47303 is the cleaner representative: it explicitly routes management requests to the service root and packages the fix in a helper rather than only normalizing the constructor input.", "best_issue_reason": null, - "best_pr_reason": "PR 47140 is the strongest representative of the cluster because it addresses the underlying base URL normalization across the CLI rather than a single call site.", + "best_pr_reason": "47303 most directly captures the user-facing fix for chat management calls and is the broader, more representative implementation.", "soft_edge_verdicts": [ { "left": "pull_request:47140", "right": "pull_request:47303", - "accept": false, - "reason": "Reject: both touch chat management URLs, but 47140 normalizes the CLI base URL globally while 47303 only reroutes the `load_model` call to the service root. They overlap in subsystem, not in the same exact code-path fix." + "accept": true, + "reason": "Same underlying chat CLI management URL problem (`/v1` base vs service root); the patches are alternative implementations of the same fix and could plausibly be combined into one PR." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The shared chat CLI management URL/service-root theme is supported by the titles and diff previews, and the overlapping files make the soft-pairing plausible. The summary stays reasonably conservative, though the differing issue targets mean the \u2018same bugfix\u2019 link should be treated as suggestive rather than certain." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -27872,7 +27917,7 @@ "soft_pairs": [ "pull_request:47165|pull_request:47289" ], - "prepared_review_unit_hash": "cc9bf615a43744fc3b87609e1b7b17fd551cb8587e193bdcef20e59955c7f4bb", + "prepared_review_unit_hash": "22fa01ba2f28cba1147651ce11b352263040833ade250bf8edab99c0b7906795", "component_budget": { "node_count": 2, "item_count": 2, @@ -27894,18 +27939,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one PR adds FSDP plan metadata across many model classes, while the other adds pipeline-parallel naive inference support plus related test/marker plumbing. They live in the same distributed-training area but solve different problems on different code paths.", - "confidence": 0.97, + "summary": "The two PRs are only loosely related through distributed-training/inference infrastructure, but they change different code paths: one adds FSDP plans to many model classes, while the other adds pipeline-parallel naive inference support and test/marker plumbing. They should not be treated as duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR as a duplicate representative: the two PRs implement different features (FSDP plan coverage vs pipeline-parallel inference support).", + "canonical_pr_reason": "PR 47165 is the more complete, non-draft distributed-model change and is the better representative of the cluster than the pipeline-parallel feature PR.", "best_issue_reason": null, - "best_pr_reason": "PR #47289 is the more self-contained representative of the cluster because it centers on one concrete feature and includes the core implementation plus test/marker updates; PR #47165 is a broad cross-model metadata sweep.", + "best_pr_reason": "PR 47165 is the strongest overall representative here because it is broader, more mature, and focused on concrete model-level FSDP plan additions rather than auxiliary pipeline-parallel scaffolding.", "soft_edge_verdicts": [ { "left": "pull_request:47165", "right": "pull_request:47289", "accept": false, - "reason": "Reject: they touch different mechanisms and code paths. #47165 adds _fsdp_plan entries to many model classes; #47289 implements pipeline parallel naive inference and related test infrastructure. Same subsystem, not the same change." + "reason": "Different features and code paths: FSDP plan annotations across model implementations vs pipeline-parallel naive inference and test markers. They are not the same underlying change." } ] }, @@ -27935,7 +27980,7 @@ "soft_pairs": [ "pull_request:47261|pull_request:47262" ], - "prepared_review_unit_hash": "1df5ff29522698efa317c5c1333d4b889cb2ab4ddbb75e94c535f3348fce28d6", + "prepared_review_unit_hash": "49046d1cea32082a35ba982144ad15477d7fd5ebb8fecba9fcc6799261dda54c", "component_budget": { "node_count": 2, "item_count": 2, @@ -27957,18 +28002,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests address unrelated code paths: trainer label-smoothing/gradient-accumulation vs. generation TopH warping. They should not be treated as duplicates.", + "summary": "The two pull requests are not duplicates: one fixes label smoothing + gradient accumulation in training, the other changes TopHLogitsWarper to honor min_tokens_to_keep in generation. Different code paths, different bugs, different subsystems.", "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 47261 is the better canonical representative because it has a clear, explicit issue target and a narrowly defined training bug fix; PR 47262 is a separate generation change.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "PR 47261 is the strongest representative of the cluster because it is issue-backed and addresses a concrete, reproducible correctness problem. PR 47262 is unrelated and should not be grouped with it as a duplicate.", "soft_edge_verdicts": [ { "left": "pull_request:47261", "right": "pull_request:47262", "accept": false, - "reason": "Different subsystems and fixes: one changes training loss scaling with gradient accumulation, the other adds/validates min_tokens_to_keep in TopHLogitsWarper. They are not the same underlying change." + "reason": "These fix different problems in different areas of the library: trainer loss scaling vs. generation token filtering. They do not appear to be the same underlying change and should not be merged as duplicates." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index a6e16e0ced6b71942c3104548a13b8ca4b6b9681..30ea567eb208cf7b3b9931b5cfe8418a1d2ab19f 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260714t120113z", + "analysis_id": "hybrid-model-20260714t180038z", "archived_artifacts": { - "hybrid": "snapshots/20260714T120113Z/analysis-runs/hybrid-model-20260714t120113z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260714T120113Z/analysis-runs/hybrid-model-20260714t120113z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-07-14T12:13:51Z", + "published_at": "2026-07-14T18:15:00Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260714T120113Z", + "snapshot_id": "20260714T180038Z", "variant": "hybrid" } diff --git a/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.json b/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..6dab861fec441e86ec0687d7f4885ef95a6cfd1a --- /dev/null +++ b/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.json @@ -0,0 +1,5228 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260714T180038Z", + "generated_at": "2026-07-14T18:14:57Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "cluster_id": "cluster-43451-14", + "summary": "Cluster of 1 issues and 13 PRs centered on issue #47321.", + "status": "open", + "confidence": 0.6, + "canonical_issue_number": 47321, + "canonical_pr_number": 47322, + "issue_numbers": [ + 47321 + ], + "pr_numbers": [ + 43451, + 46376, + 46509, + 46658, + 46836, + 47143, + 47163, + 47181, + 47216, + 47292, + 47301, + 47311, + 47322 + ], + "evidence_types": [ + "closing_reference", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43451, + "right_pr_number": 46376, + "code_similarity": 0.069, + "size_similarity": 0.215, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.176, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "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": 0.293, + "size_similarity": 0.836, + "file_overlap": 0.073, + "area_overlap": 0.17, + "patch_similarity": 0.296, + "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, + 822 + ] + ] + }, + { + "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": 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 + ], + [ + 494, + 501 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47143, + "code_similarity": 0.155, + "size_similarity": 0.475, + "file_overlap": 0.1, + "area_overlap": 0.0, + "patch_similarity": 0.197, + "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": 47181, + "code_similarity": 0.285, + "size_similarity": 0.907, + "file_overlap": 0.147, + "area_overlap": 0.021, + "patch_similarity": 0.347, + "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" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 1375, + 1382 + ] + ], + "right_ranges": [ + [ + 1249, + 1256 + ] + ] + }, + { + "filename": "src/transformers/conversion_mapping.py", + "left_ranges": [ + [ + 737, + 769 + ] + ], + "right_ranges": [ + [ + 1633, + 1653 + ] + ] + }, + { + "filename": "src/transformers/models/__init__.py", + "left_ranges": [ + [ + 290, + 296 + ] + ], + "right_ranges": [ + [ + 78, + 84 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 391, + 398 + ], + [ + 824, + 830 + ], + [ + 945, + 951 + ], + [ + 1068, + 1074 + ], + [ + 1202, + 1208 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 331, + 338 + ], + [ + 1094, + 1100 + ], + [ + 1913, + 1919 + ] + ], + "right_ranges": [ + [ + 97, + 105 + ], + [ + 1053, + 1059 + ] + ] + } + ] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47216, + "code_similarity": 0.042, + "size_similarity": 0.11, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.131, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47292, + "code_similarity": 0.002, + "size_similarity": 0.01, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47301, + "code_similarity": 0.007, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.039, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43451, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.01, + "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": 46376, + "right_pr_number": 46509, + "code_similarity": 0.142, + "size_similarity": 0.569, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.191, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 46658, + "code_similarity": 0.061, + "size_similarity": 0.18, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.17, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 46836, + "code_similarity": 0.073, + "size_similarity": 0.233, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.176, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47143, + "code_similarity": 0.109, + "size_similarity": 0.453, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.123, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47163, + "code_similarity": 0.008, + "size_similarity": 0.021, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.027, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47181, + "code_similarity": 0.075, + "size_similarity": 0.237, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.186, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47216, + "code_similarity": 0.126, + "size_similarity": 0.511, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.162, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47292, + "code_similarity": 0.01, + "size_similarity": 0.046, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.005, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47301, + "code_similarity": 0.012, + "size_similarity": 0.022, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.051, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47311, + "code_similarity": 0.005, + "size_similarity": 0.013, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.015, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46376, + "right_pr_number": 47322, + "code_similarity": 0.003, + "size_similarity": 0.01, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46509, + "right_pr_number": 46658, + "code_similarity": 0.138, + "size_similarity": 0.316, + "file_overlap": 0.053, + "area_overlap": 0.081, + "patch_similarity": 0.204, + "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": 46836, + "code_similarity": 0.147, + "size_similarity": 0.41, + "file_overlap": 0.103, + "area_overlap": 0.0, + "patch_similarity": 0.232, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "src/transformers/models/auto/auto_mappings.py", + "src/transformers/models/auto/modeling_auto.py", + "utils/check_config_attributes.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 795, + 802 + ] + ], + "right_ranges": [ + [ + 893, + 900 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ] + }, + { + "filename": "utils/check_config_attributes.py", + "left_ranges": [ + [ + 42, + 50 + ] + ], + "right_ranges": [ + [ + 237, + 243 + ] + ] + } + ] + }, + { + "left_pr_number": 46509, + "right_pr_number": 47143, + "code_similarity": 0.255, + "size_similarity": 0.796, + "file_overlap": 0.074, + "area_overlap": 0.143, + "patch_similarity": 0.159, + "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": 47181, + "code_similarity": 0.153, + "size_similarity": 0.417, + "file_overlap": 0.094, + "area_overlap": 0.0, + "patch_similarity": 0.274, + "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": [ + [ + 795, + 802 + ] + ], + "right_ranges": [ + [ + 1249, + 1256 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 1077, + 1083 + ], + [ + 1206, + 1212 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1097, + 1103 + ] + ], + "right_ranges": [ + [ + 97, + 105 + ], + [ + 1053, + 1059 + ] + ] + } + ] + }, + { + "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": 47292, + "code_similarity": 0.006, + "size_similarity": 0.026, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.005, + "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": 47311, + "code_similarity": 0.003, + "size_similarity": 0.007, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.01, + "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": 46658, + "right_pr_number": 46836, + "code_similarity": 0.22, + "size_similarity": 0.771, + "file_overlap": 0.06, + "area_overlap": 0.0, + "patch_similarity": 0.32, + "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": [ + [ + 739, + 822 + ] + ], + "right_ranges": [ + [ + 127, + 133 + ], + [ + 1495, + 1545 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ], + "right_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1085, + 1091 + ] + ], + "right_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ] + } + ] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47143, + "code_similarity": 0.124, + "size_similarity": 0.397, + "file_overlap": 0.054, + "area_overlap": 0.0, + "patch_similarity": 0.187, + "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": 0.004, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.016, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47181, + "code_similarity": 0.216, + "size_similarity": 0.758, + "file_overlap": 0.071, + "area_overlap": 0.0, + "patch_similarity": 0.284, + "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": [ + [ + 739, + 822 + ] + ], + "right_ranges": [ + [ + 1633, + 1653 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 362, + 368 + ], + [ + 591, + 599 + ], + [ + 798, + 804 + ], + [ + 884, + 891 + ], + [ + 1084, + 1090 + ], + [ + 1207, + 1213 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 1085, + 1091 + ] + ], + "right_ranges": [ + [ + 97, + 105 + ], + [ + 1053, + 1059 + ] + ] + } + ] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47216, + "code_similarity": 0.038, + "size_similarity": 0.092, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.129, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47292, + "code_similarity": 0.002, + "size_similarity": 0.008, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47301, + "code_similarity": 0.005, + "size_similarity": 0.004, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.03, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.007, + "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": 46836, + "right_pr_number": 47143, + "code_similarity": 0.155, + "size_similarity": 0.515, + "file_overlap": 0.077, + "area_overlap": 0.0, + "patch_similarity": 0.19, + "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": [ + [ + 453, + 459 + ] + ], + "right_ranges": [ + [ + 305, + 311 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ], + "right_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ], + "right_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ] + } + ] + }, + { + "left_pr_number": 46836, + "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": 46836, + "right_pr_number": 47181, + "code_similarity": 0.278, + "size_similarity": 0.983, + "file_overlap": 0.116, + "area_overlap": 0.0, + "patch_similarity": 0.31, + "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" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 893, + 900 + ] + ], + "right_ranges": [ + [ + 1249, + 1256 + ] + ] + }, + { + "filename": "src/transformers/conversion_mapping.py", + "left_ranges": [ + [ + 127, + 133 + ], + [ + 1495, + 1545 + ] + ], + "right_ranges": [ + [ + 1633, + 1653 + ] + ] + }, + { + "filename": "src/transformers/models/__init__.py", + "left_ranges": [ + [ + 453, + 459 + ] + ], + "right_ranges": [ + [ + 78, + 84 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 637, + 647 + ], + [ + 919, + 928 + ], + [ + 1130, + 1136 + ], + [ + 1260, + 1266 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 509, + 515 + ], + [ + 1119, + 1125 + ] + ], + "right_ranges": [ + [ + 97, + 105 + ], + [ + 1053, + 1059 + ] + ] + } + ] + }, + { + "left_pr_number": 46836, + "right_pr_number": 47216, + "code_similarity": 0.043, + "size_similarity": 0.119, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.127, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46836, + "right_pr_number": 47292, + "code_similarity": 0.003, + "size_similarity": 0.011, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46836, + "right_pr_number": 47301, + "code_similarity": 0.006, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.035, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46836, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.008, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46836, + "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": 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": 47181, + "code_similarity": 0.163, + "size_similarity": 0.524, + "file_overlap": 0.097, + "area_overlap": 0.0, + "patch_similarity": 0.196, + "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": [ + [ + 305, + 311 + ] + ], + "right_ranges": [ + [ + 78, + 84 + ] + ] + }, + { + "filename": "src/transformers/models/auto/auto_mappings.py", + "left_ranges": [ + [ + 412, + 418 + ], + [ + 975, + 981 + ] + ], + "right_ranges": [ + [ + 105, + 114 + ], + [ + 737, + 745 + ], + [ + 942, + 948 + ], + [ + 1027, + 1033 + ], + [ + 1161, + 1167 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 348, + 354 + ], + [ + 1960, + 1966 + ] + ], + "right_ranges": [ + [ + 97, + 105 + ], + [ + 1053, + 1059 + ] + ] + } + ] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47216, + "code_similarity": 0.061, + "size_similarity": 0.231, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.1, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47292, + "code_similarity": 0.005, + "size_similarity": 0.021, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.003, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47301, + "code_similarity": 0.006, + "size_similarity": 0.01, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.024, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47143, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.006, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.006, + "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": 47163, + "right_pr_number": 47181, + "code_similarity": 0.004, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.022, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47216, + "code_similarity": 0.014, + "size_similarity": 0.04, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.039, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47292, + "code_similarity": 0.093, + "size_similarity": 0.452, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.019, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47301, + "code_similarity": 0.192, + "size_similarity": 0.933, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.033, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47163, + "right_pr_number": 47311, + "code_similarity": 0.125, + "size_similarity": 0.607, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.025, + "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": 47181, + "right_pr_number": 47216, + "code_similarity": 0.046, + "size_similarity": 0.121, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.146, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47181, + "right_pr_number": 47292, + "code_similarity": 0.003, + "size_similarity": 0.011, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47181, + "right_pr_number": 47301, + "code_similarity": 0.007, + "size_similarity": 0.005, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.042, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47181, + "right_pr_number": 47311, + "code_similarity": 0.002, + "size_similarity": 0.003, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.008, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47181, + "right_pr_number": 47322, + "code_similarity": 0.001, + "size_similarity": 0.002, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.004, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47216, + "right_pr_number": 47292, + "code_similarity": 0.019, + "size_similarity": 0.089, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.005, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47216, + "right_pr_number": 47301, + "code_similarity": 0.02, + "size_similarity": 0.043, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.075, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47216, + "right_pr_number": 47311, + "code_similarity": 0.007, + "size_similarity": 0.024, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.016, + "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": 47292, + "right_pr_number": 47301, + "code_similarity": 0.098, + "size_similarity": 0.484, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.006, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47292, + "right_pr_number": 47311, + "code_similarity": 0.058, + "size_similarity": 0.274, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.019, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47292, + "right_pr_number": 47322, + "code_similarity": 0.045, + "size_similarity": 0.226, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.0, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47301, + "right_pr_number": 47311, + "code_similarity": 0.117, + "size_similarity": 0.567, + "file_overlap": 0.0, + "area_overlap": 0.0, + "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, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.026, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47311, + "right_pr_number": 47322, + "code_similarity": 0.165, + "size_similarity": 0.824, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.0, + "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-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-46894-5", + "summary": "Cluster of 1 issues and 4 PRs centered on issue #46894.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46894, + "canonical_pr_number": 46903, + "issue_numbers": [ + 46894 + ], + "pr_numbers": [ + 46895, + 46903, + 46958, + 47090 + ], + "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": 0.819, + "size_similarity": 0.932, + "file_overlap": 1.0, + "area_overlap": 0.783, + "patch_similarity": 0.389, + "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, + 222 + ] + ] + } + ] + }, + { + "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": 0.519, + "size_similarity": 0.433, + "file_overlap": 0.333, + "area_overlap": 0.867, + "patch_similarity": 0.192, + "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": [ + [ + 637, + 649 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 500, + 514 + ] + ], + "right_ranges": [ + [ + 500, + 512 + ] + ] + } + ] + }, + { + "left_pr_number": 46958, + "right_pr_number": 47090, + "code_similarity": 0.742, + "size_similarity": 0.852, + "file_overlap": 1.0, + "area_overlap": 0.611, + "patch_similarity": 0.387, + "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, + 222 + ] + ] + } + ] + } + ] + }, + { + "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-46693-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46693.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 46693, + "canonical_pr_number": 47215, + "issue_numbers": [ + 46693 + ], + "pr_numbers": [ + 46696, + 47215 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46696, + "right_pr_number": 47215, + "code_similarity": 0.138, + "size_similarity": 0.022, + "file_overlap": 0.2, + "area_overlap": 0.191, + "patch_similarity": 0.043, + "shared_filenames": [ + "src/transformers/modeling_flash_attention_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/modeling_flash_attention_utils.py", + "left_ranges": [ + [ + 667, + 692 + ] + ], + "right_ranges": [ + [ + 33, + 39 + ], + [ + 317, + 324 + ], + [ + 368, + 375 + ], + [ + 491, + 499 + ], + [ + 680, + 702 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43240-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 43240, + "canonical_pr_number": 43251, + "issue_numbers": [ + 43240 + ], + "pr_numbers": [ + 43251, + 43254 + ], + "evidence_types": [ + "closing_reference" + ], + "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": 0.55, + "canonical_issue_number": 46305, + "canonical_pr_number": 47162, + "issue_numbers": [ + 46305 + ], + "pr_numbers": [ + 46310, + 47162 + ], + "evidence_types": [ + "closing_reference" + ], + "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-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-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-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 evidence." + }, + { + "cluster_id": "cluster-43451-14", + "canonical_pr_number": 47322, + "duplicate_pr_numbers": [ + 43451, + 46376, + 46509, + 46658, + 46836, + 47143, + 47163, + 47181, + 47216, + 47292, + 47301, + 47311 + ], + "target_issue_number": 47321, + "reason": "PRs in cluster-43451-14 are treated as duplicates because they converge on issue #47321 with closing_reference, 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 evidence." + }, + { + "cluster_id": "cluster-46693-3", + "canonical_pr_number": 47215, + "duplicate_pr_numbers": [ + 46696 + ], + "target_issue_number": 46693, + "reason": "PRs in cluster-46693-3 are treated as duplicates because they converge on issue #46693 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-46894-5", + "canonical_pr_number": 46903, + "duplicate_pr_numbers": [ + 46895, + 46958, + 47090 + ], + "target_issue_number": 46894, + "reason": "PRs in cluster-46894-5 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." + } + ], + "best_issue": { + "cluster_id": "cluster-43451-14", + "issue_number": 47321, + "reason": "Issue #47321 is the strongest global issue candidate because it is open, belongs to a cluster with 14 artifacts, and carries 1 discussion comments plus 2 inbound references.", + "score": 244.0 + }, + "best_pr": { + "cluster_id": "cluster-43451-14", + "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 + } +} diff --git a/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..791c2f77439b385e96c8821cc1a5e028fae75734 --- /dev/null +++ b/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,28028 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260714T180038Z", + "generated_at": "2026-07-14T18:14:57Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 1, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "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:43827", + "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:44805|issue:45161", + "issue:46165|issue:46468", + "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:44841|issue:46661", + "issue:24643|issue:30277", + "issue:43716|issue:46869", + "issue:34928|issue:44805", + "issue:45685|issue:46869", + "issue:45859|issue:45902", + "issue:34928|issue:36331", + "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:34928|issue:41720", + "issue:43425|issue:44292", + "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", + "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" + ], + "prepared_review_unit_hash": "cfec7e21f83a9b79e252126344e54f65642b52412576616ed7d1c7318463b481", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 62, + "serialized_chars": 46278, + "estimated_input_tokens": 11570, + "estimated_eval_tokens": 23396 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a broad soft-similarity bucket, not a true duplicate set. The only clear duplicate pair is the two DeepSpeed ZeRO-3/BERT IndexError reports; the rest span unrelated model, docs, trainer, and device-mapping bugs.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 43596 is the best representative because it matches the only clearly duplicate issue (43638) and its title is the more general form of the shared DeepSpeed ZeRO-3/BERT IndexError.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43596 is the strongest canonical choice for the duplicate subcluster; it is the clearest, most general statement of the shared bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different Qwen3.5 problems: missing config field vs streaming chat/completions API failure." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Unrelated failures: AMD GPU support for GPT-OSS-20B vs an MPS histogram implementation error in MoE." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different models and failure modes: image dtype mismatch vs GPT-OSS on AMD GPUs." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Different bugs: autocast dtype mismatch in Phi-tiny-MoE vs GPT-OSS AMD GPU failure." + }, + { + "left": "issue:45237", + "right": "issue:46468", + "accept": false, + "reason": "Different model/example issues with no shared concrete code path." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different MoE/GPU problems; not the same bug." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch is unrelated to ModernBERT ONNX export." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Different model/export issue vs Qwen3Moe FP8 failure." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export error is unrelated to Qwen3 auto device-map cuda assert." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Table QA crash and outdated RAG docs are unrelated." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export vs import-path breakage are different problems." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are v5 docs cleanup items, but they target different removed features/pages and are not clearly the same bug." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "TRl SFT embedding unpacking and training-resume/remote-code issues are unrelated." + }, + { + "left": "issue:46165", + "right": "issue:46869", + "accept": false, + "reason": "Training resume with remote code is unrelated to multi-GPU YOLO device-map auto failure." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask/tensor shape mismatch and Gemma4 PLE device mismatch are different failures." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Both mention index errors, but one is DeepSpeed ZeRO-3/BERT initialization and the other is a mask-shape mismatch." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Same broad IndexError theme only; different underlying code paths and triggers." + }, + { + "left": "issue:44805", + "right": "issue:46294", + "accept": false, + "reason": "Mask-shape indexing error is unrelated to Qwen3.5 video token/feature mismatch." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Different runtime failures: mask/index mismatch vs GPT-OSS MoE tensor-parallel issue." + }, + { + "left": "issue:46165", + "right": "issue:46468", + "accept": false, + "reason": "Training-resume issue is unrelated to the broken Nemotron example." + }, + { + "left": "issue:44560", + "right": "issue:46165", + "accept": false, + "reason": "Video StopIteration and training resume/remote code 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": "Different DeepSpeed-related failures: collective mismatch vs BERT zero-3 index error." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same underlying DeepSpeed ZeRO-3/BERT initialization bug causing the same IndexError; these are clear duplicates." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross_entropy call and GPT-OSS tensor-parallel issue are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Different problems: detokenization bug vs LoRA save/reload mismatch." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Qwen-3-NVFP4 runtime error and LoRA save/reload mismatch are unrelated." + }, + { + "left": "issue:44841", + "right": "issue:46661", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to remote-code model_class/config_class errors." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Different DeepSpeed training errors with different symptoms and likely different root causes." + }, + { + "left": "issue:43716", + "right": "issue:46869", + "accept": false, + "reason": "Different device-mapping failures for different model families." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation-checkpointing/FSDP tensor-size recompute mismatch is unrelated to mask-shape indexing error." + }, + { + "left": "issue:45685", + "right": "issue:46869", + "accept": false, + "reason": "MPS MoE histogram error and YOLO multi-GPU device-map failure are unrelated." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Same Qwen3.5 MoE rotary-embedding area, but the reported failures are different enough that they do not clearly look like one duplicate bug." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "FSDP activation-checkpointing mismatch and CustomTrainer compute_loss signature breakage are unrelated." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "BERT zero-3 index error and GPT-OSS MoE tensor-parallel issue are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "Different models and different failure modes." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Dtype mismatch in Mistral is unrelated to GraniteMoEHybrid invalid method call." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Different models and runtime environments; not the same bug." + }, + { + "left": "issue:45874", + "right": "issue:46294", + "accept": false, + "reason": "Gemma4 reverse-embedding lookup and Qwen3.5 video chat-template mismatch are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Gemma3TextConfig v4.53/vLLM attribute issue and Mistral regex patch crash are unrelated." + }, + { + "left": "issue:45800", + "right": "issue:46515", + "accept": false, + "reason": "Both mention torch 2.4.1 compatibility, but they are separate incompatibility reports with no shared concrete bug." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Different Gemma4 failures: device mismatch vs expensive reverse embedding lookup." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "FSDP activation-checkpointing mismatch is unrelated to Qwen3 auto device-map cuda assert." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "General Torch incompatibility and Qwen-3-8B NVFP4 runtime error are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Different model families and different failure modes." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Mixtral grouped_mm tracing error is unrelated to Mistral image-preprocessor dtype mismatch." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-tiny-MoE and Qwen-3-8B NVFP4 runtime error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different models and hardware/runtime paths." + }, + { + "left": "issue:45507", + "right": "issue:45685", + "accept": false, + "reason": "Invalid method call in GraniteMoEHybrid is unrelated to MPS histogram not implemented." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Qwen NVFP4 runtime error and Qwen3.5 embedding unpacking with TRL are different bugs." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Different model/runtime failures." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT unpacking issue is unrelated to GPT-OSS tensor-parallel failure." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross_entropy call is unrelated to TRL SFT embedding unpacking." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading index error and BERT ZeRO-3 index error are similar only superficially; different model-specific roots." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are v5 docs issues, but one is about pipeline tasks and the other about the transformers run command; not clearly one duplicate change." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Phi-tiny-MoE autocast mismatch and MPS histogram error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Liger Kernel cross_entropy usage and Phi-tiny-MoE autocast mismatch are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and trainer loss-signature change are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "FSDP activation-checkpointing issue is unrelated to the GenerationMixin import breakage." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "GPT-OSS AMD GPU failure and RAG docs examples are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross_entropy issue is unrelated to tensor mask shape mismatch." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The canonical choice of issue:43596 over issue:43638 is reasonable given the nearly identical DeepSpeed ZeRO-3/BERT IndexError titles. The soft-edge verdicts are mostly cautious false negatives for clearly different bug families, and I did not see an overstatement that would require rejection." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 2, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36010", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:43039", + "issue:43296", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43596", + "issue:43638", + "issue:43792", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44512", + "issue:44560", + "issue:44589", + "issue:44841", + "issue:44863", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45237", + "issue:45362", + "issue:45399", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45800", + "issue:45812", + "issue:45823", + "issue:45901", + "issue:45920", + "issue:46015", + "issue:46294", + "issue:46304", + "issue:46402", + "issue:46661", + "issue:46772", + "issue:46869", + "issue:47228" + ], + "soft_pairs": [ + "issue:45800|issue:46304", + "issue:43296|issue:46304", + "issue:43866|issue:46402", + "issue:39290|issue:41720", + "issue:41762|issue:43596", + "issue:43901|issue:44512", + "issue:45237|issue:45901", + "issue:44589|issue:46304", + "issue:43039|issue:44560", + "issue:44841|issue:45823", + "issue:45823|issue:46294", + "issue:43975|issue:46015", + "issue:43866|issue:46772", + "issue:44863|issue:46015", + "issue:45020|issue:46304", + "issue:45812|issue:45920", + "issue:43866|issue:44863", + "issue:44292|issue:46015", + "issue:43872|issue:46304", + "issue:39290|issue:43531", + "issue:41093|issue:43638", + "issue:43792|issue:45901", + "issue:43854|issue:43866", + "issue:43425|issue:45070", + "issue:43541|issue:47228", + "issue:43975|issue:45901", + "issue:43792|issue:45070", + "issue:45362|issue:45464", + "issue:43596|issue:45800", + "issue:43541|issue:46869", + "issue:44292|issue:45901", + "issue:39290|issue:45823", + "issue:45464|issue:45685", + "issue:36010|issue:39290", + "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" + ], + "prepared_review_unit_hash": "61c6bae1e435fbe0cde27bb16809303b6b556479392ab196c9cc5280f8aca99d", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 41, + "serialized_chars": 39738, + "estimated_input_tokens": 9935, + "estimated_eval_tokens": 20126 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issue reports. The only strong duplicate signal is the AutoTokenizer wrong-token-IDs regression pair (Granite-specific vs broader multi-model report); the rest are different model, device-mapping, docs, or backend-compatibility bugs.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 45920 is the broadest and most representative report of the tokenizer regression, covering multiple model families rather than a single Granite-only case.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45920 is the best global representative because it captures the shared v5 AutoTokenizer regression with the widest impact and clearest duplicate cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45800", + "right": "issue:46304", + "accept": false, + "reason": "Both mention version/torch incompatibility, but they describe different failures and code paths." + }, + { + "left": "issue:43296", + "right": "issue:46304", + "accept": false, + "reason": "PaddleOCR-VL loading in vLLM is unrelated to the USE_DISTRIBUTED=0 PyTorch support regression." + }, + { + "left": "issue:43866", + "right": "issue:46402", + "accept": false, + "reason": "Checkpoint corruption and missing text-encoder weights are different defects." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Gemma3 sliding_window_pattern missing in vLLM is unrelated to Qwen3 auto device-map CUDA asserts." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Both are ZeRO-3 index errors, but the models and likely failure points differ; not enough evidence of the same bug." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "One is a pipeline docs mismatch, the other is obsolete run-command docs; different documentation issues." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "AMD GPU model failure and table-question-answering crash are unrelated." + }, + { + "left": "issue:44589", + "right": "issue:46304", + "accept": false, + "reason": "Float8 storage lookup failure is a different backend/runtime issue from the distributed-build regression." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross-entropy routing and Qwen3-vl video StopIteration are different code paths." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Voxtral processor failure and Gemma4 device-mapping mismatch are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46294", + "accept": false, + "reason": "Both involve model loading/input handling, but they are distinct model-specific bugs." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Incorrect detokenization and outdated RAG docs are unrelated." + }, + { + "left": "issue:43866", + "right": "issue:46772", + "accept": false, + "reason": "Checkpoint corruption is unrelated to ModelOpt compatibility." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading and outdated RAG docs do not match." + }, + { + "left": "issue:45020", + "right": "issue:46304", + "accept": false, + "reason": "Remote-code model breakage and USE_DISTRIBUTED=0 support issues are different regressions." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Both report the same v5 AutoTokenizer wrong-token-IDs regression; 45920 is the broader multi-model manifestation." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Corrupted checkpoint versus model implementation loading failure are different problems." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime failure is unrelated to RAG docs." + }, + { + "left": "issue:43872", + "right": "issue:46304", + "accept": false, + "reason": "bitsandbytes Int8Params incompatibility is unrelated to the distributed build regression." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both mention sliding_window/Qwen3, but one is Gemma3/vLLM config missing and the other is a Qwen3-MoE issue." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Mask shape mismatch and ZeRO-3 zero-size index errors are different failure modes." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper loading failure and table-question-answering crash are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Model load failure in tests and corrupted checkpoint are not the same bug." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch 2.10 incompatibility and pydantic model-field breakage are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:47228", + "accept": false, + "reason": "Different models, different compiler/tracing failures, no shared concrete code path." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization bug and table-QA crash are unrelated." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper load failure and PretrainedConfig/pydantic regression are unrelated." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both involve Qwen3.5, but one is chat crash and the other is streaming chat/completions API failure; not the same bug." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "ZeRO-3 BertModel index error and torch/version incompatibility are different issues." + }, + { + "left": "issue:43541", + "right": "issue:46869", + "accept": false, + "reason": "Torch Dynamo grouped_mm tracing failure and YOLO auto device-map failure are unrelated." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "NVFP4 runtime failure and table-question-answering crash do not share a concrete code path." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Different models and symptoms; one is missing config attr, the other is device mismatch." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Chat/completions streaming failure is unrelated to the MPS histogram_int bug." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "ImportError for GenerationMixin and Gemma3 sliding_window_pattern error are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Grouped_mm tracing failure and MPS histogram_int error are different backend issues." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Grouped_mm tracing failure and autocast dtype mismatch are not the same bug." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor failure and non-template-node compile error are unrelated." + }, + { + "left": "issue:45507", + "right": "issue:46869", + "accept": false, + "reason": "GraniteMoEHybrid invalid method call and YOLO multi-GPU auto device-map failure are different issues." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "LM-head weight tying bug and checkpoint corruption are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Mask/tensor shape mismatch and ZeRO-3 loading index error are different failures." + }, + { + "left": "issue:45399", + "right": "issue:46661", + "accept": false, + "reason": "Flash-Attn fallback gating and remote-code config_class AttributeError are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet. The only clearly supported duplicate signal is the AutoTokenizer v5 regression pair (45812/45920), and choosing 45920 as the broader canonical issue is reasonable from the titles alone. The other soft-edge rejections stay conservative and do not over-merge distinct titles into the same bug." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 3, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:39186", + "issue:40990", + "issue:41093", + "issue:42915", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43531", + "issue:43572", + "issue:43596", + "issue:43645", + "issue:43716", + "issue:43828", + "issue:43872", + "issue:43873", + "issue:43994", + "issue:44291", + "issue:44360", + "issue:44387", + "issue:44479", + "issue:44492", + "issue:44512", + "issue:44530", + "issue:44805", + "issue:44918", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45183", + "issue:45399", + "issue:45464", + "issue:45507", + "issue:45632", + "issue:45698", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45902", + "issue:46165", + "issue:46294", + "issue:46463", + "issue:46468", + "issue:46640", + "issue:46661", + "issue:46869", + "issue:47228" + ], + "soft_pairs": [ + "issue:43572|issue:45070", + "issue:45800|issue:45823", + "issue:43296|issue:43366", + "issue:43872|issue:47228", + "issue:44805|issue:44918", + "issue:43716|issue:47228", + "issue:43828|issue:45464", + "issue:43366|issue:45464", + "issue:44291|issue:45070", + "issue:43531|issue:45070", + "issue:44387|issue:45005", + "issue:46869|issue:47228", + "issue:45507|issue:47228", + "issue:41093|issue:43596", + "issue:45020|issue:45800", + "issue:43716|issue:46294", + "issue:44479|issue:46294", + "issue:43054|issue:43994", + "issue:46468|issue:46869", + "issue:43596|issue:46165", + "issue:45759|issue:45902", + "issue:46640|issue:46661", + "issue:39186|issue:44918", + "issue:43716|issue:44918", + "issue:43645|issue:45399", + "issue:40990|issue:44960", + "issue:45632|issue:45698", + "issue:42915|issue:45399", + "issue:43404|issue:45005", + "issue:44360|issue:44512", + "issue:43873|issue:45005", + "issue:44492|issue:44512", + "issue:44530|issue:45005", + "issue:44918|issue:45507", + "issue:45183|issue:46463" + ], + "prepared_review_unit_hash": "e2a62ed698009476ed068f2f1b270cd6b75c007bd740d66f652d4ebefd8280ef", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38037, + "estimated_input_tokens": 9510, + "estimated_eval_tokens": 19276 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issue reports; none of the soft pairs look like true duplicates. The only loose thematic overlap is around remote-code/model-loading regressions, but the concrete failures differ too much to merge.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45020 is the broadest, most umbrella-style report in the set: a recent-transformers regression affecting remote-code models. It best serves as the representative issue for the closest thematic subcluster without overfitting to one model or one error string.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the most generally useful and central issue among the candidates, with broad impact across custom/remote-code model loading paths and the clearest chance of subsuming nearby regressions.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Different bugs: missing pad_token_idx after the 5.0 update vs pydantic field breakage in v5.4.0." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Both are compatibility/device-related, but one is a torch/transformers version incompatibility and the other is a device mismatch in Gemma4 PLE forward." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "Different failures: PaddleOCR-VL loading with transformers 4.53.0 vs GGUF support for gpt-oss architecture." + }, + { + "left": "issue:43872", + "right": "issue:47228", + "accept": false, + "reason": "Different incompatibilities: bitsandbytes _is_hf_initialized constructor error vs a Sam3 bf16 compilation regression." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Both show shape/indexing errors, but the contexts differ: mask/tensor length mismatch vs Qwen3.5 embedding unpacking with TRL SFT." + }, + { + "left": "issue:43716", + "right": "issue:47228", + "accept": false, + "reason": "Different models and root causes: image preprocessor dtype mismatch vs a bf16 compilation regression." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Autocast dtype mismatch for Phi-tiny-MoE is not the same as Qwen3.5 streaming chat/completions failure." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF gpt-oss support is unrelated to Qwen3.5 streaming inference." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "Both involve init/model loading regressions, but one is the unexpected _is_hf_initialized argument and the other is a pydantic PretrainedConfig field breakage." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is a different model-specific issue from the v5.4 PretrainedConfig/pydantic problem." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 quantization OOM is not the same as tied-weights issues in translation models." + }, + { + "left": "issue:46869", + "right": "issue:47228", + "accept": false, + "reason": "Multi-GPU device map auto for YOLO is unrelated to Sam3 bf16 compilation." + }, + { + "left": "issue:45507", + "right": "issue:47228", + "accept": false, + "reason": "GraniteMoEHybrid invalid method calls are a different bug than a compilation regression." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are IndexErrors, but the concrete shapes and code paths differ; no evidence they are the same bug." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "A remote-code regression and a torch/transformers version incompatibility are too broad and not the same concrete failure." + }, + { + "left": "issue:43716", + "right": "issue:46294", + "accept": false, + "reason": "Both touch multimodal/video-ish paths, but one is image dtype mismatch and the other is a video feature/token count mismatch." + }, + { + "left": "issue:44479", + "right": "issue:46294", + "accept": false, + "reason": "Both concern Qwen multimodal/video models, but one is a v5.3 video input regression and the other is a specific apply_chat_template feature/token mismatch." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Siglip2 text embeddings being worse than Siglip is a model-quality report, not the same as AutoModel/pipeline nonsensical outputs." + }, + { + "left": "issue:46468", + "right": "issue:46869", + "accept": false, + "reason": "Official Nemotron example broken is unrelated to YOLO multi-GPU device map behavior." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "Different loading flows: deepspeed zero3/BertModel index error vs resuming training with local model and trust_remote_code." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Both involve Qwen multimodal configs, but one is AutoModel.from_config text_config unwrapping and the other is uninitialized inv_freq after to_empty materialization." + }, + { + "left": "issue:46640", + "right": "issue:46661", + "accept": false, + "reason": "Flash Attention 2 issue and remote-code model_class.config_class AttributeError are different code paths." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP 2-D weight RuntimeError is unrelated to Qwen3.5 embedding unpacking with TRL." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Image preprocessor dtype mismatch is not the same bug as Qwen3.5 embedding unpacking failures." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Jupyter custom model initialization breakage is different from flash-attn2 fallback being blocked by other checks." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "High perplexity on gpt-oss is a model behavior issue, not the same as a GLM5 model topic." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both involve trust_remote_code/custom modules, but one is cache-path collision and the other is loading the wrong module after save_pretrained." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure is unrelated to flash-attn2 fallback logic." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Mistral3 tied lm_head bug is a model-specific weight-tying issue, not translation-model tied-weights across v5." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "A DSA indexer discussion and docs mentioning a removed run command are different kinds of changes/problems." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization offloading behavior is not the same as tied-weights regression in translation models." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "A cache strategy typo and outdated docs about a removed command are unrelated." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weights behavior in translation models." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Qwen3.5 embedding unpacking and GraniteMoEHybrid invalid method calls are different failures." + }, + { + "left": "issue:45183", + "right": "issue:46463", + "accept": false, + "reason": "Both are validation/error-message bugs, but one is general XOR input validation and the other is T5 decoder input validation." + } + ] + }, + "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": 28, + "review_unit_index": 4, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:24643", + "issue:28530", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:41553", + "issue:41762", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43493", + "issue:43643", + "issue:43644", + "issue:43824", + "issue:43828", + "issue:43856", + "issue:43872", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44292", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44863", + "issue:44910", + "issue:44928", + "issue:44991", + "issue:45005", + "issue:45020", + "issue:45237", + "issue:45399", + "issue:45663", + "issue:45701", + "issue:45859", + "issue:45902", + "issue:45920", + "issue:45987", + "issue:46032", + "issue:46489", + "issue:46491", + "issue:46858", + "issue:46899", + "issue:46932", + "issue:47246" + ], + "soft_pairs": [ + "issue:43975|issue:44292", + "issue:44863|issue:45005", + "issue:46489|issue:46491", + "issue:39692|issue:43054", + "issue:43950|issue:44534", + "issue:36331|issue:41762", + "issue:46032|issue:47246", + "issue:43644|issue:43950", + "issue:43872|issue:44589", + "issue:46899|issue:46932", + "issue:43054|issue:45701", + "issue:45920|issue:46489", + "issue:43828|issue:44928", + "issue:38175|issue:43054", + "issue:45987|issue:46858", + "issue:43950|issue:45902", + "issue:43054|issue:43493", + "issue:44910|issue:45859", + "issue:45663|issue:45902", + "issue:42947|issue:44387", + "issue:44403|issue:44451", + "issue:24643|issue:28530", + "issue:43975|issue:45237", + "issue:44403|issue:44488", + "issue:41762|issue:42915", + "issue:43643|issue:45020", + "issue:42947|issue:43856", + "issue:44991|issue:45399", + "issue:43824|issue:43872", + "issue:41553|issue:43883" + ], + "prepared_review_unit_hash": "094e39ecfc0b9635a3af2b61177f6871281bee2dba307ec17651bc22393ab1b1", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "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 a loose bundle of unrelated Transformers issues, with only a few true duplicate subthreads. The clearest duplicate groups are around non-persistent buffer corruption, `get_seq_length()` returning a tensor instead of an int, and Gemma4 quantized embedder dtype casting.", + "confidence": 0.72, + "canonical_issue_reason": "#43950 is the strongest central representative of a concrete v5 regression: it states the non-persistent-buffer corruption bug directly and also anchors related reports that look like the same failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "#43950 is the best standalone issue to keep as the hub because it names the bug clearly, is broad enough to cover nearby duplicates, and is more specific than the umbrella-style reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Different model-specific failures; no shared code path or concrete bug." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "Checkpoint loading failure vs tied-weights regression; too different to be duplicates." + }, + { + "left": "issue:46489", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but on different models with different symptoms." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Docs/example breakage vs model quality discrepancy; not the same bug." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same v5 non-persistent-buffer corruption regression described with nearly identical wording." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Trainer callback signature issue vs DeepSpeed ZeRO-3 loading error." + }, + { + "left": "issue:46032", + "right": "issue:47246", + "accept": false, + "reason": "Related Mamba-family bugs, but different models and different recurrence paths." + }, + { + "left": "issue:43643", + "right": "issue:43950", + "accept": false, + "reason": "Remote-code config loading bug is distinct from buffer corruption." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Different bitsandbytes/storage compatibility failures." + }, + { + "left": "issue:46899", + "right": "issue:46932", + "accept": true, + "reason": "Same Gemma4 quantized vision/audio dtype-casting bug; #46932 is essentially the duplicate fix report." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 quality/implementation issue vs tokenization-version change." + }, + { + "left": "issue:45920", + "right": "issue:46489", + "accept": false, + "reason": "Broad tokenizer regression vs DeepSeek-Coder-specific tokenizer bug; not clearly the same defect." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Different dtype/attention failures on different models and code paths." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Both involve SigLIP2, but one is zero probabilities and the other is relative embedding quality; too different." + }, + { + "left": "issue:45987", + "right": "issue:46858", + "accept": true, + "reason": "Same `cache.get_seq_length()` contract bug: it returns a tensor instead of an int and breaks downstream arithmetic." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": true, + "reason": "`inv_freq` uninitialized after materialization is a concrete manifestation of the same non-persistent-buffer corruption regression." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Likely related SigLIP2 implementation concerns, but not the same concrete failure." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Flash-Attn crash from 3D `position_ids` misinterpretation vs CPU-offload incompatibility." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Different Gemma4/FSDP2 state handling issue vs non-persistent-buffer regression." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Gradient-checkpointing/LoRA inefficiency vs int4 quantization OOM; distinct problems." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic loading noise vs a specific tokenizer-load failure." + }, + { + "left": "issue:24643", + "right": "issue:28530", + "accept": false, + "reason": "DeepSpeed training error vs early-stopping metric handling; unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45237", + "accept": false, + "reason": "Tokenizer regression vs AMD GPU execution failure." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Generic load noise vs a specific model loading failure." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "ZeRO-3 loading crash vs FP8 config failure; unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "#45020 is a broad umbrella for remote-code breakage, but #43643 is a specific AutoConfig missing-fields bug." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Different memory/performance issues in different training setups." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading regression vs flash-attn2 fallback gating." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Import symbol missing vs bitsandbytes parameter-constructor incompatibility." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Bad error message for tokenizer loading vs missing tied-weights key attribute." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative about duplicates. The three accepted subthreads are supported by the titles and pair evidence, and the rejected pairs generally look distinct enough to avoid overclaiming. No PR-mergeability issue applies here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 5, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43404", + "issue:43782", + "issue:43792", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43940", + "issue:43950", + "issue:44220", + "issue:44368", + "issue:44479", + "issue:44509", + "issue:44560", + "issue:44661", + "issue:44805", + "issue:44863", + "issue:44928", + "issue:44960", + "issue:45020", + "issue:45081", + "issue:45305", + "issue:45313", + "issue:45356", + "issue:45399", + "issue:45488", + "issue:45588", + "issue:45685", + "issue:45715", + "issue:45759", + "issue:45823", + "issue:45923", + "issue:46315", + "issue:46489", + "issue:46661", + "issue:46869", + "issue:46894", + "issue:46897" + ], + "soft_pairs": [ + "issue:45081|issue:45399", + "issue:45685|issue:45823", + "issue:43782|issue:44960", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:43828|issue:45823", + "issue:41553|issue:43881", + "issue:45020|issue:45356", + "issue:42491|issue:44368", + "issue:43872|issue:46869", + "issue:45715|issue:46315", + "issue:45759|issue:46661", + "issue:45488|issue:46489", + "issue:45020|issue:46661", + "issue:43827|issue:44509", + "issue:43404|issue:43950", + "issue:43881|issue:43883", + "issue:46894|issue:46897", + "issue:43792|issue:44220", + "issue:43257|issue:45305", + "issue:36331|issue:41093", + "issue:42947|issue:45923", + "issue:43940|issue:45313", + "issue:43854|issue:44863", + "issue:44479|issue:44560", + "issue:24643|issue:30064", + "issue:41093|issue:44805", + "issue:44928|issue:45588" + ], + "prepared_review_unit_hash": "12b3656de958e58fb61d7822d7d7a391a315251f0515713e93a0d5de44e56e17", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36624, + "estimated_input_tokens": 9156, + "estimated_eval_tokens": 18568 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous model- and subsystem-specific bug reports. The only clear duplicate-like pair is the v5 docs cleanup around removed pipeline tasks; most other soft pairs are same-broad-area but different concrete failures and should stay separate.", + "confidence": 0.67, + "canonical_issue_reason": "issue:45020 is the broadest umbrella in the set, covering a cross-cutting remote_code regression rather than a single model-specific manifestation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the best global issue anchor because it is general enough to represent a broader regression class, while most other items are narrowly model- or subsystem-specific.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer regex crash for Mistral loading vs flash-attn2 fallback gating; different code paths and failures." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "MPS histogram error in MoE code vs Gemma4 device-map forward mismatch; unrelated bugs." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL weight_only load failure vs GLM5 loading issue; different models and symptoms." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Broad remote_code regression vs Gemma4 PLE device mismatch; not the same underlying bug." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES add-new-model-like failure vs remote_code loading regression; different mechanisms." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-tiny-MoE vs Gemma4 device placement issue; not duplicate." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Bad Voxtral AutoTokenizer error message vs glm-4v-9b load failure; different model-loading problems." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Generic remote_code breakage vs Kimi-K2.5 tokenizer regression; different root causes." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "LoRA compatibility break across HF versions vs a warning about tie_word_embeddings; related family, but not the same bug." + }, + { + "left": "issue:43872", + "right": "issue:46869", + "accept": false, + "reason": "bitsandbytes Int8Params constructor incompatibility vs YOLO multi-GPU device-map failure; unrelated." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "convert_ids_to_tokens slowdown vs convert_tokens_to_ids slowdown; same subsystem but different hot paths and regressions." + }, + { + "left": "issue:45759", + "right": "issue:46661", + "accept": false, + "reason": "Composite config unwrapping bug vs remote-code config_class AttributeError; similar area, but different concrete failures." + }, + { + "left": "issue:45488", + "right": "issue:46489", + "accept": false, + "reason": "LlamaTokenizer pre-tokenizer override bug affecting DeepSeek V3/R1 vs DeepSeek-Coder v1 tokenizer regression; different underlying issues." + }, + { + "left": "issue:45020", + "right": "issue:46661", + "accept": false, + "reason": "Both concern remote_code loading, but one is a broad regression and the other a specific config_class error; not enough to treat as the same bug." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both report docs still referencing removed pipeline tasks after v5; same documentation cleanup." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Mistral3 lm_head tying bug vs silent corruption of non-persistent buffers in from_pretrained; different code paths." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "GLM-4v-9b loading failure vs missing all_tied_weights_keys in MolmoForCausalLM; distinct model-specific issues." + }, + { + "left": "issue:46894", + "right": "issue:46897", + "accept": false, + "reason": "Same loss-shift pattern, but different model families (CohereASR vs Florence2); too model-specific to merge as duplicates." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper-large-v2 load failure vs a fbank feature-extraction issue; not clearly the same bug." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Qwen3 MOE weight-loading failure under ZeRO-3 vs gradient averaging under GAS with DeepSpeed; different failures." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "CustomTrainer signature mismatch vs mask/tensor shape IndexError; unrelated." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "LoRA gradient checkpointing ineffective vs missing supports_gradient_checkpointing on a remote-code variant; related theme, different bug." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both are ZeRO-3 load failures, but they affect different Qwen variants and describe different missing-weight patterns; not enough evidence of one underlying bug." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "GLM-4.7-Flash unit-test loading issue vs NemotronH checkpoint loading failure; different models and causes." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-input regressions, but one is a broad v5.3.0 model regression and the other is a StopIteration in Qwen3-vl-embedding; not the same bug." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "DeepSpeed training RuntimeError about 2-D weights vs void segmentation map processing; completely different domains." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "Both mention mask shape mismatch, but they are separate reports with different dimensions and contexts; not enough to unify." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Qwen3.5 RLHF NaN/gradient explosion issue vs flash_attention sinkless-model AttributeError; unrelated failures." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: issue:45020 is a reasonable broad anchor, the cluster is indeed mostly heterogeneous/model-specific, and the only clear duplicate-like pair identified (43827/44509) is a conservative same-bug docs cleanup match. The other soft-edge rejections stay cautious and do not overclaim duplication." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 6, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:41553", + "issue:42617", + "issue:42915", + "issue:42947", + "issue:43366", + "issue:43531", + "issue:43550", + "issue:43596", + "issue:43638", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43824", + "issue:43828", + "issue:43950", + "issue:43957", + "issue:43994", + "issue:44155", + "issue:44488", + "issue:44530", + "issue:44560", + "issue:44589", + "issue:44661", + "issue:44683", + "issue:44805", + "issue:44863", + "issue:44877", + "issue:44910", + "issue:44918", + "issue:44991", + "issue:45030", + "issue:45325", + "issue:45584", + "issue:45663", + "issue:45685", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45902", + "issue:46402", + "issue:46468", + "issue:46489", + "issue:46515", + "issue:46858", + "issue:47254" + ], + "soft_pairs": [ + "issue:43716|issue:46468", + "issue:43366|issue:43828", + "issue:44530|issue:44863", + "issue:44863|issue:46402", + "issue:43366|issue:43531", + "issue:43596|issue:46515", + "issue:41553|issue:44991", + "issue:45325|issue:45902", + "issue:43366|issue:44589", + "issue:43638|issue:46515", + "issue:45663|issue:45823", + "issue:44155|issue:45584", + "issue:42915|issue:43957", + "issue:43716|issue:44560", + "issue:43824|issue:43957", + "issue:43950|issue:45834", + "issue:42617|issue:43366", + "issue:41553|issue:44488", + "issue:44560|issue:45685", + "issue:43645|issue:43646", + "issue:44560|issue:44918", + "issue:36010|issue:42915", + "issue:44683|issue:46515", + "issue:44910|issue:46858", + "issue:43550|issue:44910", + "issue:42947|issue:47254", + "issue:45812|issue:46489", + "issue:45823|issue:46515", + "issue:44661|issue:45823", + "issue:44560|issue:44805", + "issue:38175|issue:43994", + "issue:44877|issue:45030", + "issue:43638|issue:44661", + "issue:43596|issue:44661" + ], + "prepared_review_unit_hash": "8e342b78d56c95aded481e37859fc3ff9b75e02eb0aba88a1a8266f3f7f8e6d5", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 34, + "serialized_chars": 37342, + "estimated_input_tokens": 9336, + "estimated_eval_tokens": 18928 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous cluster with one clear duplicate pair: the Transformers 5.0.0 custom model initialization reports (43645/43646). The SigLIP2 pair (38175/43994) looks related but not strong enough from the packet to treat as a definite duplicate.", + "confidence": 0.69, + "canonical_issue_reason": "Issue 43646 is the cleanest representative of the only clearly duplicated concrete bug: the Transformers 5.0.0 custom model initialization regression.", + "canonical_pr_reason": null, + "best_issue_reason": "43646 is the best anchor because it states the underlying regression directly, without the notebook-specific repro detail from 43645.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43716", + "right": "issue:46468", + "accept": false, + "reason": "Different model families and different failure modes; only a broad loading/debugging theme overlaps." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "Unrelated bugs: GGUF architecture support versus an autocast dtype mismatch." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache/group parsing crash and Nemotron checkpoint loading are different code paths." + }, + { + "left": "issue:44863", + "right": "issue:46402", + "accept": false, + "reason": "Both are loading problems, but for different models and different breakages." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "Different feature areas: GGUF model support versus Qwen3-MoE sliding-window behavior." + }, + { + "left": "issue:43596", + "right": "issue:46515", + "accept": false, + "reason": "Specific ZeRO-3/Bert indexing failure versus a broad torch-version incompatibility claim." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Both concern tokenizer/model loading, but the model and failure mechanism differ." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Different Qwen variants and different rotary/position-id bugs." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "Completely different issue classes: model support versus float8 storage lookup." + }, + { + "left": "issue:43638", + "right": "issue:46515", + "accept": false, + "reason": "The Bert/ZeRO-3 index error is not the same as a generic torch incompatibility report." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but different bugs: shared_kv_states rebuild versus device mismatch." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "Different audio/Whisper paths and different symptoms." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Different root causes: FP8 config failure versus meta-device loading regressions." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Different model families and different crash modes." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Importing a missing class is not the same bug as meta-device loading corruption." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption and OCR index errors are unrelated." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "Unrelated: a parallel script failure versus GGUF model support." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Both mention loading failures, but they affect different models and likely different code paths." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Different failures: Qwen video StopIteration versus an MPS histogram type issue." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Same underlying Transformers 5.0.0 custom-model initialization regression; 43645 just adds a notebook-specific repro context." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Both are Qwen3.5-related, but the failure mechanisms are distinct." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "Importing GenerationMixin and a FineGrainedFP8Config failure are unrelated problems." + }, + { + "left": "issue:44683", + "right": "issue:46515", + "accept": false, + "reason": "Torch-version compatibility is too broad here; the concrete bugs differ." + }, + { + "left": "issue:44910", + "right": "issue:46858", + "accept": false, + "reason": "Both are attention/cache issues, but one is 3D position_ids on Qwen3.5 and the other is device-tensor seq-length arithmetic." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Different models and different attention backends; not the same defect." + }, + { + "left": "issue:42947", + "right": "issue:47254", + "accept": false, + "reason": "Both involve PEFT/checkpointing, but one is ineffective checkpointing and the other is a ZeRO-3 CheckpointError." + }, + { + "left": "issue:45812", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions, but for different model families and likely different fixes." + }, + { + "left": "issue:45823", + "right": "issue:46515", + "accept": false, + "reason": "A Gemma4 device-placement bug is not the same as a generic torch incompatibility." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Different problems: tokenizer-mapping validation versus Gemma4 device mismatch." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different index-error contexts and different input pipelines." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": false, + "reason": "Likely related via SigLIP2 output corruption, but the packet only supports a cautious 'related' call, not a definite duplicate." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both are config validation issues, but they involve different models and different validation changes." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "ZeRO-3 Bert indexing and tokenizer-mapping validation are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Different bug class and different subsystem; only superficial error-text overlap." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The analysis is mostly cautious, but it overstates at least one soft-edge match. In particular, issue:38175 vs issue:43994 is accepted as the same underlying regression, which is stronger than the packet supports from titles alone; a more conservative call would be \"likely related\" rather than a definite duplicate. The rest of the negatives look reasonably grounded, but this positive edge is not conservative enough to accept the cluster as written." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 7, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:38175", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43632", + "issue:43644", + "issue:43645", + "issue:43708", + "issue:43720", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44075", + "issue:44164", + "issue:44291", + "issue:44387", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44661", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44960", + "issue:44977", + "issue:44991", + "issue:45276", + "issue:45292", + "issue:45313", + "issue:45335", + "issue:45357", + "issue:45488", + "issue:45507", + "issue:45672", + "issue:45812", + "issue:45834", + "issue:46050", + "issue:46459", + "issue:46468", + "issue:46512", + "issue:46846", + "issue:46899" + ], + "soft_pairs": [ + "issue:45672|issue:46899", + "issue:44661|issue:44805", + "issue:42491|issue:43720", + "issue:43632|issue:45834", + "issue:43317|issue:44756", + "issue:38175|issue:39692", + "issue:41720|issue:42915", + "issue:43644|issue:44534", + "issue:42915|issue:44756", + "issue:42491|issue:45276", + "issue:44960|issue:44977", + "issue:44451|issue:44488", + "issue:44863|issue:46846", + "issue:45313|issue:45357", + "issue:45507|issue:46468", + "issue:46459|issue:46512", + "issue:43644|issue:43645", + "issue:39692|issue:43994", + "issue:43257|issue:43854", + "issue:44488|issue:44991", + "issue:43856|issue:43873", + "issue:43854|issue:43931", + "issue:42491|issue:45335", + "issue:42491|issue:45292", + "issue:43856|issue:44387", + "issue:43708|issue:44075", + "issue:44960|issue:46050", + "issue:44164|issue:44291", + "issue:45488|issue:45812", + "issue:45276|issue:45335" + ], + "prepared_review_unit_hash": "508260a5781cd771acf2bbe1333575fe7f2bc596c2e607fa79510ad2b5624325", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36036, + "estimated_input_tokens": 9009, + "estimated_eval_tokens": 18274 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly heterogeneous; only a couple of soft pairs are true duplicates of the same concrete bug. The clearest duplicate cluster is the Transformers v5 non-persistent buffer regression, with one likely duplicate around `resize_token_embeddings` behavior.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43644 is the cleanest representative: it states the underlying v5 regression directly and has an exact duplicate in issue:44534.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43644 is the best issue to anchor on because it is broad, specific, and clearly reproducible; it also captures the shared bug more directly than the paired duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:46899", + "accept": true, + "reason": "Same Gemma4 multimodal 4-bit dtype-cast failure; both describe the same unsafe `uint8`/NF4 embedders bug." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Both are failures, but one is tokenizer-mapping logic and the other is a mask/tensor shape error; different code paths and symptoms." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Different models and different failures: qwen3_moe LoRA compatibility vs BitNet packed-weight unpacking in accelerate loading." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "Unrelated regressions: `_is_hf_initialized` in Transformers v5 vs a Kosmos2.5 OCR long-input index error." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Both involve loading/memory, but one is quantized model offload on device_map auto and the other is Strix Halo mmap OOM; not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same model family, but one is zero probabilities and the other is a broken docs example with processor mismatch/quantization failure; not the same concrete issue." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Different Qwen3 failure modes: CUDA assert on A800 vs FineGrainedFP8Config loading failure." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-exact duplicate: both report Transformers v5 filling non-persistent buffers with junk." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Different subsystems and symptoms; Qwen3Moe FP8 loading vs Strix Halo mmap memory issues." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "Both are model-specific regressions, but one is Qwen3 MOE LoRA loading and the other is Gemma4 token-embedding resizing." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Unrelated model families and failures: GLM5 vs Qwen3.5 flash-attention generation." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are tokenizer/model loading reports, but they concern different checkpoints and the evidence is too weak to call them the same bug." + }, + { + "left": "issue:44863", + "right": "issue:46846", + "accept": false, + "reason": "Different problems: NemotronH checkpoint loading vs Qwen3.5 TP-plan missing `linear_attn` entries." + }, + { + "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 keys on save; not the same bug." + }, + { + "left": "issue:46459", + "right": "issue:46512", + "accept": false, + "reason": "Both concern dtype handling, but one is loading composite checkpoints and the other is `from_config(dtype=...)` leaving config stale; different APIs and failure modes." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "They both relate to Transformers v5 regressions, but one is buffer initialization and the other is custom model initialization in notebooks." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both mention SigLIP2, but one is a docs example and the other is bad inference outputs; not clearly the same underlying defect." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different model families and loading paths: Qwen3 MOE deepspeed/accelerate conversion vs GLM-4.7-Flash unit-test loading." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading failures for different models, but the evidence does not show a shared concrete root cause." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Different issues: Qwen3 MoE training memory inefficiency vs quantization offloading not working." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Both are model-loading errors, but the affected models and failure mechanisms differ (GLM-4.7-Flash vs Qwen3-VL shape mismatch)." + }, + { + "left": "issue:42491", + "right": "issue:45335", + "accept": false, + "reason": "Different model families and different failures; Qwen3 MOE LoRA loading is unrelated to t5gemma embedding resizing." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Same as above: unrelated Qwen3 MOE loading issue vs generic resize-token-embeddings behavior." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is int4 quantization reserved-memory/OOM in v5." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Trainer checkpoint-resume math bug vs optimizer args being ignored; different code paths." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "Unrelated issues: GLM5 model work vs quantized save/load breakage after `sinq` quantization." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Both involve loading/saving, but one is extra_state handling and the other is `init_empty_weights` / `_is_hf_initialized`; not the same bug." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is LlamaTokenizer pre-tokenizer override and the other is wrong Granite token IDs; different concrete problems." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": true, + "reason": "Same underlying `resize_token_embeddings` regression: the resized embeddings are not propagated to model-specific output/decoder embeddings." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The canonical choice of issue:43644 is well supported by the exact title match with issue:44534, and the accepted soft pairs are plausible same-bug matches based on the issue titles. No PR-mergeability concerns apply here because this is an issues-only set." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 8, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42796", + "issue:42947", + "issue:43335", + "issue:43381", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43645", + "issue:43828", + "issue:43854", + "issue:43873", + "issue:43957", + "issue:44079", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44479", + "issue:44488", + "issue:44756", + "issue:44863", + "issue:44877", + "issue:44910", + "issue:44912", + "issue:44928", + "issue:44960", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45362", + "issue:45406", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45859", + "issue:46326", + "issue:46468", + "issue:46829", + "issue:46897", + "issue:46901" + ], + "soft_pairs": [ + "issue:46897|issue:46901", + "issue:45362|issue:45406", + "issue:43425|issue:43957", + "issue:44291|issue:44756", + "issue:44402|issue:44488", + "issue:45663|issue:45759", + "issue:44910|issue:45588", + "issue:41720|issue:44155", + "issue:43404|issue:43454", + "issue:39692|issue:44960", + "issue:43873|issue:44387", + "issue:42947|issue:45704", + "issue:45663|issue:45859", + "issue:44079|issue:46829", + "issue:44292|issue:44912", + "issue:43957|issue:44756", + "issue:44155|issue:45823", + "issue:36010|issue:39404", + "issue:43335|issue:45663", + "issue:44387|issue:44756", + "issue:44863|issue:46468", + "issue:44479|issue:44877", + "issue:45276|issue:45292", + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:42796|issue:46326", + "issue:45588|issue:45663", + "issue:43828|issue:43957", + "issue:43645|issue:44164", + "issue:43854|issue:45216" + ], + "prepared_review_unit_hash": "ee9a77371c88832cf0202f5635bd0baef2f4ed9f43ed918fc1ec550b185ce47d", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36784, + "estimated_input_tokens": 9196, + "estimated_eval_tokens": 18648 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous. Only two soft pairs look like true duplicates: the `resize_token_embeddings` regression pair and the training-loss double-shift pair. The remaining pairs share broad subsystems or vocabulary but do not look like the same concrete bug/change.", + "confidence": 0.88, + "canonical_issue_reason": "Issue:45276 is the strongest representative because it states the broader `resize_token_embeddings` regression, and issue:45292 appears to be the same bug from a narrower angle. The rest of the cluster is too mixed to provide a better single canonical issue.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue:45276 is the best issue representative: it is specific, clearly described, and aligns with the only clearly duplicate issue in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46897", + "right": "issue:46901", + "accept": true, + "reason": "Both describe the same training-loss double-shift problem: the model is training against shifted labels when it should not. Different models, but the underlying bug pattern matches." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different failures in different paths: one is a Qwen chat crash, the other is a Gemma4 processor `_tokenizer` attribute error in serve." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Both are version/loading regressions, but they are different root causes: Torch 2.10 incompatibility vs `meta`-device model loading failures." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "One is an `init_empty_weights` / `_is_hf_initialized` constructor mismatch; the other is an OOM workaround for mmap on Strix Halo. Not the same bug." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Both involve model loading, but the specific problems differ: vocab-size token placement vs a separate loading failure for `cjvt/sleng-bert`." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Different code paths: FSDP2 attention state rebuilding in Gemma4 vs `from_config` not unwrapping `text_config` in Qwen multimodal configs." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both touch flash attention, but the failures are distinct: 3D `position_ids` packing misinterpretation vs `s_aux=None` AttributeError for sink-less models." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Unrelated issues: a Qwen3 device-mapping CUDA assert vs an AudioFlamingo3 batched inference embedding/token leak." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Very similar symptom, but they concern different model classes and likely separate fixes; not enough to treat as the same underlying bug with high confidence." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Different models and different problem domains: SigLIP2 documentation errors vs a generic GLM5 issue with no matching bug description." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Both mention quantization and OOM-like behavior, but one is about offloading behavior and the other about increased reserved memory under int4 quantization." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Completely different bugs: gradient checkpointing with PEFT LoRA vs a T5/apex RMSNorm memory leak." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different model families and different failures: Gemma4 FSDP2 KeyError vs Qwen3.5 MoE rotary embedding CPU-offload incompatibility." + }, + { + "left": "issue:44079", + "right": "issue:46829", + "accept": false, + "reason": "Both concern output containers, but one is a generic `ModelOutput` key assignment bug and the other is a VideoPrism output field mix-up." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Both are quantization-related loading issues, but they involve different formats and symptoms (`_is_hf_initialized`/loading error vs MXFP4 fallback to bf16)." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Different root causes and different layers: `meta`-device loading regressions vs an OOM avoidance change for mmap on Strix Halo." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "Distinct problems: AudioFlamingo3 track leakage in batched inference vs Gemma4 PLE device mismatch under `device_map=\"auto\"`." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "Unrelated: an import path break for `GenerationMixin` vs Whisper pipeline `return_language` regression." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Different model/config bugs: sparse layer creation in SwitchTransformers vs Gemma4 FSDP2 attention-state rebuilding." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Both can lead to OOM, but one is reserved-memory growth under int4 quantization and the other is a platform-specific mmap workaround." + }, + { + "left": "issue:44863", + "right": "issue:46468", + "accept": false, + "reason": "Related family name, but not the same concrete bug: one cannot load NemotronH checkpoints, the other is a broken official example." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Different regressions in different areas: video-input handling for Qwen VL variants vs strict config loading for granite_speech." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same underlying regression: `resize_token_embeddings` fails to propagate correctly to output embeddings; 45276 just includes the additional `embed_tokens_per_layer` detail." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Different training-time issues: gradient checkpointing in eval mode vs Qwen3.5 RLHF instability from 3D `position_ids` and SDPA fallback." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated save/load extra-state handling vs a memory/OOM workaround." + }, + { + "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` behavior for `mistral_common` tokenizers." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Different subsystems and failures: flash attention `s_aux=None` crash vs Gemma4 FSDP2 shared KV state rebuilding." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Different failures: autocast dtype mismatch for Phi-tiny-MoE vs `meta`-device loading breaks some models." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "One is a Jupyter/custom-model initialization regression in Transformers 5.0, the other is save/from_pretrained handling of extra state." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Both are model-specific loading/save issues, but they involve different models and different APIs (`unit tests` load failure vs `save_pretrained` regression)." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative on the soft-edge calls. The two accepted pairs are well supported by the titles, and the canonical/best issue choice around the resize_token_embeddings regression is reasonable. 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": 28, + "review_unit_index": 9, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36032", + "issue:39404", + "issue:43425", + "issue:43452", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43901", + "issue:43906", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44671", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44898", + "issue:44912", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45292", + "issue:45335", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46050", + "issue:46315", + "issue:46326", + "issue:46491", + "issue:46665", + "issue:46869", + "issue:46894", + "issue:46901", + "issue:47228" + ], + "soft_pairs": [ + "issue:45684|issue:45698", + "issue:43425|issue:47228", + "issue:45292|issue:45335", + "issue:44589|issue:44756", + "issue:43646|issue:43708", + "issue:44402|issue:44451", + "issue:43425|issue:43828", + "issue:46894|issue:46901", + "issue:46315|issue:46491", + "issue:45812|issue:46315", + "issue:43425|issue:46869", + "issue:44530|issue:45084", + "issue:44683|issue:45823", + "issue:43611|issue:43645", + "issue:43425|issue:43716", + "issue:39404|issue:45701", + "issue:39404|issue:46665", + "issue:44479|issue:44805", + "issue:44589|issue:45084", + "issue:43901|issue:44509", + "issue:44164|issue:45759", + "issue:43994|issue:44079", + "issue:36032|issue:43452", + "issue:44265|issue:44898", + "issue:43906|issue:44530", + "issue:46050|issue:46326", + "issue:44991|issue:45081", + "issue:43906|issue:46326", + "issue:44671|issue:44991", + "issue:39404|issue:44912" + ], + "prepared_review_unit_hash": "22e69106b13eb217af433d24c3da1fe2b31f44766d0315c2dabcca3606fc317c", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36830, + "estimated_input_tokens": 9208, + "estimated_eval_tokens": 18672 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly distinct bugs; the only strong thematic overlap is around Transformers v5 regressions, but the concrete failure modes differ. I would anchor the cluster on the broad custom-model loading regression, not merge any of the soft pairs.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43611 is the broadest, most central regression item here: it describes a concrete v5 loading breakage affecting models via base_model_prefix, and several nearby issues are narrower variants of v5 custom-model/init/load failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best representative issue because it captures a general, user-facing Transformers v5 model-loading regression rather than a model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both concern custom-model save/load behavior, but one is about permission propagation during save_pretrained and the other about loading the wrong custom module after saving; different bugs and different code paths." + }, + { + "left": "issue:43425", + "right": "issue:47228", + "accept": false, + "reason": "Unrelated problems: torch 2.10 compatibility versus a Sam3 bf16 compilation regression between releases." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": false, + "reason": "Related API area, but one reports resize_token_embeddings not updating output_embeddings generally, while the other is a t5gemma-specific decoder.embed_tokens issue; not clearly the same concrete bug." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Completely different failures: missing Float8 storage type versus disabling mmap to avoid OOM on Strix Halo." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Different regressions: custom model initialization in notebooks versus Trainer resume_from_checkpoint/max_steps miscalculation." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Both are tokenizer-loading related, but one is an out-of-vocab special-token issue and the other is a specific model load failure for ScandiBERT; insufficient evidence of the same root cause." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Both mention dtype/torch issues, but one is torch 2.10 incompatibility and the other is an autocast dtype mismatch for a specific model; distinct bugs." + }, + { + "left": "issue:46894", + "right": "issue:46901", + "accept": false, + "reason": "Same broad double-shift pattern, but the reports are for different model families and may reflect separate implementations rather than one shared code-path bug." + }, + { + "left": "issue:46315", + "right": "issue:46491", + "accept": false, + "reason": "Different tokenizer regressions: performance slowdown in convert_tokens_to_ids versus an encode-decode leading-space round-trip behavior change." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Wrong token IDs for Granite models versus a tokenizer performance regression; not the same issue." + }, + { + "left": "issue:43425", + "right": "issue:46869", + "accept": false, + "reason": "Torch version incompatibility and YOLO multi-GPU device_map failure are unrelated." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "A Qwen3.5 PagedAttentionCache crash is not the same as a template-compilation TypeError." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Different failure modes and subsystems: compiled flex_attention on torch >=2.9 versus Gemma4 PLE device placement with device_map='auto'." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both are v5 custom-model related, but one is a base_model_prefix loading break and the other is notebook-specific model initialization behavior; not the same concrete bug." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch compatibility versus a Mistral-3 image preprocessor/model dtype mismatch." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline return_language regression and a tokenizer-version-induced tokenization change are different issues." + }, + { + "left": "issue:39404", + "right": "issue:46665", + "accept": false, + "reason": "Pipeline Whisper regression versus inconsistent from_pretrained behavior inside Python classes; distinct code paths." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression for Qwen models is not the same as a generic mask/index shape mismatch error." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Missing Float8 storage type and non-template-node compilation errors are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs-related, but they cover different removed/changed pipeline behaviors and should not be merged as one bug." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "Save/from_pretrained extra_state handling and from_config text_config unwrapping are separate serialization/configuration bugs." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Wrong SigLIP2 model outputs and ModelOutput key assignment behavior are different problems." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5 tokenizer special-token method conflict and gguf_file loading failures are unrelated tokenizer/model-loading issues." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export failures with torch_compilable_check are not the same as Perceiver non-default-resolution failures." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "An isolated reproduction of another issue is not evidence of the same underlying bug as the Qwen3.5 PagedAttentionCache crash." + }, + { + "left": "issue:46050", + "right": "issue:46326", + "accept": false, + "reason": "Quantized save/load corruption and a mistral_common continue_final_message bug are unrelated." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-loading failures, but one is a generic tokenizer load break for a specific model and the other is a Mistral regex-patching crash; different root causes." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT masked-LM prediction errors and tokenizer loading failure for EMBEDDIA/est-roberta are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:44912", + "accept": false, + "reason": "Torch 2.10 incompatibility and MXFP4 quantization falling back to bf16 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": 28, + "review_unit_index": 10, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42796", + "issue:42915", + "issue:43299", + "issue:43404", + "issue:43421", + "issue:43541", + "issue:43596", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43856", + "issue:43931", + "issue:44164", + "issue:44451", + "issue:44462", + "issue:44488", + "issue:44530", + "issue:44661", + "issue:44683", + "issue:44737", + "issue:44756", + "issue:44987", + "issue:45081", + "issue:45084", + "issue:45479", + "issue:45823", + "issue:45859", + "issue:46178", + "issue:46285", + "issue:46326", + "issue:46402", + "issue:46489", + "issue:46821", + "issue:46894", + "issue:46897", + "issue:46899", + "issue:47050" + ], + "soft_pairs": [ + "issue:45823|issue:46899", + "issue:44164|issue:46821", + "issue:43596|issue:44683", + "issue:43856|issue:44756", + "issue:45479|issue:46894", + "issue:44488|issue:44987", + "issue:43632|issue:47050", + "issue:44488|issue:45081", + "issue:36032|issue:42915", + "issue:43299|issue:43931", + "issue:45479|issue:46897", + "issue:43611|issue:43644", + "issue:42796|issue:43404", + "issue:43716|issue:45084", + "issue:44462|issue:46489", + "issue:46285|issue:46402", + "issue:44661|issue:44683", + "issue:43541|issue:45084", + "issue:44451|issue:44987", + "issue:42915|issue:44530", + "issue:32090|issue:39404", + "issue:43643|issue:44530", + "issue:44737|issue:45859", + "issue:40990|issue:43421", + "issue:36032|issue:43645", + "issue:34567|issue:40444", + "issue:43644|issue:43646", + "issue:40444|issue:46326", + "issue:42175|issue:46178" + ], + "prepared_review_unit_hash": "9751e332e97fb7131eb6d6ea760ef6d91bbda11841835afaa89b100b5eb4cd2d", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36543, + "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 a loose set of unrelated Transformers regressions, with many v5-related model/tokenizer loading or initialization bugs but no strong duplicate pairs. I would treat it as similarity clustering rather than a true duplicate set.", + "confidence": 0.66, + "canonical_issue_reason": "If one issue must represent the cluster, issue #43646 is the broadest and most generic v5 regression here: it captures the recurring custom-model initialization breakage that several other issues loosely resemble.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43646 is the best single representative because it is broad, user-facing, and closest to the repeated v5 initialization/loading failures across the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45823", + "right": "issue:46899", + "accept": false, + "reason": "Both are Gemma4 multimodal failures, but one is a device mismatch in forward and the other is a 4-bit dtype-cast bug in embedders; different code paths." + }, + { + "left": "issue:44164", + "right": "issue:46821", + "accept": false, + "reason": "Unrelated symptoms: extra_state save/load handling versus TimesFM window_size AttributeError." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Different problems in different areas: DeepSpeed ZeRO-3/BertModel init versus torch>=2.9 flex_attention compilation." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Both mention memory pain, but one is Qwen3 MoE training memory inefficiency and the other is mmap/OOM on Strix Halo; not the same bug." + }, + { + "left": "issue:45479", + "right": "issue:46894", + "accept": false, + "reason": "Different training-loss bugs: classification zero-loss degeneracy versus ASR double-shift behavior." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are model loading failures, but for different checkpoints and likely different root causes." + }, + { + "left": "issue:43632", + "right": "issue:47050", + "accept": false, + "reason": "v5 _is_hf_initialized flag breakage is unrelated to Canine fp16 ONNX attention dtype mismatch." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer-loading issues affecting different models and different failure points." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "T5 tokenizer method-name conflict is unrelated to Qwen3Moe FP8 config failure." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Closely related model family, but one is a broad Qwen3VL MoE load regression and the other is a specific checkpoint shape-mismatch load error; not clearly one concrete bug." + }, + { + "left": "issue:45479", + "right": "issue:46897", + "accept": false, + "reason": "Different model families and different loss bugs; only the broad theme of training-loss regressions is shared." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization junk." + }, + { + "left": "issue:42796", + "right": "issue:43404", + "accept": false, + "reason": "Both involve Mistral-family artifacts, but tokenizer BPE markers and lm_head weight tying are separate issues." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Mistral image-preprocessor dtype mismatch is unrelated to template-compilation TypeError." + }, + { + "left": "issue:44462", + "right": "issue:46489", + "accept": false, + "reason": "Both concern tokenizer behavior, but one ignores tokenizer.json while the other is a DeepSeek-Coder output regression on v5+." + }, + { + "left": "issue:46285", + "right": "issue:46402", + "accept": false, + "reason": "Different model-loading/structure problems for unrelated model families." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer-mapping model registration bug is unrelated to flex_attention compile failure." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch Dynamo grouped_mm tracing failure is unrelated to non-template-node compilation error." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are model-loading complaints, but for different repositories/models and likely different underlying causes." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Different Qwen3-family issues: FP8 training config versus PagedAttentionCache linear_attention crash." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer _gpu_broadcast_one NoneType error is unrelated to Whisper return_language pipeline behavior." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "AutoConfig trust_remote_code field loss is unrelated to Qwen3.5 PagedAttentionCache group-type crash." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both involve device/offload behavior, but XLNet CPU arange and Qwen3_5Moe rotary embedding offload are different code paths." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity regression on gpt-oss-20b is unrelated to runtime special-token/post-processor update behavior." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer method conflict and Jupyter custom-model initialization breakage are different bugs." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counter update and iterable-dataset multi-image finetuning failure are unrelated." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 initialization regressions, but one is junk-filled non-persistent buffers and the other is custom model initialization; different concrete bugs." + }, + { + "left": "issue:40444", + "right": "issue:46326", + "accept": false, + "reason": "Different multimodal training/tokenizer issues: multi-image IterableDataset failure versus mistral_common continue_final_message behavior." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "One is missing tensorflow in the torch extra install, the other is an invalid backend name in BACKENDS_MAPPING." + } + ] + }, + "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": 28, + "review_unit_index": 11, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28530", + "issue:34567", + "issue:36032", + "issue:42915", + "issue:43208", + "issue:43262", + "issue:43317", + "issue:43377", + "issue:43404", + "issue:43452", + "issue:43482", + "issue:43531", + "issue:43582", + "issue:43632", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43746", + "issue:43792", + "issue:43819", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44206", + "issue:44220", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44466", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44811", + "issue:44849", + "issue:44877", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45507", + "issue:45584", + "issue:46326", + "issue:46661", + "issue:46869", + "issue:46899", + "issue:47013", + "issue:47312" + ], + "soft_pairs": [ + "issue:44206|issue:44220", + "issue:43582|issue:43632", + "issue:44987|issue:44991", + "issue:45084|issue:45507", + "issue:43452|issue:43482", + "issue:44811|issue:45584", + "issue:43482|issue:47312", + "issue:43746|issue:46326", + "issue:43317|issue:43482", + "issue:43208|issue:47013", + "issue:44403|issue:45081", + "issue:43645|issue:46661", + "issue:43792|issue:46869", + "issue:43262|issue:43377", + "issue:42915|issue:45084", + "issue:43646|issue:44534", + "issue:44987|issue:45081", + "issue:43377|issue:43819", + "issue:44849|issue:45325", + "issue:42915|issue:44291", + "issue:43404|issue:44466", + "issue:28530|issue:34567", + "issue:36032|issue:43742", + "issue:44062|issue:44987", + "issue:44451|issue:45081", + "issue:43994|issue:44610", + "issue:44008|issue:46899", + "issue:42915|issue:44589", + "issue:43531|issue:44877" + ], + "prepared_review_unit_hash": "b47cda437914c50f0ba0580bf061ff14d0df87182ace07156ca9cc31ead80318", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36483, + "estimated_input_tokens": 9121, + "estimated_eval_tokens": 18498 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous. Most items are distinct bug reports with only superficial overlap around transformers v5 loading/init regressions, tokenizer/audio/model-preprocessing paths, or specific model-family issues. I do not see any true duplicate pairs among the proposed soft edges.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "43646 is the broadest representative of the recurring v5 custom-model-initialization regressions, but the cluster is not cohesive enough to treat it as a duplicate hub.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "Different audio bugs: an unsupported `center` argument in LASR feature extraction vs a failure in `_torch_extract_fbank_features()`." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Unrelated regressions: Apple Silicon allocator warmup TypeError vs `_is_hf_initialized`/model-init flag breakage." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are load failures, but they affect different models and likely different code paths; not the same bug." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Different failures: template-node compilation error vs GraniteMoEHybrid calling an invalid method." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both touch GGUF loading, but one is generic `gguf_file` handling and the other is a Qwen2.5-GGUF v5 regression; too different to merge." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Whisper decoding ignores `skip_special_tokens` vs empty-transcription failure after `align_special_tokens`; related area, different defects." + }, + { + "left": "issue:43482", + "right": "issue:47312", + "accept": false, + "reason": "Qwen2.5-GGUF loading failure vs a generic kernels 0.16 incompatibility; unrelated." + }, + { + "left": "issue:43746", + "right": "issue:46326", + "accept": false, + "reason": "PEFT local-checkpoint loading problem vs `continue_final_message` behavior for mistral_common tokenizers." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Different loading problems: device-map/offload dequantization failure vs Qwen2.5-GGUF v5 load failure." + }, + { + "left": "issue:43208", + "right": "issue:47013", + "accept": false, + "reason": "Both are xLSTM-related, but one is about three training blockers for small models and the other is a train-mode unpacking crash." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Unnecessary load-time noise vs a Mistral regex patch crash; not the same issue." + }, + { + "left": "issue:43645", + "right": "issue:46661", + "accept": false, + "reason": "Both involve model initialization/loading, but one is notebook-defined custom models and the other is remote-code models missing `config_class`." + }, + { + "left": "issue:43792", + "right": "issue:46869", + "accept": false, + "reason": "Whisper inference failure and YOLO multi-GPU `device_map=auto` failure are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug vs MIMI batching/padding-mask mismatch; different code paths." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3MoE FP8 config failure vs template-node compilation error; unrelated." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "General v5 custom-model-init regression vs non-persistent buffer corruption; too broad to be one concrete duplicate." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Different model-loading failures: `physical-intelligence/fast` vs Mistral tokenizer regex patch." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Different model bugs: MIMI batching/padding-mask issue vs DAC latent/forward mismatch with missing STE." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen bugs: hidden-states behavior vs still-image temporal position ID scaling." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3MoE FineGrainedFP8 failure vs `init_empty_weights` / `_is_hf_initialized` argument error; unrelated." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both mention `lm_head.weight`, but one is untied weights in Mistral3 and the other is device-dependent serialization inconsistency." + }, + { + "left": "issue:28530", + "right": "issue:34567", + "accept": false, + "reason": "Early stopping metric selection and token-count tracking are unrelated trainer issues." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Tokenizer `add_special_tokens` conflict vs MobileLLM load key error; unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken keyword conflict vs a model loading failure; no shared bug." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "ScandiBERT loading failure vs Mistral regex patch crash; unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different vision issues: bad SigLIP2 outputs vs OmDet processor/model input-size mismatch." + }, + { + "left": "issue:44008", + "right": "issue:46899", + "accept": false, + "reason": "Gemma 3n variable collision vs Gemma4 4-bit dtype-cast bug; different versions and code paths." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3MoE FP8 config bug vs missing Float8 storage type; unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3-MoE sliding-window issue vs GraniteSpeech strict config loading problem; 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": 28, + "review_unit_index": 12, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42796", + "issue:43089", + "issue:43296", + "issue:43322", + "issue:43381", + "issue:43425", + "issue:43531", + "issue:43540", + "issue:43611", + "issue:43632", + "issue:43716", + "issue:43746", + "issue:43792", + "issue:43872", + "issue:43883", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44464", + "issue:44479", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44811", + "issue:44991", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45812", + "issue:45923", + "issue:45987", + "issue:46315", + "issue:46491", + "issue:46710", + "issue:46821", + "issue:46899", + "issue:47254" + ], + "soft_pairs": [ + "issue:45923|issue:47254", + "issue:44464|issue:45987", + "issue:36032|issue:41720", + "issue:44265|issue:44610", + "issue:43632|issue:46821", + "issue:44479|issue:45325", + "issue:43089|issue:45987", + "issue:43322|issue:43540", + "issue:43425|issue:44403", + "issue:43883|issue:44488", + "issue:43296|issue:43531", + "issue:40444|issue:45325", + "issue:43872|issue:44291", + "issue:44220|issue:44683", + "issue:45216|issue:45357", + "issue:43611|issue:44534", + "issue:43716|issue:46899", + "issue:44991|issue:46315", + "issue:44403|issue:44991", + "issue:45812|issue:46491", + "issue:42796|issue:43746", + "issue:43381|issue:45923", + "issue:44291|issue:44589", + "issue:43792|issue:44589", + "issue:33666|issue:40444", + "issue:43632|issue:44756", + "issue:40990|issue:44811", + "issue:43883|issue:44991", + "issue:42796|issue:46710", + "issue:35707|issue:40444", + "issue:43531|issue:44291" + ], + "prepared_review_unit_hash": "ad7fba281069b7653e9f7390b9536f3d23119fd839e25fc55f2188d4da1023f0", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36369, + "estimated_input_tokens": 9093, + "estimated_eval_tokens": 18442 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mixed bag of unrelated model/tokenizer/runtime regressions. Only a couple of soft edges look like true duplicates: the `_is_hf_initialized` compatibility break and the Qwen3.5 `save_pretrained` checkpoint-key regression.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43872 is the strongest central report for the shared `_is_hf_initialized`/v5 compatibility regression: it has the most activity and inbound references, and issue:44291 is essentially the same failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43872 is the best representative issue because it states the core constructor-arg incompatibility clearly and appears to be the broadest, most reusable fix target in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45923", + "right": "issue:47254", + "accept": false, + "reason": "Different bugs: missing gradient-checkpointing support on a trust_remote_code model vs a PEFT+DeepSpeed checkpointing runtime error." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Unrelated generation problems: compiled-forward chunked decoding inconsistency vs `StaticCache.get_seq_length()` returning the wrong type/shape." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer constructor conflict and a CUDA assert during Qwen3 device mapping are different code paths." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "`torch.export`/`torch_compilable_check` failure is unrelated to the OmDet processor image-size mismatch." + }, + { + "left": "issue:43632", + "right": "issue:46821", + "accept": false, + "reason": "One is the `_is_hf_initialized` v5 regression; the other is a TimesFM `window_size` attribute bug." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Both involve Qwen vision/video behavior, but one is a broad video-input regression and the other is a specific rope-index scaling bug; not enough to treat as the same concrete issue." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation-performance overhead is a different problem from the cache length return-type bug." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Llava Next model loading segfault vs Qwen3OmniMoe video-input `ValueError`; distinct failures." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch version incompatibility and noisy transformer loading are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Missing `all_tied_weights_keys` on Molmo is unrelated to the sleng-bert tokenizer loading failure." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM and Qwen3-MoE sliding-window behavior are different bugs." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen2.5-VL related, but training with multi-image iterable datasets is a different bug from temporal position-id scaling." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Same underlying regression: both report v5 passing an unexpected `_is_hf_initialized` kwarg into model-loading/init code, with bitsandbytes as one affected path." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction issue and compiled flex-attention failure are unrelated." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same Qwen3.5 save_pretrained regression: both describe incorrect checkpoint/visual-encoder key saving after v5.4/5.5 changes." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "`base_model_prefix` loading break and junk-filled non-persistent buffers are separate v5 regressions." + }, + { + "left": "issue:43716", + "right": "issue:46899", + "accept": false, + "reason": "Both mention dtype issues in multimodal models, but they are different models and different failure mechanisms." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Tokenizer load failure vs `convert_tokens_to_ids` slowdown are different tokenizer issues." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic load noise and a tokenizer load regression are not the same bug." + }, + { + "left": "issue:45812", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different behavior (wrong IDs vs lost leading space)." + }, + { + "left": "issue:42796", + "right": "issue:43746", + "accept": false, + "reason": "Tokenizer BPE-marker handling and GraniteSpeech PEFT checkpoint loading are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Gradient-checkpointing-in-eval and missing `supports_gradient_checkpointing` are different code-path problems." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected `_is_hf_initialized` kwarg and missing Float8 storage support are different compatibility failures." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper loading failure and Float8 storage lookup error are unrelated." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and Qwen2.5-VL multi-image iterable-dataset failure are not the same bug." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "A v5 `_is_hf_initialized` regression and Strix Halo mmap OOM are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "High perplexity on a language-model eval and Whisper `batch_decode` skipping special tokens are different issues." + }, + { + "left": "issue:42796", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer BPE-marker handling and DeepSeek accuracy regression are unrelated." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with `inputs_embeds`/`past_key_values` is unrelated to Qwen2.5-VL multi-image iterable training." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and the `_is_hf_initialized` constructor-arg regression are different failures." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the two accepted soft edges are conservative clear matches: issue:43872\u2194issue:44291 share the same unexpected `_is_hf_initialized` regression, and issue:45216\u2194issue:45357 are both Qwen3.5 `save_pretrained` checkpoint/key-save regressions. The other pairs are appropriately rejected as distinct bugs." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 13, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:40444", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43421", + "issue:43454", + "issue:43482", + "issue:43493", + "issue:43525", + "issue:43606", + "issue:43611", + "issue:43644", + "issue:43646", + "issue:43701", + "issue:43818", + "issue:43856", + "issue:43873", + "issue:43994", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44462", + "issue:44466", + "issue:44509", + "issue:44534", + "issue:44568", + "issue:44829", + "issue:44877", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45701", + "issue:45812", + "issue:45834", + "issue:46402", + "issue:46821", + "issue:46909", + "issue:47217" + ], + "soft_pairs": [ + "issue:44079|issue:45663", + "issue:43482|issue:43525", + "issue:41720|issue:45081", + "issue:43421|issue:44568", + "issue:42175|issue:45542", + "issue:43296|issue:46821", + "issue:42175|issue:43296", + "issue:44462|issue:45812", + "issue:43482|issue:45538", + "issue:44509|issue:46909", + "issue:43317|issue:43856", + "issue:44079|issue:44534", + "issue:43611|issue:43646", + "issue:43377|issue:45491", + "issue:43818|issue:46402", + "issue:44265|issue:45072", + "issue:43644|issue:44079", + "issue:45325|issue:45834", + "issue:40444|issue:45381", + "issue:43818|issue:45357", + "issue:43493|issue:43994", + "issue:45325|issue:45381", + "issue:44291|issue:44877", + "issue:43454|issue:44466", + "issue:44829|issue:45479", + "issue:40444|issue:47217", + "issue:43873|issue:45701", + "issue:43701|issue:45325", + "issue:42796|issue:45381", + "issue:43606|issue:44265" + ], + "prepared_review_unit_hash": "cfeb5c69f0589d8699af65382269266a9d70c6ff49bd473c2d174a7171fc277c", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37034, + "estimated_input_tokens": 9259, + "estimated_eval_tokens": 18774 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous. A few soft edges are real duplicates around Transformers v5 tokenizer/model regressions, but most pairs are unrelated broad-similarity matches and should stay separate.", + "confidence": 0.69, + "canonical_issue_reason": "Best anchor for the cluster is issue 44462: it is a concrete, high-impact v5 AutoTokenizer regression with a clear duplicate-like counterpart (45812) and broader applicability than the narrower follow-up reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44462 is the strongest representative issue in this set because it captures a specific, user-visible regression in core tokenizer loading and generalizes beyond a single model family.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "Different bugs: ModelOutput key assignment vs Gemma4 shared_kv_states/FSDP2 behavior." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Unrelated failures: GGUF loading in v5 vs missing pad_token_id on Llama4Config." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Different code paths and symptoms: Qwen3 device mapping CUDA assert vs Mistral tokenizer regex patch crash." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Both involve special tokens, but one is runtime post-processor updates while the other is add_special_tokens behavior for a specific tokenizer." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both touch TF packaging/backend setup, but the reported failures are different and not clearly the same bug." + }, + { + "left": "issue:43296", + "right": "issue:46821", + "accept": false, + "reason": "Different models and failure modes: PaddleOCR-VL load error vs TimesFM window_size AttributeError." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "No meaningful overlap beyond broad dependency/backend concerns." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": true, + "reason": "Same underlying regression: AutoTokenizer/tokenizer.json handling causing wrong Granite token IDs in v5." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "Different issues: GGUF model loading vs CLIPTokenizer max length handling." + }, + { + "left": "issue:44509", + "right": "issue:46909", + "accept": false, + "reason": "Both are docs-related, but one is removed pipeline tasks and the other is a broken DIA example." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Different problems: quantized offload loading failure vs Qwen3 MoE training memory usage." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "Different bugs: ModelOutput key bookkeeping vs non-persistent buffer initialization." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 loading regressions, but the concrete causes differ: base_model_prefix handling vs custom model initialization." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Different models and root causes: padding-mask support in MIMI vs Gemma3 NaNs with sliding-window attention." + }, + { + "left": "issue:43818", + "right": "issue:46402", + "accept": false, + "reason": "Unrelated model-specific failures: Video-LLaVA architecture issue vs Sam3 text encoder weights not loading." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "Different subsystems: torch.export/torch_compilable_check vs bfloat16 dtype mismatch inference." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Non-persistent buffer junk vs ModelOutput key assignment are separate bugs." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both are vision/model-specific, but the actual failures are unrelated (Qwen2.5-VL RoPE indexing vs Kosmos2.5 long OCR index error)." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both mention Qwen2.5-VL/image positions, but one is finetuning with multiple images and the other is video vision_position_ids." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Different issues in different models: Video-LLaVA weights/temporal attention vs Qwen3.5 save_pretrained key serialization." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "Same underlying SigLIP2 bug: HF implementation diverges from the original and yields bad/nonsensical outputs." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Same model family, but still-image temporal position_ids and video vision_position_ids are distinct bug reports." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "Different load-time failures: init_empty_weights unexpected argument vs strict config blocking granite_speech." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both involve lm_head/tied weights, but one is generation quality and the other is save serialization differing by device." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Different training regressions with different triggers and model families." + }, + { + "left": "issue:40444", + "right": "issue:47217", + "accept": false, + "reason": "Both are VL placeholder-related, but the failing scenarios are different: multi-image finetuning vs placeholder placement in user text." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Offloading/quantization behavior and tokenization-version changes are unrelated." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL RoPE indexing." + }, + { + "left": "issue:42796", + "right": "issue:45381", + "accept": false, + "reason": "Different tokenizer/model bugs: BPE marker handling vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "Different bugs: CPU offload device mismatch vs torch.export compatibility." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-identical duplicate reports of the same non-persistent buffer junk regression in v5." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative overall. The two accepted soft edges are plausible same-bug duplicates from the titles, while the rejected pairs are mostly explained by clearly different symptoms, models, or subsystems. No obvious 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": 28, + "review_unit_index": 14, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:36331", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42617", + "issue:42796", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43262", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43454", + "issue:43550", + "issue:43572", + "issue:43582", + "issue:43606", + "issue:43720", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43828", + "issue:43873", + "issue:44186", + "issue:44265", + "issue:44479", + "issue:44534", + "issue:44568", + "issue:44610", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45584", + "issue:45759", + "issue:45823", + "issue:45920", + "issue:45923", + "issue:46050", + "issue:46209", + "issue:46305", + "issue:46315", + "issue:46821" + ], + "soft_pairs": [ + "issue:44479|issue:45823", + "issue:43828|issue:46821", + "issue:40990|issue:45584", + "issue:43262|issue:43746", + "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:43257|issue:45923", + "issue:45823|issue:46821", + "issue:42617|issue:42915", + "issue:40990|issue:42947", + "issue:43454|issue:45072", + "issue:43377|issue:44610", + "issue:42796|issue:46209", + "issue:44568|issue:45920", + "issue:41553|issue:43582", + "issue:36331|issue:42617", + "issue:43317|issue:43873", + "issue:43749|issue:44534", + "issue:43454|issue:43720", + "issue:43377|issue:44898", + "issue:43572|issue:45759", + "issue:41720|issue:43322", + "issue:46305|issue:46315", + "issue:36331|issue:42915", + "issue:44877|issue:44991", + "issue:43322|issue:46050", + "issue:41720|issue:44977" + ], + "prepared_review_unit_hash": "0828ba4d2fb3b8e23c37e8d8f960e5c4f9493601c7f3b161f73cc97cb7676d9c", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36473, + "estimated_input_tokens": 9119, + "estimated_eval_tokens": 18494 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is very heterogeneous: most items are distinct transformer regressions across tokenization, generation, loading/offloading, audio, and export paths. I found no soft pair that looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43873 is the broadest umbrella-style report in the set and best represents the offloading/quantization failure class, with multiple inbound references and discussion. The cluster is not truly duplicate-heavy, but this is the most suitable representative issue.", + "canonical_pr_reason": null, + "best_issue_reason": "43873 is the best global issue candidate because it describes the core system-level regression rather than a narrow model-specific symptom, making it the most useful triage anchor for related reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Both are multimodal/model-loading regressions, but one is a video-input regression in Qwen VL variants and the other is a Gemma4 device mismatch during forward. Different code paths and symptoms." + }, + { + "left": "issue:43828", + "right": "issue:46821", + "accept": false, + "reason": "Different models and failures: autocast dtype mismatch in Phi-tiny-MoE versus an AttributeError for TimesFM window_size." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "Perplexity regression on gpt-oss-20b is unrelated to Whisper failing on empty transcription after align_special_tokens." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio-related issues, but one is chat template sampling-rate defaulting and the other is PEFT adapter checkpoint loading. Not the same bug." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "BarkProcessor voice_preset and progressive generation with inputs_embeds/past_key_values are unrelated bugs." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Generation cache behavior and Qwen3 auto device mapping/cuda assert are different failure modes in different paths." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer crash on NER/batching is unrelated to torch.export failing on torch_compilable_check." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both involve device placement, but one is a cudaErrorAssert on Qwen3 auto device mapping and the other is a Gemma4 PLE device mismatch during forward." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Bark-small CPU offload device mismatch and DAC.from_latents mismatch with missing STE are different models and different bugs." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "AyaVision weight-tying failure and Bamba torch.compile/SDPA failure are unrelated." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "Both touch Qwen3/MoE-ish configs, but one is weight conversion under accelerate+deepspeed and the other is a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:45823", + "right": "issue:46821", + "accept": false, + "reason": "Gemma4 device mismatch and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "3d_parallel.py runtime failure is not the same as Qwen3Moe failing with FineGrainedFP8Config." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "High perplexity on a model is not the same bug as gradient checkpointing being ineffective with PEFT LoRA." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "Weight tying for AyaVision and dtype mismatches in SwitchTransformers/TimmWrapperModel are different issues." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI padding-mask inconsistency and OmDet-Turbo image-size mismatch are unrelated processor/model bugs." + }, + { + "left": "issue:42796", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer BPE marker handling and FP8 inference failure on B200 are different bugs." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "add_special_tokens BOS/EOS regression in one tokenizer is not the same as broad wrong-token-id regressions across many tokenizers." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Bad AutoTokenizer error message for Voxtral is unrelated to caching_allocator_warmup raising TypeError on Apple Silicon." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "CustomTrainer compute_loss signature breakage and 3d_parallel.py startup failure are different problems." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both involve offloading/quantization, but one is dequantized model loading with device_map=auto and the other is a broader offloading regression. Not enough evidence of the same concrete code path." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk in v5 are distinct regressions." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "AyaVision lm_head tying and BitNet packed-weight unpacking during accelerate loading are unrelated." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "MIMI batched padding-mask bug and Perceiver non-default resolution failure are different model-specific issues." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig and failure to unwrap text_config for composite Qwen multimodal configs are different config-loading bugs." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3 auto device mapping cuda assert and Llava Next segmentation fault during loading are unrelated failures." + }, + { + "left": "issue:46305", + "right": "issue:46315", + "accept": false, + "reason": "A sliding-window off-by-one bug and a slow convert_tokens_to_ids regression are not the same issue." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "CustomTrainer compute_loss keyword incompatibility and Qwen3Moe FineGrainedFP8Config failure are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict granite_speech config loading and tokenizer loading failure for est-roberta are different code paths." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Llava Next load-time segfault and quantized-model save/load breakage after sinq are different bugs." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Both are Qwen3-family issues, but one is a device-mapping cuda assert and the other is a flash-attention generation regression. 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": 28, + "review_unit_index": 15, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:26413", + "issue:28530", + "issue:33666", + "issue:36010", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43065", + "issue:43295", + "issue:43317", + "issue:43404", + "issue:43450", + "issue:43550", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43708", + "issue:43720", + "issue:43873", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44387", + "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:46402", + "issue:46612", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:46869", + "issue:46897", + "issue:47020", + "issue:47317" + ], + "soft_pairs": [ + "issue:44387|issue:45704", + "issue:40990|issue:45701", + "issue:41720|issue:42617", + "issue:43450|issue:45701", + "issue:44466|issue:45357", + "issue:46897|issue:47317", + "issue:43873|issue:44756", + "issue:26413|issue:28530", + "issue:45685|issue:46821", + "issue:44186|issue:45072", + "issue:43295|issue:45701", + "issue:43065|issue:46402", + "issue:44451|issue:47020", + "issue:36010|issue:42617", + "issue:46612|issue:46858", + "issue:43646|issue:46665", + "issue:43550|issue:43720", + "issue:43577|issue:43708", + "issue:43550|issue:43606", + "issue:44186|issue:44610", + "issue:43404|issue:45357", + "issue:43881|issue:44488", + "issue:42915|issue:43317", + "issue:46821|issue:46869", + "issue:44610|issue:44898", + "issue:41720|issue:45161", + "issue:28530|issue:33666", + "issue:44008|issue:45823", + "issue:44423|issue:45406" + ], + "prepared_review_unit_hash": "e09ab4085d731ab4f803df10085c438d1402bb20107dfaaf2840796ebf292fc6", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36441, + "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 loose similarity bucket of largely unrelated Hugging Face issues spanning Trainer resume logic, tokenizer/model loading, multimodal processors, quantization/offload, and serving/runtime failures. None of the proposed soft pairs look like the same underlying bug, so all soft edges are rejected.", + "confidence": 0.69, + "canonical_issue_reason": "issue:43873 is the most central anchor in this loose bucket: it has the most discussion/inbound reference signal and a broad, concrete failure mode that makes it the best representative issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the strongest single issue to represent the cluster because it is well-discussed, externally referenced, and specific enough to summarize the general offloading/quantization theme without overcommitting to a false duplicate merge.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Both mention memory problems, but one is int4 quantization/OOM and the other is an apex FusedRMSNorm leak in T5; different root causes." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "One is perplexity quality on gpt-oss-20b, the other is tokenizer behavior changing across versions; not the same bug." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 auto device-map CUDA assert vs a 3d_parallel.py runtime failure; unrelated models/code paths." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor output-shape bug vs tokenization changes; different components and symptoms." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both concern saving weights, but one is lm_head serialization by device and the other is incorrect visual encoder keys in Qwen3.5." + }, + { + "left": "issue:46897", + "right": "issue:47317", + "accept": false, + "reason": "Both are training-loss issues, but for different models and likely different loss code paths; too generic to merge." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantization/offload bug vs Strix Halo mmap OOM; related theme but not the same underlying failure." + }, + { + "left": "issue:26413", + "right": "issue:28530", + "accept": false, + "reason": "Checkpoint resume sample-iteration failure vs early-stopping metric lookup; separate Trainer behaviors." + }, + { + "left": "issue:45685", + "right": "issue:46821", + "accept": false, + "reason": "MPS histogram/int error in MoE vs TimesFM window_size attribute access; unrelated subsystems." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer padding/truncation crash vs bfloat16 dtype mismatch in models; different failure modes." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Processor.tokenizer regression vs tokenization changes; one is API regression, the other text output drift." + }, + { + "left": "issue:43065", + "right": "issue:46402", + "accept": false, + "reason": "Sam3 pixel decoder dummy Conv2d vs text-encoder weights not loading; distinct Sam3 bugs." + }, + { + "left": "issue:44451", + "right": "issue:47020", + "accept": false, + "reason": "Model-loading failure for ScandiBERT vs Unigram/vocab type mismatch in XLMRobertaTokenizer; separate tokenizer/model issues." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "GenerationMixin import error vs 3d_parallel.py execution failure; unrelated." + }, + { + "left": "issue:46612", + "right": "issue:46858", + "accept": false, + "reason": "Beam-search cache reorder skipped in several models vs static-cache seq-length device tensor arithmetic; different generation bugs." + }, + { + "left": "issue:43646", + "right": "issue:46665", + "accept": false, + "reason": "Custom model initialization breakage vs inconsistent from_pretrained inside Python classes; overlapping area but not the same concrete bug." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "torch.compile+SDPA failure in Bamba vs BitNet packed-weight loading bug; different model-specific paths." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Blip2 dtype not propagating vs max_steps miscalculated on checkpoint resume; unrelated loading vs Trainer logic." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Bamba compile/SDPA issue vs Bark CPU-offload device mismatch; different models and triggers." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "LayoutLMv2 NER/tokenizer crash vs OmDet-Turbo image-size mismatch; different modalities and root causes." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "lm_head weight tying bug in Mistral3 vs incorrect visual encoder keys in Qwen3.5 save_pretrained; separate serialization problems." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "glm-4v-9b loading failure vs ScandiBERT loading failure; same symptom class, but different models and likely causes." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "FineGrainedFP8Config failure vs dequantized model offload load failure; distinct quantization/offload paths." + }, + { + "left": "issue:46821", + "right": "issue:46869", + "accept": false, + "reason": "TimesFM window_size AttributeError vs YOLO multi-GPU device_map loading; unrelated." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-resolution issues, but OmDet-Turbo has a processor/model size mismatch while Perceiver has interpolation behavior failing; not identical." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Qwen3 auto device-map assert vs GPT-OSS tensor parallelism issue; different parallelization bugs." + }, + { + "left": "issue:28530", + "right": "issue:33666", + "accept": false, + "reason": "Early-stopping metric handling vs Qwen2-VL multi-GPU training; unrelated." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3n forward variable collision vs Gemma4 PLE device mismatch; different model families and code paths." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Continuous-batching multimodal crash with str.to vs Gemma4Processor _tokenizer missing; similar serving context but distinct attribute errors." + }, + { + "left": "issue:44451", + "right": "issue:47020", + "accept": false, + "reason": "Model loading failure for a specific BERT vs tokenizer constructor passing dict to Unigram; different failure stages." + } + ] + }, + "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": 28, + "review_unit_index": 16, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28530", + "issue:29159", + "issue:30064", + "issue:33357", + "issue:34634", + "issue:35707", + "issue:36032", + "issue:36331", + "issue:41720", + "issue:42915", + "issue:43316", + "issue:43322", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43572", + "issue:43716", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:44186", + "issue:44361", + "issue:44464", + "issue:44492", + "issue:44509", + "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:46291", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:46894", + "issue:47020", + "issue:47145", + "issue:47317" + ], + "soft_pairs": [ + "issue:43531|issue:43572", + "issue:42915|issue:46291", + "issue:36331|issue:41720", + "issue:44734|issue:46858", + "issue:45479|issue:47317", + "issue:44464|issue:46858", + "issue:44671|issue:44743", + "issue:44186|issue:44361", + "issue:43716|issue:46821", + "issue:45632|issue:45684", + "issue:28530|issue:30064", + "issue:46894|issue:47317", + "issue:43316|issue:44913", + "issue:45701|issue:47145", + "issue:29159|issue:36032", + "issue:30064|issue:34634", + "issue:42915|issue:43322", + "issue:43452|issue:47020", + "issue:43881|issue:44991", + "issue:33357|issue:35707", + "issue:43873|issue:45704", + "issue:43531|issue:45759", + "issue:43540|issue:43782", + "issue:43526|issue:43883", + "issue:44492|issue:44509", + "issue:44534|issue:46665", + "issue:41720|issue:43746" + ], + "prepared_review_unit_hash": "2792a65cd59705c937407b9f507ebe7af449e4ef7f30b5504e28eef411fa01ff", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35726, + "estimated_input_tokens": 8932, + "estimated_eval_tokens": 18120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated bug reports across different components; none of the soft pairs look like true duplicates. I picked the most active, broadly scoped issue as the representative canonical issue.", + "confidence": 0.86, + "canonical_issue_reason": "Issue 43873 is the most centrally connected and broadly impactful report here, with substantial discussion and inbound references; it is the best single representative of the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is a strong standalone issue: active, well-scoped, and likely to matter to multiple users. It is the best issue to anchor follow-up triage.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Different bugs: Qwen3-MoE sliding-window handling vs StableLmConfig missing pad_token_idx after v5 update." + }, + { + "left": "issue:42915", + "right": "issue:46291", + "accept": false, + "reason": "Different failure modes: Qwen3Moe FP8 loading vs pipelines import breakage from kernels version/revision requirements." + }, + { + "left": "issue:36331", + "right": "issue:41720", + "accept": false, + "reason": "Unrelated code paths: trainer compute_loss signature change vs Qwen3 auto device-mapping CUDA assert." + }, + { + "left": "issue:44734", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache logic, but the concrete failures differ: serve KV continuation indexing vs static-cache slice arithmetic on non-CUDA backends." + }, + { + "left": "issue:45479", + "right": "issue:47317", + "accept": false, + "reason": "Both are training-loss complaints, but they concern different models and no shared concrete bug pattern is established." + }, + { + "left": "issue:44464", + "right": "issue:46858", + "accept": false, + "reason": "Different problems in generation: compiled-forward inconsistency vs cache length tensor type breaking arithmetic." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "Different model-specific behaviors: CamemBERT masked LM predictions vs Qwen3 recurrent state reset with cache." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Both are tokenizer/task failures, but they affect different tokenizers and likely different root causes." + }, + { + "left": "issue:43716", + "right": "issue:46821", + "accept": false, + "reason": "Different domains: multimodal dtype mismatch vs TimesFM window_size attribute error." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Filesystem/cache-path collision is unrelated to read-only permission propagation in save_pretrained." + }, + { + "left": "issue:28530", + "right": "issue:30064", + "accept": false, + "reason": "Unrelated: early stopping metric resolution vs image processor failure on void segmentation maps." + }, + { + "left": "issue:46894", + "right": "issue:47317", + "accept": false, + "reason": "Both mention training loss, but there is no evidence they share the same underlying bug or model path." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Different configuration issues: Gemma3TextConfig API discrepancy vs GPTNeoX rotary_pct not persisting on reload." + }, + { + "left": "issue:45701", + "right": "issue:47145", + "accept": false, + "reason": "Version-dependent tokenization changes are not the same as FP8 weight_scale being skipped on load." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Tokenizer inconsistency vs T5Tokenizer add_special_tokens name conflict are different concrete bugs." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Different processors and behaviors: segmentation map processing vs BarkProcessor voice_preset handling." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "FP8 quantization loading failure and Llava Next segmentation fault are distinct model-loading bugs." + }, + { + "left": "issue:43452", + "right": "issue:47020", + "accept": false, + "reason": "Different tokenizer/model loading issues affecting different APIs and root causes." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Distinct regressions: glm-4v-9b load failure vs tokenizer loading breakage for a specific model." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "MacOS bus error on CLIP loading is unrelated to progressive generation with inputs_embeds/past_key_values." + }, + { + "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": "Qwen3-MoE sliding-window behavior is different from from_config not unwrapping text_config in composite multimodal configs." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Both are Qwen multimodal issues, but one is video input processing and the other is from_pretrained weight_only loading." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels bug is unrelated to Molmo tied-weights attribute failure." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A typo in cache strategy docs is not the same as stale docs still listing removed pipeline tasks." + }, + { + "left": "issue:44534", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffer corruption in v5 is unrelated to inconsistent from_pretrained behavior inside Python classes." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3 device-mapping CUDA assert and GraniteSpeech PEFT local-checkpoint loading are different code paths and 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": 28, + "review_unit_index": 17, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:34567", + "issue:34634", + "issue:35707", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:42886", + "issue:43257", + "issue:43329", + "issue:43381", + "issue:43388", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43606", + "issue:43644", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43844", + "issue:44062", + "issue:44077", + "issue:44451", + "issue:44589", + "issue:44610", + "issue:44625", + "issue:44843", + "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:46050", + "issue:46402", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:47254" + ], + "soft_pairs": [ + "issue:43550|issue:43653", + "issue:43257|issue:47254", + "issue:43606|issue:43720", + "issue:42490|issue:43329", + "issue:43531|issue:46821", + "issue:34567|issue:35707", + "issue:45406|issue:45464", + "issue:34634|issue:41720", + "issue:43531|issue:44589", + "issue:43606|issue:43653", + "issue:43388|issue:45059", + "issue:45759|issue:45859", + "issue:44451|issue:44991", + "issue:43540|issue:46858", + "issue:45357|issue:46402", + "issue:44077|issue:44625", + "issue:43644|issue:46665", + "issue:44987|issue:45042", + "issue:44062|issue:44451", + "issue:44898|issue:45072", + "issue:44625|issue:45479", + "issue:41720|issue:45704", + "issue:44918|issue:44936", + "issue:39692|issue:44610", + "issue:43381|issue:43844", + "issue:42886|issue:44843", + "issue:43381|issue:45704", + "issue:43746|issue:46050" + ], + "prepared_review_unit_hash": "cbabfffc2d24f75c24342af85ef6175f8f73d1fb8b6343b05e419535ce5d3821", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36312, + "estimated_input_tokens": 9078, + "estimated_eval_tokens": 18412 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is a grab-bag of unrelated Transformers bugs across tokenizers, multimodal processors, training/runtime, and model-specific loading paths. The soft pairs mostly share subsystem vocabulary but not the same concrete failure mode, so none should be merged as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44843 is the cleanest library-level regression in the set: it has a precise failure path (`AutoTokenizer.from_pretrained` -> `_patch_mistral_regex` -> unconditional `model_info()`), a clear offline-mode impact, and broader applicability than the more model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44843 is the best representative issue because it is well-scoped, reproducible, and touches a core loading path rather than a single model edge case.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "Different bugs: one is a torch.compile/SDPA failure for Bamba, the other is a tokenizer special-token registration bug for BigBird." + }, + { + "left": "issue:43257", + "right": "issue:47254", + "accept": false, + "reason": "Both involve DeepSpeed/accelerate, but one is MoE weight conversion on load and the other is a checkpointing/PEFT ZeRO-3 error; different code paths and fixes." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU offload device mismatch in Bark is unrelated to BitNet packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Path-normalization import warning is unrelated to a multimodal video token counting bug." + }, + { + "left": "issue:43531", + "right": "issue:46821", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and TimesFM `window_size` AttributeError are separate model-specific API failures." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token counting and progressive generation with embeds/past KV are different training/inference bugs." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing `_tokenizer` in serve is not the same as Qwen3.5 streaming chat/completions failure." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice preset handling and Qwen3 auto device mapping cuda assert are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Sliding-window logic for Qwen3-MoE is not the same as the Float8 storage type lookup error." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Device offload mismatch and BigBird mask-token decode failure affect different subsystems and require different fixes." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Metric gathering tuple truncation and SAM3 PCS text/bbox behavior are unrelated data-processing bugs." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Config unwrapping for composite Qwen 3.5/3.6 multimodal models is distinct from CPU-offload incompatibility in rotary embeddings." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer loading problems, but they target different failure modes and different model repos; not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:46858", + "accept": false, + "reason": "Video-input processing in Qwen3OmniMoe is unrelated to static-cache slice arithmetic on non-CUDA backends." + }, + { + "left": "issue:45357", + "right": "issue:46402", + "accept": false, + "reason": "Qwen3.5 save_pretrained key regression and Sam3Model text-encoder loading are separate serialization/loading issues." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "The patchtsmixer post_init policy change is unrelated to Qwen3.5 num_labels propagation." + }, + { + "left": "issue:43644", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffer junk initialization and inconsistent from_pretrained behavior inside Python classes are different bugs." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "A model-loading regression for `physical-intelligence/fast` is not the same as PIL image processors incorrectly requiring torchvision." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken keyword collision and loading `vesteinn/ScandiBERT` are not the same underlying issue." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver resize failure and dtype mismatches in bfloat16 inference are distinct model/runtime problems." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Missing `num_labels` propagation in Qwen3.5 config is not the same as the broader single-label zero-loss classification bug." + }, + { + "left": "issue:41720", + "right": "issue:45704", + "accept": false, + "reason": "Qwen3 auto device-map cuda assert and T5/apex FusedRMSNorm memory leak are unrelated runtime issues." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "TRL SFT unpacking failure and `trainer.evaluate()` after `train()` are different Trainer integration bugs." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "SigLIP2 doc example errors and OmDet-Turbo processor/model resolution mismatch are separate vision-model issues." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Gradient checkpointing in eval mode and abnormal gradient growth under ZeRO-3 are not the same failure." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "HF_HUB_OFFLINE cache loading and `_patch_mistral_regex` unconditionally calling `model_info()` are related only at a broad offline/loading level, but the concrete code paths differ." + }, + { + "left": "issue:43381", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing eval-mode failure and T5/apex memory leak are unrelated." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "PEFT adapters failing from local checkpoints and sinq quantization breaking after save/load are different serialization 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": 28, + "review_unit_index": 18, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:43295", + "issue:43421", + "issue:43493", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43643", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43819", + "issue:43873", + "issue:43881", + "issue:43927", + "issue:43994", + "issue:44062", + "issue:44186", + "issue:44220", + "issue:44368", + "issue:44464", + "issue:44488", + "issue:44514", + "issue:44545", + "issue:44625", + "issue:44671", + "issue:44857", + "issue:44995", + "issue:45081", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45834", + "issue:45907", + "issue:45923", + "issue:46032", + "issue:46294", + "issue:46463", + "issue:46655", + "issue:46693", + "issue:46858", + "issue:47020", + "issue:47050" + ], + "soft_pairs": [ + "issue:44368|issue:45357", + "issue:44514|issue:44545", + "issue:43421|issue:43927", + "issue:46655|issue:46693", + "issue:43531|issue:45923", + "issue:43746|issue:43873", + "issue:30064|issue:36010", + "issue:39692|issue:43493", + "issue:43421|issue:43653", + "issue:43994|issue:44671", + "issue:44464|issue:47050", + "issue:44995|issue:46858", + "issue:46032|issue:46463", + "issue:45081|issue:45356", + "issue:44857|issue:47050", + "issue:43782|issue:45381", + "issue:43421|issue:44186", + "issue:45356|issue:45907", + "issue:39692|issue:43873", + "issue:39692|issue:47050", + "issue:43526|issue:43881", + "issue:43720|issue:43819", + "issue:43596|issue:44220", + "issue:43295|issue:45834", + "issue:43550|issue:43927", + "issue:43540|issue:46294", + "issue:43550|issue:43819", + "issue:44062|issue:44488", + "issue:43653|issue:43720", + "issue:44625|issue:47050", + "issue:43531|issue:43643", + "issue:45381|issue:46693", + "issue:44488|issue:47020" + ], + "prepared_review_unit_hash": "a0c26a9b92b7dc444ac7a947c216d81b650263717850c0c1133979ece486cdf7", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37979, + "estimated_input_tokens": 9495, + "estimated_eval_tokens": 19246 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose collection of unrelated Transformers issues. The only clear duplicate pair is the two Qwen2.5-VLProcessor batched apply_chat_template crashes with padding=False.", + "confidence": 0.96, + "canonical_issue_reason": "issue:44545 is the clearest representative of the duplicate Qwen2.5-VL batched apply_chat_template crash: it names the exact method, the batched-input condition, and the padding=False trigger.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44545 is the best global issue to keep because it is the most specific and reproducible statement of the only true duplicate pair in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "Different bugs: one is a Qwen3.5 LoRA warning about tie_word_embeddings, the other is incorrect visual encoder key serialization." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same underlying bug: Qwen2.5_VLProcessor.apply_chat_template crashes on batched input when padding=False." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Both relate to tokens/config, but one is runtime post-processor updates and the other is save/load loss of custom token IDs; not the same concrete failure." + }, + { + "left": "issue:46655", + "right": "issue:46693", + "accept": false, + "reason": "Video padding calculation error vs long-context generation performance regression are different code paths and symptoms." + }, + { + "left": "issue:43531", + "right": "issue:45923", + "accept": false, + "reason": "Sliding-window handling in Qwen3-MoE is unrelated to a missing supports_gradient_checkpointing flag on a trust_remote_code model." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "PEFT local checkpoint loading and quantization offloading are separate loading/dispatch problems." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Void segmentation map processing and a GenerationMixin import error are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Documentation example errors are not the same issue as a SigLIP2 implementation discrepancy against JAX." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "One is about updating post-processors after special-token mutation; the other is a BigBird tokenizer special-token registration bug." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different model families and failures: SigLIP2 nonsensical outputs vs CamemBERT masked LM regression." + }, + { + "left": "issue:44464", + "right": "issue:47050", + "accept": false, + "reason": "Compiled chunked generation inconsistency and ONNX fp16 dtype mismatch are different problems." + }, + { + "left": "issue:44995", + "right": "issue:46858", + "accept": false, + "reason": "Stale indexer cache on second forward is not the same as device-tensor seq_length breaking static-cache arithmetic under compile." + }, + { + "left": "issue:46032", + "right": "issue:46463", + "accept": false, + "reason": "Incorrect use_cache behavior and an incorrect T5 decoder input validation error message are unrelated." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both mention tokenizer regex/fix_mistral_regex, but one crashes on backend_tokenizer access while the other is a codec-handling regression with a misleading warning." + }, + { + "left": "issue:44857", + "right": "issue:47050", + "accept": false, + "reason": "Different model/export paths: LwDetrImageLoss AMP crash vs CanineModel ONNX attention dtype mismatch." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Qwen3VL weight_only loading failure is unrelated to Qwen2.5-VL batched chat-template padding behavior." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "Deepspeed ZeRO-3/BertModel indexing and _torch_extract_fbank_features are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Processor.tokenizer regression with images is not the same as a Kosmos2.5 long-OCR index error." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "torch.compile + SDPA failure in Bamba is unrelated to DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:43540", + "right": "issue:46294", + "accept": false, + "reason": "These are different Qwen video-processing failures: one is a Qwen3OmniMoe ValueError, the other is a Qwen3.5-VL video feature/token mismatch." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "AddedToken special-argument duplication and inability to load cjvt/sleng-bert are different tokenizer bugs." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "BigBird special-token registration and BitNet packed-weight loading are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:47050", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and CanineModel ONNX dtype mismatch are separate issues." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Sliding-window behavior and missing fields in AutoConfig with trust_remote_code are unrelated." + }, + { + "left": "issue:45381", + "right": "issue:46693", + "accept": false, + "reason": "Wrong Qwen2.5-VL vision_position_ids for video input is not the same as an FA2 long-context performance regression." + }, + { + "left": "issue:44488", + "right": "issue:47020", + "accept": false, + "reason": "A model loading failure and an XLMRobertaTokenizer/Unigram constructor bug are different problems." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative. The one accepted duplicate pair, issue:44514 and issue:44545, is strongly supported by the titles and looks mergeable as a single bugfix. The other soft-edge verdicts 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": 28, + "review_unit_index": 19, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:30064", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43322", + "issue:43335", + "issue:43540", + "issue:43606", + "issue:43611", + "issue:43618", + "issue:43643", + "issue:43653", + "issue:43697", + "issue:43756", + "issue:43819", + "issue:44008", + "issue:44186", + "issue:44220", + "issue:44442", + "issue:44448", + "issue:44561", + "issue:44661", + "issue:44671", + "issue:44805", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44936", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45584", + "issue:45685", + "issue:45834", + "issue:45987", + "issue:46291", + "issue:46665", + "issue:46829", + "issue:46832", + "issue:46858", + "issue:46932", + "issue:47020" + ], + "soft_pairs": [ + "issue:46829|issue:46832", + "issue:44220|issue:44661", + "issue:41720|issue:44991", + "issue:30064|issue:36331", + "issue:43611|issue:46665", + "issue:44821|issue:45084", + "issue:45987|issue:46829", + "issue:44186|issue:44898", + "issue:43653|issue:43819", + "issue:44843|issue:45081", + "issue:43335|issue:43756", + "issue:43540|issue:44008", + "issue:44008|issue:46858", + "issue:34634|issue:36010", + "issue:43643|issue:44877", + "issue:44991|issue:47020", + "issue:45987|issue:46932", + "issue:39692|issue:45834", + "issue:43618|issue:43697", + "issue:44561|issue:45020", + "issue:44186|issue:44442", + "issue:44448|issue:44671", + "issue:45584|issue:45834", + "issue:44805|issue:44936", + "issue:44991|issue:46291", + "issue:43322|issue:45685", + "issue:43606|issue:45072" + ], + "prepared_review_unit_hash": "30c5fe24bec1c1c6c644f3a10257c7514858b39cd30233693ba96f529aa5e20d", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35401, + "estimated_input_tokens": 8851, + "estimated_eval_tokens": 17958 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: most pairs are clearly unrelated and should not be merged. The only plausible umbrella-like issue is the broad remote_code regression report, but none of the soft pairs look like true duplicates.", + "confidence": 0.41, + "canonical_issue_reason": "issue:45020 is the broadest umbrella in the set: it describes a general regression in recent transformers versions breaking `remote_code` models, which is the closest thing to a cluster-level theme. The other issues are narrower, model-specific, or entirely unrelated.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "issue:45020 best represents the cluster because it is the most general and centrally worded regression report among otherwise unrelated issues.", + "best_pr_reason": "No pull requests are present in this cluster.", + "soft_edge_verdicts": [ + { + "left": "issue:46829", + "right": "issue:46832", + "accept": false, + "reason": "Different failures in different model families: one is an output-field bug, the other is a forward-pass shape/type error." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Unrelated subsystems: audio feature extraction vs tokenizer mapping/name-generation tooling." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Different models and symptoms: CUDA assert during device mapping vs tokenizer loading failure." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Completely different code paths: image segmentation map processing vs Trainer loss signature incompatibility." + }, + { + "left": "issue:43611", + "right": "issue:46665", + "accept": false, + "reason": "Both involve loading behavior, but the concrete bugs differ: `base_model_prefix` handling vs inconsistent `from_pretrained` use inside Python classes." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "No shared underlying bug: image processor loading from URL vs template compilation failure." + }, + { + "left": "issue:45987", + "right": "issue:46829", + "accept": false, + "reason": "Static cache length contract bug is unrelated to VideoPrism output packaging." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Different tokenizer/model families and different failure modes." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Tokenizer special-token registration bug vs DAC latent reconstruction mismatch; not the same code-path." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both touch `_patch_mistral_regex`, but one is an offline-network call issue and the other is an attribute error on tokenizer internals." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Different model config bugs: sparse-layer creation vs RoPE-layer dropping." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different multimodal models and unrelated forward-pass bugs." + }, + { + "left": "issue:44008", + "right": "issue:46858", + "accept": false, + "reason": "Variable-name collision in Gemma3n is unrelated to static cache sequence-length behavior." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice preset handling is unrelated to the GenerationMixin import error." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both are config-loading issues, but one is `trust_remote_code` field loss and the other is strict config validation for a specific model." + }, + { + "left": "issue:44991", + "right": "issue:47020", + "accept": false, + "reason": "Both are tokenizer-load failures, but they affect different tokenizers and stem from different initialization bugs." + }, + { + "left": "issue:45987", + "right": "issue:46932", + "accept": false, + "reason": "Static cache return-type bug is unrelated to Gemma4 quantized embedder dtype casting." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "Documentation/example mismatch and quantization failure are unrelated to long OCR-input indexing." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Different regressions in different model outputs; no shared concrete code-path." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Both mention `remote_code` breakage, but one is specifically about removing `is_torch_fx_available` while the other is a broader regression report; not the same concrete bug." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer bugs affecting different models and input shapes." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are version-v4/v5 output discrepancies, but the model-specific failure modes are unrelated." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription failure is unrelated to Kosmos2.5 long-input indexing." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Mask-shape IndexError and post-train evaluate failure are different trainer/runtime problems." + }, + { + "left": "issue:44991", + "right": "issue:46291", + "accept": false, + "reason": "Tokenizer loading failure is unrelated to pipeline import breakage from kernels version/revision requirements." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Llava Next loading crash and MPS histogram dtype limitation are unrelated backends/problems." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "CPU offload device mismatch and bfloat16 dtype mismatch are different bugs in different models/code 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": 28, + "review_unit_index": 20, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:34634", + "issue:36331", + "issue:42175", + "issue:43388", + "issue:43421", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43644", + "issue:43645", + "issue:43716", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43782", + "issue:43818", + "issue:43825", + "issue:43827", + "issue:43931", + "issue:43994", + "issue:44008", + "issue:44291", + "issue:44479", + "issue:44610", + "issue:44671", + "issue:44811", + "issue:44977", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45183", + "issue:45381", + "issue:45479", + "issue:45563", + "issue:45701", + "issue:45834", + "issue:46032", + "issue:46294", + "issue:46463", + "issue:46512", + "issue:46665", + "issue:46697", + "issue:46728", + "issue:46846", + "issue:47276" + ], + "soft_pairs": [ + "issue:43421|issue:44811", + "issue:44479|issue:45834", + "issue:43577|issue:43818", + "issue:43550|issue:45072", + "issue:45183|issue:45479", + "issue:43644|issue:43749", + "issue:46294|issue:46463", + "issue:43782|issue:44977", + "issue:43716|issue:44008", + "issue:43582|issue:46512", + "issue:45563|issue:47276", + "issue:43825|issue:43994", + "issue:43931|issue:46846", + "issue:28218|issue:29159", + "issue:43388|issue:46032", + "issue:43540|issue:45381", + "issue:44610|issue:45072", + "issue:43645|issue:46665", + "issue:43749|issue:44291", + "issue:34634|issue:36331", + "issue:43720|issue:45072", + "issue:42175|issue:45081", + "issue:44671|issue:45701", + "issue:43577|issue:43582", + "issue:46697|issue:46728", + "issue:43825|issue:43827", + "issue:43742|issue:45020" + ], + "prepared_review_unit_hash": "61a07c097fd6e3ab966d124b74221d8fa8b68513b53b56580b95b6b509ec462f", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36239, + "estimated_input_tokens": 9060, + "estimated_eval_tokens": 18376 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are clearly different bugs or different layers of the stack. The only credible duplicate-style match is the tokenizer pair 28218/29159, where the broader slow-vs-fast inconsistency report plausibly covers the added-token spacing symptom.", + "confidence": 0.67, + "canonical_issue_reason": "29159 is the broader tokenizer report and can reasonably subsume the more specific added-token spacing bug in 28218.", + "canonical_pr_reason": null, + "best_issue_reason": "29159 is the most representative issue here: it is broad, tokenizer-focused, and the only pair that looks like a plausible duplicate relationship.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Different tokenizer/processor behaviors: post-processor updating special tokens vs batch_decode skip_special_tokens handling." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Different models and different failures: video regression in Qwen VL vs OCR index error in Kosmos2.5." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Unrelated model-loading dtype issue vs Video-LLaVA architecture/weight-sharing bug." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Different code paths and symptoms: torch.compile + SDPA failure vs bfloat16 dtype mismatch in wrappers." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Input validation error messaging is not the same as the single-label-classification zero-loss bug." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Different loading regressions: non-persistent buffer initialization vs FSDP CPU RAM efficient loading." + }, + { + "left": "issue:46294", + "right": "issue:46463", + "accept": false, + "reason": "Different model families and different problems: Qwen video chat-template mismatch vs T5 decoder input validation message." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Different failures: weight_only loading error vs flash-attention generation regression." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Different bugs in different models: image preprocessor dtype mismatch vs variable-name collision in Gemma3n forward." + }, + { + "left": "issue:43582", + "right": "issue:46512", + "accept": false, + "reason": "Apple Silicon allocator TypeError is unrelated to stale config.dtype after from_config." + }, + { + "left": "issue:45563", + "right": "issue:47276", + "accept": false, + "reason": "Both are warning-message issues, but they concern different warnings in different generation paths." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "One is a runtime error-message bug, the other is docs lagging behind v5 removals; not the same code-path problem." + }, + { + "left": "issue:43931", + "right": "issue:46846", + "accept": false, + "reason": "Different Qwen3-VL loading/shape problems; TP-plan omissions are not the same as a model weight-shape mismatch report." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": true, + "reason": "The broader slow-vs-fast tokenizer inconsistency report plausibly includes the added-token extra-space symptom." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Different logic errors: last-batch label dropping in gather_for_metrics vs incorrect use_cache results in Mamba2Mixer." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Both are video-related, but they affect different models and different bugs (processing error vs vision_position_ids bug)." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Image-size preprocessing mismatch is unrelated to bfloat16 dtype mismatch." + }, + { + "left": "issue:43645", + "right": "issue:46665", + "accept": false, + "reason": "Custom model initialization in notebooks is a different issue from from_pretrained behavior inside Python classes." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "Different loading mechanisms and different failure modes: FSDP RAM-efficient loading vs init_empty_weights unexpected argument." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice_preset is unrelated to CustomTrainer.compute_loss signature breakage." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Packed-weight unpacking during accelerate loading is a different bug from dtype mismatch in inference wrappers." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Backend installation dependency issue is unrelated to the Mistral regex patch AttributeError." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "The CamemBERT MLM regression and the generic tokenization-change report are too broad and not clearly the same underlying bug." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Different dtype/loading problems in different subsystems; no shared concrete code path." + }, + { + "left": "issue:46697", + "right": "issue:46728", + "accept": false, + "reason": "Both are image-processing edge cases, but one is over-padding and the other is incorrect patch counting." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "Key error loading one model is not the same as the broad remote_code breakage regression." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted duplicate-style pairing, 28218/29159, is reasonably supported by the issue titles and the broader/narrower relationship. The other soft-edge rejections also read as cautious and consistent with the listed topics." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 21, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36010", + "issue:36331", + "issue:42175", + "issue:42490", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43582", + "issue:43638", + "issue:43645", + "issue:43653", + "issue:43697", + "issue:43746", + "issue:43782", + "issue:43819", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:44060", + "issue:44368", + "issue:44589", + "issue:44811", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45003", + "issue:45042", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45372", + "issue:45406", + "issue:45538", + "issue:45859", + "issue:46175", + "issue:46209", + "issue:46285", + "issue:46291", + "issue:46326", + "issue:46932", + "issue:47050", + "issue:47217" + ], + "soft_pairs": [ + "issue:43782|issue:45325", + "issue:44589|issue:45042", + "issue:43262|issue:43819", + "issue:46175|issue:46932", + "issue:43653|issue:44912", + "issue:43582|issue:44857", + "issue:43262|issue:47217", + "issue:45003|issue:46285", + "issue:45372|issue:46326", + "issue:36010|issue:43645", + "issue:36010|issue:36331", + "issue:43550|issue:44912", + "issue:43531|issue:45042", + "issue:43638|issue:44936", + "issue:44060|issue:44368", + "issue:42175|issue:46291", + "issue:43257|issue:43329", + "issue:43526|issue:43697", + "issue:43452|issue:44991", + "issue:45081|issue:45406", + "issue:43819|issue:43927", + "issue:43906|issue:46209", + "issue:45084|issue:45538", + "issue:43854|issue:44977", + "issue:43746|issue:44811", + "issue:42490|issue:43257", + "issue:45859|issue:47050" + ], + "prepared_review_unit_hash": "562d2cad220aa480eba5a3651a8d3f8aadcbef13469180dfb1f51864b892d4da", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36307, + "estimated_input_tokens": 9077, + "estimated_eval_tokens": 18410 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is largely heterogeneous: the soft-similarity pairs share superficial keywords or the same broad subsystem, but they do not describe the same concrete bug/change. I would not merge any of the proposed soft edges.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue stands out; the items are mostly unrelated bug reports with only a few superficial thematic overlaps.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44060 is the closest to a representative item only because it has the strongest topical overlap with another issue in the list, but overall the cluster is too mixed to have a true canonical issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen/VL-adjacent, but one is a weight_only loading error and the other is a rope-indexing bug; different code paths." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "FP8 storage lookup failure vs PIL image processor requiring torchvision; unrelated failures." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio chat-template sampling-rate default vs DAC latent/STE mismatch; different components and symptoms." + }, + { + "left": "issue:46175", + "right": "issue:46932", + "accept": false, + "reason": "Both are Gemma4-related, but one is a huge one-hot position embedding allocation and the other is dtype casting for quantized embedders." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration bug vs MXFP4 quantization fallback; no shared bug path." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon warmup TypeError and CUDA float16 loss crash are unrelated runtime issues." + }, + { + "left": "issue:43262", + "right": "issue:47217", + "accept": false, + "reason": "Audio processor sampling-rate default is not the same as VL image-placeholder handling in user text." + }, + { + "left": "issue:45003", + "right": "issue:46285", + "accept": false, + "reason": "Unsafe sys.modules access is unrelated to CLIP text-model structural changes." + }, + { + "left": "issue:45372", + "right": "issue:46326", + "accept": false, + "reason": "Both mention mistral_common, but one is an import error for ReasoningEffort and the other is continue_final_message behavior; different problems." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "GenerationMixin import failure and Jupyter custom-model initialization breakage are distinct issues." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "Importing GenerationMixin and CustomTrainer.compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is not the same as MXFP4 quantization fallback in git-oss-20b." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding_window bug and PIL/torchvision dependency regression do not share a concrete code path." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "DeepSpeed ZeRO-3 with non-pretrained Bert IndexError is not the same as evaluate() failing after train()." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings in Qwen models, but one is a specific wrong tie warning in Qwen3-Next and the other is a separate fine-tuning warning/report; not enough evidence of one bug." + }, + { + "left": "issue:42175", + "right": "issue:46291", + "accept": false, + "reason": "Package/backend dependency issues, but one is missing TensorFlow in a torch extra and the other is kernels version/revision requirements." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3-MoE weight conversion/loading and a multimodal token-count video-branch undefined-variable bug are different failures." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and RTDetrV2 output differences are unrelated model issues." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "gguf_file handling for tokenizer/model loading is unrelated to the EMBEDDIA tokenizer load regression." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Mistral tokenizer regex patch crash and Gemma4Processor missing _tokenizer are separate attribute/path bugs." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC latent conversion mismatch and DiaConfig losing custom token IDs are different model/config bugs." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "An isolated reproduction of another issue is not the same as an FP8 inference failure on B200." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Template-node compilation error and CLIPTokenizer max_length regression are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "GLM-4.7-Flash loading test failure and Qwen3.5 flash-attention generation failure are model-specific but not the same bug." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "GraniteSpeech PEFT checkpoint loading and Whisper batch_decode skip_special_tokens behavior are different code paths." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Local-path slash import warning and Qwen3 MOE weight conversion are unrelated." + }, + { + "left": "issue:45859", + "right": "issue:47050", + "accept": false, + "reason": "Qwen3_5MoE rotary embedding CPU-offload incompatibility and Canine fp16 ONNX dtype mismatch are distinct 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": 28, + "review_unit_index": 22, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28282", + "issue:29159", + "issue:29870", + "issue:34567", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42915", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43425", + "issue:43450", + "issue:43525", + "issue:43526", + "issue:43643", + "issue:43650", + "issue:43696", + "issue:43742", + "issue:43749", + "issue:43906", + "issue:43931", + "issue:44291", + "issue:44387", + "issue:44479", + "issue:44554", + "issue:44610", + "issue:44617", + "issue:44756", + "issue:44811", + "issue:44821", + "issue:44912", + "issue:44987", + "issue:45042", + "issue:45081", + "issue:45200", + "issue:45459", + "issue:45538", + "issue:45600", + "issue:45759", + "issue:45803", + "issue:45859", + "issue:46291", + "issue:46421", + "issue:46424", + "issue:47020", + "issue:47145", + "issue:47254" + ], + "soft_pairs": [ + "issue:43643|issue:45759", + "issue:43257|issue:43696", + "issue:43526|issue:44554", + "issue:43329|issue:45600", + "issue:38175|issue:44610", + "issue:43262|issue:44811", + "issue:45200|issue:47254", + "issue:42915|issue:44821", + "issue:39692|issue:44912", + "issue:42915|issue:43742", + "issue:43329|issue:44479", + "issue:44291|issue:45042", + "issue:43329|issue:43450", + "issue:43257|issue:43931", + "issue:34567|issue:40990", + "issue:44387|issue:44617", + "issue:29159|issue:34567", + "issue:44912|issue:47145", + "issue:45459|issue:45803", + "issue:45081|issue:47020", + "issue:44987|issue:45538", + "issue:43525|issue:46291", + "issue:43425|issue:46291", + "issue:46421|issue:46424", + "issue:45859|issue:47254", + "issue:28282|issue:29870", + "issue:43749|issue:44756", + "issue:43650|issue:43906" + ], + "prepared_review_unit_hash": "1451aee02cd631e6da896464ec7e4c29e1bd610e5a6b8b67807f3e5ab7c8ecf0", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36242, + "estimated_input_tokens": 9061, + "estimated_eval_tokens": 18378 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "None of the proposed soft pairs look like true duplicates; they mostly share broad subsystems or vague symptoms (video, quantization, loading, OOM) but describe different concrete failures. I\u2019d keep the issues separate rather than merging them into duplicate clusters.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43329 is the clearest, most concrete bug report in the set: it names an exact failing helper, the bad video code path, and the undefined variables involved. It reads like a precise canonical example for this collection.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43329 is the strongest standalone issue candidate because it is specific, actionable, and tied to a clearly identified code path. It is more precise than the more generic loading/OOM reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both concern config/model-loading behavior around composite Qwen-style configs, but one is about missing fields from AutoConfig with trust_remote_code and the other is about AutoModelForCausalLM not unwrapping text_config. Different APIs and failures." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is unrelated to a GPT-oss CUDA OOM. Different models, mechanisms, and symptoms." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "BeitImageProcessorFast label reduction is a processor bug; the MPS issue is an attention correctness problem. No shared code path." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Video token counting with undefined variables is not the same as stale auto_mappings references to removed Sam3LiteText configs. Different files and failures." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 and an OmDet-Turbo input-size mismatch are different model/processor issues." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sample-rate defaulting and Whisper batch_decode ignoring skip_special_tokens are unrelated tokenizer/processor behaviors." + }, + { + "left": "issue:45200", + "right": "issue:47254", + "accept": false, + "reason": "Gemma 4 mm_token_type_ids defaulting and a PEFT/DeepSpeed checkpoint error are separate training/loading problems." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "FineGrainedFP8Config failure on Qwen3Moe and loading AutoImageProcessor from URL are different paths with no concrete overlap." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "A broken SigLIP2 documentation example is not the same bug as MXFP4 loading falling back to bf16." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure and MobileLLM key errors are distinct loading bugs." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Both involve video, but one is an undefined-variable bug in multimodal token counting while the other is a broader v5.3.0 regression across several video models. Not enough to treat as the same concrete defect." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Unexpected _is_hf_initialized handling in init_empty_weights is unrelated to PIL backend processors requiring torchvision." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "A bad video token-count helper and batched video processor shape bugs are different issues in the same broad area, but not the same underlying bug." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3 MOE weight conversion and Qwen3-VL weight-shape mismatches are separate model-loading failures." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "TrainerState token counting not updating is unrelated to high perplexity on GPT-oss-20b." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both mention OOM, but one is int4 quantization memory regression in Transformers 5.x and the other is a Sam3Video model-specific OOM. Different root causes." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency is not the same as TrainerState token accounting not updating." + }, + { + "left": "issue:44912", + "right": "issue:47145", + "accept": false, + "reason": "Both concern quantized model loading, but one is MXFP4 fallback to bf16 and the other is silent corruption from missing weight_scale application on frozen FP8 checkpoints. Different formats and bugs." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "One is exception handling around protobuf decode errors; the other is a bare except swallowing KeyboardInterrupt in Fuyu tensor conversion. Different code and behavior." + }, + { + "left": "issue:45081", + "right": "issue:47020", + "accept": false, + "reason": "Mistral regex patch crashing on tokenizer internals and XLMRobertaTokenizer passing a dict to Unigram are unrelated tokenizer initialization bugs." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Loading physical-intelligence/fast on transformers>=5.1.0 and CLIPTokenizer model_max_length being absurdly large are separate issues." + }, + { + "left": "issue:43525", + "right": "issue:46291", + "accept": false, + "reason": "Llama4Config missing pad_token_id is unrelated to the kernels package version/revision requirement when importing pipelines." + }, + { + "left": "issue:43425", + "right": "issue:46291", + "accept": false, + "reason": "Torch 2.10 incompatibility is a dependency-version issue, but not the same as kernels import failure." + }, + { + "left": "issue:46421", + "right": "issue:46424", + "accept": false, + "reason": "Lazy cache initialization causing spurious recompilation and max_cache_length only applying to Quantized Cache are different cache-behavior bugs." + }, + { + "left": "issue:45859", + "right": "issue:47254", + "accept": false, + "reason": "CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding is unrelated to PEFT/DeepSpeed checkpoint errors." + }, + { + "left": "issue:28282", + "right": "issue:29870", + "accept": false, + "reason": "PyTorch-missing AutoModel ImportError and Bart generate decoder_attention_mask unbound variable are unrelated failures." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU RAM efficient loading broken and disabling mmap on Strix Halo to avoid OOM are different memory-related issues." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "'ADD THE DATA' is not a substantive bug report, and the reproduction issue is clearly separate." + } + ] + }, + "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": 28, + "review_unit_index": 23, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:24643", + "issue:27957", + "issue:28282", + "issue:29127", + "issue:29159", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:39379", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43404", + "issue:43452", + "issue:43454", + "issue:43531", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43824", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44554", + "issue:44560", + "issue:44568", + "issue:44821", + "issue:44857", + "issue:44936", + "issue:44991", + "issue:45127", + "issue:45137", + "issue:45356", + "issue:45357", + "issue:45405", + "issue:45488", + "issue:45538", + "issue:46512", + "issue:46693", + "issue:46701", + "issue:46821", + "issue:46832", + "issue:46869", + "issue:47254" + ], + "soft_pairs": [ + "issue:28282|issue:34689", + "issue:43540|issue:46701", + "issue:43454|issue:43577", + "issue:44568|issue:45356", + "issue:33357|issue:34567", + "issue:43404|issue:45127", + "issue:42175|issue:43824", + "issue:43742|issue:44291", + "issue:44936|issue:46821", + "issue:43697|issue:45357", + "issue:45137|issue:45405", + "issue:28282|issue:29127", + "issue:39379|issue:44554", + "issue:44479|issue:46693", + "issue:43582|issue:47254", + "issue:42915|issue:45538", + "issue:44291|issue:44821", + "issue:44857|issue:46832", + "issue:45538|issue:46869", + "issue:24643|issue:29159", + "issue:44062|issue:44991", + "issue:42915|issue:43749", + "issue:43452|issue:43582", + "issue:27957|issue:28282", + "issue:44568|issue:45488", + "issue:41720|issue:43540", + "issue:43540|issue:44560", + "issue:43577|issue:46512", + "issue:43531|issue:44821" + ], + "prepared_review_unit_hash": "045a9258f53eb5f174d6f49ca8d54291af3f24902dc8752e7ad9a04d59b98da5", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35835, + "estimated_input_tokens": 8959, + "estimated_eval_tokens": 18174 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: most soft pairs only share a broad subsystem (tokenizers, loading, video, training, distributed) and do not look like the same underlying bug. I reject all soft edges. If a single anchor is needed, issue:45488 is the best representative because it is the most active open regression report and the closest to the tokenizer-regression subset, but it does not truly cover the whole cluster.", + "confidence": 0.71, + "canonical_issue_reason": "No single issue cleanly represents this mixed cluster; issue:45488 is the least-bad anchor because it is open, active, has inbound references, and matches the tokenizer-regression theme better than the rest.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45488 is the strongest standalone issue here: recent activity, multiple references, and a broad regression description make it the best issue to keep visible, even though the overall set is not a true duplicate cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Both are model-loading failures, but one is a missing-PyTorch import error and the other is a specific Llama vision loading regression; different root causes." + }, + { + "left": "issue:43540", + "right": "issue:46701", + "accept": false, + "reason": "Both involve multimodal/video pipelines, but the model families and failure modes differ too much to treat as one bug." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "One is about tied lm_head weights and garbage generation; the other is about dtype propagation in BLIP2. Shared loading theme only." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they point to different behaviors and likely different code paths." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "Completely unrelated: MacOS bus error in CLIP loading vs. a TrainerState counter not updating." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied embeddings, but one is a missing tie in Mistral3 and the other is a LoRA-merge regression after vocab extension; not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Packaging/backend availability issue vs. a missing model class import error; unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "Different loading failures with different triggers: a MobileLLM key error versus init_empty_weights passing an unexpected argument." + }, + { + "left": "issue:44936", + "right": "issue:46821", + "accept": false, + "reason": "Trainer lifecycle bug vs. TimesFM window_size AttributeError; no shared underlying defect." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RT-DETR output drift and Qwen3.5 save_pretrained key handling are different regressions." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed deque error and a minimum-version bump issue are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "One is missing PyTorch; the other is a LayoutLMv3 error-message clarity issue. Not duplicates." + }, + { + "left": "issue:39379", + "right": "issue:44554", + "accept": false, + "reason": "Different model/runtime correctness problems: mllama masking logic versus an MPS attention issue." + }, + { + "left": "issue:44479", + "right": "issue:46693", + "accept": false, + "reason": "Both are v5 / generation regressions, but one is video-input handling across Qwen VL variants and the other is FA2 long-context performance." + }, + { + "left": "issue:43582", + "right": "issue:47254", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError and a PEFT+ZeRO-3 checkpoint error are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "Qwen3Moe FP8 loading and CLIPTokenizer max length are unrelated subsystems." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "init_empty_weights argument handling and AutoImageProcessor URL loading are different bugs." + }, + { + "left": "issue:44857", + "right": "issue:46832", + "accept": false, + "reason": "Different RT-DETR-related failures in separate code paths; one is loss/AMP crashing, the other is forward logic with feature-level mismatch." + }, + { + "left": "issue:45538", + "right": "issue:46869", + "accept": false, + "reason": "Tokenizer max-length behavior and YOLO multi-GPU device mapping are unrelated." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed weight-shape runtime error and slow/fast tokenizer inconsistency are not the same issue." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "This is the closest pair, but the evidence is still too thin: one is a specific AddedToken TypeError, the other a model-specific tokenizer load failure. Could be related, but not enough to merge." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Both are training/loading-related in broad terms, but they concern different subsystems and failure modes." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "Tokenizer/from_pretrained gguf handling and Apple Silicon allocator warmup are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:28282", + "accept": false, + "reason": "XLMRoberta with Flash Attention 2 and missing-PyTorch AutoModel import are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different behaviors; shared broad theme only." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Different Qwen-family device-map versus video-input failures; not the same underlying bug." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both involve video inputs, but they are different models and different exceptions." + }, + { + "left": "issue:43577", + "right": "issue:46512", + "accept": false, + "reason": "One is BLIP2 dtype loading behavior; the other is stale config.dtype after from_config(dtype=...). Similar area, different bug." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Sliding_window behavior in Qwen3-MoE and AutoImageProcessor URL loading 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": 28, + "review_unit_index": 24, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29870", + "issue:32090", + "issue:34567", + "issue:36331", + "issue:38175", + "issue:42175", + "issue:43257", + "issue:43452", + "issue:43540", + "issue:43577", + "issue:43716", + "issue:43742", + "issue:43756", + "issue:43906", + "issue:44062", + "issue:44479", + "issue:44488", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44737", + "issue:44792", + "issue:44849", + "issue:44877", + "issue:44898", + "issue:44964", + "issue:44987", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45381", + "issue:45405", + "issue:45600", + "issue:45672", + "issue:45674", + "issue:45701", + "issue:45823", + "issue:45907", + "issue:46399", + "issue:46463", + "issue:46491", + "issue:46693", + "issue:46701", + "issue:46899", + "issue:47276" + ], + "soft_pairs": [ + "issue:44849|issue:45701", + "issue:43540|issue:44479", + "issue:42175|issue:44488", + "issue:34567|issue:36331", + "issue:44479|issue:45381", + "issue:44062|issue:45081", + "issue:44987|issue:45907", + "issue:45042|issue:45081", + "issue:38175|issue:44898", + "issue:44792|issue:44964", + "issue:43742|issue:44589", + "issue:43257|issue:43756", + "issue:43756|issue:45325", + "issue:43756|issue:46399", + "issue:45672|issue:45674", + "issue:44737|issue:46463", + "issue:46463|issue:47276", + "issue:43716|issue:46701", + "issue:45823|issue:46463", + "issue:45405|issue:45600", + "issue:43577|issue:45072", + "issue:44568|issue:46491", + "issue:43906|issue:46693", + "issue:24643|issue:29870", + "issue:44625|issue:44877", + "issue:43540|issue:46693", + "issue:43452|issue:44488", + "issue:28282|issue:32090", + "issue:46463|issue:46899" + ], + "prepared_review_unit_hash": "597e44665c281439e5717a81dc2cf025fd01633003efe358f83b6fb5223c6232", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36713, + "estimated_input_tokens": 9179, + "estimated_eval_tokens": 18614 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a grab bag of mostly unrelated Transformers issues across trainers, tokenizers, multimodal models, configs, and packaging. I don't see any soft pair that clearly shares the same underlying bug or change.", + "confidence": 0.91, + "canonical_issue_reason": "issue:46693 is the strongest single anchor here: it has the clearest regression framing, broad impact, and the most discussion/inbound references among otherwise unrelated reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46693 is the best representative issue because it is a concrete, high-signal performance regression with strong evidence of user impact and active follow-up.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 hidden-state behavior vs tokenizer/version-dependent tokenization changes; different subsystems and symptoms." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related, but one is a Qwen3OmniMoe input-processing failure and the other is a broader v5.3.0 video regression across Qwen VL models." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Packaging/backend install issue vs a specific model-loading failure; no shared code path." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "TrainerState token counter update bug vs CustomTrainer compute_loss signature breakage; separate trainer issues." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": false, + "reason": "Both mention Qwen video regressions, but the affected model families and failure modes differ enough that this is not the same bug." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer AddedToken argument conflict vs Mistral regex patch crash; unrelated tokenizer internals." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model loading failure vs list_repo_files not forwarding revision; different APIs and fixes." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Image-processor torchvision dependency regression vs a tokenizer regex AttributeError; unrelated areas." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Different image-model failures with no common concrete mechanism." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Janus test failure vs Phi-4 multimodal loading failure; same broad domain, different bugs." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "MobileLLM key error vs Float8 storage lookup error; unrelated runtime/load failures." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MoE deepspeed conversion problem vs SmolLM3 RoPE layer omission; different model families and root causes." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "SmolLM3 architecture bug vs Qwen2.5-VL temporal position scaling; not the same code path." + }, + { + "left": "issue:43756", + "right": "issue:46399", + "accept": false, + "reason": "SmolLM3 RoPE issue vs Qwen3.5 kernel registration crash; unrelated fixes." + }, + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Both are 4-bit inference issues, but one is Gemma4 uint8/torch.finfo handling and the other is BitsAndBytes skip-module exclusion logic; different root causes." + }, + { + "left": "issue:44737", + "right": "issue:46463", + "accept": false, + "reason": "XLNet CPU-device bug vs T5 decoder-input validation message; unrelated model paths." + }, + { + "left": "issue:46463", + "right": "issue:47276", + "accept": false, + "reason": "Both are diagnostics/warnings, but the affected checks and code paths differ." + }, + { + "left": "issue:43716", + "right": "issue:46701", + "accept": false, + "reason": "Mistral-3 image-preprocessor dtype mismatch vs DiffusionGemma pipeline failure; different multimodal components." + }, + { + "left": "issue:45823", + "right": "issue:46463", + "accept": false, + "reason": "Gemma4 device_map forward mismatch vs T5 validation error text; no shared bug." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "PEFT version pin issue vs stale auto_mappings reference; separate packaging/CI problems." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both involve dtype mismatches, but on different model stacks and with different failure mechanisms." + }, + { + "left": "issue:44568", + "right": "issue:46491", + "accept": false, + "reason": "Tokenizer special-token regression vs CodeLlama round-trip spacing regression; distinct tokenizer behaviors." + }, + { + "left": "issue:43906", + "right": "issue:46693", + "accept": false, + "reason": "One is a reproduction of an earlier issue, the other is a long-context FA2 performance regression; not the same bug." + }, + { + "left": "issue:24643", + "right": "issue:29870", + "accept": false, + "reason": "DeepSpeed training weight-shape error vs BART generate mask initialization bug; unrelated failures." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3.5 num_labels propagation vs strict config loading for granite_speech; different config plumbing." + }, + { + "left": "issue:43540", + "right": "issue:46693", + "accept": false, + "reason": "Video-input processing in Qwen3OmniMoe vs FA2 long-context performance regression; unrelated subsystems." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "GGUF tokenizer/model loading regression vs a specific sleng-bert load failure; too different to merge." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "Missing PyTorch dependency error vs unsupported NoneType in `_gpu_broadcast_one`; unrelated issues." + }, + { + "left": "issue:45672", + "right": "issue:46899", + "accept": false, + "reason": "Both concern Gemma4 multimodal 4-bit inference, but the reported roots differ: unsafe dtype casting in embedders vs torch.finfo on uint8 audio weights." + } + ] + }, + "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": 28, + "review_unit_index": 25, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:29159", + "issue:33357", + "issue:38175", + "issue:40444", + "issue:41762", + "issue:43295", + "issue:43452", + "issue:43531", + "issue:43643", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43906", + "issue:43994", + "issue:44361", + "issue:44442", + "issue:44466", + "issue:44493", + "issue:44737", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44857", + "issue:44898", + "issue:44913", + "issue:44936", + "issue:44991", + "issue:45005", + "issue:45216", + "issue:45357", + "issue:45372", + "issue:45375", + "issue:45588", + "issue:45663", + "issue:45701", + "issue:45704", + "issue:45920", + "issue:46050", + "issue:46121", + "issue:46170", + "issue:46468", + "issue:46821", + "issue:46829", + "issue:46846", + "issue:47000" + ], + "soft_pairs": [ + "issue:44466|issue:45005", + "issue:43906|issue:46468", + "issue:45663|issue:46846", + "issue:44913|issue:46121", + "issue:41762|issue:44936", + "issue:43697|issue:46829", + "issue:43452|issue:44843", + "issue:44737|issue:45588", + "issue:43697|issue:45216", + "issue:43720|issue:44898", + "issue:45357|issue:45920", + "issue:44493|issue:46050", + "issue:43653|issue:44361", + "issue:44361|issue:44442", + "issue:38175|issue:43295", + "issue:43688|issue:43749", + "issue:43643|issue:45375", + "issue:43531|issue:43742", + "issue:43994|issue:44898", + "issue:44821|issue:46821", + "issue:44843|issue:44991", + "issue:40444|issue:44811", + "issue:29159|issue:33357", + "issue:45372|issue:46170", + "issue:44857|issue:47000", + "issue:43688|issue:45704", + "issue:43688|issue:45701" + ], + "prepared_review_unit_hash": "9fad21d4dad710328697ca96907059847437fb3fc059356fbafa103a09e2e111", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36096, + "estimated_input_tokens": 9024, + "estimated_eval_tokens": 18304 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No strong duplicate group emerges from these issues; the soft pairs mostly share only broad subsystem similarity (tokenizers, configs, loading, vision models) but describe different failures and code paths.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45920 is the best representative: it describes a broad v5 tokenizer regression affecting multiple model families, making it the most general and cluster-like issue among otherwise mostly model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Same reason: it is the broadest and most reusable issue title in the set, with a clear regression framing and multiple impacted models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights in v5, but one is a device-dependent lm_head serialization bug and the other is a broader translation-model tied-weights issue; not the same concrete failure." + }, + { + "left": "issue:43906", + "right": "issue:46468", + "accept": false, + "reason": "One is an isolated reproduction of an older issue, the other is an example-specific Nemotron breakage; different artifacts and likely different root causes." + }, + { + "left": "issue:45663", + "right": "issue:46846", + "accept": false, + "reason": "Different models and different failure modes: Gemma-4 FSDP2 KeyError vs Qwen3.5 TP plan OOM/shape errors." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "Both involve rope/config persistence, but one is a silent default reset on reload and the other is a TypeError from a JSON-loaded list; not the same bug." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Loading Gemma3 with ZeRO-3 and trainer.evaluate() after train are unrelated execution paths with different symptoms." + }, + { + "left": "issue:43697", + "right": "issue:46829", + "accept": false, + "reason": "Different models and outputs: RTDetrV2 regression vs VideoPrism output field mix-up." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both are tokenizer-loading problems, but gguf_file breaks model/tokenizer loading generally while the other is an offline-mode regex patch bug." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU-device inefficiency and flash_attention s_aux=None crash are unrelated code paths." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "Different Qwen/RTDetr concerns: one is changed outputs at inference, the other is incorrect save_pretrained checkpoint serialization." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "BitNet accelerate-loading unpacking failure and Perceiver non-default-resolution failure are different model-specific bugs." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Qwen3.5 visual encoder key serialization is a model save_pretrained bug, while the other is a broad tokenizer-ID regression across many models." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Unexpected position-id keys and sinq quantization save/load breakage affect different serialization surfaces." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "Both are tokenizer-related, but BigBird special-token registration and MLuke task AttributeError are distinct bugs." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "MLuke task failures and AutoTokenizer loading FastSpeech2ConformerTokenizer are unrelated tokenizer issues." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Siglip2 zero-probability outputs and processor.tokenizer/image-passing regression are different model/processor failures." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Aux-loss normalization in MoE models and FSDP CPU RAM efficient loading are unrelated subsystems." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both mention missing config fields, but one is a generic trust_remote_code loading issue and the other is a specific Qwen vision config field dropped by strict parsing." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and MobileLLM-125M load key error are different model-specific issues." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Siglip2 nonsensical outputs and Perceiver interpolation failure are different vision-model regressions." + }, + { + "left": "issue:44821", + "right": "issue:46821", + "accept": false, + "reason": "AutoImageProcessor URL loading and TimesFM window_size AttributeError are unrelated loading vs model-argument bugs." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline HF hub lookup in mistral regex patch and tokenizer loading for EMBEDDIA/est-roberta are different causes." + }, + { + "left": "issue:40444", + "right": "issue:44811", + "accept": false, + "reason": "IterableDataset multi-image finetuning failure and Whisper batch_decode skip_special_tokens bug are different code paths." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency and MacOS bus error on CLIP model are unrelated." + }, + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Gemma 4 processor import failure from mistral_common and Windows ROCm torch.distributed dependency issues are different platform/library problems." + }, + { + "left": "issue:44857", + "right": "issue:47000", + "accept": false, + "reason": "LwDetrImageLoss AMP crash and RTDetrLoss infeasible cost-matrix error are different loss-function bugs." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "MoE auxiliary-loss normalization and T5 apex RMSNorm memory leak are unrelated." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Aux-loss normalization in MoE models and generic tokenization changes across versions 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": 28, + "review_unit_index": 26, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:32090", + "issue:33666", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:42898", + "issue:43278", + "issue:43295", + "issue:43452", + "issue:43504", + "issue:43525", + "issue:43550", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43749", + "issue:43756", + "issue:43761", + "issue:43782", + "issue:43927", + "issue:44361", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44484", + "issue:44561", + "issue:44568", + "issue:44625", + "issue:44704", + "issue:44743", + "issue:44977", + "issue:45356", + "issue:45405", + "issue:45440", + "issue:45542", + "issue:46050", + "issue:46178", + "issue:46402", + "issue:46491", + "issue:46693", + "issue:46701", + "issue:46829", + "issue:46832", + "issue:46932", + "issue:47013", + "issue:47254" + ], + "soft_pairs": [ + "issue:43525|issue:46701", + "issue:43646|issue:43749", + "issue:45542|issue:46178", + "issue:43295|issue:45356", + "issue:43756|issue:46402", + "issue:43653|issue:44568", + "issue:46832|issue:46932", + "issue:43761|issue:46829", + "issue:43761|issue:47013", + "issue:43653|issue:43927", + "issue:33666|issue:36010", + "issue:44464|issue:44743", + "issue:44479|issue:44625", + "issue:43452|issue:43577", + "issue:43550|issue:44361", + "issue:42898|issue:46491", + "issue:32090|issue:33666", + "issue:44466|issue:46050", + "issue:41720|issue:44464", + "issue:33666|issue:36331", + "issue:44977|issue:46693", + "issue:44464|issue:44977", + "issue:43688|issue:45440", + "issue:43504|issue:43550", + "issue:44484|issue:44704", + "issue:43278|issue:43688", + "issue:45405|issue:47254", + "issue:43782|issue:46693", + "issue:43643|issue:44561" + ], + "prepared_review_unit_hash": "9882fe2ec05e433c7b9f65425e286ec48ff57b383edde344ee639c410e2929cb", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36638, + "estimated_input_tokens": 9160, + "estimated_eval_tokens": 18576 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of model-specific regressions and API breakages. The similarity scores seem driven by shared transformer/regression vocabulary rather than true duplicate bugs, so I would not merge any of the soft-edge pairs.", + "confidence": 0.24, + "canonical_issue_reason": "Issue #43646 is the broadest representative of the set because it describes a general v5/custom-model initialization regression, but the overall cluster is too heterogeneous for a strong true canonical.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43646 is the best cluster anchor: it is generic, regression-oriented, and closest to the common theme of user-facing breakage across custom/model-loading paths.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43525", + "right": "issue:46701", + "accept": false, + "reason": "Different model families and failure modes: missing pad_token_id vs a broken image-text pipeline." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Both are regressions, but one is custom model init while the other is FSDP CPU RAM loading." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Both mention backend/import issues, but one is TensorFlow absence handling and the other is tensorflow_text mapping." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer-related regressions, but they concern unrelated code paths and different symptoms." + }, + { + "left": "issue:43756", + "right": "issue:46402", + "accept": false, + "reason": "Different models and bugs: RoPE layer dropping vs text encoder weights not loading." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both involve special tokens, but the affected tokenizers and exact failures differ." + }, + { + "left": "issue:46832", + "right": "issue:46932", + "accept": false, + "reason": "RT-DETR feature-level mismatch and Gemma4 quantized embedder casting are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:46829", + "accept": false, + "reason": "Both touch hidden_states output, but they are different models with different output-contract bugs." + }, + { + "left": "issue:43761", + "right": "issue:47013", + "accept": false, + "reason": "Similar tensor/output shape theme, but CLIPVision hidden_states and xLSTM state unpacking are different failures." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "BigBird special-token registration and DiaConfig token-id persistence are different tokenizer/config bugs." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Multi-GPU Qwen2-VL training and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Different code paths: compiled chunked generation vs recurrent-state reset in modular_qwen3_5." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression and num_labels propagation are unrelated Qwen3.5 issues." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "GGUF loading failures and BLIP2 dtype retention are separate model-loading bugs." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "torch.compile SDPA failure for Bamba is not the same as MLukeTokenizer task AttributeError." + }, + { + "left": "issue:42898", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer behavior issues, but one is a global v5 cleanup option change and the other is a CodeLlama round-trip regression." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error and Qwen2-VL multi-GPU training are not the same bug." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Both involve save/load behavior, but tied-weight serialization and sinq quantization persistence are distinct problems." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "CUDA assert with auto device mapping is unrelated to compiled-forward chunked generation inconsistencies." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Multi-GPU Qwen2-VL training and compute_loss signature mismatch are different Trainer issues." + }, + { + "left": "issue:44977", + "right": "issue:46693", + "accept": false, + "reason": "Both are generation regressions, but flash-attention failure and long-context FA2 performance regression are not the same defect." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Both affect generation, but compiled-forward inconsistency and flash-attention malfunction have different root causes." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Auxiliary-loss normalization and DeepSeekV3MoE implementation divergence are different model/math issues." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Pretrained preset loading with a legacy field is unrelated to torch.compile SDPA behavior." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "max_shard_size semantics and AutoProcessor cached_file kwargs are separate API issues." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype drift during eval and auxiliary-loss normalization are unrelated training bugs." + }, + { + "left": "issue:45405", + "right": "issue:47254", + "accept": false, + "reason": "A version pin not yet on PyPI and a PEFT/DeepSpeed checkpoint error are not duplicates." + }, + { + "left": "issue:43782", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3VL weight_only load failure and FA2 long-context performance regression are different failure modes." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both involve trust_remote_code, but one is missing config fields and the other is a removed helper import." + } + ] + }, + "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": 28, + "review_unit_index": 27, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28218", + "issue:33357", + "issue:39692", + "issue:42796", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43611", + "issue:43646", + "issue:43701", + "issue:43720", + "issue:43825", + "issue:43844", + "issue:43927", + "issue:43994", + "issue:44361", + "issue:44466", + "issue:44488", + "issue:44514", + "issue:44554", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44734", + "issue:44843", + "issue:44964", + "issue:44977", + "issue:45290", + "issue:45405", + "issue:45491", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46002", + "issue:46050", + "issue:46192", + "issue:46209", + "issue:46365", + "issue:46489", + "issue:46612", + "issue:46846", + "issue:47276" + ], + "soft_pairs": [ + "issue:44734|issue:46612", + "issue:43504|issue:43720", + "issue:44625|issue:47276", + "issue:43720|issue:43927", + "issue:43606|issue:43927", + "issue:43504|issue:44610", + "issue:43577|issue:44554", + "issue:43611|issue:46846", + "issue:43525|issue:43572", + "issue:43720|issue:44361", + "issue:44964|issue:45405", + "issue:44514|issue:45290", + "issue:43493|issue:43825", + "issue:28218|issue:33357", + "issue:44977|issue:46209", + "issue:45823|issue:46050", + "issue:45491|issue:46002", + "issue:44554|issue:46365", + "issue:39692|issue:43295", + "issue:43994|issue:46192", + "issue:44488|issue:44843", + "issue:42796|issue:44977", + "issue:43334|issue:45759", + "issue:43278|issue:43844", + "issue:43646|issue:44466", + "issue:43334|issue:43572", + "issue:43278|issue:43701", + "issue:44568|issue:46489", + "issue:44568|issue:45812", + "issue:43927|issue:44361", + "issue:43493|issue:43844" + ], + "prepared_review_unit_hash": "809b7fc3b6207bdc562f5789afadc309f6ee0d82835e28d3cb926375c2efd769", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37321, + "estimated_input_tokens": 9331, + "estimated_eval_tokens": 18918 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is thematically broad and the soft pairs mostly share only subsystem-level similarity (tokenizers, configs, multimodal loading, generation, dtype/loading regressions). I don\u2019t see any true duplicate pairs here, and there are no PRs in the set.", + "confidence": 0.95, + "canonical_issue_reason": "Issue 45812 is the best representative anchor: it describes a concrete, high-impact v4\u2192v5 tokenizer regression with clear user-visible wrong outputs and strong reporting context, making it the most canonical issue in this mixed set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45812 is the strongest overall triage candidate because it is specific, reproducible, and clearly tied to a concrete code-path regression rather than a vague platform symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44734", + "right": "issue:46612", + "accept": false, + "reason": "Both involve generation/cache behavior, but 44734 is a serve KV-continuation tensor-indexing crash while 46612 is a beam-search cache reorder omission for specific model classes. Different code paths and fixes." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "43504 is a Beit legacy-field load failure; 43720 is BitNet packed-weight unpacking during accelerate loading. Different models and root causes." + }, + { + "left": "issue:44625", + "right": "issue:47276", + "accept": false, + "reason": "44625 is missing num_labels propagation in Qwen3.5 configs; 47276 is a misleading warning message in continuous batching. No shared bug." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "BitNet packed-weight loading vs DiaConfig save/load losing custom token IDs. Same broad loading theme, but unrelated failures." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU offload device mismatch for bark-small vs config token-ID loss on save/load. Different subsystems and symptoms." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Beit legacy-field loading and OmDet-Turbo processor image-size mismatch are unrelated bugs despite both being model-loading issues." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "Blip2 dtype not propagating and an MPS attention correctness bug are different failures in different execution paths." + }, + { + "left": "issue:43611", + "right": "issue:46846", + "accept": false, + "reason": "Base-model-prefix loading regression vs Qwen3.5 tensor-parallel plan missing linear_attn entries. Different APIs and failure modes." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Both are missing token-id-style config fields, but they affect different model configs and different attributes (pad_token_id vs pad_token_idx). Not the same bug." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "Packed-weight unpacking in BitNet and MLukeTokenizer task AttributeError are unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Phi-4 multimodal load failure and an unreleased PEFT version bump are not the same underlying issue." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both crash in apply_chat_template, but one is batched input with padding=False and the other is assistant tool-call messages with no content. Different triggers and likely different fixes." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 HF-vs-JAX discrepancy is unrelated to a pipeline warning about translation support in v5." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer added-space behavior and a MacOS bus error in CLIP loading are unrelated." + }, + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 flash-attention generation failure and FP8 inference failure on B200 are different hardware/model-path issues." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Gemma4 device-map mismatch and sinq quantization save/load breakage are unrelated regressions." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Both mention NaNs, but one is Gemma3 mixed-length sliding-window attention and the other is classification-head init/checkout. Different code paths and causes." + }, + { + "left": "issue:44554", + "right": "issue:46365", + "accept": false, + "reason": "An MPS attention correctness issue is unrelated to importing Trainer being blocked by datasets." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "SigLIP2 docs example errors/quantization failure are not the same as the processor.tokenizer/image-passing regression." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "Wrong SigLIP2 outputs with AutoModel/pipeline and return_tensors being ignored with text_kwargs are different bugs." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Both are tokenizer-loading related, but one is a specific model load failure and the other is an offline-mode regression in _patch_mistral_regex. Not the same issue." + }, + { + "left": "issue:42796", + "right": "issue:44977", + "accept": false, + "reason": "Ministral BPE-marker handling and Qwen3.5 flash-attention generation are unrelated." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Both involve Qwen multimodal configs, but one is a missing pad_token_id on text config and the other is failure to unwrap text_config in AutoModelForCausalLM.from_config. Different concrete bugs." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype drift in evaluate and gradient growth under ZeRO-3 are distinct training/runtime problems." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage and inconsistent lm_head.weight serialization are different v5 regressions." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both are missing config token fields, but they affect different model families and different attribute names; not a duplicate." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Dtype mismatch in evaluation and resume_from_checkpoint key mismatch are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions, but one is add_special_tokens not adding BOS/EOS for mdeberta-v3 and the other is DeepSeek-Coder v1 BPE-marker output differences. Different concrete failures." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they affect different tokenizers and different behaviors: special-token insertion vs wrong token IDs for Granite." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig save/load custom token-ID loss and MLukeTokenizer task AttributeError are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 implementation discrepancy vs a gradient anomaly under HfDeepSpeedConfig/ZeRO-3 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": 28, + "review_unit_index": 28, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:34567", + "issue:38175", + "issue:42490", + "issue:42947", + "issue:43381", + "issue:43504", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43653", + "issue:43696", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44112", + "issue:44291", + "issue:44464", + "issue:44493", + "issue:44589", + "issue:44849", + "issue:44964", + "issue:45072", + "issue:45084", + "issue:45229", + "issue:45245", + "issue:45357", + "issue:45405", + "issue:45520", + "issue:45584", + "issue:45823", + "issue:46050", + "issue:46209", + "issue:46285", + "issue:46399", + "issue:46710", + "issue:46821", + "issue:46832", + "issue:46846", + "issue:47013", + "issue:47145", + "issue:47254" + ], + "soft_pairs": [ + "issue:43572|issue:44291", + "issue:43575|issue:45405", + "issue:46209|issue:46710", + "issue:33357|issue:33666", + "issue:43824|issue:44589", + "issue:44589|issue:45520", + "issue:43504|issue:43606", + "issue:46821|issue:46832", + "issue:44849|issue:46209", + "issue:44849|issue:45357", + "issue:44493|issue:46285", + "issue:45245|issue:46710", + "issue:43504|issue:43653", + "issue:43606|issue:44112", + "issue:45072|issue:46832", + "issue:43994|issue:44493", + "issue:44464|issue:45584", + "issue:42490|issue:43696", + "issue:43873|issue:46050", + "issue:15354|issue:38175", + "issue:43844|issue:44964", + "issue:44849|issue:47145", + "issue:42947|issue:43381", + "issue:46399|issue:46846", + "issue:47013|issue:47254", + "issue:29159|issue:32090", + "issue:45229|issue:45823", + "issue:45084|issue:45520", + "issue:28218|issue:34567" + ], + "prepared_review_unit_hash": "bb8889c19e7ca70f509b048cc3d1eb9938078535a66eb7a923e1b327dd622122", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36405, + "estimated_input_tokens": 9102, + "estimated_eval_tokens": 18460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft edge looks like a true duplicate or mergeable PR match; the candidate pairs mostly share only broad subsystem similarity (tokenizer, training, loading, OOM, etc.) while describing different concrete failures. This cluster appears heterogeneous rather than a single duplicate set.", + "confidence": 0.79, + "canonical_issue_reason": "No clear canonical issue across the set: the items span many unrelated bugs/regressions, so none safely subsumes the rest.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:29159 is the broadest tokenizer-focused report and the most plausible representative if one issue must be chosen, but it still does not explain the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Both are config/loading failures, but one is a missing `pad_token_idx` issue and the other is an unexpected `_is_hf_initialized` argument with `init_empty_weights`; different code paths." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Both mention version-related loading problems, but one is TP/OOM on a specific Qwen model and the other is an unmet PEFT minimum-version pin." + }, + { + "left": "issue:46209", + "right": "issue:46710", + "accept": false, + "reason": "Both are model-quality/performance complaints, but one is FP8 inference support on B200 and the other is an accuracy regression after a Transformers upgrade." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "One is a MacOS bus error with a community CLIP model; the other is a Qwen2-VL multi-GPU training question. Same broad area, different problem." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "Both are import/runtime type failures, but they refer to different missing symbols/storage types and different affected code paths." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "These are unrelated import/type errors: missing Float8 storage object versus a `flash_attn` KeyError on Python 3.13." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Both involve model loading/CI, but one is a legacy field in BEiT semantic segmentation and the other is a device mismatch with CPU offload on bark-small." + }, + { + "left": "issue:46821", + "right": "issue:46832", + "accept": false, + "reason": "Both are model forward/runtime errors, but one is a TimesFM `window_size` attribute issue and the other is RT-DETR feature-level mismatch." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "One is a Qwen3.5 hidden-state output bug; the other is an FP8 inference failure. No shared concrete code path." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Both concern Qwen3.5, but one is `output_hidden_states=True` and the other is incorrect visual-encoder keys in `save_pretrained`; different failures." + }, + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Both mention model structure/key mismatches, but one is a generic position-id key warning and the other is a CLIP class structure inconsistency." + }, + { + "left": "issue:45245", + "right": "issue:46710", + "accept": false, + "reason": "A category-count limit error is not the same as an accuracy regression; these are different failure modes." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "One is a BEiT legacy-field load failure; the other is BigBirdTokenizer mask-token registration/empty decode behavior." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/device related, but one is CPU offload mismatch in model loading and the other is a stale test failure. Not the same bug." + }, + { + "left": "issue:45072", + "right": "issue:46832", + "accept": false, + "reason": "Different model families and different dtype/forward-shape problems; too broad to be one underlying bug." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "One is a SigLIP2 inference quality issue; the other is an unexpected key warning around position ids. No concrete duplicate evidence." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Chunked generation with compiled forward is unrelated to Whisper empty transcription after `align_special_tokens`." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Local-path import warnings and CUDA OOM on GPT-oss-20b are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Quantization/offloading concerns do not make these the same issue: one is offloading not working as expected, the other is broken save/load after sinq quantization." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "A TorchScript `GeneratorExp` scripting limitation and unexpected zero probabilities in SigLIP2 are unrelated." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "Both are model execution failures, but one is gradient explosion with DeepSpeed/ZeRO-3 and the other is a model-loading import failure for Phi-4 multimodal." + }, + { + "left": "issue:44849", + "right": "issue:47145", + "accept": false, + "reason": "Qwen3.5 hidden-state output bug versus FP8 checkpoint loading missing `weight_scale`; different subsystems and symptoms." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is ineffective with PEFT LoRA and the other says it cannot be used in eval mode; not the same concrete bug." + }, + { + "left": "issue:46399", + "right": "issue:46846", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is a missing decorator for `apply_rotary_pos_emb` registration and the other is an incomplete TP plan causing OOM/shape errors." + }, + { + "left": "issue:47013", + "right": "issue:47254", + "accept": false, + "reason": "Both involve checkpointing/state handling, but one is an xLSTM train-mode unpacking crash and the other is a PEFT + DeepSpeed ZeRO-3 checkpoint error." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency is unrelated to a Trainer `_gpu_broadcast_one` NoneType type error." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Both are large-model inference/multi-device problems, but one is CUDA OOM for Gemma4 and the other is a device mismatch in Gemma4 PLE with `device_map=\"auto\"`." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Compilation of non-template nodes and a `flash_attn` import KeyError are different failure classes." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer added-space behavior and TrainerState token-count tracking are unrelated 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": 28, + "review_unit_index": 29, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:29870", + "issue:29942", + "issue:30333", + "issue:37428", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43065", + "issue:43295", + "issue:43450", + "issue:43493", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43701", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43937", + "issue:43957", + "issue:44186", + "issue:44423", + "issue:44554", + "issue:44610", + "issue:44734", + "issue:44869", + "issue:44977", + "issue:45072", + "issue:45198", + "issue:45200", + "issue:45229", + "issue:45406", + "issue:45446", + "issue:45520", + "issue:45588", + "issue:45759", + "issue:45823", + "issue:46050", + "issue:46121", + "issue:46178", + "issue:46192", + "issue:46369", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821" + ], + "soft_pairs": [ + "issue:45200|issue:45229", + "issue:45588|issue:46050", + "issue:44423|issue:44734", + "issue:43493|issue:43701", + "issue:43937|issue:46701", + "issue:41720|issue:46710", + "issue:44554|issue:45446", + "issue:37428|issue:43957", + "issue:44869|issue:46752", + "issue:43065|issue:46369", + "issue:43295|issue:43450", + "issue:44977|issue:46701", + "issue:29942|issue:30333", + "issue:44977|issue:46710", + "issue:45072|issue:45198", + "issue:45588|issue:45759", + "issue:45406|issue:46121", + "issue:43572|issue:46821", + "issue:43295|issue:43873", + "issue:43540|issue:45200", + "issue:44186|issue:46192", + "issue:45823|issue:46192", + "issue:43295|issue:44610", + "issue:43572|issue:43957", + "issue:43450|issue:43575", + "issue:43550|issue:43844", + "issue:37428|issue:43824", + "issue:45520|issue:46178", + "issue:45200|issue:45823", + "issue:29870|issue:29942", + "issue:42490|issue:43493", + "issue:43295|issue:43493", + "issue:39692|issue:46192", + "issue:42915|issue:43824" + ], + "prepared_review_unit_hash": "86d9086a7c49d5de2b1ba484ebf91d289352ea872b98193318690fbb3454d05d", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37960, + "estimated_input_tokens": 9490, + "estimated_eval_tokens": 19236 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous and the soft-similarity pairs do not look like true duplicates. I found no safe issue-to-issue merges, so the cluster should be split rather than consolidated.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43295 is the most representative anchor because it describes a concrete regression in processor/tokenizer behavior with broader impact than the other items, which are mostly unrelated model-, backend-, or pipeline-specific bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43295 is the best single issue to keep as the cluster representative, but only as an anchor for triage; it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Different Gemma4 problems: text-only fine-tuning defaults vs multi-GPU inference OOM." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Different failures: flash-attention sinkless-model crash vs sinq save/load breakage." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve/runtime crashes, but they hit different code paths and symptoms." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "Unrelated: SigLIP2 implementation discrepancy vs checkpoint resume key mismatch." + }, + { + "left": "issue:43937", + "right": "issue:46701", + "accept": false, + "reason": "Different models and issues: invalid GenerationConfig vs image-text pipeline failure." + }, + { + "left": "issue:41720", + "right": "issue:46710", + "accept": false, + "reason": "Different model-specific regressions: A800 auto-device-map crash vs DeepSeek accuracy regression." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "Different backend bugs: MPS attention correctness vs flex_attention import version check." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "Both are loading/import issues, but they involve different APIs and failure modes." + }, + { + "left": "issue:44869", + "right": "issue:46752", + "accept": false, + "reason": "Different text-processing crashes: Whisper timestamp decoding vs empty chat template input." + }, + { + "left": "issue:43065", + "right": "issue:46369", + "accept": false, + "reason": "Different SAM3 issues: dummy layer implementation vs memory overflow in a text variant." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Both involve processors, but one is a tokenizer regression and the other is batched video shape handling." + }, + { + "left": "issue:44977", + "right": "issue:46701", + "accept": false, + "reason": "Different multimodal failures affecting different models and pipelines." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "No overlap: Flash Attention 2 test failures vs MLFlow reporting on training failure." + }, + { + "left": "issue:44977", + "right": "issue:46710", + "accept": false, + "reason": "Different regression targets: flash-attention generation bug vs accuracy drop after version upgrade." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "Different subsystem bugs: dtype mismatch inference vs save/tokenization failure." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Different root causes: flash_attention sink handling vs multimodal config unwrapping." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Different areas: Gemma4 processor private tokenizer attr vs RoPE validation list handling." + }, + { + "left": "issue:43572", + "right": "issue:46821", + "accept": false, + "reason": "Different config/argument bugs in unrelated models and code paths." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Processor/tokenizer regression is unrelated to quantization offloading behavior." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Different multimodal inputs: Qwen3Omni video processing vs Gemma4 mm_token_type_ids defaulting." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "Different tokenizer behaviors: NER/batched padding crash vs return_tensors precedence bug." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Device-mapping mismatch is unrelated to return_tensors being ignored with text_kwargs." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Both are processor-related, but they concern different models and incompatible input-shape bugs." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "Unrelated config issues: missing pad_token_idx vs meta-device loading breakage." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Video processor batch-shape bug is unrelated to tensor-parallel OOM loading." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "Different training/inference problems: torch.compile SDPA failure vs ZeRO-3 gradient growth." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and likely different fixes." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Different backend-import problems with no shared code path." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but separate bugs: missing mm_token_type_ids default vs PLE device mismatch." + }, + { + "left": "issue:29870", + "right": "issue:29942", + "accept": false, + "reason": "Unrelated issues: Bart generate() UnboundLocalError vs Flash Attention test failures." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "Local-path trailing slash warning is unrelated to SigLIP2 implementation discrepancy." + }, + { + "left": "issue:43295", + "right": "issue:43493", + "accept": false, + "reason": "Processor/tokenizer regression and SigLIP2 implementation mismatch are different problems." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 documentation errors are unrelated to the return_tensors precedence bug." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Qwen2.5-VL import error are different model-specific 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": 28, + "review_unit_index": 30, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:37428", + "issue:41720", + "issue:43262", + "issue:43278", + "issue:43334", + "issue:43335", + "issue:43425", + "issue:43450", + "issue:43452", + "issue:43525", + "issue:43550", + "issue:43606", + "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" + ], + "soft_pairs": [ + "issue:43334|issue:43525", + "issue:43278|issue:43856", + "issue:28218|issue:35141", + "issue:44521|issue:45290", + "issue:44466|issue:45127", + "issue:43824|issue:43975", + "issue:45381|issue:46209", + "issue:44737|issue:46894", + "issue:43262|issue:43450", + "issue:43975|issue:45520", + "issue:43653|issue:44112", + "issue:43452|issue:46512", + "issue:43950|issue:46285", + "issue:44561|issue:46234", + "issue:44442|issue:45198", + "issue:45200|issue:46899", + "issue:44479|issue:46192", + "issue:43819|issue:44112", + "issue:44898|issue:45412", + "issue:43335|issue:43550", + "issue:45200|issue:45325", + "issue:45200|issue:46932", + "issue:41720|issue:46209", + "issue:44610|issue:44857", + "issue:43819|issue:45083", + "issue:37428|issue:43425", + "issue:43606|issue:45083" + ], + "prepared_review_unit_hash": "01ffcfd6f2f636907eb96d73781880afc2b4d92103f99de8201b569aaa901885", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35639, + "estimated_input_tokens": 8910, + "estimated_eval_tokens": 18076 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: the soft pairs mostly share a subsystem name or a vague regression theme, but they describe different code paths and different fixes. I would not merge any of the proposed soft edges.", + "confidence": 0.33, + "canonical_issue_reason": "Best representative is issue:44521 because it is open, has the most engagement/inbound refs, and targets a central user-facing API (`apply_chat_template`) with broad multimodal impact. That said, the cluster is not a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue:44521 is the strongest single issue to anchor the cluster: active, high-discussion, and on a shared API path that can attract related fixes. It is more representative than the narrower model-specific or regression-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both are missing `pad_token_id`, but they affect different configs/models (`Qwen3VLTextConfig` vs `Llama4Config`) and likely need separate fixes." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Different bugs: dtype mismatch in evaluation vs memory inefficiency in Qwen3 MoE training." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing after added tokens is unrelated to embedding reinitialization during `post_init`." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both touch `apply_chat_template`, but one is about multimodal assistant masks and the other about tool-call messages with no content; different failure modes and fixes." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Serialization inconsistency of `lm_head.weight` is distinct from LoRA merge collapse with extended vocab." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "Import error for a Qwen VL class is unrelated to a detokenization regression in DeepSeek Coder." + }, + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Qwen2.5-VL video position IDs bug is unrelated to FP8 inference failing on B200." + }, + { + "left": "issue:44737", + "right": "issue:46894", + "accept": false, + "reason": "XLNet CPU positional encoding bug and CohereASR double-shift training loss are different code paths." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling-rate default and batched video shape handling are different processors and bugs." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Detokenization regression and Python 3.13 import-utils KeyError are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "Tokenizer special-token registration issue is unrelated to a stale CI test." + }, + { + "left": "issue:43452", + "right": "issue:46512", + "accept": false, + "reason": "`gguf_file` loading failure and `from_config(dtype=...)` stale config dtype are separate initialization problems." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Non-persistent buffer corruption on load is unrelated to CLIP text-model structural inconsistency." + }, + { + "left": "issue:44561", + "right": "issue:46234", + "accept": false, + "reason": "Missing utility after v5 removal and missing `is_falcon_mamba_ssm_available` are different import symbol regressions." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "FastSpeech2Conformer tokenizer loading and Wav2Vec2 save/tokenization failures are not the same bug." + }, + { + "left": "issue:45200", + "right": "issue:46899", + "accept": false, + "reason": "Defaulting `mm_token_type_ids` to zeros is unrelated to unsafe dtype casts in quantized embedders." + }, + { + "left": "issue:44479", + "right": "issue:46192", + "accept": false, + "reason": "Video regression in multimodal processors and `return_tensors` being ignored are separate API issues." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC latent conversion mismatch and GraniteSpeech CI test staleness are unrelated." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver non-default resolution failure is unrelated to RT-DETR memory release behavior." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction and Bamba SDPA compile failure are distinct model issues." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "`mm_token_type_ids` defaulting and Qwen2.5-VL rope-index scaling are different multimodal bugs." + }, + { + "left": "issue:45200", + "right": "issue:46932", + "accept": false, + "reason": "Default zero `mm_token_type_ids` is not the same as quantized embedder dtype casting." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Different hardware/runtime failures in different models: A800 auto device mapping vs B200 FP8 inference." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output size mismatch and AMP/CUDA loss crash are unrelated." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC latent mismatch and Qwen3 Omni helper length computation are different bugs." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "Missing flash-attention import helper and Torch 2.10 incompatibility are not the same underlying issue." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU offload device mismatch in Bark is unrelated to Qwen3 Omni feature-length calculations." + } + ] + }, + "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": 28, + "review_unit_index": 31, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28282", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:42490", + "issue:43335", + "issue:43388", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43827", + "issue:43854", + "issue:43901", + "issue:44112", + "issue:44206", + "issue:44351", + "issue:44448", + "issue:44466", + "issue:44617", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44857", + "issue:45083", + "issue:45412", + "issue:45520", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45759", + "issue:45859", + "issue:46170", + "issue:46285", + "issue:46424", + "issue:46431", + "issue:46512", + "issue:46566", + "issue:46697", + "issue:46821", + "issue:46858" + ], + "soft_pairs": [ + "issue:44792|issue:45563", + "issue:43335|issue:43653", + "issue:43827|issue:43901", + "issue:44857|issue:45593", + "issue:46512|issue:46821", + "issue:44206|issue:45584", + "issue:45759|issue:46285", + "issue:28282|issue:33357", + "issue:38175|issue:45412", + "issue:29942|issue:32090", + "issue:43388|issue:43526", + "issue:43335|issue:43504", + "issue:43611|issue:44466", + "issue:44855|issue:45083", + "issue:46424|issue:46858", + "issue:43550|issue:44112", + "issue:43335|issue:45083", + "issue:43335|issue:44112", + "issue:43335|issue:43606", + "issue:46431|issue:46566", + "issue:45520|issue:46170", + "issue:44448|issue:44743", + "issue:37428|issue:44351", + "issue:44617|issue:46697", + "issue:42490|issue:43854", + "issue:34689|issue:35141", + "issue:45588|issue:45859" + ], + "prepared_review_unit_hash": "78e6b97ab87741163e91808a6762fe58cebaf94ce369a0d64d6ace57fcea4645", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36255, + "estimated_input_tokens": 9064, + "estimated_eval_tokens": 18384 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated issue reports with superficial overlap in Transformers regressions, model loading, cache handling, and docs/tests. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected. The closest thematic hub is issue #43335, but it still does not absorb the others into one duplicate cluster.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue #43335 is the closest hub in this set because a few other reports are loosely about model/config initialization, but the overall group is too heterogeneous for a true canonical duplicate.", + "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 warning in paged generate()." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Sparse-layer config bug vs BigBirdTokenizer mask-token registration/empty decode bug; different code paths." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs issues, but they target different API deprecations and different pages." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "AMP/CUDA crash in LwDetrImageLoss vs D-FINE not using auxiliary losses; unrelated behavior." + }, + { + "left": "issue:46512", + "right": "issue:46821", + "accept": false, + "reason": "Stale config.dtype after from_config vs window_size AttributeError; different APIs and failures." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Unsupported center arg in LaserFeatureExtractor vs Whisper empty-transcription failure after align_special_tokens." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Different model families and bugs: Qwen multimodal from_config unwrap vs CLIP text-model structure inconsistency." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Missing PyTorch ImportError vs MacOS bus error for community CLIP; different failure modes." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "Unexpected zero probabilities in Siglip2 vs RT-DETR memory release bug; unrelated." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures vs NoneType broadcast error in Trainer; different subsystems." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both involve labels, but one is gather_for_metrics dropping tuple elements and the other is BeitImageProcessorFast reduce_labels returning one label." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Sparse-layer zero-count config bug vs BeitForSemanticSegmentation legacy-field loading issue." + }, + { + "left": "issue:43611", + "right": "issue:44466", + "accept": false, + "reason": "base_model_prefix loading regression vs inconsistent lm_head serialization; separate code paths." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "TorchScript parse failure from comment placement vs helper _get_feat_extract_output_lengths behavior; unrelated." + }, + { + "left": "issue:46424", + "right": "issue:46858", + "accept": false, + "reason": "max_cache_length only honored for quantized cache vs device-tensor seq_length breaking arithmetic; both cache-related but different bugs." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "torch.compile SDPA failure in Bamba vs stale GraniteSpeech CI test; not the same bug." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Different functionality: sparse-layer config creation vs feature-extractor length helper behavior." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "Model config bug vs stale CI test; no shared underlying defect." + }, + { + "left": "issue:43335", + "right": "issue:43606", + "accept": false, + "reason": "Sparse-layer config bug vs Bark CPU offload device mismatch; different models and code paths." + }, + { + "left": "issue:46431", + "right": "issue:46566", + "accept": false, + "reason": "Missing chat template error vs meta-device weights under device_map='auto'; different loading issues." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "Python 3.13 flash_attn import key error vs Windows ROCm torch.distributed dependency issue." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Pegasus v4/v5 output difference vs Qwen3 recurrent-state reset with cache; different model-specific bugs." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different underlying causes." + }, + { + "left": "issue:44617", + "right": "issue:46697", + "accept": false, + "reason": "CUDA OOM in Sam3Video vs Swin2SR over-padding images; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Trailing-slash local model import warning vs GLM-4.7-Flash unit-test loading failure; different triggers." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Llama 3.2 Vision loading regression vs post_init reinitializing resized embeddings; not the same defect." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "flash_attention.py s_aux=None AttributeError vs Qwen3_5MoeTextRotaryEmbedding CPU offload incompatibility; different problems." + } + ] + }, + "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": 28, + "review_unit_index": 32, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "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:38175", + "issue:39692", + "issue:42490", + "issue:43278", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43701", + "issue:43720", + "issue:43819", + "issue:43825", + "issue:44112", + "issue:44206", + "issue:44360", + "issue:44448", + "issue:44493", + "issue:44514", + "issue:44521", + "issue:44561", + "issue:44779", + "issue:44857", + "issue:45072", + "issue:45083", + "issue:45127", + "issue:45245", + "issue:45290", + "issue:45310", + "issue:45459", + "issue:45478", + "issue:45593", + "issue:45823", + "issue:46234", + "issue:46326", + "issue:47000" + ], + "soft_pairs": [ + "issue:45290|issue:46326", + "issue:43819|issue:44493", + "issue:39692|issue:43526", + "issue:43606|issue:44857", + "issue:29127|issue:32090", + "issue:44514|issue:44521", + "issue:44206|issue:45823", + "issue:43278|issue:43825", + "issue:39692|issue:44493", + "issue:28218|issue:32090", + "issue:43720|issue:44112", + "issue:10000|issue:15354", + "issue:43701|issue:45083", + "issue:45310|issue:45478", + "issue:44448|issue:44779", + "issue:26413|issue:31501", + "issue:45127|issue:45245", + "issue:44857|issue:45072", + "issue:32090|issue:33357", + "issue:43550|issue:45083", + "issue:42490|issue:43575", + "issue:38175|issue:44493", + "issue:44360|issue:44561", + "issue:45459|issue:46234", + "issue:45593|issue:47000", + "issue:27957|issue:34689", + "issue:29942|issue:35141", + "issue:29870|issue:32090" + ], + "prepared_review_unit_hash": "2f0a53fc8bccf759cf65ec3c21ab931b7d57e3807a54decfc2e99db1123070b4", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36086, + "estimated_input_tokens": 9022, + "estimated_eval_tokens": 18300 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a collection of unrelated issue reports that only share broad subsystem vocabulary. I found one clear duplicate pair about the Qwen3.5 MoE from_pretrained regression; the rest should stay separate.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45310 is the best canonical issue because issue:45478 repeats the same Qwen3.5 MoE from_pretrained failure with the same version qualifier, and 45310 appears to be the earlier, fuller report.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the strongest representative for the only confirmed duplicate pair in this cluster, with a clear, specific title and earliest occurrence among the matching reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45290", + "right": "issue:46326", + "accept": false, + "reason": "Both involve chat-template/tokenizer behavior, but one is a crash with tool-call assistant messages and the other is a continue_final_message bug for mistral_common tokenizers; not the same underlying defect." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "Different problems: DAC latent reconstruction vs widespread unexpected position_id keys during model loading." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Docs/example errors for SigLIP2 are unrelated to BeitImageProcessorFast reduce_labels returning the wrong label shape." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "Both are runtime failures, but one is CPU offload device mismatch and the other is float16 AMP loss crash; different code paths and symptoms." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "Clarifying a LayoutLMv3 error message is unrelated to a Trainer broadcast TypeError." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both touch apply_chat_template, but one is a batched-padding crash and the other is incorrect assistant_masks for multimodal inputs; too different to merge." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Feature-extractor center-arg regression and Gemma4 PLE device mismatch are distinct failures." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype drift in evaluate is unrelated to a pipeline error-message regression." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 docs/model mismatch and unexpected position-id keys are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing bug and Trainer broadcast TypeError are different subsystems and failures." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "Packed-weight loading in BitNet and a stale GraniteSpeech CI test are not the same bug." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBertModel issue and torch.jit.script GeneratorExp incompatibility are unrelated." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to qwen3_omni_moe feature-extraction length behavior." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same concrete bug: transformers>=5.4.0 causing Qwen3.5 MoE from_pretrained errors; the titles match closely and 45478 is a later repeat." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Different v4-v5 behavior reports, but one is Pegasus generation output and the other is Deepseek tokenizer correctness." + }, + { + "left": "issue:26413", + "right": "issue:31501", + "accept": false, + "reason": "Resume-from-checkpoint epoch_iterator failure and FineWeb training not starting are too unspecific and not clearly the same underlying issue." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/extended-vocab collapse is unrelated to a category-cardinality runtime error." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Different dtype-related crashes in different models/code paths; not plausibly one merged fix." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer NoneType broadcast error and MacOS CLIP bus error are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba-9B-v2 is unrelated to qwen3_omni_moe helper behavior." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Local-path import warning and Qwen2-57B-A14B-Instruct OOM are unrelated load-time issues." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 is a model-specific inference bug, not a position-id key regression." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "A discussion about a missing ReLU in the DSA indexer is unrelated to removal of is_torch_fx_available breaking trust_remote_code models." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf error handling and missing import_utils helper are unrelated." + }, + { + "left": "issue:45593", + "right": "issue:47000", + "accept": false, + "reason": "Different object-detection loss issues with different failure modes; not the same concrete bug." + }, + { + "left": "issue:27957", + "right": "issue:34689", + "accept": false, + "reason": "Flash Attention 2 support for XLMRoberta and a Llama 3.2 Vision model-loading regression are unrelated." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention 2 CI failures and output embedding reinitialization on resize are unrelated." + }, + { + "left": "issue:29870", + "right": "issue:32090", + "accept": false, + "reason": "BART decoder_attention_mask initialization bug and Trainer broadcast NoneType error are different code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analyst output is grounded in the packet and stays conservative on the soft-edge verdicts. The Qwen3.5 MoE pair (issue:45310 / issue:45478) is a clear duplicate match with nearly identical titles and matching failure mode, and the other pairs are not overstated as duplicates." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 33, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29870", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:37428", + "issue:39692", + "issue:40444", + "issue:41720", + "issue:42175", + "issue:42490", + "issue:43257", + "issue:43278", + "issue:43299", + "issue:43329", + "issue:43381", + "issue:43526", + "issue:43575", + "issue:43701", + "issue:43873", + "issue:43927", + "issue:44186", + "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:43278|issue:43329", + "issue:37428|issue:42175", + "issue:44464|issue:46424", + "issue:45290|issue:45381", + "issue:43257|issue:43299", + "issue:29870|issue:33290", + "issue:44855|issue:46399", + "issue:44448|issue:44977", + "issue:43278|issue:43381", + "issue:40444|issue:43526", + "issue:32090|issue:33290", + "issue:25251|issue:32090", + "issue:43927|issue:45698", + "issue:45920|issue:46192", + "issue:44977|issue:45229", + "issue:45290|issue:46655", + "issue:44530|issue:47276", + "issue:44561|issue:44779", + "issue:42490|issue:43873", + "issue:43873|issue:45229", + "issue:43381|issue:43701", + "issue:29127|issue:34567", + "issue:46421|issue:46858", + "issue:44448|issue:44466", + "issue:44871|issue:44977", + "issue:33357|issue:39692", + "issue:44186|issue:45198", + "issue:45588|issue:45823", + "issue:43278|issue:43575", + "issue:41720|issue:45229" + ], + "prepared_review_unit_hash": "2ae52087baa20118412a36bd9dcf48bcd04ccc4f7e80a08e5b1921ea2f38c7ce", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37062, + "estimated_input_tokens": 9266, + "estimated_eval_tokens": 18788 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated Transformers issue reports that only look superficially similar by subsystem (tokenizers, cache, generation, trainer, vision-language, quantization). I do not see any true duplicate pairs among the soft candidates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45920 is the broadest, highest-signal regression report in the set, with multiple affected model families and follow-up references, so it is the most representative issue if one must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 has the widest blast radius and strongest evidence of downstream impact; it is the best single issue to anchor the cluster, even though the cluster is not actually duplicate-heavy.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Different failures: training/eval dtype drift for embeddings vs an undefined helper in multimodal token counting." + }, + { + "left": "issue:37428", + "right": "issue:42175", + "accept": false, + "reason": "Completely unrelated: flash-attention import error vs missing TensorFlow in a torch-only install." + }, + { + "left": "issue:44464", + "right": "issue:46424", + "accept": false, + "reason": "One is inconsistent generation under compiled forward; the other is cache length config being ignored for non-quantized caches." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Different code paths: chat template tool-call handling vs video vision_position_ids in Qwen2.5-VL." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MOE, but one is accelerate/deepspeed weight conversion and the other is a v5 model-loading breakage; not the same bug." + }, + { + "left": "issue:29870", + "right": "issue:33290", + "accept": false, + "reason": "UnboundLocalError in BART generation is unrelated to an OOM from Adafactor with DeepSpeed." + }, + { + "left": "issue:44855", + "right": "issue:46399", + "accept": false, + "reason": "Both mention registration/decorators, but the causes differ: Python 3.13 JIT parsing vs missing decorator on rotary embedding registration." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Pegasus v4/v5 output drift is a serialization/behavior regression; Qwen3.5 flash-attention failure is a separate generation bug." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Embedding dtype changes in eval are unrelated to gradient checkpointing being disallowed in eval mode." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "Different components and symptoms: multi-image IterableDataset finetuning failure vs BeitImageProcessorFast label reduction bug." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "GPU broadcast NoneType error is not the same underlying problem as optimizer memory blow-up/OOM." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting change and Trainer broadcast failure are different bugs in different execution paths." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both are save/load-related, but one loses custom token IDs in config while the other reloads the wrong custom module." + }, + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "Wrong token IDs from AutoTokenizer is a real tokenizer regression; silently ignored return_tensors is an API-argument precedence issue." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Flash-attention generation failure and multi-GPU OOM are unrelated symptoms with different root causes." + }, + { + "left": "issue:45290", + "right": "issue:46655", + "accept": false, + "reason": "Chat-template tool-call crash and video padding miscalculation are different multimodal preprocessing bugs." + }, + { + "left": "issue:44530", + "right": "issue:47276", + "accept": false, + "reason": "Cache group-type crash on Qwen3.5 is unrelated to a misleading synced_gpus warning message." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Trust_remote_code breakage from a removed utility is not the same as DeepSeek tokenizer output regression." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Local-path import warning and quantization offloading behavior are separate loader/offload issues." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Quantization offloading not working is not the same as Gemma4 multi-GPU inference OOM." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Eval-mode gradient checkpointing restriction and resume_from_checkpoint key mismatch are different trainer concerns." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 box-content error messaging is unrelated to TrainerState token accounting not updating." + }, + { + "left": "issue:46421", + "right": "issue:46858", + "accept": false, + "reason": "Both touch cache/compile, but one is spurious recompilation from lazy cache init and the other is device-tensor seq-length arithmetic breakage." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Pegasus output differences across versions and lm_head serialization differing by device are distinct regressions." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma eos_token_id config inconsistency is unrelated to Qwen3.5 flash-attention generation failures." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "MacOS bus error on a community CLIP model and SigLIP2 docs/example mistakes are not the same underlying problem." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2 NER/padding crashes and Wav2Vec2 save_pretrained/tokenization failures affect different tokenizers and code paths." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Flash-attention crashing on s_aux=None and Gemma4 PLE device mismatch are separate runtime bugs." + }, + { + "left": "issue:43278", + "right": "issue:43575", + "accept": false, + "reason": "Training/eval dtype drift is unrelated to tensor-parallel OOM when loading Qwen2-57B-A14B-Instruct." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Gemma4 multi-GPU OOM are different multi-GPU failures with different roots." + } + ] + }, + "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": 28, + "review_unit_index": 34, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:29127", + "issue:29870", + "issue:29942", + "issue:30064", + "issue:33357", + "issue:34689", + "issue:35141", + "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:44361", + "issue:44387", + "issue:44442", + "issue:44479", + "issue:44568", + "issue:44743", + "issue:44977", + "issue:45198", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45712", + "issue:45803", + "issue:46121", + "issue:46693", + "issue:47217", + "issue:47276" + ], + "soft_pairs": [ + "issue:43976|issue:45440", + "issue:44361|issue:45198", + "issue:29942|issue:30064", + "issue:43611|issue:46121", + "issue:45198|issue:45803", + "issue:44206|issue:44479", + "issue:43412|issue:45588", + "issue:29870|issue:34689", + "issue:44568|issue:45491", + "issue:44077|issue:47276", + "issue:43653|issue:44743", + "issue:43756|issue:45440", + "issue:42915|issue:42994", + "issue:44977|issue:45563", + "issue:43653|issue:44442", + "issue:44265|issue:45198", + "issue:43065|issue:43994", + "issue:43575|issue:43701", + "issue:29127|issue:33357", + "issue:42490|issue:43381", + "issue:45491|issue:46693", + "issue:43065|issue:45412", + "issue:29127|issue:35141", + "issue:43575|issue:44387", + "issue:43994|issue:45412", + "issue:45372|issue:45588", + "issue:43526|issue:47217", + "issue:45440|issue:45593", + "issue:43334|issue:43531", + "issue:45459|issue:45712" + ], + "prepared_review_unit_hash": "ff5ebf186777ee67b8353f4740de7fbf1a28953ba8221dd8fc7c3b0ed3009357", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36832, + "estimated_input_tokens": 9208, + "estimated_eval_tokens": 18672 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are a grab-bag of unrelated Transformer bugs across tokenizers, vision, quantization, export, attention, and loading paths. None of the soft pairs look like the same underlying issue, so no pair should be merged.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44387 is the strongest standalone representative: a concrete, version-scoped regression with clear impact (CUDA reserved memory/OOM) and a well-defined code path.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44387 is the best issue to keep as the cluster representative because it is specific, actionable, and broadly impactful compared with the other unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python-version loading failure vs DeepSeek model implementation divergence; different models and code paths." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLuke tokenizer AttributeError on tasks vs Wav2Vec2 save/tokenization failure; unrelated subsystems." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 CI failures vs image processor void-segmentation handling; no shared bug." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "Base-model-prefix loading regression vs rope-validation type error; different functionality." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure vs Fuyu bare-except error masking; different root causes." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Unsupported center arg in a feature extractor vs Qwen video-input regression; unrelated." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export/erfinv issue vs flash_attention s_aux=None crash; different code paths." + }, + { + "left": "issue:29870", + "right": "issue:34689", + "accept": false, + "reason": "BART generate decoder mask bug vs Llama 3.2 Vision loading regression; unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer special-token addition bug vs Gemma3 NaN embeddings; different failure modes." + }, + { + "left": "issue:44077", + "right": "issue:47276", + "accept": false, + "reason": "PatchTSMixer post_init behavior vs a misleading warning message; not the same bug." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "BigBird tokenizer decode/special-token issue vs Qwen recurrent-state reset bug; unrelated." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "SmolLM3 RoPE-layer mismatch vs DeepSeekV3MoE divergence; different models and fixes." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Qwen3MoE FP8 config failure vs quantized-model saving failure; both quantization-related but not the same defect." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug vs stale paged-generate warning; unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token bug vs AutoTokenizer loading failure for FastSpeech2ConformerTokenizer; different problems." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure on torch_compilable_check vs Wav2Vec2 save/tokenization issue; unrelated." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder vs Siglip2 nonsense outputs; different models and symptoms." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Tensor-parallel load OOM vs resume_from_checkpoint key mismatch; unrelated loading/training bugs." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 box-info messaging issue vs MacOS bus error loading a CLIP model; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Trailing-slash import warning vs gradient-checkpointing eval-mode restriction; different behaviors." + }, + { + "left": "issue:45491", + "right": "issue:46693", + "accept": false, + "reason": "Gemma3 mixed-length NaNs vs long-context FA2 performance regression; both attention-related but not the same concrete bug." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Dummy decoder layer cleanup vs RT-DETR memory-release bug; unrelated." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "LayoutLMv3 error-message clarity vs token-embedding reinitialization on resize; different code paths." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "TP load OOM for a specific model vs int4 quantization reserved-memory regression; different causes and fixes." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "Siglip2 bad outputs vs RT-DETR memory leak; unrelated." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 processor import error from mistral_common vs flash_attention crash on s_aux=None; unrelated." + }, + { + "left": "issue:43526", + "right": "issue:47217", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug vs VL image-placeholder handling failure; different components." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer protobuf exception handling vs dummy-class cleanup/check_repo issue; unrelated." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Qwen3-VL pad_token_id AttributeError vs Qwen3-MoE sliding_window bug; different model/config defects." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer protobuf exception handling vs dummy-class cleanup/check_repo issue; 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": 28, + "review_unit_index": 35, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "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:43646", + "issue:43761", + "issue:43824", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44077", + "issue:44206", + "issue:44279", + "issue:44442", + "issue:44462", + "issue:44514", + "issue:44625", + "issue:44857", + "issue:45072", + "issue:45245", + "issue:45250", + "issue:45372", + "issue:45381", + "issue:45406", + "issue:45412", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45834", + "issue:45859", + "issue:46121", + "issue:46369", + "issue:46399", + "issue:46640", + "issue:46693", + "issue:46701", + "issue:46821", + "issue:47312" + ], + "soft_pairs": [ + "issue:45406|issue:46701", + "issue:43421|issue:45491", + "issue:44442|issue:45072", + "issue:44625|issue:45584", + "issue:45250|issue:46640", + "issue:45372|issue:45406", + "issue:45412|issue:46369", + "issue:45245|issue:45563", + "issue:43526|issue:45834", + "issue:43066|issue:44462", + "issue:45372|issue:45459", + "issue:27957|issue:33290", + "issue:27957|issue:29127", + "issue:33290|issue:34689", + "issue:33290|issue:35141", + "issue:43526|issue:46693", + "issue:43824|issue:44279", + "issue:43994|issue:46369", + "issue:43381|issue:43873", + "issue:42915|issue:43931", + "issue:44857|issue:45588", + "issue:45588|issue:46821", + "issue:46121|issue:46399", + "issue:43316|issue:47312", + "issue:43646|issue:44077", + "issue:44514|issue:45381", + "issue:43761|issue:45859", + "issue:43262|issue:44206" + ], + "prepared_review_unit_hash": "f84be0d7e18c05671abb459d826a9b6ecddd9c16715974499a7ed02227cee295", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35913, + "estimated_input_tokens": 8979, + "estimated_eval_tokens": 18214 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a noisy, mostly heterogeneous issue cluster. The soft pairs are only superficially related, so I would not merge any of them. The strongest standalone issue in the set is the FA2 regression report with the clearest repro and discussion.", + "confidence": 0.31, + "canonical_issue_reason": "Issue 46693 is the most concrete and actionable report in the set: it describes a specific Flash Attention 2 long-context regression after 5.2.0, with substantial discussion and clear symptoms.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46693 is the best issue to keep as the representative one because it is well-scoped, technically specific, and has the strongest evidence of a real underlying bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45406", + "right": "issue:46701", + "accept": false, + "reason": "Both are Gemma-related, but one is a processor attribute crash and the other is a pipeline/output issue; different code paths and root causes." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "One is a tokenizer post-processor update bug, the other is a GPU NaN embedding issue; unrelated failures." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "AutoTokenizer loading a tokenizer class and dtype mismatches in inference are unrelated problems." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Different models and different bugs: num_labels propagation vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:45250", + "right": "issue:46640", + "accept": false, + "reason": "Both mention Flash Attention 2, but they are too broad and do not clearly describe the same concrete bug." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Both touch Gemma 4 loading, but one is an import dependency issue and the other is a missing _tokenizer attribute crash." + }, + { + "left": "issue:45412", + "right": "issue:46369", + "accept": false, + "reason": "Both are memory-related, but they affect different models and different failure modes." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-count validation and a stale generation warning are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "Different vision processors and different symptoms; no shared code-path problem is evident." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer decoder type mismatch and ignoring tokenizer.json are distinct tokenizer-loading issues." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Both mention tokenizers, but one is a missing dependency import and the other is an exception-handling bug hiding real errors." + }, + { + "left": "issue:27957", + "right": "issue:33290", + "accept": false, + "reason": "XLM-RoBERTa Flash Attention support and Adafactor OOM in DeepSpeed are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:29127", + "accept": false, + "reason": "Flash Attention support and an error-message clarity issue in LayoutLMv3 are unrelated." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Optimizer OOM and model-loading breakage for a vision model are different bugs." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Adafactor memory usage and embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:46693", + "accept": false, + "reason": "Different models and different bug classes: label reduction vs Flash Attention performance regression." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "A missing export/import and a vague dependency issue do not look like the same underlying problem." + }, + { + "left": "issue:43994", + "right": "issue:46369", + "accept": false, + "reason": "Wrong model outputs and memory overflow are unrelated symptoms." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing in eval mode and quantization offloading behavior are different concerns." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Both involve Qwen models, but the FP8 config failure and weight-shape mismatch are different code paths." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "Float16 AMP loss crash and a Flash Attention sinkless-model AttributeError are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:46821", + "accept": false, + "reason": "Different subsystems: flash-attention integration vs TimesFM window_size handling." + }, + { + "left": "issue:46121", + "right": "issue:46399", + "accept": false, + "reason": "Rope-validation TypeError and a kernel registration decorator error are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:47312", + "accept": false, + "reason": "Configuration API discrepancy and kernel compatibility are different issues." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Custom model initialization breakage is broader; the patchtsmixer post_init change is a specific model-level constraint, not the same bug." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL, but one is batched chat-template padding and the other is video vision_position_ids; different failures." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "Hidden_states being None and CPU offload incompatibility are unrelated model behaviors." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and a feature extractor arg crash are different audio-preprocessing 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": 28, + "review_unit_index": 36, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:28282", + "issue:29870", + "issue:42175", + "issue:42490", + "issue:42617", + "issue:43316", + "issue:43493", + "issue:43525", + "issue:43550", + "issue:43575", + "issue:43643", + "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:44811", + "issue:44855", + "issue:44871", + "issue:44898", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:44987", + "issue:45137", + "issue:45216", + "issue:45245", + "issue:45325", + "issue:45440", + "issue:45759", + "issue:45907", + "issue:46178", + "issue:46234", + "issue:46506", + "issue:46697" + ], + "soft_pairs": [ + "issue:44442|issue:44898", + "issue:44743|issue:44811", + "issue:44977|issue:45245", + "issue:45325|issue:46697", + "issue:43720|issue:44442", + "issue:42490|issue:44743", + "issue:42175|issue:46234", + "issue:44077|issue:44479", + "issue:27957|issue:29870", + "issue:43873|issue:44871", + "issue:43493|issue:43827", + "issue:43643|issue:44913", + "issue:43316|issue:45759", + "issue:44964|issue:45440", + "issue:44062|issue:45907", + "issue:43316|issue:43525", + "issue:25251|issue:28282", + "issue:43749|issue:43931", + "issue:44987|issue:46178", + "issue:44855|issue:45137", + "issue:42617|issue:43575", + "issue:43550|issue:43701", + "issue:43931|issue:45216", + "issue:44190|issue:44496", + "issue:43701|issue:46506", + "issue:43575|issue:43742", + "issue:43761|issue:44079" + ], + "prepared_review_unit_hash": "ce4c1eac178814d55974bd777e98a99991b512847b015651618002d4d5b6b658", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35867, + "estimated_input_tokens": 8967, + "estimated_eval_tokens": 18190 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issue reports spanning different models, APIs, and regressions. None of the soft pairs look like the same underlying bug or change, so all duplicate-style links should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Different components and failure modes: tokenizer loading vs image classification interpolation; no shared code path." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen recurrent-state cache bug vs Whisper processor decode option handling; unrelated models and APIs." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention generation bug for Qwen3.5 vs categorical cardinality runtime limit; clearly different issues." + }, + { + "left": "issue:45325", + "right": "issue:46697", + "accept": false, + "reason": "Qwen2.5-VL rope position scaling bug vs Swin2SR over-padding in image processor; unrelated vision preprocessing paths." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "BitNet packed-weight loading failure is unrelated to FastSpeech2ConformerTokenizer loading." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local path trailing-slash warning vs Qwen recurrent-state reset; different subsystems and symptoms." + }, + { + "left": "issue:42175", + "right": "issue:46234", + "accept": false, + "reason": "Packaging/backend dependency issue vs missing import_utils helper; not the same bug." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "PatchTSMixer post_init validation has nothing to do with Qwen video-input regression." + }, + { + "left": "issue:27957", + "right": "issue:29870", + "accept": false, + "reason": "Flash Attention 2 support for XLMRoberta is unrelated to BART decoder_attention_mask generate error." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior is not the same as Gemma eos_token_id configuration mismatch." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 HF/JAX numerical discrepancy is unrelated to docs still referencing removed pipeline()." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "AutoConfig trust_remote_code missing fields vs GPTNeoX rotary_pct reload persistence; different config bugs." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Gemma3TextConfig API discrepancy vs AutoModelForCausalLM text_config unwrapping for Qwen multimodal configs; related theme but not same bug." + }, + { + "left": "issue:44964", + "right": "issue:45440", + "accept": false, + "reason": "Phi-4 multimodal loading failure and DeepseekV3MoE divergence are distinct model-specific issues." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken special-arg duplication is unrelated to list_repo_files revision forwarding in CI." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3TextConfig API mismatch is not the same as Llama4Config missing pad_token_id." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k output nesting bug is unrelated to AutoModel import requiring PyTorch." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "FSDP CPU RAM efficient loading regression is unrelated to Qwen3-VL weight-shape mismatch." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "Loading a specific repo with transformers>=5.1.0 is unrelated to tensorflow_text backend mapping." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Torch.jit indentation parsing issue is unrelated to DeepSpeed ZeRO3 deque pop error." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "3d_parallel.py execution failure and Qwen2-57B-A14B-Instruct tp OOM are different problems." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "Bamba torch.compile/SDPA bug is unrelated to resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Model weight-shape mismatch when loading differs from Qwen3.5 save_pretrained checkpoint corruption." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Local dataset loading in image classification script is unrelated to unrecognized model_type in config.json." + }, + { + "left": "issue:43701", + "right": "issue:46506", + "accept": false, + "reason": "Checkpoint key mismatch is unrelated to CPU/CUDA output differences in PPDocLayoutV3." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "TP OOM on Qwen2-57B-A14B-Instruct is not the same as MobileLLM-125M key error." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment bug affect different layers 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": 28, + "review_unit_index": 37, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:27957", + "issue:32090", + "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:43606", + "issue:43646", + "issue:43701", + "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:46234", + "issue:46291", + "issue:46365", + "issue:46421", + "issue:46640", + "issue:46829", + "issue:46932", + "issue:47217", + "issue:47276" + ], + "soft_pairs": [ + "issue:43606|issue:43701", + "issue:44442|issue:45803", + "issue:27957|issue:32090", + "issue:42490|issue:47276", + "issue:43232|issue:43388", + "issue:46829|issue:46932", + "issue:43761|issue:44188", + "issue:43295|issue:44871", + "issue:43316|issue:43716", + "issue:43295|issue:43825", + "issue:44871|issue:45584", + "issue:42907|issue:45698", + "issue:16998|issue:22355", + "issue:44521|issue:47217", + "issue:44077|issue:44534", + "issue:43646|issue:44913", + "issue:46234|issue:46291", + "issue:43452|issue:44496", + "issue:44871|issue:46050", + "issue:44464|issue:46421", + "issue:43404|issue:43441", + "issue:43352|issue:46640", + "issue:44162|issue:44792", + "issue:45803|issue:46365", + "issue:43065|issue:44617", + "issue:43540|issue:44514" + ], + "prepared_review_unit_hash": "9c7284c371883b5f015ffc298c4935c41946f2f7362121863568d7d264b6425a", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35322, + "estimated_input_tokens": 8831, + "estimated_eval_tokens": 17918 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the soft pairs mostly share broad Transformers keywords (multimodal, generation, save/load, FlashAttention) but point to different concrete failures and code paths. I would not merge any of the proposed pairs as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #44521 is the strongest representative only by cluster prominence: it has the most inbound references and broad multimodal chat-templating scope. That said, the overall cluster does not form a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "#44521 is the most active and broadly referenced issue in the set, so it is the best single issue to anchor triage if one must be chosen.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "Different failures: CPU offload device mismatch in Bark vs checkpoint key mismatch during resume_from_checkpoint." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading failure for FastSpeech2Conformer is unrelated to Fuyu's bare-except error handling bug." + }, + { + "left": "issue:27957", + "right": "issue:32090", + "accept": false, + "reason": "FlashAttention2 support for XLM-Roberta is unrelated to Trainer's NoneType broadcast error." + }, + { + "left": "issue:42490", + "right": "issue:47276", + "accept": false, + "reason": "Both mention warnings, but one is about local-path import warnings and the other about synced_gpus messaging in continuous batching." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs syncing and metric gathering are different code paths with different symptoms." + }, + { + "left": "issue:46829", + "right": "issue:46932", + "accept": false, + "reason": "VideoPrism hidden-state output bug is distinct from Gemma4 quantized embedder dtype casting." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression is unrelated to torch.compile attention-kernel divergence." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Custom processor/tokenizer access regression is different from Gemma-3 eos_token_id config inconsistency." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "Gemma3TextConfig API mismatch and Mistral-3 image-preprocessor dtype mismatch are separate issues." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Processor/tokenizer regression is not the same bug as an inaccurate pipeline error message." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Gemma-3 eos_token config mismatch is unrelated to Whisper empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Model save/load problems, but one is dequantized Ministral/Devstral saving and the other is wrong custom-module loading after save_pretrained." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "DeBERTa model_max_length question and missing transformers.onnx module are unrelated." + }, + { + "left": "issue:44521", + "right": "issue:47217", + "accept": false, + "reason": "Both are multimodal chat-template related, but one is zero assistant masks and the other is failure when the image placeholder appears in user text; not the same concrete bug." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "PatchTSMixer post_init restriction and non-persistent buffer junk are different v5 issues." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Custom model initialization breakage is unrelated to GPTNeoX rotary_pct not persisting on reload." + }, + { + "left": "issue:46234", + "right": "issue:46291", + "accept": false, + "reason": "Missing import_utils helper and kernels revision/version import failure are unrelated dependency/import bugs." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "gguf_file loading regression is not the same as an unrecognized model due to missing model_type in config." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Gemma-3 eos_token configuration mismatch is distinct from sinq quantized model save/load breakage." + }, + { + "left": "issue:44464", + "right": "issue:46421", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency is different from cache lazy init causing spurious recompilation." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Weight tying bug in Mistral3ForConditionalGeneration is unrelated to FlashAttention support failure in Ministral-3." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "A specific unsupported-model FlashAttention limitation is not enough to treat this as the same bug as a vague generic Flash Attention 2 issue." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 workflow breakage and janus image-generation test failure are unrelated model-specific problems." + }, + { + "left": "issue:45803", + "right": "issue:46365", + "accept": false, + "reason": "Bare-except error swallowing in Fuyu preprocessing is unrelated to Trainer import failure caused by datasets." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and Sam3Video CUDA OOM are different problems; one is architecture cleanup, the other is resource usage." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Qwen3OmniMoe video-input value error and Qwen2_5_VLProcessor batched chat-template crash are different models 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": 28, + "review_unit_index": 38, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:35141", + "issue:41950", + "issue:42490", + "issue:43388", + "issue:43475", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43572", + "issue:43650", + "issue:43673", + "issue:43688", + "issue:43931", + "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:46178", + "issue:46202", + "issue:46234", + "issue:46369", + "issue:46424", + "issue:46693", + "issue:46697", + "issue:46894", + "issue:47217" + ], + "soft_pairs": [ + "issue:43572|issue:46178", + "issue:44610|issue:46369", + "issue:45381|issue:47217", + "issue:44242|issue:46894", + "issue:43504|issue:45698", + "issue:45542|issue:46234", + "issue:45632|issue:46424", + "issue:44792|issue:44977", + "issue:44534|issue:44913", + "issue:44792|issue:45600", + "issue:44479|issue:47217", + "issue:43525|issue:43531", + "issue:25251|issue:27957", + "issue:45803|issue:46170", + "issue:46082|issue:46697", + "issue:43931|issue:44060", + "issue:45127|issue:45200", + "issue:35141|issue:42490", + "issue:46693|issue:47217", + "issue:43650|issue:46202", + "issue:41950|issue:43475", + "issue:43673|issue:44743", + "issue:43388|issue:43688", + "issue:44855|issue:45588", + "issue:44190|issue:44964", + "issue:44792|issue:45405" + ], + "prepared_review_unit_hash": "27519010fae07f64358665c9a5ab63e5a7c8f6cf933ab0a3960557553a5952c0", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35586, + "estimated_input_tokens": 8897, + "estimated_eval_tokens": 18050 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the items cover many unrelated bugs across configs, loading, pipelines, caching, and model-specific regressions. None of the soft pairs look like true duplicates or the same fixable code-path change.", + "confidence": 0.96, + "canonical_issue_reason": "issue:46693 is the most substantiated and broadly impactful report in the set, with the most discussion and multiple inbound references; it is the strongest representative if one issue must anchor the cluster, though the set is not a true duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46693 is the best standalone issue candidate because it has the richest report quality and clear regression scope (FA2 long-context performance after 5.2.0).", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "Different failures: missing config field in StableLmConfig vs an import_utils backend-mapping error for tensorflow_text." + }, + { + "left": "issue:44610", + "right": "issue:46369", + "accept": false, + "reason": "Different model paths and symptoms: OmDet-Turbo processor size mismatch vs SAM 3 LiteText memory overflow." + }, + { + "left": "issue:45381", + "right": "issue:47217", + "accept": false, + "reason": "Both are VL-related, but one is wrong vision_position_ids for video input while the other is image placeholder handling in user text." + }, + { + "left": "issue:44242", + "right": "issue:46894", + "accept": false, + "reason": "Both mention loss, but the concrete bug is different: missing load-balancing loss vs a training-loss double-shift issue in CohereASR/Moonshine pattern." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both involve loading, but one is a legacy field in BEiT preset loading and the other is wrong custom module resolution after save_pretrained/from_pretrained." + }, + { + "left": "issue:45542", + "right": "issue:46234", + "accept": false, + "reason": "Different import problems: tf backend installation handling vs a missing falcon-mamba availability helper in import_utils." + }, + { + "left": "issue:45632", + "right": "issue:46424", + "accept": false, + "reason": "Different cache subsystems: trust_remote_code cache-path collision vs cache max_cache_length semantics for quantized cache." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Different concrete bugs: a Janus image-generation test failure vs Qwen3.5 flash-attention generation breakage." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Unrelated issues: non-persistent buffer initialization junk vs GPTNeoXConfig rotary_pct not persisting on reload." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "Different targets: Janus generate-images test failure vs stale auto_mappings references to removed Sam3LiteText configs." + }, + { + "left": "issue:44479", + "right": "issue:47217", + "accept": false, + "reason": "Both concern VL models, but one is a v5.3.0 video-input regression and the other is image placeholder placement in user text." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Different model-specific bugs: missing pad_token_id on Llama4Config vs sliding_window handling in Qwen3-MoE." + }, + { + "left": "issue:25251", + "right": "issue:27957", + "accept": false, + "reason": "Completely different areas: pipeline top_k output nesting vs XLM-R with Flash Attention 2." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Different classes of bugs: bare except swallowing interrupts/errors vs Windows ROCm import/type dependency failure." + }, + { + "left": "issue:46082", + "right": "issue:46697", + "accept": false, + "reason": "Different code paths: LlamaConfig validation for head_dim vs Swin2SR image over-padding in the processor." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Both are model-loading/weights issues, but one is a shape mismatch on Qwen3-VL while the other is a tied-weights warning bug in Qwen3-Next." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "Different bugs: LoRA merge with extended vocab causing collapse vs Gemma 4 mm_token_type_ids defaulting for text-only fine-tuning." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Different subjects entirely: embedding reinitialization after resizing vs trailing-slash import warning for local model paths." + }, + { + "left": "issue:46693", + "right": "issue:47217", + "accept": false, + "reason": "Different regressions: FA2 long-context performance degradation vs VL image-placeholder failure." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "No shared bug pattern: a vague 'ADD THE DATA' request vs Gemma token roundtrip failure on \u2581." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Different bugs in different subsystems: video-classification pipeline processor selection vs missing fpn_position_embeddings on Sam3VisionEncoderOutput." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both involve cache/state, but the concrete failures differ: missing GenerationMixin cache in chunked prefill vs recurrent states resetting for seq_len>1." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Different training/eval bugs: gather_for_metrics dropping tuple labels vs auxiliary loss normalization in OLMoE/GPT Oss." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Completely different failure modes: Python 3.13 parsing/IndentationError on import vs AttributeError in flash_attention for sink-less models." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Different surfaces: local dataset loading in a training script vs loading a multimodal model with latest transformers." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "Unrelated: Janus image-generation test failure vs dependency version bump for peft." + } + ] + }, + "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": 28, + "review_unit_index": 39, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:25251", + "issue:29870", + "issue:35141", + "issue:39692", + "issue:41720", + "issue:43065", + "issue:43295", + "issue:43335", + "issue:43526", + "issue:43550", + "issue:43644", + "issue:43827", + "issue:43854", + "issue:43994", + "issue:44190", + "issue:44242", + "issue:44479", + "issue:44610", + "issue:44617", + "issue:44734", + "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:46617", + "issue:46655", + "issue:46697", + "issue:46832", + "issue:46858", + "issue:46897", + "issue:46932", + "issue:47013", + "issue:47312", + "issue:47321" + ], + "soft_pairs": [ + "issue:29870|issue:35141", + "issue:45072|issue:46932", + "issue:43827|issue:43994", + "issue:45245|issue:46858", + "issue:44734|issue:45406", + "issue:43335|issue:44855", + "issue:43644|issue:44913", + "issue:46399|issue:47312", + "issue:44869|issue:46832", + "issue:44617|issue:45584", + "issue:45588|issue:47013", + "issue:45632|issue:46617", + "issue:41720|issue:45245", + "issue:43827|issue:47321", + "issue:45381|issue:46369", + "issue:43550|issue:44190", + "issue:39692|issue:47321", + "issue:25251|issue:29870", + "issue:45563|issue:45584", + "issue:44242|issue:46897", + "issue:43854|issue:44792", + "issue:43065|issue:44610", + "issue:44933|issue:45362", + "issue:43335|issue:44190", + "issue:44479|issue:44734", + "issue:45381|issue:46655", + "issue:39692|issue:44190", + "issue:44617|issue:46369", + "issue:43295|issue:45276", + "issue:43526|issue:46697" + ], + "prepared_review_unit_hash": "33e9619f0be2850dafb51aca8e7ab38f19961bdf77ba9dca8397dd14901b40b0", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36321, + "estimated_input_tokens": 9081, + "estimated_eval_tokens": 18418 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All proposed soft links look like false positives: these issues span unrelated bugs across generation, processors, configs, docs, serving, and model-specific regressions.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits this cluster; the soft pairs do not describe the same underlying bug, and the set is too heterogeneous to collapse around one issue.", + "canonical_pr_reason": null, + "best_issue_reason": "If forced to pick a clean standalone report, issue 43827 is the most self-contained, but it is not a good cluster representative because the cluster is not a duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29870", + "right": "issue:35141", + "accept": false, + "reason": "Different bugs: BART generate() decoder mask init vs token-embedding resize/post_init reinitialization." + }, + { + "left": "issue:45072", + "right": "issue:46932", + "accept": false, + "reason": "Both involve dtype/quantization, but they affect different models and code paths; not the same concrete bug." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "One is a docs reference to removed pipeline APIs; the other is a runtime SigLIP2 inference issue." + }, + { + "left": "issue:45245", + "right": "issue:46858", + "accept": false, + "reason": "Completely different failures: category-count limit vs static-cache seq-length tensor arithmetic." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "Both hit transformers serve, but the crashes come from different code paths and different root causes." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "Unrelated issues: SwitchTransformers sparse-layer construction vs a Python 3.13 IndentationError in DebertaV2Model." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "One is buffer initialization corruption in v5.0.0; the other is GPTNeoX config reload persistence." + }, + { + "left": "issue:46399", + "right": "issue:47312", + "accept": false, + "reason": "Kernel-related, but one is a missing decorator registration bug and the other is a broad incompatibility report." + }, + { + "left": "issue:44869", + "right": "issue:46832", + "accept": false, + "reason": "Whisper timestamp decoding crash and RT-DETR feature-level handling are unrelated." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Different models and failures: Sam3Video OOM vs Whisper empty-transcription generation bug." + }, + { + "left": "issue:45588", + "right": "issue:47013", + "accept": false, + "reason": "Different code paths: flash-attention auxiliary-state handling vs xLSTM unpacking behavior in train mode." + }, + { + "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 report." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Unrelated: Qwen3 auto device mapping cuda assert vs an embedding/category cardinality error." + }, + { + "left": "issue:43827", + "right": "issue:47321", + "accept": false, + "reason": "Docs pipeline-removal issue is unrelated to AMD quark class updates." + }, + { + "left": "issue:45381", + "right": "issue:46369", + "accept": false, + "reason": "Different SAM/Qwen video-path regression vs SAM3 litetext memory overflow." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "torch.compile SDPA failure and local dataset loading script error are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:47321", + "accept": false, + "reason": "SigLIP2 docs/model mismatch is unrelated to AMD quark class update timing." + }, + { + "left": "issue:25251", + "right": "issue:29870", + "accept": false, + "reason": "Pipeline top_k nesting behavior is unrelated to BART decoder_attention_mask initialization." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Paged generate warning logic and Whisper empty-transcription failure are different bugs." + }, + { + "left": "issue:44242", + "right": "issue:46897", + "accept": false, + "reason": "Both concern training loss, but they are different model-specific loss bugs with different mechanics." + }, + { + "left": "issue:43854", + "right": "issue:44792", + "accept": false, + "reason": "Different failures: GLM-4.7-Flash test loading vs Janus image generation test failure." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Sam3PixelDecoder dummy Conv2d and OmDet-Turbo input-size mismatch are unrelated." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Missing import from image_utils is unrelated to Qwen3.5 chat crashes." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "SwitchTransformers config bug and image-classification dataset loading are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Both are regressions, but one is video-input handling and the other is serving KV-cache indexing." + }, + { + "left": "issue:45381", + "right": "issue:46655", + "accept": false, + "reason": "Both involve video processors, but the bugs are different: vision_position_ids vs padding calculation." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "SigLIP2 docs example problems do not match the local dataset loading failure." + }, + { + "left": "issue:44617", + "right": "issue:46369", + "accept": false, + "reason": "Both mention memory pressure, but they are different SAM3 models and different failure modes." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Processor.tokenizer regression is unrelated to Gemma4 embedding-resize propagation." + }, + { + "left": "issue:43526", + "right": "issue:46697", + "accept": false, + "reason": "Different image-processor bugs: label reduction vs over-padding on divisible sizes." + } + ] + }, + "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": 28, + "review_unit_index": 40, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:27957", + "issue:32090", + "issue:35141", + "issue:42175", + "issue:42796", + "issue:43317", + "issue:43404", + "issue:43531", + "issue:43540", + "issue:43643", + "issue:43644", + "issue:43673", + "issue:43854", + "issue:43994", + "issue:44112", + "issue:44246", + "issue:44279", + "issue:44462", + "issue:44496", + "issue:44521", + "issue:44734", + "issue:44737", + "issue:44779", + "issue:44938", + "issue:44964", + "issue:45003", + "issue:45127", + "issue:45216", + "issue:45276", + "issue:45290", + "issue:45357", + "issue:45381", + "issue:45542", + "issue:45584", + "issue:45600", + "issue:45632", + "issue:45803", + "issue:45907", + "issue:46234", + "issue:46421", + "issue:46489", + "issue:46752", + "issue:46858", + "issue:47145", + "issue:47217", + "issue:47312", + "issue:47321" + ], + "soft_pairs": [ + "issue:45127|issue:45357", + "issue:43673|issue:46421", + "issue:44779|issue:46489", + "issue:45276|issue:45584", + "issue:47145|issue:47321", + "issue:43540|issue:44734", + "issue:44112|issue:45600", + "issue:43854|issue:44964", + "issue:43643|issue:44496", + "issue:43317|issue:47312", + "issue:44462|issue:45907", + "issue:43994|issue:47321", + "issue:44521|issue:45381", + "issue:43644|issue:47321", + "issue:45290|issue:46752", + "issue:27957|issue:35141", + "issue:44246|issue:45003", + "issue:43673|issue:46858", + "issue:44737|issue:45803", + "issue:44112|issue:44964", + "issue:44964|issue:45600", + "issue:44938|issue:45542", + "issue:42175|issue:44279", + "issue:32090|issue:35141", + "issue:43404|issue:45216", + "issue:42796|issue:44521", + "issue:44496|issue:47217", + "issue:45632|issue:46234", + "issue:43531|issue:45600" + ], + "prepared_review_unit_hash": "863ec8cb3dfc192419b857b61f5c64db56155b05600c0d53caf96e396351f4b6", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35896, + "estimated_input_tokens": 8974, + "estimated_eval_tokens": 18204 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issues; only the DeepSeek tokenizer regression pair looks like the same underlying bug.", + "confidence": 0.84, + "canonical_issue_reason": "issue:46489 is the clearest representative of the duplicate DeepSeek tokenizer regression, and it explicitly ties the failure to the earlier PR gap.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46489 has the most specific repro framing and the strongest linkage to the prior fix, so it is the best issue anchor here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "Different regressions: LoRA/extended-vocab collapse vs incorrect visual-encoder keys on save_pretrained." + }, + { + "left": "issue:43673", + "right": "issue:46421", + "accept": false, + "reason": "Both involve cache/chunked prefill, but the concrete failures and code paths differ." + }, + { + "left": "issue:44779", + "right": "issue:46489", + "accept": true, + "reason": "Same DeepSeek tokenizer regression in v5+, with 46489 reading like a follow-up on the same bug/fix gap." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Unrelated: gemma4 resize_token_embeddings vs Whisper empty-transcription generation." + }, + { + "left": "issue:47145", + "right": "issue:47321", + "accept": false, + "reason": "Different areas entirely: FP8 checkpoint loading vs AMD quark class metadata." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Different symptoms and paths: video input parsing vs serve/KV-cache continuation indexing." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI-related, but they describe unrelated failures in different tests/code." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are model-loading failures, but for different models and different causes." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Different config-loading bugs: missing fields with trust_remote_code vs missing model_type." + }, + { + "left": "issue:43317", + "right": "issue:47312", + "accept": false, + "reason": "Different bugs: dequantized model offload loading vs kernel compatibility." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Different repo/file-resolution bugs: tokenizer.json selection vs list_repo_files revision forwarding." + }, + { + "left": "issue:43994", + "right": "issue:47321", + "accept": false, + "reason": "Unrelated: SigLIP2 inference correctness vs AMD quark class update." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both are multimodal, but assistant mask generation and vision_position_ids are different bugs." + }, + { + "left": "issue:43644", + "right": "issue:47321", + "accept": false, + "reason": "Unrelated: buffer initialization corruption vs AMD quark class update." + }, + { + "left": "issue:45290", + "right": "issue:46752", + "accept": false, + "reason": "Both crash in apply_chat_template, but on different inputs and different failure modes." + }, + { + "left": "issue:27957", + "right": "issue:35141", + "accept": false, + "reason": "Different model/features: Flash Attention 2 support vs embedding reinitialization after resize." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Different issues: import slowness vs unsafe sys.modules access." + }, + { + "left": "issue:43673", + "right": "issue:46858", + "accept": false, + "reason": "Both are cache-related, but one is missing cache state and the other is device-tensor slice arithmetic." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "Different bugs: XLNet device placement vs Fuyu exception handling." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "Unrelated CI test failure vs model-loading regression." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Different areas: Phi-4 multimodal loading vs removed config references in auto_mappings." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Different environment/backend issues: Python 3.14 import failure vs missing TensorFlow backend." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "Different dependency problems with different packaging symptoms." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer broadcast NoneType error is unrelated to embedding resize/reinit." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Different model-save/load bugs: lm_head tying vs incorrect checkpoint contents." + }, + { + "left": "issue:42796", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer BPE-marker handling is unrelated to multimodal assistant-mask generation." + }, + { + "left": "issue:44496", + "right": "issue:47217", + "accept": false, + "reason": "Different config/model-loading failures: missing model_type vs image-placeholder handling." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "Different trust_remote_code cache-path issue vs missing import_utils symbol." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Different bugs: sliding_window behavior in Qwen3-MoE vs stale CI config references." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft-edge pair, issue:44779 vs issue:46489, is well supported by both titles referring to the same DeepSeek tokenizer regression across v5+. The remaining false verdicts are cautious and do not overstate linkage. No mergeability concerns apply here because this is an issue-pair evaluation, not a PR-pair merge decision." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 41, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36246", + "issue:38617", + "issue:41628", + "issue:41950", + "issue:42907", + "issue:43295", + "issue:43299", + "issue:43335", + "issue:43381", + "issue:43441", + "issue:43479", + "issue:43526", + "issue:43582", + "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:44792", + "issue:44933", + "issue:44987", + "issue:45042", + "issue:45083", + "issue:45216", + "issue:45335", + "issue:45375", + "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" + ], + "soft_pairs": [ + "issue:43335|issue:45593", + "issue:36246|issue:41628", + "issue:36246|issue:38617", + "issue:44496|issue:45042", + "issue:43295|issue:45335", + "issue:45083|issue:46506", + "issue:44295|issue:46506", + "issue:41950|issue:44016", + "issue:43976|issue:44987", + "issue:43673|issue:46424", + "issue:43582|issue:46932", + "issue:44683|issue:45446", + "issue:44496|issue:45759", + "issue:43526|issue:43901", + "issue:46082|issue:46121", + "issue:43673|issue:44336", + "issue:43381|issue:45593", + "issue:44016|issue:44220", + "issue:42907|issue:43441", + "issue:44016|issue:44062", + "issue:43650|issue:44220", + "issue:44327|issue:46752", + "issue:44792|issue:46424", + "issue:43441|issue:45588", + "issue:44279|issue:46515", + "issue:43976|issue:44279", + "issue:43824|issue:44933", + "issue:45335|issue:45584", + "issue:43479|issue:45375", + "issue:43299|issue:45216", + "issue:43901|issue:44496", + "issue:43901|issue:47217" + ], + "prepared_review_unit_hash": "43c09fa5e7621de44b3e60f07f5af79478b189f4a91127abacc07cc9da96970e", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37308, + "estimated_input_tokens": 9327, + "estimated_eval_tokens": 18910 + }, + "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. The soft pairs are only superficially similar (imports, loading regressions, cache issues, multimodal configs), but they do not look like the same underlying code-path bug, so I rejected all candidate merges.", + "confidence": 0.22, + "canonical_issue_reason": "issue:43824 is the strongest standalone report in the set: it is a concrete import failure for a public class, has discussion and inbound references, and is actionable without needing broader context.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43824 is the best overall issue to anchor this cluster because it has the clearest repro signal and strongest external interest among the listed bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "Different models and failure modes: sparse-layer config creation vs missing auxiliary losses." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and likely different export regressions." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Unrelated missing names from different modules; no shared code path is evident." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Model-loading/config resolution vs image processor torchvision dependency; not the same bug." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Different subsystems: processor/tokenizer regression vs embedding resize behavior." + }, + { + "left": "issue:45083", + "right": "issue:46506", + "accept": false, + "reason": "Different bugs: qwen3_omni_moe feature-length helper vs CPU/CUDA divergence in PPDocLayoutV3." + }, + { + "left": "issue:44295", + "right": "issue:46506", + "accept": false, + "reason": "Position_ids buffer access issue is unrelated to cross-device output differences." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "Pipeline image-processor lookup bug and notebook syntax error are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "A Python-version compatibility problem is not the same as a model-loading failure for one repo." + }, + { + "left": "issue:43673", + "right": "issue:46424", + "accept": false, + "reason": "Both involve cache, but one is missing cache support in generation and the other is a max-length config limitation." + }, + { + "left": "issue:43582", + "right": "issue:46932", + "accept": false, + "reason": "Apple Silicon allocator TypeError and Gemma4 dtype casting are different runtime issues." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Compiled flex_attention failure on new torch versions is separate from a version-check/import bug." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Both are config/model-loading related, but one is an unrecognized model and the other is unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to a text-classification docs/behavior mismatch." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "LlamaConfig head_dim validation and rope-parameter JSON handling are different validation paths." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Generation cache behavior and ANSI output in loading_report do not share the same underlying problem." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing eval-mode restriction is unrelated to missing auxiliary losses." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error and fbank feature extraction are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Both mention Ministral, but saving dequantized models and FlashAttention loading failures are different code paths." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax and AddedToken keyword conflict are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "A placeholder issue title and an audio feature extraction bug are not the same change or bug." + }, + { + "left": "issue:44327", + "right": "issue:46752", + "accept": false, + "reason": "QA span decoding top-k bounds error and empty chat-template input error are different edge cases." + }, + { + "left": "issue:44792", + "right": "issue:46424", + "accept": false, + "reason": "Janus image-generation test failure is not the same as a cache-length configuration bug." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "FlashAttention failure for Ministral-3 and s_aux=None crash in flash_attention integration are distinct problems." + }, + { + "left": "issue:44279", + "right": "issue:46515", + "accept": false, + "reason": "General dependency issue and torch 2.4.1 incompatibility are not the same defect." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python support regression and dependency issue are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Different missing imports from different modules; both are import errors but not the same symbol or path." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize behavior and Whisper empty-transcription generation failure are different bugs." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Default multimodal config initialization and missing deepstack_visual_indexes handling are different config issues." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3VL loading regression and Qwen3.5 save_pretrained regression are opposite directions and different code paths." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Docs/behavior mismatch in a pipeline is unrelated to an unrecognized model config error." + }, + { + "left": "issue:43901", + "right": "issue:47217", + "accept": false, + "reason": "Text-classification docs mismatch and VL image-placeholder 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": 28, + "review_unit_index": 42, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:39692", + "issue:41950", + "issue:42796", + "issue:43295", + "issue:43299", + "issue:43381", + "issue:43577", + "issue:43644", + "issue:43650", + "issue:43653", + "issue:43854", + "issue:43901", + "issue:43931", + "issue:43976", + "issue:44016", + "issue:44075", + "issue:44242", + "issue:44279", + "issue:44291", + "issue:44351", + "issue:44368", + "issue:44462", + "issue:44466", + "issue:44683", + "issue:44779", + "issue:45020", + "issue:45081", + "issue:45200", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45357", + "issue:45405", + "issue:45542", + "issue:45588", + "issue:45593", + "issue:45812", + "issue:45823", + "issue:46082", + "issue:46123", + "issue:46209", + "issue:46369", + "issue:46399", + "issue:46512", + "issue:46614", + "issue:47000", + "issue:47085", + "issue:47145" + ], + "soft_pairs": [ + "issue:41950|issue:45020", + "issue:43295|issue:44779", + "issue:44279|issue:45405", + "issue:39692|issue:43901", + "issue:46614|issue:47085", + "issue:46123|issue:46369", + "issue:42796|issue:43650", + "issue:43577|issue:45216", + "issue:44779|issue:45812", + "issue:44075|issue:44368", + "issue:43976|issue:45405", + "issue:44466|issue:45276", + "issue:41950|issue:45823", + "issue:43381|issue:47000", + "issue:43299|issue:45357", + "issue:43653|issue:45593", + "issue:46082|issue:46399", + "issue:46082|issue:46512", + "issue:44351|issue:45278", + "issue:44462|issue:45081", + "issue:43644|issue:47145", + "issue:43299|issue:43854", + "issue:44242|issue:45200", + "issue:43650|issue:46209", + "issue:43931|issue:47145", + "issue:44016|issue:44291", + "issue:44683|issue:45542", + "issue:45588|issue:47000" + ], + "prepared_review_unit_hash": "861efadf39029a645eb776f1209d8156f2a8f9dadf945ba3ce31afee107ffc30", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36259, + "estimated_input_tokens": 9065, + "estimated_eval_tokens": 18386 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous issue cluster with no strong duplicate family; the soft pairs mostly share only broad subsystem or version-regression themes, not the same concrete bug. No PRs are present. I\u2019d keep issue #45020 as the most representative standalone issue because it is a broad, high-signal regression report.", + "confidence": 0.72, + "canonical_issue_reason": "Issue #45020 is the best canonical-style anchor here: it describes a clear, broad regression affecting recent transformers versions and model loading, which is more representative than the narrower model-specific, docs-only, or dependency-only reports in the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #45020 is the strongest single issue to keep: it is concrete, high-impact, and broadly useful as a reference for version-induced breakages, unlike the other items that are either unrelated or too specialized to serve as a cluster representative.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Different bugs: video-classification pipeline processor lookup vs remote_code model-loading regressions." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Different code paths: processor.tokenizer regression vs Deepseek tokenizer output corruption." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Dependency-related reports, but one is a vague dependency issue and the other is an unreleased PEFT version bump." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs-related, but they describe different example/API problems and not the same underlying bug." + }, + { + "left": "issue:46614", + "right": "issue:47085", + "accept": false, + "reason": "Stale espeak backend state is unrelated to Rust tokenizer thread-safety regressions." + }, + { + "left": "issue:46123", + "right": "issue:46369", + "accept": false, + "reason": "Final partial-batch handling and model memory overflow are distinct failures." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer marker handling and a placeholder/non-issue report are not the same bug." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Both involve model-state correctness, but one is dtype propagation in Blip2 and the other is Qwen3.5 save_pretrained serialization." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and likely different fixes." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Optimizer argument handling and a tie_word_embeddings warning are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version compatibility and a PEFT version bump are separate dependency problems." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "lm_head serialization inconsistency is not the same as Gemma4 embedding resizing not propagating." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Different subsystems: pipeline processor lookup vs Gemma4 device placement during forward." + }, + { + "left": "issue:43381", + "right": "issue:47000", + "accept": false, + "reason": "Gradient checkpointing in eval mode and RTDetr loss infeasibility are unrelated training/runtime bugs." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Related model family, but the failures are different save/load paths and not mergeable as one fix." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer special-token registration and D-FINE auxiliary loss logic are unrelated." + }, + { + "left": "issue:46082", + "right": "issue:46399", + "accept": false, + "reason": "LlamaConfig validation and Qwen3.5 kernel registration are different issues." + }, + { + "left": "issue:46082", + "right": "issue:46512", + "accept": false, + "reason": "One is config validation, the other is stale dtype metadata after from_config." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "Both mention imports, but one is a missing symbol and the other is a broad post-upgrade import breakage." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json and a Mistral regex patch crash are different tokenizer-loading failures." + }, + { + "left": "issue:43644", + "right": "issue:47145", + "accept": false, + "reason": "Both concern checkpoint/state loading, but one is junk non-persistent buffers and the other is missing FP8 weight_scale application." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Different model-loading failures affecting different model families and code paths." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Load-balancing loss omission and mm_token_type_ids defaulting are different model behaviors." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "A placeholder issue and an FP8 inference-on-B200 bug are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:47145", + "accept": false, + "reason": "Weight-shape mismatch on Qwen3-VL is unrelated to FP8 compressed-tensors weight_scale loading." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error and init_empty_weights argument handling are completely different problems." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and a missing TensorFlow backend dependency are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:47000", + "accept": false, + "reason": "flash_attention s_aux=None crash and RTDetrLoss infeasible cost matrix are distinct failures in different components." + } + ] + }, + "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": 28, + "review_unit_index": 43, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "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:43873", + "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:45081", + "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:45907", + "issue:46851", + "issue:47000", + "issue:47145", + "issue:47217" + ], + "soft_pairs": [ + "issue:45081|issue:45907", + "issue:44336|issue:46851", + "issue:42994|issue:43317", + "issue:43873|issue:47145", + "issue:43441|issue:45834", + "issue:45859|issue:46851", + "issue:45103|issue:45372", + "issue:43479|issue:43525", + "issue:33290|issue:45656", + "issue:44623|issue:44749", + "issue:44246|issue:47217", + "issue:42175|issue:43976", + "issue:44671|issue:44779", + "issue:44261|issue:44263", + "issue:44623|issue:45397", + "issue:44016|issue:45823", + "issue:44186|issue:47000", + "issue:45542|issue:45800", + "issue:43525|issue:43881", + "issue:43673|issue:43761", + "issue:43784|issue:45542", + "issue:43901|issue:44246", + "issue:44188|issue:44336", + "issue:43525|issue:43643", + "issue:45600|issue:45712", + "issue:43575|issue:47145", + "issue:42796|issue:43441", + "issue:45325|issue:46851", + "issue:44589|issue:45362" + ], + "prepared_review_unit_hash": "f516b9719948c4fb3e8c00dfcf72a8f81c220d273b41b8b02e61def7901c3e46", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36331, + "estimated_input_tokens": 9083, + "estimated_eval_tokens": 18422 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a true duplicate set; it\u2019s a loose semantic bucket with many unrelated model-, tokenizer-, config-, and packaging-level issues. All soft-edge pairs should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "issue:43873 is the broadest and most active open issue in the list, so it is the least-specific representative if one must be chosen, though the overall cluster is heterogeneous.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best representative by scope and activity, but it still does not unify the cluster into a single bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral tokenizer regex crash vs repository file listing revision forwarding; unrelated code paths and symptoms." + }, + { + "left": "issue:44336", + "right": "issue:46851", + "accept": false, + "reason": "ANSI output formatting bug vs Qwen3.5 variable-length position_ids bug; no shared underlying defect." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Saving quantized models vs loading dequantized models with offload are related only at a high level; different failure modes." + }, + { + "left": "issue:43873", + "right": "issue:47145", + "accept": false, + "reason": "Both touch quantization/offload, but one is generic offloading behavior and the other is FP8 frozen checkpoint scaling; not the same bug." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Ministral FlashAttention failure and Kosmos2.5 OCR index error are unrelated model-specific crashes." + }, + { + "left": "issue:45859", + "right": "issue:46851", + "accept": false, + "reason": "Both involve Qwen3.5 positional logic, but one is CPU offload incompatibility and the other is variable-length input handling; too different to merge." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring future-annotations crash vs mistral_common import error in Gemma 4 loading; unrelated subsystems." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Different config classes and fields (Phi4Multimodal vision/audio defaults vs Llama4 pad_token_id); not the same bug." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Deepspeed OOM on optimizer state vs optimizer step being called twice; distinct training-loop bugs." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "processor.save_pretrained missing files vs a performance regression during filtering after upgrade; unrelated." + }, + { + "left": "issue:44246", + "right": "issue:47217", + "accept": false, + "reason": "Import latency complaint vs VL image-placeholder failure; different behavior and code paths." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Packaging backend omission vs Python version compatibility regression; unrelated release/setup issues." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "CamemBERT masked LM regression vs Deepseek tokenizer regression; both v5 regressions but different components and fixes." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "LayerNorm epsilon precision issue vs torch.split return-value issue; no shared underlying bug." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Missing processor save files vs gemma-4 zero3 from_pretrained; different failure modes." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error vs Gemma4 device mismatch under device_map=auto; unrelated." + }, + { + "left": "issue:44186", + "right": "issue:47000", + "accept": false, + "reason": "LayoutLMv2 tokenizer padding/truncation crash vs RTDetrLoss infeasible cost matrix; different areas entirely." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorFlow backend packaging issue vs torch/transformers compatibility issue; not the same bug." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Llama4 missing pad_token_id vs glm-4v-9b loading failure; different model/config defects." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "GenerationMixin cache missing during chunked prefill vs CLIPVisionModel hidden_states regression; unrelated regression locations." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Import NameError in sentence-transformers integration vs undefined tf backend packaging; no common code path." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs still mentioning return_all_scores vs intermittent import latency; one is documentation, the other performance." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "torch.compile attention-kernel divergence vs ANSI code generation in loading_report; unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Missing config field in Llama4 vs missing fields from AutoConfig trust_remote_code; both config-related but different root causes." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "auto_mappings stale config references vs dummy_pt_objects leftovers; both cleanup/CI issues but not the same artifact set." + }, + { + "left": "issue:43575", + "right": "issue:47145", + "accept": false, + "reason": "TP OOM on large model load vs FP8 weight_scale omission; both involve memory/quantization but different bugs." + }, + { + "left": "issue:42796", + "right": "issue:43441", + "accept": false, + "reason": "Tokenizer BPE marker handling vs FlashAttention failure in a Ministral model; tokenizer vs attention backend, different fixes." + }, + { + "left": "issue:45325", + "right": "issue:46851", + "accept": false, + "reason": "Qwen2.5-VL temporal position scaling bug vs Qwen3.5 variable-length position_ids bug; similar theme, different model logic." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage lookup error vs Qwen3.5 chat crash; unrelated failure 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": 28, + "review_unit_index": 44, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:42491", + "issue:42617", + "issue:42915", + "issue:43299", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43723", + "issue:43824", + "issue:43976", + "issue:44162", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44373", + "issue:44485", + "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:46506", + "issue:46614", + "issue:46655", + "issue:46665", + "issue:46726", + "issue:46752", + "issue:46821", + "issue:46829", + "issue:46851", + "issue:47145", + "issue:47176", + "issue:47177", + "issue:47217", + "issue:47312" + ], + "soft_pairs": [ + "issue:43824|issue:43976", + "issue:44336|issue:44373", + "issue:47176|issue:47177", + "issue:43723|issue:44162", + "issue:44246|issue:44496", + "issue:45083|issue:46851", + "issue:44871|issue:47145", + "issue:46291|issue:46399", + "issue:44297|issue:45397", + "issue:42617|issue:43976", + "issue:46399|issue:46851", + "issue:46655|issue:46726", + "issue:45003|issue:47217", + "issue:42491|issue:43299", + "issue:45072|issue:46829", + "issue:43526|issue:45003", + "issue:43531|issue:45362", + "issue:45103|issue:46399", + "issue:45561|issue:45563", + "issue:43540|issue:46752", + "issue:45588|issue:46851", + "issue:44295|issue:45397", + "issue:44246|issue:44871", + "issue:45003|issue:46821", + "issue:46614|issue:46665", + "issue:44336|issue:44485", + "issue:43550|issue:46506", + "issue:44242|issue:45593", + "issue:42915|issue:47312" + ], + "prepared_review_unit_hash": "36201e158f61c22f9daa3abb7ef7bbf96445801727c3f537d58b814874c5f52c", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35663, + "estimated_input_tokens": 8916, + "estimated_eval_tokens": 18088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous issue reports; no soft pair looks like the same underlying bug/change closely enough to merge. The closest subgroup is around Qwen3-MoE/model-loading compatibility, but even those are distinct symptoms and code paths.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 42491 is the broadest and most central compatibility report in the set: a Qwen3-MoE LoRA model trained on hf4.x no longer works on hf5.x. It best represents the small Qwen3-MoE loading-regression theme, though the overall cluster is not truly cohesive.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 42491 is the strongest representative issue because it describes a concrete, user-facing regression tied to the main branch / version transition, and it is broader than the other Qwen-specific variants. Still, it should not be treated as a duplicate of the unrelated issues here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Different problems: missing model import vs Python version support regression." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "Unrelated: terminal ANSI formatting vs a docstring typo." + }, + { + "left": "issue:47176", + "right": "issue:47177", + "accept": false, + "reason": "Both are security issues, but different attack surfaces and fixes: shell-command bypass vs archive path traversal." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading failure and ESM2 breakage are unrelated subsystems and symptoms." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Performance/import slowness is unrelated to unrecognized model config loading." + }, + { + "left": "issue:45083", + "right": "issue:46851", + "accept": false, + "reason": "Both involve Qwen-family models, but different code paths: feature-length helper vs position_ids handling." + }, + { + "left": "issue:44871", + "right": "issue:47145", + "accept": false, + "reason": "Different model/config issues: Gemma eos_token_id mismatch vs FP8 weight_scale loading." + }, + { + "left": "issue:46291", + "right": "issue:46399", + "accept": false, + "reason": "Both mention kernels, but they fail in different places for different reasons." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer config save mismatch is unrelated to ZeRO3 from_pretrained loading." + }, + { + "left": "issue:42617", + "right": "issue:43976", + "accept": false, + "reason": "3d_parallel.py runtime failure is unrelated to Python-version packaging support." + }, + { + "left": "issue:46399", + "right": "issue:46851", + "accept": false, + "reason": "Different Qwen3.5 failures: kernel decorator registration vs variable-length position_ids." + }, + { + "left": "issue:46655", + "right": "issue:46726", + "accept": false, + "reason": "Both are vision-related, but one is padding math and the other is backend availability/error messaging." + }, + { + "left": "issue:45003", + "right": "issue:47217", + "accept": false, + "reason": "Sys.modules access bug is unrelated to VL image-placeholder handling." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Same broad family, but different models and failure modes: LoRA hf4.x/hf5.x compatibility vs Qwen3VL MoE loading." + }, + { + "left": "issue:45072", + "right": "issue:46829", + "accept": false, + "reason": "Different inference bugs in different model classes: dtype mismatch vs output field misassignment." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "ImageProcessor label reduction bug is unrelated to modeling_utils sys.modules access." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Sliding-window Qwen3-MoE bug and Qwen3.5 chat crash are not the same underlying defect." + }, + { + "left": "issue:45103", + "right": "issue:46399", + "accept": false, + "reason": "Auto-docstring annotation handling and kernel registration are unrelated." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "Test xdist file-race bug is unrelated to a stale generate warning." + }, + { + "left": "issue:43540", + "right": "issue:46752", + "accept": false, + "reason": "Video-input ValueError and empty-chat-list IndexError are different code paths." + }, + { + "left": "issue:45588", + "right": "issue:46851", + "accept": false, + "reason": "Flash-attention sinkless-model crash is unrelated to variable-length position_ids." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Buffer read error with position_ids is unrelated to gemma-4 ZeRO3 loading." + }, + { + "left": "issue:44246", + "right": "issue:45593", + "accept": false, + "reason": "Import slowness is unrelated to D-FINE auxiliary-loss logic." + }, + { + "left": "issue:45003", + "right": "issue:46821", + "accept": false, + "reason": "Safe sys.modules access and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:46614", + "right": "issue:46665", + "accept": false, + "reason": "Phonemizer backend state bug is unrelated to from_pretrained behavior inside Python classes." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI code emission and GLM-5 RoPE implementation are different areas and bugs." + }, + { + "left": "issue:43550", + "right": "issue:46506", + "accept": false, + "reason": "torch.compile SDPA failure and CPU/CUDA result mismatch are distinct issues." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Missing load-balancing loss and missing auxiliary losses are similar phrasing but different models and mechanisms." + }, + { + "left": "issue:42915", + "right": "issue:47312", + "accept": false, + "reason": "Qwen3MoE FP8 config failure is unrelated to kernels 0.16 incompatibility." + } + ] + }, + "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": 28, + "review_unit_index": 45, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36246", + "issue:41950", + "issue:42757", + "issue:42886", + "issue:43257", + "issue:43295", + "issue:43352", + "issue:43475", + "issue:43540", + "issue:43596", + "issue:43901", + "issue:44038", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44220", + "issue:44295", + "issue:44351", + "issue:44451", + "issue:44466", + "issue:44479", + "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: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:42886|issue:46132", + "issue:44779|issue:45292", + "issue:43901|issue:45003", + "issue:43540|issue:46655", + "issue:43257|issue:45137", + "issue:41950|issue:44220", + "issue:46726|issue:47217", + "issue:36246|issue:42757", + "issue:36246|issue:44351", + "issue:36246|issue:43596", + "issue:45375|issue:45759", + "issue:44479|issue:46655", + "issue:45292|issue:45701", + "issue:44451|issue:46752", + "issue:43295|issue:45292", + "issue:44162|issue:45200", + "issue:45137|issue:45600", + "issue:44295|issue:44749", + "issue:43352|issue:45250", + "issue:44466|issue:45335", + "issue:46170|issue:47321", + "issue:44060|issue:44077", + "issue:45003|issue:45561", + "issue:43475|issue:46369", + "issue:44877|issue:45375", + "issue:44995|issue:45200", + "issue:44038|issue:46506", + "issue:44479|issue:46170", + "issue:44987|issue:45459" + ], + "prepared_review_unit_hash": "a01e317b0cabddff92db36d4ada940ee96bd6079afcac538f8eedfad24a723ce", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35703, + "estimated_input_tokens": 8926, + "estimated_eval_tokens": 18108 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is a loose semantic cluster, not a duplicate set: it mixes unrelated import errors, config issues, tokenizer regressions, video-processor bugs, DeepSpeed failures, and model-specific regressions. I don\u2019t see any soft pair that looks like the same underlying bug or a PR-worthy consolidation target.", + "confidence": 0.92, + "canonical_issue_reason": "Issue 44479 is the best anchor because it is a concrete, user-facing regression with a clear affected family (Qwen2.5-VL/Qwen3-VL/Qwen3.5/Qwen3.5-MoE) and a specific code-path class (video input handling). It is the most broadly scoped and representative bug report among the candidates.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44479 is the strongest standalone triage target: it has the clearest regression framing, the broadest immediate impact within one bug family, and enough specificity to guide a fix without being overly generic.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline cache loading in tokenizers and AttentionInterface.register behavior changes are unrelated code paths and different failures." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "DeepSeek tokenizer output corruption and resize_token_embeddings not updating output embeddings are different bugs in different subsystems." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch for return_all_scores has nothing to do with unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:43540", + "right": "issue:46655", + "accept": false, + "reason": "Both involve video handling, but one is a Qwen3OmniMoe input validation failure and the other is a SmolVLM video padding calculation bug." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "DeepSpeed-related, but one is Qwen3 MoE weight conversion under accelerate+deepspeed and the other is a ZeRO3 deque/indexing crash." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification image-processor lookup and fbank feature extraction are different pipeline/audio code paths." + }, + { + "left": "issue:46726", + "right": "issue:47217", + "accept": false, + "reason": "Missing vision backend import handling is unrelated to VL placeholder parsing in user text." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Different missing symbols from different packages: qwen2_5_vl image processor export vs huggingface_hub is_offline_mode." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they concern different exported APIs and different fixes." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "ImportError for a VL image processor and a DeepSpeed/BertModel IndexError are unrelated failures." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "One is strict config dropping a missing field; the other is AutoModelForCausalLM not unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:44479", + "right": "issue:46655", + "accept": false, + "reason": "Both are video-related, but the underlying bugs differ: Qwen VL regression vs SmolVLM padding logic." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Embedding resize behavior and version-dependent tokenization are unrelated bugs." + }, + { + "left": "issue:44451", + "right": "issue:46752", + "accept": false, + "reason": "Model loading failure for ScandiBERT and apply_chat_template crashing on empty input are different user flows." + }, + { + "left": "issue:43295", + "right": "issue:45292", + "accept": false, + "reason": "Processor.tokenizer regression with images is unrelated to resize_token_embeddings not updating output embeddings." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaults are different model-specific issues." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "ZeRO3 initialization crash and stale auto_mappings references to removed configs are separate DeepSpeed/config problems." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Registering position_ids as a buffer causing read errors is unrelated to a slowdown after upgrading transformers." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "A model not supporting Flash Attention 2.0 and the generic Flash Attention 2.0 issue are not the same concrete bug." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Inconsistent lm_head.weight serialization by device is different from t5gemma resize_token_embeddings not affecting decoder.embed_tokens." + }, + { + "left": "issue:46170", + "right": "issue:47321", + "accept": false, + "reason": "Windows/ROCm import problems from torch.distributed dependency are unrelated to an AMD quark class not being updated." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Incorrect tied-weights warning in Qwen3-Next and patchtsmixer post_init policy are different model/config issues." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access in modeling_utils and a pytest-xdist race on captured_info.txt are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:46369", + "accept": false, + "reason": "SAM 3 Video missing fpn_position_embeddings and sam3-litetext memory overflow are different failures in different model variants." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Strict config rejection for granite_speech and a missing field in Qwen3_5MoeVisionConfig are different schema issues." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa stale indexer cache crash and Gemma 4 mm_token_type_ids defaulting are unrelated bugs." + }, + { + "left": "issue:44038", + "right": "issue:46506", + "accept": false, + "reason": "Qwen3-VL-Moe regression and CPU/CUDA result divergence in PPDocLayoutV3 are not the same code-path problem." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "Video-input regression in Qwen VL models and Windows/ROCm distributed import failure are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Loading failure for a specific model and tokenizer error masking when protobuf is absent are different 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": 28, + "review_unit_index": 46, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "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:44162", + "issue:44188", + "issue:44242", + "issue:44361", + "issue:44368", + "issue:44483", + "issue:44623", + "issue:44749", + "issue:44857", + "issue:44877", + "issue:44977", + "issue:44995", + "issue:45229", + "issue:45397", + "issue:45412", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45656", + "issue:45685", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46132", + "issue:46170", + "issue:46202", + "issue:46421", + "issue:46431", + "issue:46710", + "issue:46726", + "issue:46851", + "issue:47068", + "issue:47145", + "issue:47312" + ], + "soft_pairs": [ + "issue:44242|issue:44857", + "issue:44623|issue:46726", + "issue:29870|issue:30333", + "issue:44877|issue:46170", + "issue:44995|issue:45588", + "issue:44368|issue:44977", + "issue:45397|issue:45656", + "issue:42199|issue:46851", + "issue:43653|issue:44162", + "issue:43475|issue:43540", + "issue:42994|issue:47312", + "issue:42886|issue:45229", + "issue:45685|issue:46431", + "issue:45412|issue:46123", + "issue:43723|issue:46710", + "issue:43065|issue:43408", + "issue:43931|issue:44368", + "issue:44188|issue:46202", + "issue:44368|issue:45563", + "issue:45987|issue:46082", + "issue:43906|issue:46202", + "issue:46132|issue:47145", + "issue:44188|issue:46421", + "issue:44483|issue:46202", + "issue:45593|issue:47068", + "issue:44361|issue:44749" + ], + "prepared_review_unit_hash": "d367508cb85e5834681694e352c64da60d53e09e323ab0b312e3512e2598e0e9", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34913, + "estimated_input_tokens": 8729, + "estimated_eval_tokens": 17714 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All candidate soft pairs are false positives: the items are heterogeneous and do not describe the same underlying bug/change. No PRs are present.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:44242 is the strongest standalone bug report here: it is concrete, reproducible, and has the clearest follow-up signal among the issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Both are detection/loss-related, but one is a missing MoE auxiliary loss and the other is an AMP/CUDA crash in LwDetrImageLoss; different failures and code paths." + }, + { + "left": "issue:44623", + "right": "issue:46726", + "accept": false, + "reason": "One is about save_pretrained missing files, the other about AutoProcessor failing when a vision backend is absent; related area, not the same bug." + }, + { + "left": "issue:29870", + "right": "issue:30333", + "accept": false, + "reason": "Completely unrelated: BART generate() UnboundLocalError versus MLflow reporting a failed training job as still running." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Different subsystems and symptoms: strict config loading for granite_speech versus Windows/ROCm import dependency issues." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Both are runtime crashes, but one is a stale cache/indexer issue in GLM-MoE and the other is a flash-attention AttributeError on sink-less models." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "One is a harmless warning about tie_word_embeddings during LoRA fine-tuning; the other is a generation correctness issue with flash-attention." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Both mention deepspeed, but one is gemma-4 from_pretrained under zero3 and the other is double optimizer stepping; different concrete bugs." + }, + { + "left": "issue:42199", + "right": "issue:46851", + "accept": false, + "reason": "Different models and failures: DETR cardinality logic versus Qwen3.5 variable-length position_ids handling." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer special-token registration bug versus ESM2 being broken; no shared underlying code-path." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "SAM3 video attribute missing versus Qwen3OmniMoe video-input validation error; unrelated model paths." + }, + { + "left": "issue:42994", + "right": "issue:47312", + "accept": false, + "reason": "Quantized model saving failure is unrelated to kernel version incompatibility." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and multi-GPU Gemma inference OOM are different issues." + }, + { + "left": "issue:45685", + "right": "issue:46431", + "accept": false, + "reason": "MPS histogram/int type error in MoE code versus chat template absence in a processor; different causes." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "Memory not released on deletion versus partial-batch is_last handling in a pipeline; unrelated bugs." + }, + { + "left": "issue:43723", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer loading issue versus accuracy regression after upgrade; same project area, not the same defect." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "A dummy Conv2d implementation and a model-type warning are not the same bug." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch versus a Qwen3.5 warning about tied embeddings; different failure modes." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to Gemma pipeline roundtrip behavior on \u2581." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "A config warning about tied embeddings is unrelated to a stale num_return_sequences warning in paged generate()." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "StaticCache return-type contract bug versus LlamaConfig head_dim validation; distinct code paths." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "An isolated reproduction of an external issue is not the same as a Gemma pipeline roundtrip failure." + }, + { + "left": "issue:46132", + "right": "issue:47145", + "accept": false, + "reason": "AttentionInterface.register behavior change is unrelated to FP8 compressed-tensors loading without weight_scale." + }, + { + "left": "issue:44188", + "right": "issue:46421", + "accept": false, + "reason": "Different compile/cache problems: attention-kernel divergence versus spurious recompilation with chunked prefill." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "API request rejection and Gemma tokenizer roundtrip failure are unrelated." + }, + { + "left": "issue:45593", + "right": "issue:47068", + "accept": false, + "reason": "Both involve loss computation, but one is missing auxiliary losses and the other is a potential host sync in a Python conditional." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "Tokenizer AttributeError on tasks versus a performance regression after upgrading Transformers; 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": 28, + "review_unit_index": 47, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:39692", + "issue:42994", + "issue:43065", + "issue:43232", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43701", + "issue:43723", + "issue:43784", + "issue:43994", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44373", + "issue:44623", + "issue:44749", + "issue:44871", + "issue:44964", + "issue:45103", + "issue:45137", + "issue:45230", + "issue:45310", + "issue:45375", + "issue:45397", + "issue:45478", + "issue:45561", + "issue:45656", + "issue:46123", + "issue:46132", + "issue:46326", + "issue:46399", + "issue:46463", + "issue:46726", + "issue:46752", + "issue:46821", + "issue:46851", + "issue:46858", + "issue:46877", + "issue:47000", + "issue:47020", + "issue:47217" + ], + "soft_pairs": [ + "issue:45103|issue:46858", + "issue:46752|issue:47020", + "issue:43994|issue:46726", + "issue:43701|issue:44749", + "issue:44297|issue:44623", + "issue:43065|issue:46123", + "issue:44749|issue:45656", + "issue:43646|issue:44162", + "issue:44964|issue:45137", + "issue:44623|issue:45656", + "issue:43475|issue:46821", + "issue:43475|issue:43643", + "issue:43784|issue:46399", + "issue:42994|issue:43482", + "issue:44373|issue:46851", + "issue:39692|issue:46726", + "issue:44871|issue:46132", + "issue:44188|issue:46851", + "issue:46463|issue:46877", + "issue:28282|issue:30990", + "issue:45561|issue:47217", + "issue:45375|issue:45478", + "issue:46877|issue:47000", + "issue:45230|issue:45397", + "issue:44038|issue:44623", + "issue:45310|issue:45375", + "issue:43723|issue:46326", + "issue:42994|issue:43525", + "issue:43232|issue:43673" + ], + "prepared_review_unit_hash": "ad5d4eb5ca5bceb30ee33b635191253c08443d5021927ddab5deeb5fdc697cb6", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36160, + "estimated_input_tokens": 9040, + "estimated_eval_tokens": 18336 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a semantic grab-bag of Transformer v5-era regressions and model-specific bugs; there is one clear duplicate issue pair, but most soft edges are not the same underlying problem.", + "confidence": 0.39, + "canonical_issue_reason": "issue:43646 is the broadest representative report in the set: it describes Transformers 5.0.0 breaking custom model initialization, which matches the general v5-regression pattern seen across many of the other issues.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is also the best single issue representative because it is broader than the model-specific reports and captures the common upgrade-breakage theme of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45103", + "right": "issue:46858", + "accept": false, + "reason": "Different subsystems: auto_docstring annotation parsing vs cache sequence-length arithmetic under torch.compile." + }, + { + "left": "issue:46752", + "right": "issue:47020", + "accept": false, + "reason": "Empty-list chat templating crash is unrelated to XLMRobertaTokenizer passing the wrong type to Unigram." + }, + { + "left": "issue:43994", + "right": "issue:46726", + "accept": false, + "reason": "Both involve vision models/processors, but one is a bad AutoModel/pipeline result and the other is an AutoProcessor backend-missing error." + }, + { + "left": "issue:43701", + "right": "issue:44749", + "accept": false, + "reason": "resume_from_checkpoint key mismatch is unrelated to a training slowdown after upgrading Transformers." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both touch save_pretrained, but tokenizer_class mismatch and processor missing files are different serialization bugs." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "Sam3PixelDecoder dummy Conv2d is unrelated to MaskGenerationPipeline dropping the final partial batch." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Performance regression after upgrade is not the same as deepspeed causing optimizer.step to run twice." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Generic custom model initialization breakage is too different from ESM2-specific model breakage." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal loading failure and DeepSpeed ZeRO3 deque underflow are unrelated code paths." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Processor save_pretrained file emission is unrelated to deepspeed double optimizer-step behavior." + }, + { + "left": "issue:43475", + "right": "issue:46821", + "accept": false, + "reason": "SAM 3 video missing field on vision output is unrelated to TimesFM window_size AttributeError." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "Missing fpn_position_embeddings and trust_remote_code AutoConfig field loss are different bugs." + }, + { + "left": "issue:43784", + "right": "issue:46399", + "accept": false, + "reason": "Import-time NameError in sentence-transformers is unrelated to Qwen3.5 rotary-position registration failure." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different failure modes." + }, + { + "left": "issue:44373", + "right": "issue:46851", + "accept": false, + "reason": "Wrong docstring for position_ids is documentation-only; variable-length position_ids is a runtime model bug." + }, + { + "left": "issue:39692", + "right": "issue:46726", + "accept": false, + "reason": "SigLIP2 example errors are a documentation/example issue, not the same as missing vision-backend processor handling." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Gemma eos_token_id inconsistency and AttentionInterface.register behavior change are distinct configuration/API issues." + }, + { + "left": "issue:44188", + "right": "issue:46851", + "accept": false, + "reason": "torch.compile attention-kernel divergence is unrelated to Qwen3.5 variable-length position_ids handling." + }, + { + "left": "issue:46463", + "right": "issue:46877", + "accept": false, + "reason": "T5 decoder-input validation messaging and auto_docstring stdout logging are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "PyTorch-missing ImportError and Sentence Transformers loading hang are different root causes." + }, + { + "left": "issue:45561", + "right": "issue:47217", + "accept": false, + "reason": "pytest-xdist race in testing utils is unrelated to VL image placeholder failures." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": true, + "reason": "These are the same Qwen3.5 MoE from_pretrained error, with identical titles and likely the same underlying regression." + }, + { + "left": "issue:46877", + "right": "issue:47000", + "accept": false, + "reason": "doc-lint diagnostics printing at import time is unrelated to RTDetrLoss infeasible cost matrix failure." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "A generic bug report is not the same underlying issue as gemma-4 zero3 from_pretrained failure." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3 VL bug and processor.save_pretrained missing files are not the same code-path problem." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is a broad from_pretrained error and the other is a missing config field under @strict; not enough to treat as the same bug." + }, + { + "left": "issue:43723", + "right": "issue:46326", + "accept": false, + "reason": "AutoTokenizer loading regression and mistral_common continue_final_message behavior are different tokenizer issues." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized model saving failure and Llama4Config missing pad_token_id are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "sync_gpus generation state handling and chunked_prefill cache absence are different generation bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the lone accepted pair (issue:45375/issue:45478) is a clearly supported duplicate by title. The other soft-edge rejections stay conservative and do not overclaim shared root cause from the available evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 48, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:33290", + "issue:35141", + "issue:36010", + "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:44222", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44661", + "issue:44743", + "issue:44792", + "issue:44857", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45137", + "issue:45310", + "issue:45356", + "issue:45405", + "issue:45406", + "issue:45561", + "issue:45593", + "issue:45712", + "issue:45753", + "issue:46202", + "issue:46701", + "issue:46877", + "issue:47059", + "issue:47068" + ], + "soft_pairs": [ + "issue:44869|issue:45356", + "issue:41628|issue:44933", + "issue:42371|issue:45092", + "issue:43873|issue:44945", + "issue:36010|issue:41628", + "issue:44661|issue:47059", + "issue:43278|issue:45137", + "issue:45081|issue:45103", + "issue:44857|issue:46877", + "issue:45406|issue:46202", + "issue:46202|issue:46701", + "issue:45593|issue:46877", + "issue:44038|issue:45310", + "issue:41628|issue:44351", + "issue:44222|issue:45310", + "issue:44945|issue:45753", + "issue:43232|issue:44792", + "issue:45561|issue:47059", + "issue:42175|issue:44933", + "issue:44743|issue:47068", + "issue:45405|issue:46877", + "issue:42831|issue:44222", + "issue:30333|issue:33290", + "issue:42175|issue:45712", + "issue:43784|issue:43824", + "issue:42673|issue:43704", + "issue:42898|issue:44568", + "issue:44496|issue:45561", + "issue:43976|issue:44933", + "issue:30990|issue:35141" + ], + "prepared_review_unit_hash": "03de24922232bfe7d939356807d8e2ac912e495312196e8b8617241485acdc47", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36129, + "estimated_input_tokens": 9033, + "estimated_eval_tokens": 18322 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous issue noise. The clearest duplicate pair is the repeated Qwen3ForCausalLM VRAM leak report; the other pairs look related only at a broad subsystem level, not the same concrete defect.", + "confidence": 0.83, + "canonical_issue_reason": "issue:42673 is the best canonical issue candidate for the duplicate pair because it is the earlier, more discussed report with the exact same title and bug scenario as issue:43704.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the strongest anchor for the one clear duplicate pair: it has the same title and underlying multi-dataloader VRAM leak problem as issue:43704, and it has more discussion context.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Both mention tokenizer/decoding behavior, but one is a Whisper timestamp crash and the other is a Kimi tokenizer regression; different failures and code paths." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Both are import-related, but they involve different missing symbols/modules and do not point to the same concrete bug." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 control settings and InternVL2 meta-init incompatibility are unrelated issues." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior and pipeline-parallel output corruption are different problems." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different APIs and with no clear shared root cause." + }, + { + "left": "issue:44661", + "right": "issue:47059", + "accept": false, + "reason": "`add-new-model-like` mapping logic and a CLI/huggingface_hub compatibility failure are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift in eval and a DeepSpeed ZeRO3 deque error are distinct runtime issues." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Tokenizer regex patching and auto-docstring annotation handling fail in different subsystems." + }, + { + "left": "issue:44857", + "right": "issue:46877", + "accept": false, + "reason": "An AMP crash in LwDetrImageLoss and import-time docstring diagnostics are unrelated." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Both involve Gemma-related processing, but one is a processor attribute crash and the other is a tokenizer roundtrip failure." + }, + { + "left": "issue:46202", + "right": "issue:46701", + "accept": false, + "reason": "Gemma pipeline roundtrip failure and image-text-to-text pipeline breakage are different bugs." + }, + { + "left": "issue:45593", + "right": "issue:46877", + "accept": false, + "reason": "D-FINE auxiliary loss behavior is unrelated to auto_docstring logging." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen3 MoE, but the versions, model variants, and failure modes differ." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports with no concrete shared fix." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save/load behavior and Qwen3.5 MoE from_pretrained errors are not the same bug." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output issues and a Qwen3_5 infinite loop on a specific image are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus and a Janus image-generation test failure are different issues." + }, + { + "left": "issue:45561", + "right": "issue:47059", + "accept": false, + "reason": "pytest-xdist file races and CLI incompatibility with huggingface_hub are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Backend dependency installation and an image_utils import failure are separate packaging/API problems." + }, + { + "left": "issue:44743", + "right": "issue:47068", + "accept": false, + "reason": "Recurrent-state reset in modular_qwen3_5.py and a DeepSpeed host-sync concern are different code paths." + }, + { + "left": "issue:45405", + "right": "issue:46877", + "accept": false, + "reason": "Minimum PEFT version bump and auto_docstring stdout noise are unrelated." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Both reference FP8, but one is an accuracy issue and the other is save_pretrained behavior; not the same defect." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting on failed jobs and Adafactor OOM in DeepSpeed are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Torch-extra backend packaging and leftover dummy PT objects are different maintenance issues." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors on newer transformers, but they are missing different exported names and do not share a concrete bug." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Identical title and the same VRAM leak scenario in multiple dataloader threads indicate a direct duplicate." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Both involve tokenizer behavior in v5, but cleanup spacing and special-token insertion are different bugs." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Model config recognition failure and pytest-xdist file races are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python-version compatibility and an image_utils import error are unrelated." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly grounded, but it overstates the evidence for the canonical issue. In the packet, issue:42673 has more discussion activity than issue:43704, but it does not have more inbound references (both show 0). The claim that this is the \u201conly true duplicate pair\u201d is also stronger than what the packet alone proves; the safe conclusion is that it is the clearest duplicate pair among the listed candidates, not that other duplicates are impossible." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 28, + "review_unit_index": 49, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28282", + "issue:30333", + "issue:34689", + "issue:36331", + "issue:39692", + "issue:42222", + "issue:42673", + "issue:42898", + "issue:43116", + "issue:43295", + "issue:43479", + "issue:43723", + "issue:43874", + "issue:44016", + "issue:44060", + "issue:44075", + "issue:44079", + "issue:44186", + "issue:44190", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44448", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44843", + "issue:44869", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45200", + "issue:45310", + "issue:45600", + "issue:45656", + "issue:45812", + "issue:46077", + "issue:46123", + "issue:46231", + "issue:46693", + "issue:46829", + "issue:46909", + "issue:47059", + "issue:47085" + ], + "soft_pairs": [ + "issue:44843|issue:47085", + "issue:44568|issue:47085", + "issue:43479|issue:46829", + "issue:42898|issue:45812", + "issue:44623|issue:45310", + "issue:44186|issue:47085", + "issue:42673|issue:46693", + "issue:43723|issue:44568", + "issue:39692|issue:46909", + "issue:45600|issue:47059", + "issue:43479|issue:44869", + "issue:44016|issue:46231", + "issue:43479|issue:44079", + "issue:42673|issue:47085", + "issue:46077|issue:46123", + "issue:30333|issue:34689", + "issue:27957|issue:30333", + "issue:42898|issue:44779", + "issue:44060|issue:45125", + "issue:42898|issue:43295", + "issue:42898|issue:44448", + "issue:30333|issue:36331", + "issue:42222|issue:45200", + "issue:22355|issue:28282", + "issue:44295|issue:45656", + "issue:43874|issue:45092", + "issue:44734|issue:45103", + "issue:43116|issue:44190", + "issue:44222|issue:44297", + "issue:44075|issue:46231" + ], + "prepared_review_unit_hash": "780f99d8f68988366e498b72a99bc62878f3e79d272ee3a84ee7777cf69987e2", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36313, + "estimated_input_tokens": 9079, + "estimated_eval_tokens": 18414 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, model-loading, training, docs, and serving issues. None of the soft pairs look like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "No single issue is a true duplicate anchor here; if forced to pick one, issue 47085 is the best current triage target because it is open and active, but the cluster itself is not coherent.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 47085 is the most actionable current report in the set (open, recent, and discussed), though it is not truly representative of the unrelated cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44843", + "right": "issue:47085", + "accept": false, + "reason": "Offline AutoTokenizer/Mistral hub lookup bug vs tokenizer thread-safety regression; different code paths and symptoms." + }, + { + "left": "issue:44568", + "right": "issue:47085", + "accept": false, + "reason": "mDeBERTa add_special_tokens BOS/EOS regression is unrelated to Rust tokenizer thread safety." + }, + { + "left": "issue:43479", + "right": "issue:46829", + "accept": false, + "reason": "Phi4MultimodalConfig default-init bug and VideoPrism output-field mismatch are unrelated model bugs." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Both are v5 tokenizer regressions, but one is cleanup-space behavior and the other is Granite token-ID generation; not the same bug." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "processor.save_pretrained file omission and Qwen3.5 MoE from_pretrained failure are unrelated save/load problems." + }, + { + "left": "issue:44186", + "right": "issue:47085", + "accept": false, + "reason": "LayoutLMv2Tokenizer NER/padding crash is unrelated to thread-safety in fast tokenizers." + }, + { + "left": "issue:42673", + "right": "issue:46693", + "accept": false, + "reason": "VRAM leak in Qwen3ForCausalLM multiple dataloader threads is not the same as FA2 long-context performance regression." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Generic AutoTokenizer v5 loading issue vs a specific mDeBERTa add_special_tokens regression; different failures." + }, + { + "left": "issue:39692", + "right": "issue:46909", + "accept": false, + "reason": "Both are broken doc examples, but for different models/examples and different error classes." + }, + { + "left": "issue:45600", + "right": "issue:47059", + "accept": false, + "reason": "Removed-config auto_mappings CI breakage is unrelated to huggingface_hub CLI completion breakage." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config defaulting bug and Whisper timestamp decode crash are unrelated." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "Notebook syntax error and training API doc parameter mismatch are different documentation issues." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Bidirectional-attention masking bug and final-partial-batch pipeline bookkeeping bug are different generation issues." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow reporting on failed training and Llama 3.2 Vision loading regression are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 with XLMRoberta and MLflow failure reporting are unrelated subsystems." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "General v5 cleanup-space behavior change and Deepseek tokenizer incorrect IDs are not the same tokenizer bug." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-Next tied-weights warning and missing _tp_plan for Qwen3.5 MoE are different model-architecture issues." + }, + { + "left": "issue:42898", + "right": "issue:43295", + "accept": false, + "reason": "Token cleanup behavior change is unrelated to processor.tokenizer/image-passing regression." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Pegasus output differences between v4/v5 are a model-specific regression, not the clean_up_tokenization_spaces issue." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "Failed-job MLflow reporting and CustomTrainer.compute_loss signature change are unrelated training issues." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken vitpose models and Gemma 4 mm_token_type_ids defaults are unrelated multimodal model bugs." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Missing transformers.onnx module and missing PyTorch dependency are different import errors." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "position_ids buffer access error and deepspeed double optimizer-step bug are unrelated." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Missing image-processor method in GLM46V and meta-init incompatibility in InternVL2 are different multimodal-loading failures." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "serve KV-cache continuation indexing crash and auto_docstring future-annotations crash are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Classification example empty-results bug and local-dataset loading failure are different example-script issues." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "FP8 save_pretrained MoE issue and tokenizer_class metadata mismatch are unrelated save-pretrained bugs." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "SGD args ignored and documentation parameter-name mismatch 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": 28, + "review_unit_index": 50, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:30990", + "issue:38175", + "issue:41628", + "issue:42757", + "issue:42796", + "issue:42831", + "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:45357", + "issue:45405", + "issue:45656", + "issue:45712", + "issue:45834", + "issue:46082", + "issue:46132", + "issue:46209", + "issue:46305", + "issue:46424", + "issue:46431", + "issue:46614", + "issue:46762", + "issue:46932", + "issue:47059", + "issue:47276" + ], + "soft_pairs": [ + "issue:45405|issue:45712", + "issue:44448|issue:46132", + "issue:42796|issue:43723", + "issue:42831|issue:44623", + "issue:44297|issue:45656", + "issue:43479|issue:43644", + "issue:44933|issue:45712", + "issue:45245|issue:46132", + "issue:44222|issue:44623", + "issue:46209|issue:46431", + "issue:44373|issue:44485", + "issue:44297|issue:45310", + "issue:41628|issue:42757", + "issue:43723|issue:44977", + "issue:43122|issue:46209", + "issue:43504|issue:47059", + "issue:46614|issue:46932", + "issue:43232|issue:47276", + "issue:45071|issue:46082", + "issue:44360|issue:46762", + "issue:44222|issue:44484", + "issue:43232|issue:45834", + "issue:44190|issue:46082", + "issue:30990|issue:38175", + "issue:44263|issue:45092", + "issue:44295|issue:44623", + "issue:44327|issue:47059", + "issue:43232|issue:46424", + "issue:44263|issue:44373", + "issue:43475|issue:46305", + "issue:45092|issue:45357" + ], + "prepared_review_unit_hash": "e4329a170ed0dec0b832f3a08aae2cecda6081ec8128cae695f6fa19762bef31", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36102, + "estimated_input_tokens": 9026, + "estimated_eval_tokens": 18308 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly broad Transformers regressions that overlap only at the subsystem level; none of the soft pairs look like the same concrete bug/change, so all soft edges should be rejected.", + "confidence": 0.87, + "canonical_issue_reason": "issue:43723 is the broadest core-API bug here: a v5 tokenizer-loading failure with clear regression framing, making it the best representative among the tokenizer-related reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:47059 has the strongest cross-cutting signal and most inbound references, with a concrete dependency-compatibility failure that is easy to reproduce and triage.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "Unrelated problems: a PEFT version bump not yet on PyPI vs leftover dummy classes leaking into repo checks." + }, + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "Both touch model behavior, but one is Pegasus output drift across v4/v5 and the other is an AttentionInterface registration side effect." + }, + { + "left": "issue:42796", + "right": "issue:43723", + "accept": false, + "reason": "Tokenizer-specific BPE marker handling vs a general AutoTokenizer.from_pretrained loading failure; different bugs." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FP8 accuracy regression and processor.save_pretrained missing files are separate code paths and failure modes." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer metadata mismatch on save_pretrained is unrelated to optimizer stepping twice under deepspeed." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Config initialization bug for multimodal defaults vs a buffers-initialization/data corruption issue." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "One is a missing import from image_utils; the other is dummy classes leaking into namespace and checks." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category-count limit and AttentionInterface.register behavior are different functional issues." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "FP8 MOE save_pretrained failure vs processor save_pretrained omitting files; not the same bug." + }, + { + "left": "issue:46209", + "right": "issue:46431", + "accept": false, + "reason": "B200 FP8 inference failure and missing chat template are unrelated runtime/configuration issues." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "A wrong docstring and a RoPE implementation discussion are not the same underlying defect." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save_pretrained class-name mismatch vs Qwen3.5 MoE from_pretrained error; different surfaces and causes." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but from different packages and unrelated symbols." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading in v5 is unrelated to Qwen3.5 flash-attention generation behavior." + }, + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenization drift across versions and FP8 inference on B200 do not share a concrete code path." + }, + { + "left": "issue:43504", + "right": "issue:47059", + "accept": false, + "reason": "A legacy model-loading field issue is unrelated to a huggingface_hub CLI completion attr error." + }, + { + "left": "issue:46614", + "right": "issue:46932", + "accept": false, + "reason": "Stale phonemizer backend state and Gemma4 quantized embedder dtype casting are different bugs." + }, + { + "left": "issue:43232", + "right": "issue:47276", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus and a misleading warning message are not the same issue." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "PretrainedConfig type-checking regression vs LlamaConfig head_dim validation; related area, different defect." + }, + { + "left": "issue:44360", + "right": "issue:46762", + "accept": false, + "reason": "DSA indexer missing ReLU and Minimax attention numerical differences are distinct implementation problems." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "A save_pretrained FP8 bug and a question about default shard size are not duplicates." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation sync_gpus kwargs handling vs long-OCR-input index error in Kosmos2.5 are separate issues." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Local dataset loading in an example script and LlamaConfig head_dim validation are unrelated." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Sentence Transformers loading hang and zero probabilities in SigLIP2 are different model-specific failures." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return semantics in a DSA indexer and remote-code checkpoint incompatibility with meta init are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading a buffer after registration and missing files in processor.save_pretrained are different failure modes." + }, + { + "left": "issue:44327", + "right": "issue:47059", + "accept": false, + "reason": "QA decode_spans top_k bounds error and CLI completion attr mismatch are unrelated subsystems." + }, + { + "left": "issue:43232", + "right": "issue:46424", + "accept": false, + "reason": "Generation kwargs after sync_gpus and cache max length behavior are separate features/bugs." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "A runtime torch.split/indexer bug and a docstring correction do not describe the same change." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM 3 Video missing attribute and MoonshineStreaming off-by-one mask logic are different model code paths." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Meta-initialization incompatibility for remote-code checkpoints vs incorrect visual encoder keys on save_pretrained are different problems." + } + ] + }, + "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": 28, + "review_unit_index": 51, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28218", + "issue:39401", + "issue:41628", + "issue:42796", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43122", + "issue:43408", + "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:45381", + "issue:45406", + "issue:45600", + "issue:45753", + "issue:46002", + "issue:46050", + "issue:46305", + "issue:46328", + "issue:46369", + "issue:46402", + "issue:46655", + "issue:46726", + "issue:46909", + "issue:47059", + "issue:47217" + ], + "soft_pairs": [ + "issue:42907|issue:46050", + "issue:46369|issue:46909", + "issue:46305|issue:46655", + "issue:44279|issue:46328", + "issue:22355|issue:27957", + "issue:42831|issue:45310", + "issue:43986|issue:46726", + "issue:44263|issue:44360", + "issue:45381|issue:46305", + "issue:43408|issue:46402", + "issue:41628|issue:43824", + "issue:42831|issue:44038", + "issue:44964|issue:47059", + "issue:44261|issue:45092", + "issue:44483|issue:45406", + "issue:44829|issue:46002", + "issue:22355|issue:28218", + "issue:43576|issue:43784", + "issue:44188|issue:44373", + "issue:43784|issue:45600", + "issue:42831|issue:45753", + "issue:43644|issue:46305", + "issue:43756|issue:44060", + "issue:39401|issue:44568", + "issue:43986|issue:47217", + "issue:42886|issue:42994", + "issue:42796|issue:43122" + ], + "prepared_review_unit_hash": "26ce5c3a09a5f1d06678b1c02a0df124c93e6ea5f9ebce9961138efd72020b85", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 34978, + "estimated_input_tokens": 8745, + "estimated_eval_tokens": 17746 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is not a true duplicate cluster: the items span many unrelated tokenizer, model-loading, quantization, CLI, and docs bugs. All soft edges look like superficial similarity rather than the same underlying defect.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue stands out because the set is heterogeneous and does not center on one shared bug or code path.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue is a strong global representative; the pairings are mostly topic-level similarity, not duplicate reports of one defect.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both are about quantized model saving, but they target different quantization flows and failure modes (dequantized Ministral/Devstral vs sinq save/load)." + }, + { + "left": "issue:46369", + "right": "issue:46909", + "accept": false, + "reason": "One is a sam3-litetext memory overflow bug; the other is a broken DIA documentation example. Different artifact types and code paths." + }, + { + "left": "issue:46305", + "right": "issue:46655", + "accept": false, + "reason": "Both involve video processing, but one is a MoonshineStreaming sliding-window off-by-one and the other is a SmolVLM padding-calculation bug." + }, + { + "left": "issue:44279", + "right": "issue:46328", + "accept": false, + "reason": "A generic dependency issue and a pad() performance regression are unrelated problems." + }, + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "Import failure for transformers.onnx and XLMRoberta Flash Attention 2 are unrelated bugs." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy drift and Qwen3.5 Moe from_pretrained errors are different issues in different subsystems." + }, + { + "left": "issue:43986", + "right": "issue:46726", + "accept": false, + "reason": "Both mention missing vision backends, but one is a video-model AutoProcessor crash without torchvision and the other is an unrecognized image processor path; not the same concrete failure." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "GlmMoeDsaIndexer torch.split handling and a missing ReLU in the DSA indexer are different implementation bugs." + }, + { + "left": "issue:45381", + "right": "issue:46305", + "accept": false, + "reason": "Qwen2.5-VL video vision_position_ids and MoonshineStreaming mask off-by-one are unrelated code paths." + }, + { + "left": "issue:43408", + "right": "issue:46402", + "accept": false, + "reason": "A sam3_video vs sam3_tracker config warning is not the same as Sam3Model text-encoder weights not loading." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Different missing exports: AutoImageProcessor versus Qwen2_5_VLForConditionalGeneration." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "FineGrainedFP8 accuracy issues and Qwen3-VL-Moe loading bugs are unrelated." + }, + { + "left": "issue:44964", + "right": "issue:47059", + "accept": false, + "reason": "Phi-4 multimodal loading and a huggingface_hub CLI compatibility failure are different problems." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Both are model-init related, but one is a missing rms_norm_eps precision issue and the other is a remote-code/meta-init incompatibility." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "An API request rejection and a Gemma4Processor missing _tokenizer crash are unrelated." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Flash-Attention-3 training degeneration and NaNs after checkout/init have different symptoms and likely different root causes." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "No module named transformers.onnx is unrelated to tokenizer spacing after added tokens." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "A broken transformers env CLI command and a sentence-transformers import NameError are different regressions." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "torch.compile attention-kernel branching and a wrong docstring are not the same bug." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import-time NameError in sentence-transformers and stale auto_mappings references are unrelated." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "An FP8 accuracy issue and an infinite loop in Qwen3_5 on a specific image are different failures." + }, + { + "left": "issue:43644", + "right": "issue:46305", + "accept": false, + "reason": "Non-persistent buffer junk in Transformers 5.0 and a MoonshineStreaming mask off-by-one are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Smollm3 RoPE-layer dropping and Qwen3-Next tied-weight warnings are both model-architecture issues, but not the same concrete defect." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer bugs, but one is wrong offset mapping in Qwen3 and the other is missing BOS/EOS insertion in mdeberta-v3." + }, + { + "left": "issue:43986", + "right": "issue:47217", + "accept": false, + "reason": "AutoProcessor failing without torchvision and VL models failing on placeholders in user text are different code-path problems." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model saving are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43122", + "accept": false, + "reason": "Both mention tokenization, but one is a Ministral BPE-marker bug and the other is a version-to-version tokenization change; not clearly 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": 28, + "review_unit_index": 52, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:29159", + "issue:30990", + "issue:33357", + "issue:39379", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42886", + "issue:42907", + "issue:43066", + "issue:43122", + "issue:43329", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43540", + "issue:43646", + "issue:43824", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44117", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44488", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45278", + "issue:45290", + "issue:45381", + "issue:46002", + "issue:46172", + "issue:46326", + "issue:46369", + "issue:46762", + "issue:46909", + "issue:47059" + ], + "soft_pairs": [ + "issue:39379|issue:46172", + "issue:42886|issue:43066", + "issue:43408|issue:46369", + "issue:42757|issue:45278", + "issue:27957|issue:30990", + "issue:43646|issue:46002", + "issue:45381|issue:46909", + "issue:43540|issue:46909", + "issue:30990|issue:33357", + "issue:44295|issue:44297", + "issue:42907|issue:44964", + "issue:43329|issue:47059", + "issue:42907|issue:44945", + "issue:44625|issue:44829", + "issue:43976|issue:44945", + "issue:43122|issue:46326", + "issue:41628|issue:45278", + "issue:39401|issue:44521", + "issue:39401|issue:44488", + "issue:44117|issue:44704", + "issue:42371|issue:44263", + "issue:42907|issue:43937", + "issue:44704|issue:45290", + "issue:22355|issue:29159", + "issue:43122|issue:44521", + "issue:43408|issue:43475", + "issue:42907|issue:43452", + "issue:44360|issue:44485", + "issue:44038|issue:44297", + "issue:43824|issue:44351", + "issue:46172|issue:46762", + "issue:44038|issue:44295" + ], + "prepared_review_unit_hash": "41ed60119f7d91753547b8b5a6caff202d1da7053e88d753344fd6903722ebc2", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36769, + "estimated_input_tokens": 9193, + "estimated_eval_tokens": 18642 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is largely heterogeneous: most items are separate bugs/features in different subsystems (imports, tokenizers, multimodal models, GLM/SAM, docs). The soft edges mostly reflect superficial overlap in keywords rather than the same underlying defect, so I reject all candidate duplicates.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45278 is the broadest and most general report here, centered on a regression causing multiple import failures after a version update; it best represents the small import-error subset of the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best single issue to keep because it is broad, concrete, and likely to attract related import-regression reports, whereas the others are narrower and more clearly distinct.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39379", + "right": "issue:46172", + "accept": false, + "reason": "Different models and different bugs: mllama mask handling vs DeepSeek tensor transpose logic." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Both are tokenizer-related, but one is offline cache loading and the other is decoder-type metadata in v5." + }, + { + "left": "issue:43408", + "right": "issue:46369", + "accept": false, + "reason": "Unrelated SAM3 issues: a model-type warning versus a memory overflow in a different model variant." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "Both mention imports, but one is a specific huggingface_hub symbol regression and the other is a broad transformers upgrade breakage." + }, + { + "left": "issue:27957", + "right": "issue:30990", + "accept": false, + "reason": "Different symptoms and code paths: Flash Attention 2 behavior in XLM-RoBERTa vs a loading hang in Sentence Transformers." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model init breakage and NaNs in classification heads are different failure modes." + }, + { + "left": "issue:45381", + "right": "issue:46909", + "accept": false, + "reason": "Qwen2.5-VL video position IDs and a DIA documentation example are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:46909", + "accept": false, + "reason": "Video-input processing bug in Qwen3OmniMoe is not the same as a broken docs example." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Different models/platforms and distinct failures: loading stuck vs MacOS bus error." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "One is buffer access for position_ids; the other is tokenizer config metadata saved incorrectly." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Dequantized save failure and multimodal model load failure are separate issues." + }, + { + "left": "issue:43329", + "right": "issue:47059", + "accept": false, + "reason": "Video multimodal-token counting bug is unrelated to CLI completion breakage in huggingface_hub." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Different areas and different failure modes: saving dequantized models vs pipeline-parallel output corruption." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Qwen3.5 label propagation and Flash Attention 3 training degeneration are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility and pipeline-parallel output correctness are not the same bug." + }, + { + "left": "issue:43122", + "right": "issue:46326", + "accept": false, + "reason": "Both involve tokenization, but one is version-to-version tokenization drift and the other is continue_final_message behavior with mistral_common." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and different regression contexts." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Wrong offset mappings and zero assistant masks are distinct tokenizer/chat-template bugs." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Tokenizer offset mapping bug and model-loading failure for cjvt/sleng-bert are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Mapping-name None handling and AutoProcessor kwargs forwarding are different call sites and bugs." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 API guidance and a torch.split indexer issue are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Dequantized save failure and invalid GenerationConfig for GLM-5 are separate defects." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "AutoProcessor/cached_file kwargs forwarding is not the same as chat-template crashes on tool-call messages." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "Missing transformers.onnx module and slow/fast tokenizer inconsistency are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Tokenization drift across versions is not the same as multimodal assistant_mask generation." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Both are SAM3-related, but one is a type warning while the other is a missing attribute in vision encoder output." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Dequantized save regression and gguf_file/from_pretrained loading breakage are different code paths." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "DSA indexer activation discussion and GLM-5 RoPE implementation are not the same issue." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe load bug and tokenizer save metadata mismatch are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Two import errors, but for different exported symbols and likely different regressions." + }, + { + "left": "issue:46172", + "right": "issue:46762", + "accept": false, + "reason": "Different numerical issues in different models: a tensor transpose bug versus attention output mismatch." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe loading and position_ids buffer access 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": 28, + "review_unit_index": 53, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "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:43116", + "issue:43122", + "issue:43232", + "issue:43278", + "issue:43493", + "issue:43611", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44188", + "issue:44222", + "issue:44351", + "issue:44483", + "issue:44485", + "issue:44625", + "issue:44779", + "issue:44938", + "issue:44945", + "issue:44995", + "issue:45072", + "issue:45200", + "issue:45216", + "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:47217", + "issue:47312" + ], + "soft_pairs": [ + "issue:45216|issue:45478", + "issue:44038|issue:44222", + "issue:43116|issue:45200", + "issue:43122|issue:44779", + "issue:22355|issue:29127", + "issue:39401|issue:46614", + "issue:43278|issue:44945", + "issue:42175|issue:44938", + "issue:42222|issue:47312", + "issue:36296|issue:43116", + "issue:42757|issue:44351", + "issue:39379|issue:43232", + "issue:43493|issue:45753", + "issue:44188|issue:46328", + "issue:46231|issue:46877", + "issue:28348|issue:46640", + "issue:42898|issue:46614", + "issue:43611|issue:46002", + "issue:44483|issue:47217", + "issue:43976|issue:45446", + "issue:43937|issue:44485", + "issue:42673|issue:45072", + "issue:45357|issue:45478", + "issue:38617|issue:45278", + "issue:44995|issue:46697", + "issue:44625|issue:46002" + ], + "prepared_review_unit_hash": "c32e51b0a0e53f5ee0b561115c95988a1a6df58bd144e508d9627e5ff85e84e8", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 34957, + "estimated_input_tokens": 8740, + "estimated_eval_tokens": 17736 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft pairs look like true duplicates; they mostly share broad subsystems or release/version themes but describe different concrete bugs. The set is heterogeneous, so the best issue anchor is only a loose umbrella, not a real duplicate canonical.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45278 is the broadest regression-style report in the set and the closest umbrella for the import/version-breakage subset.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the least specific and most representative of a cluster-wide regression theme, but it does not actually unify the rest of the issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 pretrained flows, but one is a save_pretrained regression and the other is a from_pretrained loading error; different code paths and symptoms." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Both mention MoE/transformers, but one is a Qwen3-VL-Moe bug and the other is an FP8 save_pretrained issue; not the same defect." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Unrelated problems: example-script multi-label classification returns empty results versus Gemma 4 mm_token_type_ids defaulting behavior." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but for different models and different failure modes (generic tokenization change vs DeepSeek-specific incorrect output)." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "No meaningful overlap: an import error for transformers.onnx versus a LayoutLMv3 error-message clarity issue." + }, + { + "left": "issue:39401", + "right": "issue:46614", + "accept": false, + "reason": "Different tokenizer implementations and bugs: Qwen3 offset_mapping versus Wav2Vec2PhonemeCTCTokenizer stale phonemizer language state." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "One is a dtype regression between train/eval, the other is incorrect output under pipeline parallelism; different failure mechanisms." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Both are import/startup-related, but one is a packaging/backend dependency issue and the other is a Python 3.14 compatibility failure." + }, + { + "left": "issue:42222", + "right": "issue:47312", + "accept": false, + "reason": "Completely different: broken vitpose model files versus kernel-version incompatibility." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor parallel training bug and multi-label example-script behavior are unrelated." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Both are ImportErrors, but for different missing symbols in different packages; likely separate version skew or deprecation issues." + }, + { + "left": "issue:39379", + "right": "issue:43232", + "accept": false, + "reason": "Both involve generation, but one is mllama mask handling and the other is sync_gpus/_update_model_kwargs_for_generation timing; not the same bug." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Different model families and failure types: SigLIP2 implementation mismatch versus Qwen3_5 infinite loop on a specific image." + }, + { + "left": "issue:44188", + "right": "issue:46328", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to a SequenceFeatureExtractor.pad performance issue." + }, + { + "left": "issue:46231", + "right": "issue:46877", + "accept": false, + "reason": "Documentation parameter-name mismatch and import-time docstring diagnostics are separate tooling/doc bugs." + }, + { + "left": "issue:28348", + "right": "issue:46640", + "accept": false, + "reason": "Both mention Flash Attention, but one is a GPT2LMHeadModel support request and the other is an unspecified issue; not the same concrete change." + }, + { + "left": "issue:42898", + "right": "issue:46614", + "accept": false, + "reason": "Tokenization-space cleanup behavior in v5 is unrelated to a phonemizer backend caching bug." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Model loading with base_model_prefix attribute versus NaNs in classification heads are different problems." + }, + { + "left": "issue:44483", + "right": "issue:47217", + "accept": false, + "reason": "Both touch chat/completions-style workflows, but one is request validation on the API and the other is multimodal placeholder placement in VL models." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Both are version-check complaints, but they affect different compatibility gates and different code paths." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "GLM-5 generation config validation and RoPE implementation are distinct issues." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leakage in Qwen3ForCausalLM threads versus dtype mismatches in inference are unrelated." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5 save/load regressions, but one is incorrect visual encoder key serialization and the other is a from_pretrained loading error; too little evidence of a single root cause." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "The latter is a broad import-regression report, but the former is a specific missing-symbol ImportError; not enough evidence they are the same defect." + }, + { + "left": "issue:44995", + "right": "issue:46697", + "accept": false, + "reason": "A stale cache crash on second forward pass is unrelated to image over-padding in an image processor." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "num_labels propagation in Qwen3.5 config and NaNs in classification heads are different configuration/initialization 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": 28, + "review_unit_index": 54, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:33453", + "issue:39379", + "issue:39401", + "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:44075", + "issue:44117", + "issue:44291", + "issue:44355", + "issue:44466", + "issue:44610", + "issue:44704", + "issue:44779", + "issue:44936", + "issue:45071", + "issue:45083", + "issue:45092", + "issue:45405", + "issue:45446", + "issue:45478", + "issue:45563", + "issue:45588", + "issue:46002", + "issue:46231", + "issue:46328", + "issue:46710", + "issue:46726", + "issue:46762", + "issue:46829", + "issue:46909", + "issue:46981", + "issue:47116" + ], + "soft_pairs": [ + "issue:43066|issue:44779", + "issue:45083|issue:45446", + "issue:44038|issue:45478", + "issue:43116|issue:43493", + "issue:44075|issue:44936", + "issue:44610|issue:46909", + "issue:44355|issue:46231", + "issue:43618|issue:46829", + "issue:43066|issue:46710", + "issue:42371|issue:43874", + "issue:46726|issue:46981", + "issue:43116|issue:43825", + "issue:42673|issue:42796", + "issue:43992|issue:44704", + "issue:46328|issue:47116", + "issue:43827|issue:45446", + "issue:39379|issue:43688", + "issue:33453|issue:39401", + "issue:43824|issue:46981", + "issue:43673|issue:46328", + "issue:45092|issue:46762", + "issue:45405|issue:45446", + "issue:44291|issue:45071", + "issue:43064|issue:43673", + "issue:44117|issue:45588", + "issue:43937|issue:45563", + "issue:44466|issue:46002" + ], + "prepared_review_unit_hash": "219ba301a1720bd04e3b01820886436b92fd2b35a731beb0a109b4dd2085edb6", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35106, + "estimated_input_tokens": 8777, + "estimated_eval_tokens": 17810 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: many items are separate v5 regressions or unrelated model/component bugs. The only loosely recurring theme is tokenizer-related regressions, but the soft pairs still don\u2019t look like true duplicates under a strict same-root-cause standard.", + "confidence": 0.66, + "canonical_issue_reason": "Issue 43066 is the clearest representative of the tokenizer-regression theme in this set and is specific enough to anchor the cluster, even though many neighboring issues are not actually duplicates of it.", + "canonical_pr_reason": null, + "best_issue_reason": "43066 is the best single issue to represent the cluster because it is a concrete v5 tokenizer bug and sits closest to the small subset of tokenizer-related reports; the rest are too diverse to justify a different canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they describe different concrete failures and different models; not enough evidence they stem from the same code path." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Different subsystems and failure modes: a qwen3_omni_moe helper vs a flex_attention version check bug." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3-style MoE loading errors, but they target different model variants and distinct from_pretrained failures." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "A classification example returning empty results is unrelated to a SigLIP2 implementation discrepancy." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Both are trainer/optimizer related, but one is about SGD args being ignored and the other is an evaluate-after-train failure." + }, + { + "left": "issue:44610", + "right": "issue:46909", + "accept": false, + "reason": "Processor input-size mismatch and a broken documentation example are unrelated." + }, + { + "left": "issue:44355", + "right": "issue:46231", + "accept": false, + "reason": "Running compiled Python files and a docs parameter-name typo are not the same bug." + }, + { + "left": "issue:43618", + "right": "issue:46829", + "accept": false, + "reason": "Both concern output fields, but they affect different models and different returned attributes; not the same concrete defect." + }, + { + "left": "issue:43066", + "right": "issue:46710", + "accept": false, + "reason": "A tokenizer-decoder regression and an accuracy regression in DeepSeek-R1 are too broad to treat as the same issue." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 configuration guidance and a missing image-processor method are unrelated." + }, + { + "left": "issue:46726", + "right": "issue:46981", + "accept": false, + "reason": "Both involve AutoProcessor, but one is about missing vision backend handling and the other is about repeated 429 retries." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Both mention pipeline/example behavior, but one is a classification script bug and the other is a pipeline API/error-message mismatch." + }, + { + "left": "issue:42673", + "right": "issue:42796", + "accept": false, + "reason": "A VRAM leak in multi-threaded causal LM use is unrelated to a tokenizer BPE marker bug." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "One is a missing weight during UMT5 encoder loading; the other is AutoProcessor not forwarding kwargs to cached_file." + }, + { + "left": "issue:46328", + "right": "issue:47116", + "accept": false, + "reason": "A padding performance issue and a streaming pipeline regression are not the same underlying bug." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() and a flex_attention version check 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 bugs." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "A generic tokenizer-loading regression and Qwen3 offset_mapping are only superficially related." + }, + { + "left": "issue:43824", + "right": "issue:46981", + "accept": false, + "reason": "An import error for a model class and AutoProcessor retry behavior on 429s are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:46328", + "accept": false, + "reason": "Missing generation cache and SequenceFeatureExtractor padding inefficiency are unrelated." + }, + { + "left": "issue:45092", + "right": "issue:46762", + "accept": false, + "reason": "Meta-init incompatibility for InternVL2 and numerical attention differences in Minimax are not the same bug." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A released-version dependency bump and a PyTorch version check bug are unrelated." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Both are v5 loading/configuration regressions, but they concern different failure points and different APIs." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Wrong optimizer state handling in FSDP2/PEFT is unrelated to a missing GenerationMixin cache in chunked prefill." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "A tokenizer mapping assumption and a flash_attention AttributeError are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "An invalid GenerationConfig and a stale warning in paged generate() are different issues." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "Serialization of tied weights depending on device is unrelated to NaNs in classification heads after checkout/init." + } + ] + }, + "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": 28, + "review_unit_index": 55, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:38617", + "issue:39379", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42617", + "issue:43122", + "issue:43208", + "issue:43408", + "issue:43441", + "issue:43450", + "issue:43742", + "issue:43937", + "issue:43986", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44380", + "issue:44410", + "issue:44496", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44877", + "issue:44977", + "issue:45071", + "issue:45081", + "issue:45216", + "issue:45310", + "issue:45341", + "issue:45561", + "issue:45632", + "issue:46077", + "issue:46209", + "issue:46231", + "issue:46298", + "issue:46328", + "issue:46512", + "issue:46752", + "issue:46762", + "issue:46894", + "issue:46897", + "issue:46981", + "issue:47000", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:42222|issue:43441", + "issue:44263|issue:46762", + "issue:42222|issue:47000", + "issue:44380|issue:46077", + "issue:39379|issue:46897", + "issue:42222|issue:46752", + "issue:44117|issue:45081", + "issue:43937|issue:44977", + "issue:42617|issue:45341", + "issue:45071|issue:46512", + "issue:43450|issue:43986", + "issue:39401|issue:44779", + "issue:43408|issue:43742", + "issue:44410|issue:45310", + "issue:44829|issue:45216", + "issue:43208|issue:46894", + "issue:45632|issue:46981", + "issue:45071|issue:47059", + "issue:45341|issue:46328", + "issue:44704|issue:47110", + "issue:45341|issue:45561", + "issue:44877|issue:46231", + "issue:38617|issue:41628", + "issue:39401|issue:46209", + "issue:43122|issue:46298", + "issue:43937|issue:44496", + "issue:42371|issue:44261", + "issue:44261|issue:46762", + "issue:38617|issue:44220" + ], + "prepared_review_unit_hash": "9f8ba4e98084a331476daf5d9bc1bbeb35e5c9ed8252536c639218b3ace3a485", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "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 a lexical bucket of unrelated Transformers issues. None of the soft edges look like the same underlying bug or change, so all should stay separate.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 42222 is the most discussion-heavy concrete regression in the set and the closest thing to a hub, but the cluster is still too heterogeneous to treat it as a true canonical duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 42222 is the best representative issue here because it is concrete, impactful, and has the most activity among the candidates, even though it does not actually duplicate the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "VitPose model breakage and Ministral FlashAttention failure are unrelated bugs." + }, + { + "left": "issue:44263", + "right": "issue:46762", + "accept": false, + "reason": "torch.split indexing behavior and Minimax numerical mismatch are different code paths." + }, + { + "left": "issue:42222", + "right": "issue:47000", + "accept": false, + "reason": "VitPose loading breakage is unrelated to RTDetr loss infeasibility." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT2 attention scaling and Gemma4 bidirectional masks are distinct model-specific bugs." + }, + { + "left": "issue:39379", + "right": "issue:46897", + "accept": false, + "reason": "mllama mask handling and Florence2 double-shift training loss are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:46752", + "accept": false, + "reason": "VitPose model breakage does not match apply_chat_template empty-list crash." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer registry None handling and Mistral regex patch crash differ materially." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "Invalid GenerationConfig and flash-attention generation regression are different failures." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "3d_parallel.py runtime failure is unrelated to a testing_utils documentation bug." + }, + { + "left": "issue:45071", + "right": "issue:46512", + "accept": false, + "reason": "PretrainedConfig type checking and stale config.dtype are not the same bug." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video processor shape and torchvision-less AutoProcessor crash differ." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Qwen3 offset_mapping corruption and Deepseek tokenization regression are separate issues." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "sam3_video/sam3_tracker warning is unrelated to the MobileLLM key error." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing QKV/gate projections and Qwen3.5 MoE from_pretrained failure are different bugs." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "FlashAttention-3 training degeneration and save_pretrained checkpoint regression do not match." + }, + { + "left": "issue:43208", + "right": "issue:46894", + "accept": false, + "reason": "xLSTM small-model blockers and CohereASR double-shift loss bug are unrelated." + }, + { + "left": "issue:45632", + "right": "issue:46981", + "accept": false, + "reason": "Cache-path collision and repeated 429 retries are different AutoProcessor problems." + }, + { + "left": "issue:45071", + "right": "issue:47059", + "accept": false, + "reason": "Config type-checking regression and CLI completion breakage are unrelated." + }, + { + "left": "issue:45341", + "right": "issue:46328", + "accept": false, + "reason": "Testing utils bug and SequenceFeatureExtractor.pad performance issue are separate." + }, + { + "left": "issue:44704", + "right": "issue:47110", + "accept": false, + "reason": "Cached_file kwargs forwarding and special_tokens list handling are different code paths." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "Generic testing_utils note and xdist captured_info.txt race are not the same defect." + }, + { + "left": "issue:44877", + "right": "issue:46231", + "accept": false, + "reason": "Strict config loading and a docs parameter-name mismatch are unrelated." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Missing exported symbol layer_type_validation and AutoImageProcessor import failure are distinct." + }, + { + "left": "issue:39401", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer offset mapping and FP8 inference on B200 are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:46298", + "accept": false, + "reason": "Tokenization regression across versions and noisy logging are not the same issue." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GenerationConfig invalidity and unrecognized model_type are different load-time failures." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API warning and rms_norm_eps precision bug are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:46762", + "accept": false, + "reason": "LayerNorm epsilon precision issue and Minimax attention mismatch are different bugs." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "Import error in configuration_utils is unrelated to fbank feature extraction." + } + ] + }, + "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": 28, + "review_unit_index": 56, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:27957", + "issue:41628", + "issue:42371", + "issue:42831", + "issue:43012", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43208", + "issue:43352", + "issue:43441", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43673", + "issue:43874", + "issue:43931", + "issue:43986", + "issue:44038", + "issue:44075", + "issue:44077", + "issue:44117", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44360", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44829", + "issue:44933", + "issue:44987", + "issue:45003", + "issue:45071", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45464", + "issue:45478", + "issue:45588", + "issue:45656", + "issue:46172", + "issue:46752", + "issue:46897", + "issue:46981", + "issue:47116" + ], + "soft_pairs": [ + "issue:44933|issue:45278", + "issue:44484|issue:45310", + "issue:42371|issue:42831", + "issue:44075|issue:45656", + "issue:44483|issue:45464", + "issue:43097|issue:44493", + "issue:44410|issue:45478", + "issue:43066|issue:43122", + "issue:44263|issue:44485", + "issue:44077|issue:46981", + "issue:43931|issue:44410", + "issue:43673|issue:45341", + "issue:46752|issue:47116", + "issue:41628|issue:45071", + "issue:43931|issue:44038", + "issue:43874|issue:44261", + "issue:43012|issue:43577", + "issue:43441|issue:44829", + "issue:27957|issue:43352", + "issue:43208|issue:46897", + "issue:43097|issue:43452", + "issue:44484|issue:45478", + "issue:44263|issue:46172", + "issue:44360|issue:46172", + "issue:43531|issue:45341", + "issue:44117|issue:44987", + "issue:44261|issue:44485", + "issue:43986|issue:45341", + "issue:45003|issue:45341", + "issue:43097|issue:45588", + "issue:44246|issue:46981" + ], + "prepared_review_unit_hash": "74b60a1a0cc547a0872cd761da7389d387b8f7fdc14ad4059fd8ac8f9adcb1a7", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36419, + "estimated_input_tokens": 9105, + "estimated_eval_tokens": 18466 + }, + "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, with a few loose v5/tokenizer/import regressions and many model-specific bugs. I would not merge any of the soft pairs; they describe different underlying failures or code paths.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 45278 is the broadest representative of the cluster\u2019s dominant theme: post-upgrade Transformers v5 breakages causing import/load regressions. The cluster is heterogeneous, so this is only a loose canonical choice.", + "canonical_pr_reason": null, + "best_issue_reason": "45278 is the most general and cluster-like issue among mostly narrow, model-specific reports. It best captures the shared upgrade-regression theme without being tied to one niche model.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "Both involve import failures, but 44933 is a specific nonexistent image_utils import while 45278 is a broad version-upgrade import regression. Not the same bug." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "Different problems: save_pretrained shard sizing vs Qwen3.5 MoE from_pretrained loading failure." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 configuration warnings/settings and FineGrainedFP8 accuracy issues are different precision-related topics, not one concrete defect." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD arguments being ignored is unrelated to optimizer.step being called twice under deepspeed." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Both touch chat/completions, but one is a request acceptance/regression issue and the other is a streaming failure on a specific model. Different code paths." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Removal of tie_embeddings_and_encoder_decoder and unexpected position-id keys are unrelated breakages." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Different Qwen-family model loading failures with different missing/mismatched components." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Both involve tokenizer changes in v5, but one is wrong decoder type and the other is changed tokenization output; not the same concrete bug." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "torch.split return-value handling in GlmMoeDsaIndexer is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:44077", + "right": "issue:46981", + "accept": false, + "reason": "patchtsmixer post_init policy and repeated 429 retries in AutoProcessor are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Model weight-shape mismatch and missing attn projections are different Qwen-related model bugs." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Generation cache missing in chunked_prefill and a testing_utils bug are unrelated." + }, + { + "left": "issue:46752", + "right": "issue:47116", + "accept": false, + "reason": "Empty-list apply_chat_template crash and streaming pipeline regression are different API failures." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "AutoImageProcessor import failure and PreTrainedConfig type-checking breakage are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both are Qwen3-VL-related, but one is a weight-shape mismatch while the other is a from_pretrained error in a different model variant." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Missing image patch counting and an rms_norm_eps precision issue are not the same bug." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "A compile-time float32 warning and Blip2 dtype fields staying float32 are related only at a very high level, not as one fixable defect." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both mention FlashAttention, but one is a model support/error path and the other is a training-quality regression with flash_attention_3." + }, + { + "left": "issue:27957", + "right": "issue:43352", + "accept": false, + "reason": "One is enabling Flash Attention 2 for XLMRoberta, the other is a model not yet supporting Flash Attention 2.0." + }, + { + "left": "issue:43208", + "right": "issue:46897", + "accept": false, + "reason": "xLSTM training bugs and Florence2 double-shift loss bug are unrelated model-specific issues." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "A removed config attribute and gguf_file/from_pretrained failures are different loader/API problems." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "max_shard_size behavior is unrelated to Qwen3.5 MoE loading errors." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "GlmMoeDsaIndexer split handling and deepseek-v4 transpose/dtype bug are distinct implementation errors." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "Missing ReLU in DSA indexer and deepseek-v4 tensor transpose behavior are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and a testing_utils bug do not share the same underlying defect." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer mapping returning None and a physical-intelligence/fast load failure are only loosely related at the auto-tokenizer layer, not the same bug." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "rms_norm_eps precision issue and GLM-5 RoPE implementation are separate model internals." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "AutoProcessor/torchvision crash and testing_utils bug are unrelated." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "sys.modules access bug in modeling_utils and testing_utils bug are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Removed tie_embeddings_and_encoder_decoder and flash_attention s_aux=None crash are different failures." + }, + { + "left": "issue:44246", + "right": "issue:46981", + "accept": false, + "reason": "Import latency and repeated 429 retries are different performance/network behaviors, not duplicates." + } + ] + }, + "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": 28, + "review_unit_index": 57, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:38617", + "issue:42757", + "issue:42890", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43208", + "issue:43502", + "issue:43550", + "issue:43673", + "issue:43873", + "issue:43874", + "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:44704", + "issue:44936", + "issue:44938", + "issue:45103", + "issue:45375", + "issue:45440", + "issue:45753", + "issue:46050", + "issue:46123", + "issue:46172", + "issue:46178", + "issue:46298", + "issue:46365", + "issue:46369", + "issue:46421", + "issue:46655", + "issue:46664", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821", + "issue:47116", + "issue:47312" + ], + "soft_pairs": [ + "issue:45753|issue:46369", + "issue:45440|issue:46172", + "issue:43097|issue:46050", + "issue:43986|issue:47312", + "issue:44410|issue:45375", + "issue:43874|issue:44263", + "issue:43673|issue:46123", + "issue:43097|issue:43873", + "issue:44038|issue:44410", + "issue:43012|issue:46821", + "issue:44062|issue:44117", + "issue:38617|issue:42757", + "issue:44117|issue:46701", + "issue:43064|issue:44315", + "issue:46655|issue:46664", + "issue:46298|issue:46752", + "issue:46172|issue:46710", + "issue:43986|issue:44484", + "issue:38617|issue:44351", + "issue:42890|issue:46298", + "issue:44704|issue:45103", + "issue:44936|issue:46365", + "issue:43502|issue:43992", + "issue:43208|issue:43550", + "issue:44451|issue:47116", + "issue:44380|issue:46421", + "issue:44938|issue:46178", + "issue:43064|issue:47116", + "issue:42907|issue:45753" + ], + "prepared_review_unit_hash": "a3d27c83dce933b3aa8193f04f08db83bf65e6ca16caeb41a45a18828cc0cfe9", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36020, + "estimated_input_tokens": 9005, + "estimated_eval_tokens": 18266 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a duplicate set: the items are a grab bag of unrelated Transformers issues (imports, model-specific bugs, training regressions, processor/API behavior, and a few versioning problems). The soft-similarity scores are driven by broad vocabulary overlap, not a shared underlying fix.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45753", + "right": "issue:46369", + "accept": false, + "reason": "Both mention model behavior, but one is a Qwen3-specific infinite loop on an image and the other is a Trainer import issue tied to datasets; no shared code path." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Different models and failures: DeepseekV3 implementation divergence vs a deepseek-v4 tensor transpose/dtype bug." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "One is a v5 API removal complaint, the other is a quantization save/load regression; unrelated underlying problems." + }, + { + "left": "issue:43986", + "right": "issue:47312", + "accept": false, + "reason": "Video AutoProcessor crash without torchvision is unrelated to a kernels 0.16 compatibility issue." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen-family vision/model config issues, but one is missing projections in a layer and the other is a silently dropped config field; not the same bug." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Both touch GLM-related code, but one is a missing image processor method and the other concerns torch.split return values in a different component." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Both involve chunked prefill terminology, but one is missing GenerationMixin cache in v5 and the other is a pipeline batching bug that drops results." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Unrelated: tie_embeddings API removal vs offloading failing with quantization." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Different model families and failures; one is a generic transformers 5.0 issue with Qwen3-VL-Moe, the other is missing projections in qwen3next layer 0." + }, + { + "left": "issue:43012", + "right": "issue:46821", + "accept": false, + "reason": "A PyTorch bfloat16 precision warning is not the same as a TimesFM window_size AttributeError." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizers AddedToken construction bug vs tokenizer mapping returning None during from_pretrained; similar area but different failure modes." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they refer to different symbols from different packages and likely different dependency/version breaks." + }, + { + "left": "issue:44117", + "right": "issue:46701", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES/ from_pretrained logic is unrelated to the DiffusionGemmaGenerationOutput image-text pipeline failure." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Different training-stack problems: optimizer state placement with FSDP2/PEFT vs Liger Kernel not applied with model_init." + }, + { + "left": "issue:46655", + "right": "issue:46664", + "accept": false, + "reason": "Both are SmolVLM video-processor issues, but one is padding calculation and the other is num_frames semantics; separate bugs." + }, + { + "left": "issue:46298", + "right": "issue:46752", + "accept": false, + "reason": "Noisy logging and apply_chat_template IndexError are unrelated." + }, + { + "left": "issue:46172", + "right": "issue:46710", + "accept": false, + "reason": "Deepseek-v4 tensor transpose bug is not the same as an accuracy regression after upgrading Transformers." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "AutoProcessor/torchvision crash is unrelated to max_shard_size default behavior in save_pretrained." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import-name failures, but one is missing layer_type_validation and the other is missing HybridCache; different symbols and likely different version mismatches." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "Seed-related test flakiness is not the same as a noisy logging message." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "Passing kwargs to cached_file and auto_docstring crashing under future annotations are unrelated code paths." + }, + { + "left": "issue:44936", + "right": "issue:46365", + "accept": false, + "reason": "trainer.evaluate() after train() failing is unrelated to an import conflict caused by datasets when importing Trainer." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "local_files_only leaking API requests is a networking/cache bug; UMT5Encoder missing embed_tokens.weight is a model loading/state-dict issue." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs for small models and Bamba torch.compile/SDPA failure are different model-specific issues." + }, + { + "left": "issue:44451", + "right": "issue:47116", + "accept": false, + "reason": "Loading a specific model failing and a streaming pipeline regression in v5 are too broad and do not point to the same bug." + }, + { + "left": "issue:44380", + "right": "issue:46421", + "accept": false, + "reason": "GPT2 attention scaling being ignored and cache lazy initialization causing recompilation are separate runtime behaviors." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 import failure and tensorflow_text backend registration error are different compatibility problems." + }, + { + "left": "issue:43064", + "right": "issue:47116", + "accept": false, + "reason": "FSDP2/PEFT optimizer-state corruption and streaming pipeline regression are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Saving dequantized Ministrals/Devstrals and Qwen3 infinite-loop-on-image are different model bugs with no shared fix." + } + ] + }, + "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": 28, + "review_unit_index": 58, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:42175", + "issue:42757", + "issue:42831", + "issue:42890", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43526", + "issue:43550", + "issue:43582", + "issue:43756", + "issue:43824", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44077", + "issue:44162", + "issue:44222", + "issue:44230", + "issue:44261", + "issue:44368", + "issue:44380", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44655", + "issue:44734", + "issue:45478", + "issue:45479", + "issue:45584", + "issue:45712", + "issue:46132", + "issue:46172", + "issue:46298", + "issue:46365", + "issue:47059", + "issue:47116", + "issue:47276" + ], + "soft_pairs": [ + "issue:42907|issue:44655", + "issue:43012|issue:43550", + "issue:45712|issue:46298", + "issue:43066|issue:43906", + "issue:43012|issue:44368", + "issue:43825|issue:44655", + "issue:43992|issue:44484", + "issue:43827|issue:47116", + "issue:43352|issue:43976", + "issue:45584|issue:46365", + "issue:44038|issue:44230", + "issue:44734|issue:47116", + "issue:44162|issue:46172", + "issue:44380|issue:46132", + "issue:43352|issue:44380", + "issue:43208|issue:43526", + "issue:43097|issue:46298", + "issue:44077|issue:46365", + "issue:44380|issue:45479", + "issue:36683|issue:43582", + "issue:43012|issue:47276", + "issue:42757|issue:43824", + "issue:42175|issue:46365", + "issue:44222|issue:45478", + "issue:42890|issue:47059", + "issue:43064|issue:44380", + "issue:43756|issue:44485", + "issue:42831|issue:44261", + "issue:36010|issue:42757", + "issue:44483|issue:44514" + ], + "prepared_review_unit_hash": "10ceace297638a28c73058fe3da40121a58c3685e84c2dd0c290590cb5433e7a", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36554, + "estimated_input_tokens": 9139, + "estimated_eval_tokens": 18534 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is very heterogeneous: the soft pairs mostly share only broad vocabulary (imports, save/load, warnings, pipeline, model families) but not the same concrete failure mode. I found no strong duplicate pair, and there are no PRs to triage here.", + "confidence": 0.19, + "canonical_issue_reason": "issue:36683 is the most actively discussed and clearly scoped runtime bug in the set; among unrelated items, it is a reasonable representative because it has the strongest standalone signal and no better duplicate anchor appears.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:36683 is the best overall issue candidate because it is concrete, reproducible, and heavily discussed; it is also more representative than the many one-off environment/model-specific reports here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both are save-related failures, but one is dequantized model saving and the other is Pipeline.save_pretrained; different code paths and likely different bugs." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "One is a PyTorch warning during bfloat16 compilation, the other is a torch.compile SDPA failure in Bamba; not the same concrete issue." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Left is a repo hygiene / dummy class export issue; right is a logging message complaint. No shared bug." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "The second is only an isolated reproduction reference; there is not enough evidence it is the same tokenizer-decoder bug." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "A compilation precision warning is unrelated to a tie_word_embeddings warning in Qwen fine-tuning." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Pipeline error-message behavior and Pipeline object save_pretrained support are different concerns." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing embed_tokens weight in UMT5 loading is not the same as save_pretrained max_shard_size behavior." + }, + { + "left": "issue:43827", + "right": "issue:47116", + "accept": false, + "reason": "Docs still referencing pipeline() and a streaming pipeline regression are different problems." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Flash Attention support for one model and Python 3.9/3.10 compatibility are unrelated." + }, + { + "left": "issue:45584", + "right": "issue:46365", + "accept": false, + "reason": "Whisper empty transcription failure and an import-trainer dependency issue do not match." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Generic Qwen3-VL-Moe breakage in v5 and fp8 inference support are not the same concrete defect." + }, + { + "left": "issue:44734", + "right": "issue:47116", + "accept": false, + "reason": "A /v1/responses KV-cache indexing crash is not the same as a streaming pipeline regression." + }, + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "ESM2 breakage and a DeepSeek tensor transpose bug are different model-specific failures." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "GPT2 attention scaling being ignored and AttentionInterface.register changing behavior are related to attention, but not the same bug." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "FlashAttention unsupported for Nemotron and GPT2 attention scaling ignored are different code paths and symptoms." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:46298", + "accept": false, + "reason": "A removed config option and a noisy logging message are not the same issue." + }, + { + "left": "issue:44077", + "right": "issue:46365", + "accept": false, + "reason": "patchtsmixer post_init validation and Trainer import breakage due to datasets are unrelated." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention scaling ignored under SDPA/FlashAttention is unrelated to degenerate zero-loss classification behavior." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size and an Apple Silicon TypeError in caching allocator warmup are unrelated." + }, + { + "left": "issue:43012", + "right": "issue:47276", + "accept": false, + "reason": "A bfloat16 compile precision warning is not the same as a misleading synced_gpus warning in continuous batching." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different packages and different fix paths." + }, + { + "left": "issue:42175", + "right": "issue:46365", + "accept": false, + "reason": "Backend selection under pip install extras and a Trainer import conflict with datasets are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained failure and a Qwen3.5 MoE from_pretrained error are different directions and likely different bugs." + }, + { + "left": "issue:42890", + "right": "issue:47059", + "accept": false, + "reason": "A flaky integration test due to missing set_seed calls and a CLI incompatibility with huggingface_hub are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "FSDP2/PEFT optimizer-state corruption and GPT2 attention scaling under SDPA are different failures." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE layer dropping and GLM-5 RoPE implementation are both RoPE-related but not the same bug." + }, + { + "left": "issue:42831", + "right": "issue:44261", + "accept": false, + "reason": "FineGrainedFP8 accuracy drift and MLA q_a_layernorm epsilon handling are separate precision issues." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and different source packages." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "A chat completions API acceptance failure and a batched chat-template crash are different code 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": 28, + "review_unit_index": 59, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:29870", + "issue:33453", + "issue:36296", + "issue:36683", + "issue:38617", + "issue:42994", + "issue:43010", + "issue:43064", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43525", + "issue:43575", + "issue:43611", + "issue:43618", + "issue:43761", + "issue:43867", + "issue:43976", + "issue:43992", + "issue:44075", + "issue:44077", + "issue:44355", + "issue:44380", + "issue:44393", + "issue:44492", + "issue:44625", + "issue:44655", + "issue:44855", + "issue:44908", + "issue:44938", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45310", + "issue:45375", + "issue:45593", + "issue:46077", + "issue:46202", + "issue:46365", + "issue:46512", + "issue:46851", + "issue:46877", + "issue:47000", + "issue:47059", + "issue:47110", + "issue:47317", + "issue:47321" + ], + "soft_pairs": [ + "issue:43618|issue:43761", + "issue:38617|issue:43502", + "issue:45276|issue:46077", + "issue:43299|issue:43867", + "issue:43618|issue:44380", + "issue:42994|issue:44655", + "issue:44655|issue:47321", + "issue:44625|issue:46365", + "issue:43611|issue:44655", + "issue:44625|issue:46077", + "issue:43352|issue:45593", + "issue:36683|issue:43475", + "issue:44355|issue:46877", + "issue:36683|issue:44077", + "issue:43976|issue:44938", + "issue:46077|issue:46512", + "issue:43867|issue:45216", + "issue:44938|issue:47059", + "issue:43618|issue:45216", + "issue:36683|issue:47110", + "issue:36683|issue:43525", + "issue:29870|issue:36296", + "issue:44908|issue:46851", + "issue:43992|issue:47317", + "issue:44355|issue:44855", + "issue:43010|issue:45278", + "issue:43064|issue:44492", + "issue:44075|issue:46365", + "issue:43992|issue:45310", + "issue:43867|issue:46202", + "issue:44393|issue:47000", + "issue:29870|issue:33453", + "issue:36296|issue:43575", + "issue:43867|issue:45375", + "issue:43618|issue:44077" + ], + "prepared_review_unit_hash": "73cca387a9e84d8698db9f08233bf4023a5b11aa42e77399c0d4cacda15d0824", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 37767, + "estimated_input_tokens": 9442, + "estimated_eval_tokens": 19140 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose grab-bag of Transformers v5 regression reports, mostly model-specific load/save/config issues plus a few unrelated runtime/import bugs. None of the soft pairs look like the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 43611 is the broadest and most representative loading-regression report here: it describes a general v5 model instantiation failure, which is closer to the shared theme than the narrower model-specific attribute bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "43611 is the best single representative because it is a clear, general loading regression with a concrete failure mode, whereas the other candidates are mostly narrower, model-specific, or unrelated. ", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP regressions, but one is missing attentions assignment and the other loses hidden_states; different outputs and different failure modes." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "ImportError for a missing symbol is unrelated to local_files_only still triggering network calls." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Same model family, but resize_token_embeddings not updating embeddings is distinct from bidirectional attention still building causal masks." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Qwen3VL MoE load breakage and sorted-state_dict load errors are different concrete loading problems." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "CLIP attentions missing is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure is distinct from Pipeline save_pretrained failure." + }, + { + "left": "issue:44655", + "right": "issue:47321", + "accept": false, + "reason": "Pipeline serialization and AMD quark class updates are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:46365", + "accept": false, + "reason": "Qwen3.5 num_labels propagation has nothing to do with Trainer import failing because of datasets." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Model loading via base_model_prefix is a different code path from saving Pipeline objects." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Different models and different bugs: num_labels propagation vs attention mode handling." + }, + { + "left": "issue:43352", + "right": "issue:45593", + "accept": false, + "reason": "Flash Attention support warning for Nemotron is unrelated to D-FINE auxiliary losses when denoising is off." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Missing vocab_size in Gemma3Config and missing fpn_position_embeddings in SAM3 output are different model-specific attribute regressions." + }, + { + "left": "issue:44355", + "right": "issue:46877", + "accept": false, + "reason": "Compiled-file import errors are not the same as auto_docstring printing doc-lint diagnostics at import time." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Gemma3 config missing vocab_size and patchtsmixer post_init policy are unrelated config issues." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility reports, but they target different versions and different breakages." + }, + { + "left": "issue:46077", + "right": "issue:46512", + "accept": false, + "reason": "Attention-mask semantics in Gemma4 are unrelated to from_config(dtype=...) leaving config.dtype stale." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Sorting state_dict causing load errors is distinct from Qwen3.5 save_pretrained producing the wrong checkpoint." + }, + { + "left": "issue:44938", + "right": "issue:47059", + "accept": false, + "reason": "Python 3.14 loading failure is unrelated to the CLI completion attribute error from huggingface_hub 1.22.0." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIP attentions not assigned and Qwen3.5 checkpoint saving are different subsystems and symptoms." + }, + { + "left": "issue:36683", + "right": "issue:47110", + "accept": false, + "reason": "Gemma3 missing vocab_size and special_tokens being a list instead of a mapping are unrelated attribute-shape issues." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Missing vocab_size in Gemma3Config is not the same bug as Llama4Config missing pad_token_id." + }, + { + "left": "issue:29870", + "right": "issue:36296", + "accept": false, + "reason": "Bart generate UnboundLocalError and tensor-parallel training are different code paths and failure modes." + }, + { + "left": "issue:44908", + "right": "issue:46851", + "accept": false, + "reason": "Scheduler kwargs being ignored is unrelated to Qwen3.5 variable-length position_ids handling." + }, + { + "left": "issue:43992", + "right": "issue:47317", + "accept": false, + "reason": "Missing embed_tokens.weight in UMT5Encoder is unrelated to PPFormulaNet training-loss behavior." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Both involve import-time failures, but compiled Python file errors and @torch.jit.script parsing with comments are different root causes." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "Cache update no_grad decoration and broad import errors after upgrade are not the same bug." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "FSDP2/PEFT optimizer-state corruption is unrelated to a typo in cache strategy names." + }, + { + "left": "issue:44075", + "right": "issue:46365", + "accept": false, + "reason": "SGD args being ignored is unrelated to Trainer import failure caused by datasets." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "UMT5 missing embed_tokens.weight and Qwen3.5 MoE from_pretrained errors are different model-specific loading failures." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "Sorted-state_dict load errors and Gemma pipeline roundtrip failures are different serialization problems." + }, + { + "left": "issue:44393", + "right": "issue:47000", + "accept": false, + "reason": "Qwen3-VL 2D box hallucination/output errors are unrelated to RTDetrLoss cost-matrix infeasibility." + }, + { + "left": "issue:29870", + "right": "issue:33453", + "accept": false, + "reason": "Bart generation UnboundLocalError and tokenizer-loading regression are different bugs." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug and Qwen2-57B-A14B OOM on load are distinct issues." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "State_dict ordering load error and missing Qwen3.5 vision config field are not the same root cause." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIP attentions regression and patchtsmixer post_init policy change 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": 28, + "review_unit_index": 60, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "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:44222", + "issue:44246", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44734", + "issue:44749", + "issue:44964", + "issue:45071", + "issue:45125", + "issue:45357", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45561", + "issue:45850", + "issue:46132", + "issue:46285", + "issue:46617", + "issue:46664", + "issue:47110", + "issue:47317" + ], + "soft_pairs": [ + "issue:41669|issue:44246", + "issue:42890|issue:45561", + "issue:43010|issue:43976", + "issue:43867|issue:45357", + "issue:43010|issue:45850", + "issue:28282|issue:33453", + "issue:44222|issue:45125", + "issue:43618|issue:46285", + "issue:44360|issue:45440", + "issue:42548|issue:44351", + "issue:45125|issue:45478", + "issue:45125|issue:45375", + "issue:43502|issue:44062", + "issue:43525|issue:47317", + "issue:42890|issue:44112", + "issue:42890|issue:44964", + "issue:42898|issue:46664", + "issue:44734|issue:46617", + "issue:45071|issue:47317", + "issue:43723|issue:44556", + "issue:44387|issue:44556", + "issue:43010|issue:43065", + "issue:43824|issue:46617", + "issue:43576|issue:46132", + "issue:43502|issue:44485", + "issue:43874|issue:46617", + "issue:45850|issue:47110", + "issue:30333|issue:33453", + "issue:44355|issue:44749", + "issue:30990|issue:31795", + "issue:43010|issue:43232" + ], + "prepared_review_unit_hash": "1ce4a46ce9616d0e53a8a56028da4969eb9d9a6ba4be42fb3146b3b9cb7c6b6d", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36590, + "estimated_input_tokens": 9148, + "estimated_eval_tokens": 18552 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous transformer issues. The only clear duplicate-style match is the import-time slowdown pair; the rest share broad subsystems but not the same concrete bug.", + "confidence": 0.77, + "canonical_issue_reason": "issue:41669 is the most concrete central bug in the cluster: it identifies a specific import-time performance regression and matches the closest soft duplicate (import takes long).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:41669 is the best representative issue because it is specific, actionable, and anchors the only strong same-bug pair in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41669", + "right": "issue:44246", + "accept": true, + "reason": "Both point to the same import-performance regression: transformers import is slow, with 41669 naming import * usage as the cause." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "Both are test/CI-related, but one is a missing set_seed flake and the other is an xdist file race; different bugs." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "One is a no_grad decoration proposal for cache updates; the other is a Python version compatibility failure." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Both involve model loading/saving, but the concrete failures differ: sorted state_dict loading vs incorrect visual encoder keys." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "One is a specific generation/cache mutation issue; the other is a vague broad complaint about many models breaking." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "Both are loading/import regressions, but one is a missing PyTorch dependency import error and the other is tokenizer loading regression." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "Both mention Qwen/MoE, but one is FP8 save_pretrained behavior and the other is missing tensor-parallel metadata." + }, + { + "left": "issue:43618", + "right": "issue:46285", + "accept": false, + "reason": "Both are CLIP-related structural issues, but they concern different classes and different API surface changes." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "Different code paths and symptoms: a DSA indexer ReLU discussion versus a DeepSeekV3MoE implementation divergence." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different exported symbols and likely different broken exports." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5-MoE related, but the reported failures are different concrete issues with different causes." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Related model family, but one is missing _tp_plan and the other is a config field being dropped; not the same bug." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "Different failure modes: unwanted network access versus AddedToken argument conflict." + }, + { + "left": "issue:43525", + "right": "issue:47317", + "accept": false, + "reason": "Unrelated domains: Llama4 config attribute error versus a FormulaNet training-loss bug." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/test issues, but one is seed-related nondeterminism and the other is a stale device override regression." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Different failures: flaky integration tests versus inability to load a specific model." + }, + { + "left": "issue:42898", + "right": "issue:46664", + "accept": false, + "reason": "Tokenization-space behavior change is unrelated to SmolVLM video frame semantics." + }, + { + "left": "issue:44734", + "right": "issue:46617", + "accept": false, + "reason": "KV-cache continuation crash and trust_remote_code cache FileNotFoundError are different code paths." + }, + { + "left": "issue:45071", + "right": "issue:47317", + "accept": false, + "reason": "Type-checking regression versus a model-specific training-loss bug; no same underlying change." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both involve loading after v5 upgrades, but one is tokenizer loading and the other is checkpoint reload compatibility; too broad to merge." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Memory/OOM regression under int4 quantization is unrelated to checkpoint reload failure." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "No shared bug: cache update no_grad decoration versus a dummy Conv2d in Sam3PixelDecoder." + }, + { + "left": "issue:43824", + "right": "issue:46617", + "accept": false, + "reason": "Missing import for a specific Qwen class is not the same as cache-based trust_remote_code FileNotFoundError." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "A broken CLI env command and AttentionInterface.register behavior change are different issues." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "local_files_only network leakage is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:43874", + "right": "issue:46617", + "accept": false, + "reason": "Missing image-processor method and cache FileNotFoundError are unrelated." + }, + { + "left": "issue:45850", + "right": "issue:47110", + "accept": false, + "reason": "A broad compatibility complaint is not the same as a specific special_tokens attribute error." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow reporting on failed training is unrelated to tokenizer loading regression." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled-Python-file errors and upgrade-induced slowdown during filtering are different problems." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers loading hang and forward-argument documentation confusion are not the same bug." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Both involve generation internals, but one is cache update no_grad and the other is post-sync_gpus kwarg handling." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analyst summary is grounded in the packet and stays conservative on the soft edges. The accepted 41669/44246 pair is a plausible same-bug match on import-time slowdown, and the rejected pairs are mostly supported by clear differences in failure mode or scope. 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": 28, + "review_unit_index": 61, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:42548", + "issue:42757", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43525", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43827", + "issue:43867", + "issue:43901", + "issue:44038", + "issue:44222", + "issue:44230", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44961", + "issue:45071", + "issue:45125", + "issue:45362", + "issue:45375", + "issue:45468", + "issue:45850", + "issue:46095", + "issue:46143", + "issue:46144", + "issue:46209", + "issue:46617", + "issue:46664", + "issue:46726", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:46664|issue:46726", + "issue:43502|issue:44222", + "issue:43299|issue:44556", + "issue:30333|issue:36296", + "issue:29870|issue:31795", + "issue:43827|issue:44961", + "issue:33453|issue:34689", + "issue:41628|issue:42548", + "issue:44222|issue:44230", + "issue:46143|issue:46144", + "issue:43502|issue:44961", + "issue:44355|issue:45071", + "issue:43576|issue:45071", + "issue:43576|issue:43824", + "issue:43576|issue:47059", + "issue:42757|issue:43576", + "issue:44038|issue:45125", + "issue:31795|issue:38617", + "issue:36296|issue:43901", + "issue:44315|issue:44360", + "issue:43475|issue:46617", + "issue:44351|issue:46617", + "issue:44230|issue:46095", + "issue:45375|issue:47110", + "issue:44485|issue:44961", + "issue:44315|issue:46209", + "issue:45362|issue:45850", + "issue:44961|issue:45468", + "issue:43525|issue:47110", + "issue:43673|issue:44492", + "issue:43352|issue:44961", + "issue:43867|issue:47110" + ], + "prepared_review_unit_hash": "2274bc1c8c7853ab3d2acef1199cd8ead0f20759ac640d45f537cf806c4ce248", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36090, + "estimated_input_tokens": 9023, + "estimated_eval_tokens": 18302 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly noisy and only loosely related by Transformers/v5 breakage themes. The strongest representative is the broad v5 regression report, but the soft pairs are not true duplicates and should stay separate.", + "confidence": 0.36, + "canonical_issue_reason": "issue:45850 is the broadest representative of the cluster\u2019s main theme: widespread v5 compatibility regressions affecting many models.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 best captures the cluster\u2019s shared user pain around v5 breaking multiple model-loading paths, even though the cluster is not a tight duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46664", + "right": "issue:46726", + "accept": false, + "reason": "Both involve processors, but one is video-frame semantics and the other is missing-vision-backend handling; different bugs." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "local_files_only network requests and FP8 save_pretrained are unrelated code paths." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both are v5 loading regressions, but they affect different models and likely different root causes." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLFlow reporting on failed jobs is unrelated to tensor-parallel training behavior." + }, + { + "left": "issue:29870", + "right": "issue:31795", + "accept": false, + "reason": "A runtime generate() error is not the same as documentation confusion." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs referencing pipeline() and a one-word issue title are not the same underlying bug." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Different loading regressions affecting different models and likely different mechanisms." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Import failures for different symbols do not indicate one shared duplicate." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "One is FP8 save_pretrained behavior; the other is FP8 inference support." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "kwargs propagation and attention-interface compliance are related but distinct implementation bugs." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "local_files_only network leakage has nothing to do with the unrelated issue title." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file execution errors and PretrainedConfig type checking are separate concerns." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "CLI env command breakage is unrelated to config type-checking regressions." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "A broken env CLI command is not the same as an import error for a model class." + }, + { + "left": "issue:43576", + "right": "issue:47059", + "accept": false, + "reason": "Both are CLI-related, but one is a transformers env command issue and the other is a huggingface_hub completion API mismatch." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing is_offline_mode import and env CLI breakage are unrelated failures." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "General Qwen3-VL-Moe breakage is not the same as missing _tp_plan for tensor parallelism." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Documentation confusion and an ImportError for layer_type_validation are different classes of issues." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training bugs do not duplicate text-classification docs behavior." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel application and a DSA indexer ReLU discussion are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:46617", + "accept": false, + "reason": "SAM 3 Video attribute missing and trust_remote_code cache FileNotFound are different failure modes." + }, + { + "left": "issue:44351", + "right": "issue:46617", + "accept": false, + "reason": "Missing HybridCache import is not the same as cache-file lookup failure." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "FP8 inference support and a weight-conversion-script question are not duplicates." + }, + { + "left": "issue:45375", + "right": "issue:47110", + "accept": false, + "reason": "A dropped config field and a special_tokens list error are unrelated." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation and a title-only issue are not the same bug." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "model_init Liger Kernel application and FP8 inference on B200 are different problems." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "A specific Qwen3.5-35B chat crash is not the same as a broad v5 breakage meta-issue." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "The placeholder title does not match the Gemma-4 audio positional encoding bug." + }, + { + "left": "issue:43525", + "right": "issue:47110", + "accept": false, + "reason": "Llama4Config missing pad_token_id and a special_tokens keys error are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Missing cache in chunked_prefill and a typo in cache strategies are not the same defect." + }, + { + "left": "issue:43352", + "right": "issue:44961", + "accept": false, + "reason": "Nemotron Flash Attention support and an unrelated title are not duplicates." + }, + { + "left": "issue:43867", + "right": "issue:47110", + "accept": false, + "reason": "State_dict sorting load errors and special_tokens handling 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": 28, + "review_unit_index": 62, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43976", + "issue:44077", + "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:44829", + "issue:44861", + "issue:44908", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45230", + "issue:45341", + "issue:45381", + "issue:45542", + "issue:45753", + "issue:45850", + "issue:46018", + "issue:46172", + "issue:46218", + "issue:46328", + "issue:46655", + "issue:46664", + "issue:47059", + "issue:47176" + ], + "soft_pairs": [ + "issue:44273|issue:45542", + "issue:44485|issue:46664", + "issue:44623|issue:45230", + "issue:44273|issue:45003", + "issue:45381|issue:46664", + "issue:41669|issue:43673", + "issue:43525|issue:45850", + "issue:44393|issue:45753", + "issue:44393|issue:45071", + "issue:44466|issue:44556", + "issue:43572|issue:45850", + "issue:44077|issue:44861", + "issue:44556|issue:45092", + "issue:42371|issue:44492", + "issue:45850|issue:47059", + "issue:44492|issue:45850", + "issue:44246|issue:44273", + "issue:43408|issue:44315", + "issue:41669|issue:46218", + "issue:44492|issue:46328", + "issue:42890|issue:44908", + "issue:43976|issue:44393", + "issue:44295|issue:45230", + "issue:43352|issue:44393", + "issue:44297|issue:45230", + "issue:44704|issue:44908", + "issue:43576|issue:44908", + "issue:43576|issue:47176", + "issue:44908|issue:45071", + "issue:44655|issue:44861", + "issue:41669|issue:45341", + "issue:41669|issue:44351", + "issue:43976|issue:46218", + "issue:31795|issue:33453", + "issue:43976|issue:47176", + "issue:46018|issue:46172", + "issue:43519|issue:46655", + "issue:44315|issue:44829", + "issue:44485|issue:47176" + ], + "prepared_review_unit_hash": "daa0e7990b44d21c59096dad7e3e6467658c3b88e3fdb3f955e1c7c75ef98cec", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 39, + "serialized_chars": 38873, + "estimated_input_tokens": 9719, + "estimated_eval_tokens": 19694 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is not a true duplicate cluster: the issues span import/lazy-loading, tokenizer/processor save-load regressions, model-specific config errors, performance problems, and a security bypass. No soft pair looks like the same concrete bug/change closely enough to merge.", + "confidence": 0.67, + "canonical_issue_reason": "issue:44273 is the best representative if one issue must anchor the set: it is open, actively discussed, and the broadest import/lazy-loading regression among the candidates.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44273 has the strongest cluster-central signal and the most active discussion, so it is the most suitable issue to keep as the representative tracker.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Different problems: lazy loading vs a TensorFlow backend/tensorboard install error." + }, + { + "left": "issue:44485", + "right": "issue:46664", + "accept": false, + "reason": "Unrelated model bugs: GLM-5 RoPE implementation vs SmolVLM video frame semantics." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Too vague on the right side; no evidence both describe the same save_pretrained processor bug." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Related import area, but one is lazy loading failure and the other is unsafe sys.modules access; not clearly the same bug." + }, + { + "left": "issue:45381", + "right": "issue:46664", + "accept": false, + "reason": "Both are video-processing issues, but they concern different models and different calculations." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-performance issue vs generation cache missing during chunked prefill; different code paths and symptoms." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Specific missing config attribute vs a broad complaint that many models are broken after v5." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Different Qwen3-VL failures: bounding-box output vs an infinite loop on one image." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Model-specific vision bug vs a generic PretrainedConfig type-checking regression." + }, + { + "left": "issue:44466", + "right": "issue:44556", + "accept": false, + "reason": "Serialization inconsistency of tied weights vs inability to reload old checkpoints; related theme, but not the same concrete bug." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "Specific StableLmConfig pad token regression vs a broad v5 compatibility complaint." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Different failures: optional post_init policy vs AttributeError in tied-weight-key handling." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both involve v5 checkpoint compatibility, but one is general reload failure and the other is an InternVL2 meta-init incompatibility." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 settings issue vs a typo in cache strategy text; unrelated." + }, + { + "left": "issue:45850", + "right": "issue:47059", + "accept": false, + "reason": "Broad v5 model breakage vs a CLI incompatibility with huggingface_hub 1.22.0." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A cache-strategy typo does not match a broad model-breakage regression." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Import slowdown sometimes vs lazy loading malfunction; adjacent theme, but not same bug." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "Model type warning vs Liger Kernel not applied; unrelated subsystems." + }, + { + "left": "issue:41669", + "right": "issue:46218", + "accept": false, + "reason": "Import-star cleanup/performance issue vs missing exported class; not the same underlying defect." + }, + { + "left": "issue:44492", + "right": "issue:46328", + "accept": false, + "reason": "Typo in cache strategy naming vs inefficient padding conversion; unrelated." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Test flakiness from missing set_seed calls vs scheduler kwargs being ignored." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version support issue vs a Qwen3-VL hallucination/bbox bug." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Concrete buffer/position_ids error vs an unspecified bug report." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention support limitation vs a Qwen3-VL output bug." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save_pretrained mismatch vs an unspecified bug report." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "Processor kwargs forwarding vs scheduler kwargs handling; different APIs." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "Broken env CLI command vs inverse_sqrt scheduler kwargs handling." + }, + { + "left": "issue:43576", + "right": "issue:47176", + "accept": false, + "reason": "CLI command breakage vs a shell-operator allowlist bypass; different security and functionality concerns." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler configuration bug vs PretrainedConfig type-checking regression." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Pipeline save_pretrained problem vs tied-weight-key AttributeError; unrelated save/load paths." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Large import slowdown issue vs a small testing_utils bug." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star/lazy-load performance regression vs missing HybridCache export; not the same bug." + }, + { + "left": "issue:43976", + "right": "issue:46218", + "accept": false, + "reason": "Python compatibility issue vs missing Qwen3_5 import; unrelated." + }, + { + "left": "issue:43976", + "right": "issue:47176", + "accept": false, + "reason": "Python version support vs shell command allowlist bypass; different domains." + }, + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "Related to deepseek-v4, but one is a gating logic complaint and the other is a dtype/transpose bug." + }, + { + "left": "issue:43519", + "right": "issue:46655", + "accept": false, + "reason": "Different video bugs: timestamp calculation vs padding calculation." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Training-time Liger Kernel application issue vs flash_attention_3 causing degenerate training." + }, + { + "left": "issue:44485", + "right": "issue:47176", + "accept": false, + "reason": "RoPE implementation discussion vs a shell command substitution bypass; 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": 28, + "review_unit_index": 63, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "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:43717", + "issue:43976", + "issue:44230", + "issue:44263", + "issue:44273", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44861", + "issue:44908", + "issue:44998", + "issue:45230", + "issue:45250", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45850", + "issue:46095", + "issue:46144", + "issue:46218", + "issue:46894", + "issue:47059", + "issue:47110", + "issue:47176", + "issue:47177" + ], + "soft_pairs": [ + "issue:44360|issue:45468", + "issue:45468|issue:46894", + "issue:43576|issue:44273", + "issue:45230|issue:45310", + "issue:43298|issue:46095", + "issue:43976|issue:47177", + "issue:43519|issue:44485", + "issue:44485|issue:47177", + "issue:44273|issue:44908", + "issue:43519|issue:44908", + "issue:31795|issue:34689", + "issue:43576|issue:46218", + "issue:41669|issue:43704", + "issue:44485|issue:45468", + "issue:45850|issue:47177", + "issue:45850|issue:47176", + "issue:44861|issue:47110", + "issue:43704|issue:44485", + "issue:42371|issue:43704", + "issue:43352|issue:47177", + "issue:43502|issue:43519", + "issue:43606|issue:45468", + "issue:42913|issue:43066", + "issue:45850|issue:46218", + "issue:43519|issue:44623", + "issue:44263|issue:45468", + "issue:44861|issue:47059", + "issue:43704|issue:45341", + "issue:43352|issue:46218", + "issue:44351|issue:46218", + "issue:28348|issue:45250", + "issue:44861|issue:45290", + "issue:44908|issue:44998", + "issue:44998|issue:47177", + "issue:44998|issue:47176", + "issue:44492|issue:44998", + "issue:44230|issue:44623", + "issue:28348|issue:43352", + "issue:43618|issue:44998", + "issue:31515|issue:38617", + "issue:44230|issue:45310", + "issue:29870|issue:31515", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:43298|issue:44230", + "issue:43717|issue:46144" + ], + "prepared_review_unit_hash": "acf7b2061bd324d230419e2b71f5e818e92d4841cbd86a29120d446afb20d904", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 47, + "serialized_chars": 40900, + "estimated_input_tokens": 10225, + "estimated_eval_tokens": 20706 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are unrelated issue reports; only the tokenizer v5 regression pair appears to be a plausible duplicate. No PRs are present.", + "confidence": 0.68, + "canonical_issue_reason": "issue:43066 is the most concrete/root-cause-oriented report in the only likely duplicate pair, so it works best as the canonical issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43066 names the specific tokenizer decoder-type regression in v5, making it the most actionable representative of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "Different subsystems: DSA indexer vs Gemma-4 audio positional encoding." + }, + { + "left": "issue:45468", + "right": "issue:46894", + "accept": false, + "reason": "Unrelated model components and bugs; no shared code path." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "CLI env command breakage vs lazy-loading behavior are unrelated." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Generic placeholder report vs a specific Qwen3.5 MoE loading error; not enough to treat as same bug." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Continuous batching audio support and DeepSeek weight conversion are unrelated feature requests." + }, + { + "left": "issue:43976", + "right": "issue:47177", + "accept": false, + "reason": "Python version compatibility issue vs tar/zip-slip security bug." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Timestamp calculation in Qwen3VL vs GLM-5 RoPE implementation." + }, + { + "left": "issue:44485", + "right": "issue:47177", + "accept": false, + "reason": "RoPE math bug vs archive extraction security issue." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading vs inverse_sqrt scheduler kwargs are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Different model/processor bug and scheduler bug." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion vs Llama 3.2 Vision loading regression." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "Broken env command and missing Qwen3_5 module import are distinct." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-star cleanup/performance vs VRAM leak in Qwen3ForCausalLM." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Different model internals and different failure modes." + }, + { + "left": "issue:45850", + "right": "issue:47177", + "accept": false, + "reason": "Broad v5 breakage complaint vs shell/archive security vulnerabilities." + }, + { + "left": "issue:45850", + "right": "issue:47176", + "accept": false, + "reason": "Broad model-regression complaint vs command-substitution allowlist bypass." + }, + { + "left": "issue:44861", + "right": "issue:47110", + "accept": false, + "reason": "Tied-weight key crash vs special_tokens list handling bug." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak in Qwen3ForCausalLM vs GLM-5 RoPE implementation." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 configuration advice vs Qwen3 memory leak." + }, + { + "left": "issue:43352", + "right": "issue:47177", + "accept": false, + "reason": "Flash Attention support gap vs command injection/file-write security issues." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "local_files_only network leak vs timestamp miscalculation." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch vs Gemma-4 audio encoding issue." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": true, + "reason": "Both point to the same tokenizer v5 regression; the decoder-type bug plausibly explains the v4/v5 behavior change." + }, + { + "left": "issue:45850", + "right": "issue:46218", + "accept": false, + "reason": "General v5 compatibility complaint vs missing Qwen3_5 import." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Processor timestamp bug vs missing processor save_pretrained files." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "GLM MoE split/indexing bug vs audio positional encoding bug." + }, + { + "left": "issue:44861", + "right": "issue:47059", + "accept": false, + "reason": "Tied-weight key AttributeError vs CLI completion breakage with huggingface_hub." + }, + { + "left": "issue:43704", + "right": "issue:45341", + "accept": false, + "reason": "Runtime VRAM leak vs a testing_utils bug." + }, + { + "left": "issue:43352", + "right": "issue:46218", + "accept": false, + "reason": "Unsupported Flash Attention 2.0 model vs missing import/module for Qwen3_5." + }, + { + "left": "issue:44351", + "right": "issue:46218", + "accept": false, + "reason": "HybridCache import failure vs Qwen3_5 module import failure." + }, + { + "left": "issue:28348", + "right": "issue:45250", + "accept": false, + "reason": "Specific GPT2LMHeadModel FA2 support request vs a generic Flash Attention 2.0 issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Weight-key AttributeError vs chat-template/tool-call crash." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs bug vs unrelated non-technical placeholder issue." + }, + { + "left": "issue:44998", + "right": "issue:47177", + "accept": false, + "reason": "Unemployment placeholder vs security vulnerability." + }, + { + "left": "issue:44998", + "right": "issue:47176", + "accept": false, + "reason": "Unemployment placeholder vs shell-operator allowlist bypass." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "Cache-strategy typo vs unrelated placeholder issue." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request vs processor save_pretrained file omission." + }, + { + "left": "issue:28348", + "right": "issue:43352", + "accept": false, + "reason": "Both mention Flash Attention 2.0, but one is GPT2-specific support and the other is a different model's unsupported path." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression vs unrelated placeholder issue." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow from_pretrained loading vs ImportError for layer_type_validation." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 inference support request vs Qwen3.5 MoE loading error." + }, + { + "left": "issue:29870", + "right": "issue:31515", + "accept": false, + "reason": "BART generate UnboundLocalError vs slow checkpoint loading." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLFlow job-finalization bug vs slow from_pretrained loading." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Loading performance issue vs tensor-parallel training bug." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Slow checkpoint loading vs forward-API documentation confusion." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Audio continuous batching limitation vs FP8 support request." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba init_weights usage vs RoFormer attention-interface issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: only issues are present, and the sole accepted soft edge (42913/43066) is a conservative duplicate call based on closely related tokenizer-v5 descriptions. The rest of the soft-edge rejections 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": 28, + "review_unit_index": 64, + "review_unit_count": 64, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-524", + "nodes": [ + "issue:37934", + "issue:42936", + "issue:43240", + "issue:43668", + "issue:43813", + "issue:44031", + "issue:46018", + "issue:46144" + ], + "soft_pairs": [ + "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": "0707202ad3d81cead153ea2784e043490a923c4d03e91c897dcc68c2019b2048", + "component_budget": { + "node_count": 524, + "item_count": 524, + "soft_pair_count": 1931, + "serialized_chars": 1001067, + "estimated_input_tokens": 250267, + "estimated_eval_tokens": 500790 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 11, + "serialized_chars": 7771, + "estimated_input_tokens": 1943, + "estimated_eval_tokens": 4142 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: the issues cover unrelated models, tokenizer warnings, config typing, a typo, a loss-kwargs propagation bug, and an attention-interface mismatch. None of the soft pairs looks like the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "If a single issue must anchor the cluster, issue 42936 is the most concrete and discussion-heavy runtime bug, but it is not actually a duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 42936 is the strongest representative only because it is a clear behavioral bug with the most discussion; the rest are mostly unrelated, one-off reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Different bugs in different models: Mask2Former ignore_value handling vs DeepSeek-V4 shared-expert gating." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "Config type-hint problem vs tokenizer regex warning after a version upgrade; unrelated failure modes." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Mask2Former ignore_value usage is unrelated to kwargs not being forwarded to a loss function." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss-kwargs plumbing bug and RoFormer attention-interface issue affect different code paths." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Different subsystems and symptoms: model ignore_value behavior vs tokenizer regex warning." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "No shared underlying bug; one is a loss-parameter issue, the other is an attention implementation concern." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Mask2Former loss handling and a typo in quantization code are not duplicates." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Incorrect type hint in ModernBERTConfig is unrelated to a typo in a different conversion path." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Loss kwargs forwarding and config annotation accuracy are separate issues." + }, + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "A question about Llama4TextL2Norm semantics is unrelated to a ModernBERTConfig type hint bug." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Different models and different defects: unused ignore_value vs incorrect norm_eps typing." + } + ] + }, + "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": 28, + "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": "8f85390143372ef7f6576cd3af9794c63b155bd07693e324001322ffb3eedc8e", + "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 not duplicates of one another: one is about tokenizer truncation behavior, one is a feature request for ReduceLROnPlateau-like scheduler behavior, and one is about extracting word start/end timestamps from Wav2Vec2 outputs.", + "confidence": 0.98, + "canonical_issue_reason": "Issue #10519 is the most concrete and narrowly scoped of the set, so it is the best representative issue if one must be chosen, though it is not actually the same problem as the others.", + "canonical_pr_reason": null, + "best_issue_reason": "#10519 is the best issue to anchor the cluster because it is specific and actionable; the other two are separate feature requests in different areas.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Tokenizer truncation behavior and Wav2Vec2 word-boundary timing are unrelated problems in different code paths." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "A truncation-option request for tokenizers is unrelated to a scheduler-feature request." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "These are separate feature requests: learning-rate plateau scheduling versus speech word time offsets." + } + ] + }, + "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": 28, + "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": "c2faf18d4344a4ae2c8299aac85704516576eafc8ab61e40cbe9b89f5faa3f1f", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2479, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2479, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These three issues do not describe the same bug: one is about TAPAS tokenization mismatching TensorFlow, one is about a private-model download/auth error message, and one is a generic model-loading failure for bert-base-uncased. No duplicate cluster is supported here.", + "confidence": 0.89, + "canonical_issue_reason": "Issue #16618 is the broadest model-loading report and the closest thing to a representative issue in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the most generally phrased and likely most reusable anchor, while #13244 and #15980 are clearly separate concerns.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tokenization parity for TAPAS is unrelated to a generic model-load failure for bert-base-uncased." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Private-model auth/error-message behavior is a different problem from failing to load a public model." + } + ] + }, + "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": 28, + "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-26350-2", + "nodes": [ + "issue:26350", + "issue:33260" + ], + "soft_pairs": [ + "issue:26350|issue:33260" + ], + "prepared_review_unit_hash": "52cc011f3c96df745ad6ae9ab5630c850f8c3a2c1bbe33fbf93f63c45cb50f7d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate community contribution requests for different capabilities: Flash Attention 2 support across more architectures vs GGUF support across more architectures. They are not duplicate reports of the same underlying issue.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue: the issues target different features and code paths, so neither can represent the other.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a strong global representative for the pair because they are unrelated requests; if forced to pick by activity alone, issue 26350 has the longer discussion history, but it is not a duplicate of 33260.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different feature requests with different targets (Flash Attention 2 architecture support vs GGUF architecture support); not the same underlying 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": 5, + "component_count": 28, + "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-27088-10", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370", + "issue:46222", + "issue:46226", + "issue:46931", + "issue:47299" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:47299", + "issue:27088|issue:38435", + "issue:28936|issue:43834", + "issue:27088|issue:28936", + "issue:28936|issue:44365", + "issue:28936|issue:38435", + "issue:28936|issue:44370", + "issue:38435|issue:44370", + "issue:38435|issue:44365", + "issue:38435|issue:43834", + "issue:38435|issue:47299", + "issue:38435|issue:46226", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:43834|issue:47299", + "issue:43834|issue:46226", + "issue:43834|issue:46931", + "issue:44365|issue:44370", + "issue:44365|issue:47299", + "issue:44365|issue:46226", + "issue:44365|issue:46931", + "issue:44370|issue:47299", + "issue:44370|issue:46226", + "issue:44370|issue:46931", + "issue:46222|issue:47299", + "issue:44370|issue:46222", + "issue:44365|issue:46222", + "issue:43834|issue:46222", + "issue:46222|issue:46226", + "issue:46226|issue:46931", + "issue:46226|issue:47299", + "issue:46931|issue:47299" + ], + "prepared_review_unit_hash": "6364a7874badb5b21681623ddacd740668e8d6acdc257c532cb05d8b33ad095e", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 35, + "serialized_chars": 18659, + "estimated_input_tokens": 4665, + "estimated_eval_tokens": 9586 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 35, + "serialized_chars": 18659, + "estimated_input_tokens": 4665, + "estimated_eval_tokens": 9586 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly repeated i18n-doc template issues and some noisy/spam-like variants. They are textually similar, but they are not true duplicates because they refer to different languages or unrelated gibberish requests.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43834 is the cleanest generic template instance and best represents the repeated i18n-doc translation pattern, but it is still not a duplicate of the language-specific issues.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43834 is the best cluster representative: it is the most generic and centrally similar to the other template-like translation issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Both are i18n-doc translation requests, but one is Turkish-specific and the other is a generic template; same category, not the same issue." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Different language targets; the similarity comes from the shared template, not a shared underlying request." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Turkish-specific request vs generic placeholder template; not the same concrete issue." + }, + { + "left": "issue:27088", + "right": "issue:47299", + "accept": false, + "reason": "These are not the same request; one is a Turkish docs translation issue, the other is a noisy/spam-like generic translation title." + }, + { + "left": "issue:27088", + "right": "issue:38435", + "accept": false, + "reason": "Turkish and Romanian translation requests are separate issues, despite matching template language." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish translation request vs generic template issue; similar wording only." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Turkish and Spanish docs translations are distinct requests, not duplicates." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Different language targets; same template, different underlying issue." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Spanish vs Romanian translation requests are separate issues." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Same broad translation theme, but not the same language-specific issue." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian-specific request and a generic template issue are not duplicates." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Romanian translation request vs another template instance; similarity is superficial." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Different concrete request; one is language-specific, the other is generic." + }, + { + "left": "issue:38435", + "right": "issue:47299", + "accept": false, + "reason": "Romanian translation issue and spam-like generic translation title do not match the same underlying change." + }, + { + "left": "issue:38435", + "right": "issue:46226", + "accept": false, + "reason": "These are not the same translation request; the titles are only loosely similar." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "Both are placeholder-style translation issues, but they are separate instances rather than one duplicate." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "Highly similar templates, but no evidence they describe the same concrete task." + }, + { + "left": "issue:43834", + "right": "issue:47299", + "accept": false, + "reason": "Generic translation template vs spam-like variant; not the same issue." + }, + { + "left": "issue:43834", + "right": "issue:46226", + "accept": false, + "reason": "These look like repeated template noise, not one mergeable duplicate issue." + }, + { + "left": "issue:43834", + "right": "issue:46931", + "accept": false, + "reason": "Different language/request content; the overlap is only in template structure." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "Both are generic translation-template issues, but they appear to be separate instances, not the same bug/change." + }, + { + "left": "issue:44365", + "right": "issue:47299", + "accept": false, + "reason": "Template similarity is high, but the underlying request text is not the same." + }, + { + "left": "issue:44365", + "right": "issue:46226", + "accept": false, + "reason": "No clear shared concrete request beyond generic translation wording." + }, + { + "left": "issue:44365", + "right": "issue:46931", + "accept": false, + "reason": "Different content and no evidence of a shared underlying issue." + }, + { + "left": "issue:44370", + "right": "issue:47299", + "accept": false, + "reason": "Generic translation template vs noisy variant; not a duplicate." + }, + { + "left": "issue:44370", + "right": "issue:46226", + "accept": false, + "reason": "These are separate template-like issues, not one specific task." + }, + { + "left": "issue:44370", + "right": "issue:46931", + "accept": false, + "reason": "No indication they describe the same concrete translation request." + }, + { + "left": "issue:46222", + "right": "issue:47299", + "accept": false, + "reason": "Both are noisy translation-like issues, but the titles/content do not match as the same request." + }, + { + "left": "issue:44370", + "right": "issue:46222", + "accept": false, + "reason": "Template similarity only; one is generic English, the other is unrelated Thai-like text." + }, + { + "left": "issue:44365", + "right": "issue:46222", + "accept": false, + "reason": "Surface similarity from repeated template wording, but not the same underlying issue." + }, + { + "left": "issue:43834", + "right": "issue:46222", + "accept": false, + "reason": "Generic translation template vs unrelated Thai text; too weak to merge." + }, + { + "left": "issue:46222", + "right": "issue:46226", + "accept": false, + "reason": "Both are noisy/non-English variants, but there is no clear proof of the same concrete request." + }, + { + "left": "issue:46226", + "right": "issue:46931", + "accept": false, + "reason": "Different titles and no shared concrete task beyond superficial similarity." + }, + { + "left": "issue:46226", + "right": "issue:47299", + "accept": false, + "reason": "Spam-like or template-like wording is similar, but the actual request is not the same." + }, + { + "left": "issue:46931", + "right": "issue:47299", + "accept": false, + "reason": "These are unrelated translation-like issues with only weak textual overlap." + } + ] + }, + "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": 28, + "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-36718-21", + "nodes": [ + "issue:36718", + "issue:36817", + "issue:36879", + "issue:37015", + "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:36879|issue:37015", + "issue:37015|issue:42503", + "issue:41115|issue:43837", + "issue:41115|issue:44121", + "issue:41115|issue:42503", + "issue:41115|issue:45027", + "issue:41211|issue:43671", + "issue:42503|issue:43837", + "issue:42503|issue:43671", + "issue:42503|issue:42954", + "issue:42738|issue:45027", + "issue:42503|issue:42738", + "issue:42738|issue:42954", + "issue:42738|issue:46653", + "issue:42954|issue:44205", + "issue:42954|issue:43671", + "issue:42954|issue:45177", + "issue:42954|issue:46653", + "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:43837|issue:45027", + "issue:37015|issue:45027", + "issue:43837|issue:45177", + "issue:45322|issue:46133", + "issue:42503|issue:46653", + "issue:45027|issue:46653" + ], + "prepared_review_unit_hash": "b8d753ce1f0fb2d1cd2aabf936063e48bc3626cdc975c26214eb728496fa8ed1", + "component_budget": { + "node_count": 21, + "item_count": 21, + "soft_pair_count": 48, + "serialized_chars": 28616, + "estimated_input_tokens": 7154, + "estimated_eval_tokens": 14564 + }, + "budget": { + "node_count": 21, + "item_count": 21, + "soft_pair_count": 48, + "serialized_chars": 28616, + "estimated_input_tokens": 7154, + "estimated_eval_tokens": 14564 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This is a loose cluster of model-request issues, not true duplicates. They share a generic 'add support/config/model' pattern, but the concrete models, tasks, and code paths are mostly unrelated, so every soft edge should be rejected.", + "confidence": 0.96, + "canonical_issue_reason": "If a single anchor is needed, issue 42503 ('Add ModernVBERT models') is the broadest and most representative model-addition request in the set, but it does not subsume the others as duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 42503 is the most umbrella-like and central to the cluster\u2019s recurring theme of adding model support/configuration.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36718", + "right": "issue:47004", + "accept": false, + "reason": "Related CV keypoint requests, but RoMa matcher and LoMa detector/matcher are different concrete model additions." + }, + { + "left": "issue:36718", + "right": "issue:41115", + "accept": false, + "reason": "Different domains and changes: keypoint matching vs adding MiniCPM3 architecture support." + }, + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Both are model-addition requests, but EuroBert config and DEIMv2 are unrelated concrete models." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert config and Qwen3-TTS support are unrelated model families and code paths." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "Different model requests: EuroBert config vs RF-DETR support." + }, + { + "left": "issue:36817", + "right": "issue:42954", + "accept": false, + "reason": "EuroBert config and XiaoMi MiMo are distinct model additions, not the same change." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "Both are vision-model requests, but RF-DETR and DEIMv2 are separate concrete integrations." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "RF-DETR detection support and LFM2.5 Audio support are unrelated." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR model support and the weight-sparse transformer request address different code paths." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are unrelated model additions." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR and Qwen3-TTS are distinct model support requests." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT support and Qwen3-ASR support are different model integrations." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT model addition and the circuit-sparsity transformer request are not the same underlying change." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT and LFM2.5 Audio 1.5B are unrelated model requests." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are separate model-addition issues sharing only the generic template." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are different concrete model additions, not duplicates." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are unrelated." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 model architecture and circuitgpt support do not share the same code-path change." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT are distinct model requests." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 support and Voxtral TTS support are unrelated model integrations." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 support and Qwen3-TTS support are different model additions." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT models and Qwen3-ASR support are unrelated concrete changes." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are separate model requests." + }, + { + "left": "issue:42503", + "right": "issue:42954", + "accept": false, + "reason": "ModernVBERT models and XiaoMi MiMo are distinct additions with no shared underlying bug." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like RoPE support and Voxtral TTS support are unrelated." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "Both are broad model-related requests, but one asks for ModernVBERT support and the other for RoPE handling in BERT-like models." + }, + { + "left": "issue:42738", + "right": "issue:42954", + "accept": false, + "reason": "BERT-like models with RoPE and XiaoMi MiMo are different model concerns." + }, + { + "left": "issue:42738", + "right": "issue:46653", + "accept": false, + "reason": "BERT-like RoPE support and the Mirko Privitera issue are unrelated; the latter is not a model request." + }, + { + "left": "issue:42954", + "right": "issue:44205", + "accept": false, + "reason": "XiaoMi MiMo and SAM3-LiteText are different model additions." + }, + { + "left": "issue:42954", + "right": "issue:43671", + "accept": false, + "reason": "MiMo and Qwen3-TTS are unrelated model families." + }, + { + "left": "issue:42954", + "right": "issue:45177", + "accept": false, + "reason": "MiMo and DeepSeek-OCR2 are distinct requests." + }, + { + "left": "issue:42954", + "right": "issue:46653", + "accept": false, + "reason": "XiaoMi MiMo and the Mirko Privitera issue are unrelated." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Qwen3-TTS and Qwen3-ASR are related by vendor and modality, but they are different concrete support changes and should not be merged." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Both involve audio models, but Qwen3-TTS and Voxtral TTS are different model integrations." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-TTS support and circuitgpt support are unrelated." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS and DeepSeek-OCR2 are different model requests." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Both are audio-related, but Qwen3-ASR and LFM2.5 Audio 1.5B are separate concrete model supports." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR support and EUPE support are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-ASR support and the weight-sparse transformer request are different code paths." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are distinct model support requests." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT support and Voxtral TTS support are unrelated model additions." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model requests." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "EUPE and TIPSv2 are both vision-language research models, but they are separate concrete additions." + }, + { + "left": "issue:42503", + "right": "issue:46653", + "accept": false, + "reason": "ModernVBERT models and the Mirko Privitera issue are unrelated; the latter is not a model request." + }, + { + "left": "issue:45027", + "right": "issue:46653", + "accept": false, + "reason": "Voxtral TTS support and the Mirko Privitera 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": 28, + "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": "2608ef635619629315066ddcfc213b58426237a38c057cfcda16267c956799ba", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1506, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1506, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one is a docs request about adding Pruna, the other is a vision RoPE refactor. The soft match should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "No canonical issue: the issues describe unrelated work (documentation/feature listing vs a vision model refactor).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46443 is the more substantive technical issue; issue:38740 is a documentation request and not a good duplicate anchor for this cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38740", + "right": "issue:46443", + "accept": false, + "reason": "Different topics and scopes: a docs addition request versus a vision RoPE refactor; no shared 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": 28, + "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": "20efbfdefd64bdcca6557f20fa91741c46f901f55c5bb1df02dcae3bccd62fb5", + "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 look like two different feature requests, not duplicates: one asks for a block decoding setting, the other asks for DeepSeek v3.2 support.", + "confidence": 0.93, + "canonical_issue_reason": "Issue #43037 is the more specific, user-facing request and is the better representative of the cluster if one must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "#43037 is the best issue candidate because it clearly states a concrete model-support request, whereas #41084 is more generic and ambiguous.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "The titles describe different underlying changes: block decoding configuration vs support for a specific model version. No shared target or code-path evidence suggests duplication." + } + ] + }, + "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": 28, + "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": "6a550ff73ad490d306cc553c5a8b91812c5cfb360d7c7d536b050b1591a76fa5", + "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": "The two issues describe different feature requests in different areas: one is about enabling TiledFusedLogitsLoss/FLCE via `from_pretrained()`, while the other is about integrating TiledMLP for lower memory usage. They do not appear to be duplicates or the same underlying change.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41306", + "right": "issue:41826", + "accept": false, + "reason": "Different requested functionality and code paths: FLCE configuration in `from_pretrained()` vs. a TiledMLP memory-footprint integration. Shared ML/model-efficiency theme is too broad to treat as 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": 10, + "component_count": 28, + "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": "dff91ce0c664ed7506f4bfc626ee53cc8f794ef0c61efde14caa39ebc881d9c1", + "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": "The two issues are related only at a high level: both concern `Trainer` not handling custom trainers well, but they describe different problems and likely different fixes. I would not merge them as duplicates.", + "confidence": 0.87, + "canonical_issue_reason": "Issue #43598 is the broader and more central customizability complaint about `Trainer` behavior, so it works better as the representative issue in this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "#43598 is the best cluster representative because it captures the more general Trainer extensibility concern, whereas #43600 is a narrower FLOPS-estimation case.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "They share the theme of custom trainers, but one is about `remove_unused_column` customizability and the other is about FLOPS estimation logic. Different code paths and different fixes, so 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": 11, + "component_count": 28, + "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": "358e64c354e7671df4cb5810651e7f2485ed375887bd0732fd148cbfcd038c8c", + "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 very high level (Qwen auto-class support) but they target different model families and different auto-class behaviors, so they should not be treated as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "issue:43684 is the broader, more central registration request and is a better representative of the Qwen auto-model support theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43684 is the best single issue to keep as the representative because it covers a wider API registration gap rather than a narrower task-specific extension.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying requests: Qwen3-Omni auto-registration for general/conditional generation vs Qwen3.5 sequence classification support. Same ecosystem, but not 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": 12, + "component_count": 28, + "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": "3ccb2674d358644f097cd1ce5c4cf43770fabf212b4a6b3a6b0e6d4768d4858f", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The three \"huggingface \" issues look related by title and timing, but the packet does not provide enough concrete evidence to confirm duplicates; \"Fett\" appears unrelated. I would use issue 43845 as a provisional representative only because it is the earliest of the similar issues, not because duplicate status is proven.", + "confidence": 0.59, + "canonical_issue_reason": "issue:43845 is the earliest of the similar \"huggingface \" issues, so it is the best provisional canonical representative if they are later confirmed to be duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43845 is the strongest single issue candidate because it is earliest among the closely similar \"huggingface \" reports, though the evidence is only metadata-level similarity.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "The titles are different and the packet provides no body text, links, or shared targets showing the same bug." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "\"Fett\" and \"huggingface \" do not clearly point to the same underlying issue, and there is no concrete linking evidence." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "The titles are unrelated and the packet gives no explicit or structural evidence that they are duplicates." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": false, + "reason": "The titles are nearly identical, but the packet only shows soft similarity plus close timestamps; that is not enough to confirm a duplicate." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "These two issues look similar by title and timing, but there is no body text or other concrete evidence proving the same report." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "Near-identical titles alone are insufficient here; the packet does not show explicit links, shared targets, or other proof of duplication." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The grounding is mostly okay, but the soft-edge verdicts are too strong for the evidence in the packet. The packet only shows near-identical titles, similar body lengths, and close creation times; it does not show body text, explicit links, or other concrete proof that these are duplicates. In particular, the claims about a \"title/body pattern\" and that the huggingface issues \"strongly suggest duplicate reports\" overstate the evidence. Since all three huggingface pairs are only supported by soft_similarity and the deterministic_accept flag is false, those edges should be treated more conservatively." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 28, + "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": "fd13b095b8bc320c77fad5f92e9d3816e76fa0c79a539ba7f31a7bb602be7c57", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 12, + "serialized_chars": 7361, + "estimated_input_tokens": 1841, + "estimated_eval_tokens": 3938 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 12, + "serialized_chars": 7361, + "estimated_input_tokens": 1841, + "estimated_eval_tokens": 3938 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is dominated by spammy or extremely generic issues with no concrete shared bug report; the titles and minimal bodies do not support duplicate merging. The only somewhat substantive item is the generic documentation suggestion, but it still does not match the others.", + "confidence": 0.97, + "canonical_issue_reason": "issue:44418 is the only item with a non-spam, legible title and slightly more plausible issue content; the rest look like low-signal spam placeholders.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44418 is the best representative issue in this set, but only as a generic placeholder\u2014not because it is a true duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both look like spam placeholders; no concrete shared bug or change is described." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Both look like spam placeholders with no substantive overlap beyond generic similarity." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "A spam placeholder and a generic documentation suggestion are not the same underlying issue." + }, + { + "left": "issue:44371", + "right": "issue:47036", + "accept": false, + "reason": "No shared bug/change; one is spam-like and the other is a vague community tip." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder vs. unrelated named-topic issue; no duplicate evidence." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "A generic docs suggestion and a spam placeholder are not the same underlying report." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "No concrete overlap; the titles point to unrelated low-signal issues." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Both are spam placeholders, but there is no evidence they report the same specific problem." + }, + { + "left": "issue:44572", + "right": "issue:47036", + "accept": false, + "reason": "The items are unrelated in content and intent; similarity is only superficial." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "No shared bug or code path is described, so they should not be merged." + }, + { + "left": "issue:44573", + "right": "issue:47036", + "accept": false, + "reason": "The issues do not describe the same underlying problem." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Only generic similarity is present; no duplicate-level evidence." + } + ] + }, + "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": 28, + "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": "c90f232a04742d55c9bcd9fc918a6a904c648986e70561012c140beacdd71d88", + "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": "The two closed issues are highly similar and #45145 is the best canonical candidate, but the available evidence only supports a soft similarity match rather than a confirmed duplicate.", + "confidence": 0.78, + "canonical_issue_reason": "#45145 is the better canonical target because it was created earlier and has an inbound reference; however, the packet does not prove the two reports are identical beyond strong title/body similarity.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is the strongest single representative because it is earlier and already referenced, while #45151 appears closely related but not definitively the same report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "The issues have identical titles and similar lengths, but the packet only provides soft-similarity evidence. That is not enough to confirm they describe the exact same underlying issue or exclude scope differences." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The canonical choice (#45145) is grounded by the earlier timestamp and inbound reference, but the duplicate/merge verdict is too strong for the evidence given. The packet only shows a soft-similarity match with identical titles and similar body lengths; it does not directly establish identical content or the absence of scope differences. Please soften the language and avoid presenting the pair as confirmed duplicates." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 15, + "component_count": 28, + "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": 28, + "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": "c5437194a67a708343378d870f8d5a9422e0be379376a67755d4a03170696374", + "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": "Two closed issues with unrelated-looking personal-name titles and no body content; there is no evidence they describe the same underlying problem.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good canonical representative: both are bare, distinct-name entries with no description or shared technical context.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "The titles reference different people and dates, with no body, links, or technical details suggesting the same underlying bug or report." + } + ] + }, + "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": 28, + "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": "72a0e84faac6331a6d79c72bfa71456961085c12cb6a05ff489ffe9bd0cbf1b4", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are just separate name/date entries with no shared subject matter, body, or linked code context. They do not look like duplicates or the same underlying bug/report.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good canonical candidate; both are unrelated standalone entries with no evidence of a shared problem or duplicate relationship.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "Different names and different dates; no overlap in content, targets, or evidence of 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": 18, + "component_count": 28, + "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": "65816f935ab5762d1027f0063d21928f9e38f712f6d1f76e94c7941d8680e219", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1542, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1542, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are unrelated: one requests a new Qwen3_5 token-classification/value-model feature, while the other asks for FP8 training support. There is no evidence they describe the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "issue #46295 is the better cluster anchor because it is the broader, more generally scoped request; #45810 is a model-specific feature request that does not align with FP8 training support.", + "canonical_pr_reason": null, + "best_issue_reason": "issue #46295 is the most suitable issue to represent the cluster since it is more general and likely to attract broader discussion, whereas #45810 is narrowly about adding one model head.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45810", + "right": "issue:46295", + "accept": false, + "reason": "Different topics and code paths: adding Qwen3_5ForTokenClassification for a value model vs enabling FP8 training support. They do not look like 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": "issue", + "component_index": 19, + "component_count": 28, + "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": "351b2400ac403bf973088c758aeef969368a0da609df8c1693d0dafc4d3bc8df", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5726, + "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 contains one loosely related dataset post and two highly similar issue pairs, but the packet only provides soft_similarity evidence. That is not enough to assert deterministic duplicate pairs or a single canonical issue.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Same broad finance/data theme, but 45869 is a dataset/sample bundle post and 45870 is an example request; not the same concrete issue." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Related domain only; 45869 is about a dataset bundle, while 45871 is an example request." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": false, + "reason": "Very similar titles and body size, but the packet only shows soft_similarity and no stronger linkage, so this is not safe to treat as a confirmed duplicate." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Both mention DataForge and finance, but one is a prediction example and the other is a scraping-service request; different underlying change/request." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Same broad project, but prediction example vs scraping-service request are different concrete issues." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Different concrete request types: prediction example versus scraping-service request." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Different concrete request types: prediction example versus scraping-service request." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": false, + "reason": "Highly similar titles/body size, but only soft_similarity evidence is present; insufficient to confirm a duplicate pair." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet only supports soft similarity, not exact duplication. The analyst overstates the evidence by calling 45870/45871 and 45872/45873 \"exact duplicate\" pairs and by selecting a best issue on that basis. The conservative reading is that these are highly similar issue pairs, but the packet does not provide deterministic accept evidence or any stronger linkage. 45869 being only loosely related is fine, but the summary and verdicts are too strong overall." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 20, + "component_count": 28, + "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": "9c5a8219604400fa091d729cf9230a8fe8551cd1f6f1ee7ccd013493f8df95d7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1640, + "estimated_input_tokens": 410, + "estimated_eval_tokens": 1076 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1640, + "estimated_input_tokens": 410, + "estimated_eval_tokens": 1076 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are thematically related, but the packet only supports title-level similarity. There isn\u2019t enough evidence to treat them as the same regression or the same failure mode, so the soft duplicate edge should be rejected.", + "confidence": 0.56, + "canonical_issue_reason": "Issue 46439 is the broader and more general report: static-cache early initialization corrupting linear-attention layers via shape corruption.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46439 is the best canonical issue for the cluster because it describes the underlying static-cache/linear-attention problem more generally, while 46441 appears to be a more model-specific crash report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46439", + "right": "issue:46441", + "accept": false, + "reason": "Reject: the packet only shows related static-cache/linear-attention wording. One issue reports wrong-shape `RuntimeError` during early initialization, the other reports a Qwen3.5 `KeyError: 'linear_attention'`; without shared targets, files, or body details, this is not strong enough to call them the same bug." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The pairing is not supported strongly enough by the packet. The titles suggest related static-cache / linear-attention crashes, but the analyst overstates the evidence by treating them as the same regression and same failure mode. With only title-level similarity and no body details or shared targets/files, the soft-edge verdict should be more conservative (e.g., undecided rather than accepted)." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 21, + "component_count": 28, + "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": "1467a789812560af1fbbe88c2234b200aa5717c3b021c80eaef6ca27df159854", + "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": "These two issues are exact duplicates: same title, same underlying Wav2Vec2CTCTokenizer v5 special-token regression, and identical evidence. The open issue is the better canonical target.", + "confidence": 0.99, + "canonical_issue_reason": "issue #46552 is the same report but still open and already has inbound references, so it should serve as the canonical issue.", + "canonical_pr_reason": null, + "best_issue_reason": "#46552 is the best issue to keep because it is open, identical in content, and already attracting references; #46551 can be treated as the duplicate/closed copy.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46551", + "right": "issue:46552", + "accept": true, + "reason": "Exact duplicate: identical title and reported bug, with matching text indicating the same tokenizer regression." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both issues have identical titles, equal body length, and perfect soft-similarity/Jaccard overlap, while #46552 is open and has inbound references. The duplicate/canonical judgment is conservative and supported by the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 22, + "component_count": 28, + "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": "3de6dcc3830f10b8802b6d12e24742cc5eefc8006c6ccb91403c9138a1d0d13f", + "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 do not look like duplicates: one reports a tensor-shape mismatch during object detection training, while the other reports out-of-memory after several epochs in RT-DETR v2. They point to different failure modes and likely different code paths.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46657 is the more concrete, narrowly described bug report, but it should not be treated as the same problem as 46692.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46657", + "right": "issue:46692", + "accept": false, + "reason": "Reject: the titles and likely symptoms differ substantially\u2014tensor size mismatch vs OOM after epochs\u2014and there is no evidence they stem from the same underlying 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": 28, + "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": "25f7310998229ca1f2570f455c15705cfa12dfe1268f9abd0c5fb78bcc904108", + "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 failure modes in different loading paths: one is a ZeRO-3/zero.Init partitioning bug causing OOM on multi-GPU load, the other is a 4-bit device_map/max_memory dispatch error that incorrectly falls back to CPU. They are not duplicates and should not be merged.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 46822 is the better representative of the cluster because it is an open, specific regression about ZeRO-3 zero.Init failing to partition a composite submodule and causing OOM during multi-GPU model loading.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46822 is the best issue to keep as the cluster anchor: it describes a concrete code-path problem, is still open, and is more likely to need follow-up than the already-closed 46823.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46822", + "right": "issue:46823", + "accept": false, + "reason": "Reject: the titles describe unrelated bugs in different subsystems and loading mechanisms. One is ZeRO-3 partitioning during zero.Init, the other is 4-bit dispatch with device_map/max_memory falling back to CPU despite fitting. They do not appear to share the same underlying defect." + } + ] + }, + "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": 28, + "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-46906-11", + "nodes": [ + "issue:46906", + "issue:46940", + "issue:46948", + "issue:46969", + "issue:46985", + "issue:46998", + "issue:47037", + "issue:47049", + "issue:47053", + "issue:47091", + "issue:47102" + ], + "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" + ], + "prepared_review_unit_hash": "83e1454d2d470d72fd10903330395786c364e339ed4198a720b770c92ef0869f", + "component_budget": { + "node_count": 11, + "item_count": 11, + "soft_pair_count": 34, + "serialized_chars": 18418, + "estimated_input_tokens": 4605, + "estimated_eval_tokens": 9466 + }, + "budget": { + "node_count": 11, + "item_count": 11, + "soft_pair_count": 34, + "serialized_chars": 18418, + "estimated_input_tokens": 4605, + "estimated_eval_tokens": 9466 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a series of similarly titled, date-stamped Serge integration triage issues. They read like repeated daily reports rather than the same underlying bug, so I rejected all soft duplicate edges. No PRs are present.", + "confidence": 0.93, + "canonical_issue_reason": "issue:46906 is the earliest dated entry in the series and has the longest body, making it the safest canonical representative.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:47037 has the highest discussion activity, so it is the most informative single issue to inspect in this cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46906", + "right": "issue:46940", + "accept": false, + "reason": "Same triage template, but these are separate daily reports on different dates; no evidence they describe the same failure." + }, + { + "left": "issue:46906", + "right": "issue:46948", + "accept": false, + "reason": "Date-stamped triage entries from different days; title similarity alone is not enough to merge them." + }, + { + "left": "issue:46906", + "right": "issue:46969", + "accept": false, + "reason": "These look like distinct daily integration triage reports, not the same underlying bug." + }, + { + "left": "issue:46906", + "right": "issue:46985", + "accept": false, + "reason": "Reject: recurring triage item for a different date, with no concrete shared defect evidence." + }, + { + "left": "issue:46906", + "right": "issue:47037", + "accept": false, + "reason": "Both are generic Serge integration triage reports, but they are separate dated instances rather than one bug." + }, + { + "left": "issue:46940", + "right": "issue:46948", + "accept": false, + "reason": "Same repeating triage series, but each issue is tied to a different day and likely a different run." + }, + { + "left": "issue:46940", + "right": "issue:46969", + "accept": false, + "reason": "No concrete shared code-path failure; these are distinct dated triage entries." + }, + { + "left": "issue:46940", + "right": "issue:46985", + "accept": false, + "reason": "Reject: same broad subsystem and title pattern, but not the same underlying issue." + }, + { + "left": "issue:46940", + "right": "issue:46998", + "accept": false, + "reason": "These appear to be separate daily triage reports, not duplicates of one another." + }, + { + "left": "issue:46948", + "right": "issue:46969", + "accept": false, + "reason": "Different date-stamped triage reports; nothing here indicates the same bug instance." + }, + { + "left": "issue:46948", + "right": "issue:46985", + "accept": false, + "reason": "Reject: recurring integration triage entries, but no shared concrete failure is shown." + }, + { + "left": "issue:46948", + "right": "issue:46998", + "accept": false, + "reason": "Same generic title, but they are separate daily reports rather than duplicate issues." + }, + { + "left": "issue:46969", + "right": "issue:46985", + "accept": false, + "reason": "These are sequential triage reports, not one underlying defect report." + }, + { + "left": "issue:46969", + "right": "issue:47037", + "accept": false, + "reason": "Different dated Serge integration triage items; title overlap is too generic to merge." + }, + { + "left": "issue:46969", + "right": "issue:47091", + "accept": false, + "reason": "Reject: same recurring triage template, but no evidence of identical failure context." + }, + { + "left": "issue:46985", + "right": "issue:46998", + "accept": false, + "reason": "Different daily triage entries; not enough signal that they track the same bug." + }, + { + "left": "issue:46985", + "right": "issue:47037", + "accept": false, + "reason": "Same subsystem and template, but these look like separate daily failures or investigations." + }, + { + "left": "issue:46985", + "right": "issue:47049", + "accept": false, + "reason": "Reject: dated triage reports for different runs, not a single issue." + }, + { + "left": "issue:46985", + "right": "issue:47091", + "accept": false, + "reason": "No concrete matching failure details; these are distinct triage artifacts." + }, + { + "left": "issue:46985", + "right": "issue:47102", + "accept": false, + "reason": "These are separate date-based triage reports, not the same underlying bug." + }, + { + "left": "issue:46998", + "right": "issue:47037", + "accept": false, + "reason": "Generic title similarity only; the issues are from different dates and likely different incidents." + }, + { + "left": "issue:46998", + "right": "issue:47049", + "accept": false, + "reason": "Reject: recurring integration triage items, but no evidence they describe the same code-path problem." + }, + { + "left": "issue:46998", + "right": "issue:47053", + "accept": false, + "reason": "Different daily triage reports; title match is too broad to call them duplicates." + }, + { + "left": "issue:46998", + "right": "issue:47091", + "accept": false, + "reason": "These appear to be separate daily Serge triage issues, not one bug." + }, + { + "left": "issue:47037", + "right": "issue:47049", + "accept": false, + "reason": "Same series, but each is a dated triage entry; no shared concrete defect is shown." + }, + { + "left": "issue:47037", + "right": "issue:47053", + "accept": false, + "reason": "Reject: similar titles, but they are distinct daily reports rather than the same failure." + }, + { + "left": "issue:47037", + "right": "issue:47091", + "accept": false, + "reason": "No evidence of the same underlying bug; these are separate triage instances." + }, + { + "left": "issue:47037", + "right": "issue:47102", + "accept": false, + "reason": "Different dated triage issues in a recurring series, not duplicates." + }, + { + "left": "issue:47049", + "right": "issue:47053", + "accept": false, + "reason": "These are adjacent daily triage reports, but there is no concrete sign they refer to the same incident." + }, + { + "left": "issue:47049", + "right": "issue:47091", + "accept": false, + "reason": "Reject: recurring Serge integration triage entries, not the same underlying bug report." + }, + { + "left": "issue:47049", + "right": "issue:47102", + "accept": false, + "reason": "Different dates and no shared failure details; treat as separate issues." + }, + { + "left": "issue:47053", + "right": "issue:47091", + "accept": false, + "reason": "Same template and subsystem, but not enough evidence to merge them." + }, + { + "left": "issue:47053", + "right": "issue:47102", + "accept": false, + "reason": "Reject: distinct daily triage reports with no demonstrated common root cause." + }, + { + "left": "issue:47091", + "right": "issue:47102", + "accept": false, + "reason": "These look like separate day-by-day triage entries, not 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": 25, + "component_count": 28, + "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": "33356f13194da80e612623ed6b5c6abff5242172fa310bd1426c01462cf58b3f", + "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 issues are clear duplicates: both request the same RWKV7 model support. Keep the more active report as the canonical issue.", + "confidence": 0.98, + "canonical_issue_reason": "Issue #46984 is the better canonical choice because it has more discussion activity, suggesting it is the more developed or representative report.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #46984 is the best single issue to keep because it appears to have slightly more engagement while describing the same request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46983", + "right": "issue:46984", + "accept": true, + "reason": "Same title and same requested feature (RWKV7 model support); they describe the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: both issues have the same title, and the only activity difference favors #46984 (discussion_activity 2 vs 0). The duplicate verdict is conservative enough given the strong similarity signal." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 26, + "component_count": 28, + "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": "66075f18a85e4141256cef6ee0d9f7a96b04712a6edbb528a8a7ae0d748c99fa", + "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": "The two closed issues are thematically similar (both about AI/LLM-driven vulnerability discovery), but they do not read like the same concrete bug or actionable request. They appear to be separate discussion/proposal posts rather than duplicates of one underlying problem.", + "confidence": 0.86, + "canonical_issue_reason": "No single issue clearly subsumes the other; both are broad, overlapping topics but not the same specific report or request.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #47015 is the more directly framed concept/topic and is the better representative if one item must be picked for the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:47007", + "right": "issue:47015", + "accept": false, + "reason": "Shared high-level subject matter only. The titles suggest different angles ('lessons from using advanced LLMs' vs 'novel application of transformers'), not the same underlying issue 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": 27, + "component_count": 28, + "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-47120-7", + "nodes": [ + "issue:47120", + "issue:47182", + "issue:47227", + "issue:47256", + "issue:47265", + "issue:47279", + "issue:47309" + ], + "soft_pairs": [ + "issue:47120|issue:47182", + "issue:47120|issue:47309", + "issue:47120|issue:47256", + "issue:47120|issue:47279", + "issue:47120|issue:47265", + "issue:47182|issue:47309", + "issue:47182|issue:47256", + "issue:47182|issue:47265", + "issue:47182|issue:47279", + "issue:47227|issue:47256", + "issue:47227|issue:47309", + "issue:47227|issue:47279", + "issue:47182|issue:47227", + "issue:47227|issue:47265", + "issue:47256|issue:47309", + "issue:47256|issue:47265", + "issue:47256|issue:47279", + "issue:47265|issue:47279", + "issue:47265|issue:47309", + "issue:47279|issue:47309" + ], + "prepared_review_unit_hash": "273a5e2ba9963daf52e8be996b5b851c69f83b927130500d6e000265bbe42fd0", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 20, + "serialized_chars": 11160, + "estimated_input_tokens": 2790, + "estimated_eval_tokens": 5836 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 20, + "serialized_chars": 11160, + "estimated_input_tokens": 2790, + "estimated_eval_tokens": 5836 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are highly similar daily serge integration-triage issues, but the date-stamped titles indicate separate tracking tickets rather than duplicates of one underlying bug. No PRs are present.", + "confidence": 0.87, + "canonical_issue_reason": "issue:47279 is the strongest anchor in the series: it is closed, has the highest inbound references, and looks like the most central instance of the recurring triage thread. The date-stamped issues still appear to be separate daily tracker tickets, so this is only a loose representative choice.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:47279 is the best representative issue in the cluster because it is the most referenced and central of the closed daily triage tickets. It is not a true duplicate target, but it is the most suitable anchor if one must be chosen.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:47120", + "right": "issue:47182", + "accept": false, + "reason": "Same template and subsystem, but different dated triage tickets; these read like separate daily tracking issues rather than one bug." + }, + { + "left": "issue:47120", + "right": "issue:47309", + "accept": false, + "reason": "Date-stamped triage entries for different days; no evidence they describe the same underlying failure." + }, + { + "left": "issue:47120", + "right": "issue:47256", + "accept": false, + "reason": "Looks like another daily integration triage ticket, not a duplicate of the 2026-07-07 tracker." + }, + { + "left": "issue:47120", + "right": "issue:47279", + "accept": false, + "reason": "Same recurring triage series, but distinct day-specific issue; not the same underlying bug report." + }, + { + "left": "issue:47120", + "right": "issue:47265", + "accept": false, + "reason": "Different daily tracker entry in the same series, with no concrete shared bug description shown." + }, + { + "left": "issue:47182", + "right": "issue:47309", + "accept": false, + "reason": "Both are similarly titled triage tickets, but the date separation suggests independent daily reports." + }, + { + "left": "issue:47182", + "right": "issue:47256", + "accept": false, + "reason": "Same broad integration-triage pattern, but not enough to treat as one duplicate issue." + }, + { + "left": "issue:47182", + "right": "issue:47265", + "accept": false, + "reason": "These are separate dated triage issues rather than one concrete bug ticket." + }, + { + "left": "issue:47182", + "right": "issue:47279", + "accept": false, + "reason": "Recurring tracker entry, but each date appears to be its own triage issue." + }, + { + "left": "issue:47227", + "right": "issue:47256", + "accept": false, + "reason": "Date-specific triage tickets; same title pattern does not imply the same underlying failure." + }, + { + "left": "issue:47227", + "right": "issue:47309", + "accept": false, + "reason": "Different daily integration-failure triage reports, not a single duplicate issue." + }, + { + "left": "issue:47227", + "right": "issue:47279", + "accept": false, + "reason": "Same recurring series, but no shared bug details indicating a mergeable duplicate." + }, + { + "left": "issue:47182", + "right": "issue:47227", + "accept": false, + "reason": "These appear to be successive daily triage logs, not the same issue." + }, + { + "left": "issue:47227", + "right": "issue:47265", + "accept": false, + "reason": "No concrete evidence of the same bug; just the same triage template on different dates." + }, + { + "left": "issue:47256", + "right": "issue:47309", + "accept": false, + "reason": "Different dated tracker issues in the same recurring integration-failure triage series." + }, + { + "left": "issue:47256", + "right": "issue:47265", + "accept": false, + "reason": "Separate daily triage ticket, not the same underlying bug report." + }, + { + "left": "issue:47256", + "right": "issue:47279", + "accept": false, + "reason": "Same pattern and subsystem, but different day-specific triage issue." + }, + { + "left": "issue:47265", + "right": "issue:47279", + "accept": false, + "reason": "These are recurring triage logs; no sign they should be merged as duplicates." + }, + { + "left": "issue:47265", + "right": "issue:47309", + "accept": false, + "reason": "Different day-stamped integration triage issues, not one shared bug." + }, + { + "left": "issue:47279", + "right": "issue:47309", + "accept": false, + "reason": "Same series, but the date and tracker context indicate separate issues rather than duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 28, + "component_count": 28, + "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": "949d2ba5cf49dfeb5510fd59366e439118e5e9f9e8db89f27c1a02c70c1df745", + "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": "#47138 and #47139 appear to be duplicate issue reports: they have identical titles and the similarity signal is maximal, but the issue bodies are not visible here, so the match should be treated as likely rather than certain. #47138 is the best canonical issue because it was created first and has slightly more inbound references.", + "confidence": 0.92, + "canonical_issue_reason": "#47138 is the earliest-created report and has more inbound references, so it is the best canonical issue to keep.", + "canonical_pr_reason": null, + "best_issue_reason": "#47138 is the strongest single issue to track this bug because it appears to duplicate #47139 while having slightly more linkage/support.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:47138", + "right": "issue:47139", + "accept": true, + "reason": "Identical titles and a 1.0 similarity signal make this look like a duplicate issue report, though the bodies are not visible here so this is best stated as an apparent duplicate." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet supports a strong similarity/likely-duplicate assessment, but the analyst states it too definitively. We only know the titles are identical and the similarity score is 1.0; we do not see the bodies, so phrases like \"same content,\" \"clear duplicate,\" and \"same underlying bug\" overstate the evidence. The canonical-choice rationale is grounded, but the soft-edge verdict and summary should be phrased more conservatively (e.g. \"appears to be a duplicate\")." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 34, + "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-40546-76", + "nodes": [ + "pull_request:43251", + "pull_request:43254", + "pull_request:44775", + "pull_request:46196", + "pull_request:46266", + "pull_request:46376", + "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:46990", + "pull_request:46995", + "pull_request:47003", + "pull_request:47069", + "pull_request:47077", + "pull_request:47078", + "pull_request:47082", + "pull_request:47090", + "pull_request:47093", + "pull_request:47108", + "pull_request:47163", + "pull_request:47166", + "pull_request:47173", + "pull_request:47179", + "pull_request:47181", + "pull_request:47184", + "pull_request:47202", + "pull_request:47204", + "pull_request:47210", + "pull_request:47212", + "pull_request:47216", + "pull_request:47243", + "pull_request:47253", + "pull_request:47268", + "pull_request:47275", + "pull_request:47284", + "pull_request:47292", + "pull_request:47295", + "pull_request:47301", + "pull_request:47302", + "pull_request:47304", + "pull_request:47318", + "pull_request:47323" + ], + "soft_pairs": [ + "pull_request:43251|pull_request:43254", + "pull_request:46903|pull_request:47090", + "pull_request:46903|pull_request:46958", + "pull_request:46958|pull_request:46974", + "pull_request:47268|pull_request:47284", + "pull_request:46954|pull_request:46958", + "pull_request:47173|pull_request:47284", + "pull_request:47202|pull_request:47204", + "pull_request:47284|pull_request:47292", + "pull_request:47108|pull_request:47202", + "pull_request:47204|pull_request:47243", + "pull_request:47077|pull_request:47078", + "pull_request:46266|pull_request:46957", + "pull_request:47077|pull_request:47212", + "pull_request:47173|pull_request:47184", + "pull_request:46376|pull_request:47202", + "pull_request:47166|pull_request:47173", + "pull_request:47077|pull_request:47275", + "pull_request:44775|pull_request:47275", + "pull_request:46990|pull_request:47253", + "pull_request:46995|pull_request:47323", + "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:46395|pull_request:47181", + "pull_request:47003|pull_request:47204", + "pull_request:47275|pull_request:47302", + "pull_request:47093|pull_request:47275", + "pull_request:46196|pull_request:47210", + "pull_request:46995|pull_request:47163", + "pull_request:46912|pull_request:47082", + "pull_request:47163|pull_request:47173", + "pull_request:44775|pull_request:47301", + "pull_request:47204|pull_request:47295", + "pull_request:46196|pull_request:46995", + "pull_request:47302|pull_request:47304", + "pull_request:47077|pull_request:47301", + "pull_request:44775|pull_request:47077", + "pull_request:46376|pull_request:46995", + "pull_request:47069|pull_request:47318" + ], + "prepared_review_unit_hash": "0d824e590206cc512162542d6364f2ad1d0a95da6beeec13951e0091ba1a9f44", + "component_budget": { + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 44, + "serialized_chars": 122076, + "estimated_input_tokens": 30519, + "estimated_eval_tokens": 61294 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most soft edges are false positives across unrelated docs, model additions, tests, or adjacent subsystems. The only plausible duplicate-like pair is the CohereASR loss double-shift follow-up, where 47090 is the best representative.", + "confidence": 0.84, + "canonical_issue_reason": null, + "canonical_pr_reason": "No issue artifacts here; among the PRs, 47090 is the cleanest representative of the CohereASR loss-double-shift fix because it is narrowly scoped and preserves the shared loss helper path.", + "best_issue_reason": null, + "best_pr_reason": "47090 is the strongest standalone PR: it targets the exact CohereASR training-loss bug, keeps trainer-compatible loss handling, and is narrower than the broader predecessor PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": false, + "reason": "Same target issue, but the patches are just two overlapping implementations of the same helper change; not a clear duplicate PR pair to merge." + }, + { + "left": "pull_request:46903", + "right": "pull_request:47090", + "accept": true, + "reason": "Same CohereASR loss double-shift bug in the same forward path; 47090 is a narrower follow-up of the same fix and could be consolidated into one PR." + }, + { + "left": "pull_request:46903", + "right": "pull_request:46958", + "accept": false, + "reason": "Both touch CohereASR loss handling, but the fixes diverge materially and 46903 bundles an extra PPFormulaNet change, so this is not a clean duplicate pair." + }, + { + "left": "pull_request:46958", + "right": "pull_request:46974", + "accept": false, + "reason": "Unrelated models and bugs: CohereASR loss logic versus a MusicGen generate typo." + }, + { + "left": "pull_request:47268", + "right": "pull_request:47284", + "accept": false, + "reason": "Different models and test failures; only superficial test-edit overlap." + }, + { + "left": "pull_request:46954", + "right": "pull_request:46958", + "accept": false, + "reason": "Pytest-v9 module attribute fix versus CohereASR loss bug; unrelated code paths." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47284", + "accept": false, + "reason": "Both are test-only changes, but they address different models and failures." + }, + { + "left": "pull_request:47202", + "right": "pull_request:47204", + "accept": false, + "reason": "Both touch docker/build tooling, but one removes a CI image builder while the other adds exporter deps; not the same change." + }, + { + "left": "pull_request:47284", + "right": "pull_request:47292", + "accept": false, + "reason": "Different model test updates for different hardware expectations; not the same bug." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47202", + "accept": false, + "reason": "Doc-builder pin comment cleanup versus docker-image builder removal; unrelated." + }, + { + "left": "pull_request:47204", + "right": "pull_request:47243", + "accept": false, + "reason": "Exporter dependency packaging versus Executorch exporter fixes; adjacent area but distinct changes." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47078", + "accept": false, + "reason": "Japanese docs additions/translation versus wording cleanup; same locale, but not the same underlying change." + }, + { + "left": "pull_request:46266", + "right": "pull_request:46957", + "accept": false, + "reason": "Model-conversion mappings for dinos versus finegrained FP8 decode/kernel optimization; unrelated." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47212", + "accept": false, + "reason": "Japanese docs translation versus KV-cache docs update; no shared bug or feature." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47184", + "accept": false, + "reason": "Separate test maintenance PRs for different models and accelerators." + }, + { + "left": "pull_request:46376", + "right": "pull_request:47202", + "accept": false, + "reason": "Continuous batching multimodal support versus docker builder removal; different subsystems and goals." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47173", + "accept": false, + "reason": "Quantization test extensions versus XPU expectation fixes; not the same issue." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47275", + "accept": false, + "reason": "Japanese docs translation versus new model support; unrelated." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47275", + "accept": false, + "reason": "Docs-only distributed training update versus new model support; not a duplicate change." + }, + { + "left": "pull_request:46990", + "right": "pull_request:47253", + "accept": false, + "reason": "Both are distributed-training related, but one is FSDP orchestration/loading and the other is tensor-parallel lm_head validation." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47323", + "accept": false, + "reason": "Different kernel availability and deepgemm multi-device fixes; shared accelerator context only." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47301", + "accept": false, + "reason": "New model support versus assisted-decoding docs for MTP; unrelated." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47275", + "accept": false, + "reason": "Compressed-tensors FP8 versus EfficientVit-SAM model support; different features." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47179", + "accept": false, + "reason": "Different model families and unrelated rotary/test changes." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47202", + "accept": false, + "reason": "SDPA dispatch support versus CI docker builder removal; different layers of the project." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47284", + "accept": false, + "reason": "Paged-attention XPU test coverage versus Mimo v2 flash model fixes; not the same underlying bug." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47284", + "accept": false, + "reason": "RoPE tests versus Mimo model fixes; unrelated." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47202", + "accept": false, + "reason": "Unlimited OCR model addition versus docker builder removal; no overlap in bug/change." + }, + { + "left": "pull_request:46395", + "right": "pull_request:47181", + "accept": false, + "reason": "Two separate new-model PRs for different architectures; same broad pattern, but not a shared change." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47204", + "accept": false, + "reason": "OpenVINO exporter feature versus exporter dependencies in torch-light; not the same code-path problem." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47302", + "accept": false, + "reason": "Model-addition PR versus tokenizer doc formatting cleanup; unrelated." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47275", + "accept": false, + "reason": "OpenPangu-V2 model support versus EfficientVit-SAM model support; different model integrations." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47210", + "accept": false, + "reason": "Both concern attention optimizations, but one is sdpa dispatch for models and the other is packed-row varlen routing; distinct bugs." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47163", + "accept": false, + "reason": "Flash-linear-attention availability on MLU versus paged-attention XPU tests; different concerns." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47082", + "accept": false, + "reason": "RoPE test improvements versus RoMa model support; unrelated." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47173", + "accept": false, + "reason": "Paged-attention XPU coverage versus XPU expectation updates for other model tests; not the same PR change." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47301", + "accept": false, + "reason": "Docs for distributed training versus assisted-decoding docs for MTP; unrelated." + }, + { + "left": "pull_request:47204", + "right": "pull_request:47295", + "accept": false, + "reason": "Exporter deps in torch-light versus DiffusionGemma mask casting; no shared bug." + }, + { + "left": "pull_request:46196", + "right": "pull_request:46995", + "accept": false, + "reason": "Different attention/kernel features: sdpa dispatch versus flash-linear-attention availability." + }, + { + "left": "pull_request:47302", + "right": "pull_request:47304", + "accept": false, + "reason": "Tokenizer backend docs versus MPS graph-cache docs; unrelated documentation changes." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47301", + "accept": false, + "reason": "Japanese docs translation versus MTP docs; different documentation topics." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47077", + "accept": false, + "reason": "Distributed-training docs versus Japanese DistilBERT translation; no overlap." + }, + { + "left": "pull_request:46376", + "right": "pull_request:46995", + "accept": false, + "reason": "Continuous batching multimodal support versus flash-linear-attention availability; different features." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47318", + "accept": false, + "reason": "Config property typing versus flash-attn fallback test gating; unrelated changes." + } + ] + }, + "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": 34, + "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-40546-76", + "nodes": [ + "pull_request:43451", + "pull_request:44517", + "pull_request:44775", + "pull_request:46196", + "pull_request:46266", + "pull_request:46267", + "pull_request:46376", + "pull_request:46395", + "pull_request:46836", + "pull_request:46895", + "pull_request:46912", + "pull_request:46957", + "pull_request:46990", + "pull_request:46995", + "pull_request:47069", + "pull_request:47077", + "pull_request:47078", + "pull_request:47082", + "pull_request:47093", + "pull_request:47108", + "pull_request:47113", + "pull_request:47143", + "pull_request:47163", + "pull_request:47166", + "pull_request:47173", + "pull_request:47175", + "pull_request:47179", + "pull_request:47181", + "pull_request:47204", + "pull_request:47210", + "pull_request:47216", + "pull_request:47218", + "pull_request:47243", + "pull_request:47253", + "pull_request:47275", + "pull_request:47291", + "pull_request:47292", + "pull_request:47294", + "pull_request:47298", + "pull_request:47301", + "pull_request:47302", + "pull_request:47304", + "pull_request:47306", + "pull_request:47311", + "pull_request:47318", + "pull_request:47320", + "pull_request:47322" + ], + "soft_pairs": [ + "pull_request:47294|pull_request:47320", + "pull_request:46995|pull_request:47181", + "pull_request:46267|pull_request:47181", + "pull_request:47093|pull_request:47181", + "pull_request:44517|pull_request:46376", + "pull_request:47166|pull_request:47292", + "pull_request:46836|pull_request:47275", + "pull_request:46266|pull_request:47301", + "pull_request:44775|pull_request:47304", + "pull_request:44775|pull_request:47181", + "pull_request:47093|pull_request:47143", + "pull_request:47082|pull_request:47210", + "pull_request:47093|pull_request:47163", + "pull_request:46266|pull_request:47216", + "pull_request:47108|pull_request:47301", + "pull_request:47082|pull_request:47108", + "pull_request:46376|pull_request:47291", + "pull_request:47082|pull_request:47275", + "pull_request:47179|pull_request:47210", + "pull_request:46196|pull_request:47179", + "pull_request:46912|pull_request:46990", + "pull_request:46990|pull_request:47181", + "pull_request:47078|pull_request:47275", + "pull_request:46395|pull_request:47311", + "pull_request:47311|pull_request:47318", + "pull_request:47175|pull_request:47179", + "pull_request:47210|pull_request:47302", + "pull_request:47077|pull_request:47302", + "pull_request:47082|pull_request:47216", + "pull_request:47163|pull_request:47179", + "pull_request:46990|pull_request:47306", + "pull_request:47108|pull_request:47302", + "pull_request:46376|pull_request:47318", + "pull_request:46836|pull_request:47218", + "pull_request:47210|pull_request:47304", + "pull_request:46376|pull_request:47218", + "pull_request:46957|pull_request:47216", + "pull_request:47173|pull_request:47311", + "pull_request:47108|pull_request:47216", + "pull_request:46912|pull_request:47078", + "pull_request:43451|pull_request:46990", + "pull_request:47179|pull_request:47301", + "pull_request:47113|pull_request:47204", + "pull_request:47143|pull_request:47318", + "pull_request:47306|pull_request:47318", + "pull_request:47218|pull_request:47311", + "pull_request:47218|pull_request:47304", + "pull_request:47166|pull_request:47218", + "pull_request:47216|pull_request:47302", + "pull_request:46266|pull_request:46376", + "pull_request:47243|pull_request:47253", + "pull_request:46912|pull_request:47294", + "pull_request:47143|pull_request:47320", + "pull_request:47175|pull_request:47294", + "pull_request:46895|pull_request:47306", + "pull_request:46196|pull_request:46376", + "pull_request:46196|pull_request:47301", + "pull_request:47253|pull_request:47294", + "pull_request:47069|pull_request:47113", + "pull_request:47275|pull_request:47304", + "pull_request:44517|pull_request:47311", + "pull_request:47078|pull_request:47322", + "pull_request:44517|pull_request:47301", + "pull_request:44517|pull_request:47181", + "pull_request:47078|pull_request:47243", + "pull_request:47113|pull_request:47304", + "pull_request:43451|pull_request:46266", + "pull_request:43451|pull_request:44517", + "pull_request:46912|pull_request:47291", + "pull_request:47069|pull_request:47302", + "pull_request:47243|pull_request:47294", + "pull_request:46895|pull_request:47298" + ], + "prepared_review_unit_hash": "4499e173318c595d028d3f5a62260fe9b7e76e32092315b30b25681bb2ccff08", + "component_budget": { + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 72, + "serialized_chars": 137274, + "estimated_input_tokens": 34319, + "estimated_eval_tokens": 68894 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose bag of unrelated pull requests: model additions, documentation updates, test expectation tweaks, and infrastructure fixes. None of the soft pairs look like the same underlying change, so all should be rejected as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "No single PR is a safe cluster representative because the items span unrelated code paths and topics; there is no coherent duplicate set here.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47294", + "right": "pull_request:47320", + "accept": false, + "reason": "Causal-conv1d kernel helper work vs response-schema deprecation; different subsystems and goals." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47181", + "accept": false, + "reason": "Flash-linear-attention availability change vs new Cosmos3 Edge model support; unrelated." + }, + { + "left": "pull_request:46267", + "right": "pull_request:47181", + "accept": false, + "reason": "Serve chat parsing refactor vs model implementation add; no shared bug/change." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47181", + "accept": false, + "reason": "Two different model integrations (OpenPangu-V2 vs Cosmos3 Edge), not duplicates." + }, + { + "left": "pull_request:44517", + "right": "pull_request:46376", + "accept": false, + "reason": "Qwen3 TTS model addition vs continuous-batching multimodal support; different change sets." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47292", + "accept": false, + "reason": "Both are test updates, but for different models and hardware targets." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47275", + "accept": false, + "reason": "Unlimited OCR model support vs EfficientViT-SAM model support; distinct artifacts." + }, + { + "left": "pull_request:46266", + "right": "pull_request:47301", + "accept": false, + "reason": "RoPE modernization/tests vs assisted-decoding docs; unrelated." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47304", + "accept": false, + "reason": "Parallelism docs restructuring vs MPS graph-cache docs; both docs-only but unrelated topics." + }, + { + "left": "pull_request:44775", + "right": "pull_request:47181", + "accept": false, + "reason": "Docs navigation change vs model support PR; no same underlying change." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47143", + "accept": false, + "reason": "OpenPangu-V2 model support vs NeuCodec model support; both model adds but not the same one." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47302", + "accept": false, + "reason": "Packed-SDPA routing code vs tokenizer fallback docs; different layers entirely." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47163", + "accept": false, + "reason": "Model integration PR vs XPU paged-attention test add; unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:47216", + "accept": false, + "reason": "RoPE tests/code vs compressed-tensors fp8 changes; not the same issue." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47301", + "accept": false, + "reason": "Doc-builder pin comment tweak vs assisted-decoding docs update; unrelated docs maintenance." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47275", + "accept": false, + "reason": "RoMa keypoint matcher vs EfficientViT-SAM model support; different models and code paths." + }, + { + "left": "pull_request:46376", + "right": "pull_request:47291", + "accept": false, + "reason": "Continuous batching multimodal support vs deprecating CB config; related area, but not the same change." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47210", + "accept": false, + "reason": "Vision model add vs attention-kernel routing; no concrete overlap." + }, + { + "left": "pull_request:47179", + "right": "pull_request:47210", + "accept": false, + "reason": "Granite-SWA model support vs SDPA varlen packing; unrelated." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47179", + "accept": false, + "reason": "SDPA dispatch for wav2vec2-family audio models vs Granite-SWA model support; different scopes." + }, + { + "left": "pull_request:46912", + "right": "pull_request:46990", + "accept": false, + "reason": "RoPE test expansion vs FSDP orchestration; no shared bug or feature." + }, + { + "left": "pull_request:46990", + "right": "pull_request:47181", + "accept": false, + "reason": "FSDP orchestration changes vs Cosmos3 Edge model support; unrelated." + }, + { + "left": "pull_request:47078", + "right": "pull_request:47304", + "accept": false, + "reason": "Japanese doc wording cleanup vs MPS graph-cache docs; different documentation changes." + }, + { + "left": "pull_request:46395", + "right": "pull_request:47311", + "accept": false, + "reason": "Param2MoE model implementation vs Gemma4 H100 expectation update; unrelated." + }, + { + "left": "pull_request:47311", + "right": "pull_request:47318", + "accept": false, + "reason": "Gemma4 test expectations vs flash-attn availability helper; unrelated test infra changes." + }, + { + "left": "pull_request:47175", + "right": "pull_request:47179", + "accept": false, + "reason": "Olmo3 CI expectation tweak vs Granite-SWA model support; not the same artifact." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47304", + "accept": false, + "reason": "Packed SDPA implementation vs docs-only MPS cache note; different layers." + }, + { + "left": "pull_request:46376", + "right": "pull_request:47218", + "accept": false, + "reason": "Continuous-batching code vs image-loading refactor in tests; no duplicate bug." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47218", + "accept": false, + "reason": "Unlimited OCR model add vs test image-loading refactor; unrelated." + }, + { + "left": "pull_request:47243", + "right": "pull_request:47253", + "accept": false, + "reason": "Exporter fixes vs tensor-parallel LM-head validation; different code paths." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47291", + "accept": false, + "reason": "Compressed-tensors FP8 work vs CB config deprecation; unrelated." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47311", + "accept": false, + "reason": "XPU Expectations test fixes vs Gemma4 H100 expectations; different test targets." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47302", + "accept": false, + "reason": "Doc-builder pin comment vs tokenizer-backend fallback docs; unrelated docs changes." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47301", + "accept": false, + "reason": "Quantization feature work vs assisted-decoding docs; not the same change." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47304", + "accept": false, + "reason": "Audio requirements fix vs MPS docs update; unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46266", + "accept": false, + "reason": "Molmo2 model add vs DINO/DINOv2 modernization; distinct feature areas." + }, + { + "left": "pull_request:43451", + "right": "pull_request:44517", + "accept": false, + "reason": "Molmo2 model support vs Qwen3 TTS support; both model adds, but different models and code paths." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47291", + "accept": false, + "reason": "RoPE tests vs continuous-batching config deprecation; unrelated." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47302", + "accept": false, + "reason": "`num_labels` typing fix vs tokenizer backend docs; unrelated." + }, + { + "left": "pull_request:47243", + "right": "pull_request:47294", + "accept": false, + "reason": "Exporter determinism fix vs kernel-function integration; different domains." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47306", + "accept": false, + "reason": "CohereASR training-loss fix vs Neuron experiment scripts; unrelated." + }, + { + "left": "pull_request:46196", + "right": "pull_request:46376", + "accept": false, + "reason": "Audio SDPA support vs multimodal continuous batching; same broad area but different concrete changes." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47301", + "accept": false, + "reason": "Audio attention dispatch vs assisted-decoding docs; no overlap." + }, + { + "left": "pull_request:47253", + "right": "pull_request:47294", + "accept": false, + "reason": "Tensor-parallel validation vs causal_conv1d kernel functions; unrelated." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47113", + "accept": false, + "reason": "Config typing tweak vs audio requirements fix; separate issues." + }, + { + "left": "pull_request:47275", + "right": "pull_request:47304", + "accept": false, + "reason": "EfficientViT-SAM model support vs MPS graph-cache docs; unrelated." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47301", + "accept": false, + "reason": "Qwen3 TTS model add vs assisted-decoding docs; different topics." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47181", + "accept": false, + "reason": "Qwen3 TTS vs Cosmos3 Edge model support; distinct model additions." + }, + { + "left": "pull_request:47078", + "right": "pull_request:47322", + "accept": false, + "reason": "Japanese docs wording vs AMD Quark config fix; unrelated docs vs code change." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47304", + "accept": false, + "reason": "Audio requirements change vs MPS docs cleanup; not the same bug/change." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46266", + "accept": false, + "reason": "Model addition vs Dino modernization/conversion mapping; unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:44517", + "accept": false, + "reason": "Different model families and different generated code; not duplicates." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47291", + "accept": false, + "reason": "RoPE test updates vs CB config deprecation; no common fix." + }, + { + "left": "pull_request:47179", + "right": "pull_request:47301", + "accept": false, + "reason": "Granite-SWA support vs assisted-decoding docs; unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47204", + "accept": false, + "reason": "Audio feature/proc requirements fix vs Dockerfile exporter dependency additions; unrelated." + }, + { + "left": "pull_request:47143", + "right": "pull_request:47318", + "accept": false, + "reason": "NeuCodec model add vs flash-attn availability helper; different functionality." + }, + { + "left": "pull_request:47306", + "right": "pull_request:47318", + "accept": false, + "reason": "Neuron experiment scripts vs flash-attn test helper; unrelated." + }, + { + "left": "pull_request:47218", + "right": "pull_request:47311", + "accept": false, + "reason": "Image-loading helper refactor in tests vs Gemma4 expectation update; different test maintenance." + }, + { + "left": "pull_request:47218", + "right": "pull_request:47304", + "accept": false, + "reason": "Image-loading helper refactor vs MPS docs cleanup; unrelated." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47218", + "accept": false, + "reason": "Quantization test updates vs image-loading refactor; no shared issue." + }, + { + "left": "pull_request:47216", + "right": "pull_request:47302", + "accept": false, + "reason": "Compressed-tensors fp8 changes vs tokenizer backend docs; unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:46376", + "accept": false, + "reason": "Dino modernization vs continuous-batching multimodal support; different areas." + }, + { + "left": "pull_request:47243", + "right": "pull_request:47294", + "accept": false, + "reason": "Executorch exporter fixes vs causal_conv1d kernel functions; unrelated." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47294", + "accept": false, + "reason": "RoPE testing changes vs kernel helper PR; unrelated." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47298", + "accept": false, + "reason": "CohereASR loss fix vs MI450 image decoding fallback changes; 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": 34, + "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-40546-76", + "nodes": [ + "pull_request:41855", + "pull_request:42668", + "pull_request:43251", + "pull_request:43424", + "pull_request:43451", + "pull_request:43823", + "pull_request:43995", + "pull_request:44827", + "pull_request:45170", + "pull_request:45213", + "pull_request:45655", + "pull_request:46048", + "pull_request:46376", + "pull_request:46509", + "pull_request:46658", + "pull_request:46836", + "pull_request:46895", + "pull_request:46912", + "pull_request:46957", + "pull_request:46958", + "pull_request:47003", + "pull_request:47069", + "pull_request:47077", + "pull_request:47090", + "pull_request:47113", + "pull_request:47143", + "pull_request:47163", + "pull_request:47166", + "pull_request:47173", + "pull_request:47175", + "pull_request:47181", + "pull_request:47184", + "pull_request:47210", + "pull_request:47216", + "pull_request:47243", + "pull_request:47253", + "pull_request:47275", + "pull_request:47291", + "pull_request:47292", + "pull_request:47294", + "pull_request:47301", + "pull_request:47302", + "pull_request:47304", + "pull_request:47306", + "pull_request:47311", + "pull_request:47322", + "pull_request:47323" + ], + "soft_pairs": [ + "pull_request:46957|pull_request:47301", + "pull_request:47294|pull_request:47306", + "pull_request:47113|pull_request:47175", + "pull_request:47113|pull_request:47243", + "pull_request:46658|pull_request:47302", + "pull_request:46895|pull_request:47323", + "pull_request:47113|pull_request:47275", + "pull_request:46658|pull_request:47275", + "pull_request:47069|pull_request:47175", + "pull_request:46912|pull_request:47253", + "pull_request:47302|pull_request:47322", + "pull_request:47069|pull_request:47323", + "pull_request:46957|pull_request:47302", + "pull_request:43451|pull_request:46957", + "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:43823|pull_request:45655", + "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:47301|pull_request:47302", + "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:47166", + "pull_request:47163|pull_request:47184", + "pull_request:46509|pull_request:47181", + "pull_request:46509|pull_request:47163", + "pull_request:46376|pull_request:46509", + "pull_request:47291|pull_request:47322", + "pull_request:47163|pull_request:47292", + "pull_request:46376|pull_request:47181", + "pull_request:47216|pull_request:47301", + "pull_request:43451|pull_request:46836", + "pull_request:47143|pull_request:47181", + "pull_request:46376|pull_request:47143", + "pull_request:46509|pull_request:46836", + "pull_request:46509|pull_request:47301", + "pull_request:46376|pull_request:47301", + "pull_request:46836|pull_request:47143", + "pull_request:47163|pull_request:47181", + "pull_request:46376|pull_request:47163", + "pull_request:47181|pull_request:47301", + "pull_request:47163|pull_request:47311", + "pull_request:46658|pull_request:47143", + "pull_request:46658|pull_request:47322", + "pull_request:47143|pull_request:47163", + "pull_request:47181|pull_request:47311", + "pull_request:46376|pull_request:47311", + "pull_request:47143|pull_request:47322", + "pull_request:43451|pull_request:47243" + ], + "prepared_review_unit_hash": "e96ccc8e8ffb7338b5617c127147779d498fbc11454ee76986cd2c8258a83b2a", + "component_budget": { + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 72, + "serialized_chars": 131684, + "estimated_input_tokens": 32921, + "estimated_eval_tokens": 66098 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a heterogeneous cluster: model additions, docs, tests, exporter/kernel fixes, and one real duplicate subgroup. The only clear duplicate set is the CohereASR training-loss double-shift trio (46895/46958/47090); the rest are unrelated and should not be merged as duplicates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical for the whole cluster; the only true duplicate subgroup is the CohereASR training-loss double-shift trio, and 47090 is the clearest representative of that subgroup.", + "best_issue_reason": null, + "best_pr_reason": "47090 is the best representative of the confirmed duplicate subgroup: it has the clearest title and explicitly describes the same CohereASR double-shift fix as 46895 and 46958.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46957", + "right": "pull_request:47301", + "accept": false, + "reason": "Different changes: finegrained FP8 decode/kernels vs MTP docs. Same repo area, not the same bug or mergeable fix." + }, + { + "left": "pull_request:47294", + "right": "pull_request:47306", + "accept": false, + "reason": "Kernel-native causal_conv1d plumbing is unrelated to Neuron training experiments; no shared concrete code-path problem." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47175", + "accept": false, + "reason": "Audio feature/proc requirement fixes and OLMo3 CI expectation updates are unrelated test fixes." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47243", + "accept": false, + "reason": "Audio feature/proc requirement fixes do not match ExecuTorch exporter flattening fixes." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47302", + "accept": false, + "reason": "Step3p7 model build/conversion work is unrelated to tokenizer-backend docs fallback wording." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47323", + "accept": false, + "reason": "CohereASR loss-shift fix and DeepGEMM multi-device support are different subsystems and different bugs." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47275", + "accept": false, + "reason": "Audio requirement fixes and new EfficientViTSam model support are unrelated." + }, + { + "left": "pull_request:46658", + "right": "pull_request:47275", + "accept": false, + "reason": "Step3p7 flash conversion and EfficientViTSam model addition are unrelated model work." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47175", + "accept": false, + "reason": "A configuration typing fix for num_labels is unrelated to OLMo3 test expectation changes." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47253", + "accept": false, + "reason": "RoPE test updates across models do not match tensor-parallel plan handling in lm_heads." + }, + { + "left": "pull_request:47302", + "right": "pull_request:47322", + "accept": false, + "reason": "Docs wording for tokenizers fallback is unrelated to AMD Quark config class changes." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47323", + "accept": false, + "reason": "num_labels None handling and DeepGEMM multi-device fixes are unrelated." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47302", + "accept": false, + "reason": "FP8 decode/kernel optimization and tokenizer docs fallback are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46957", + "accept": false, + "reason": "Molmo2 model addition is unrelated to finegrained FP8 batching/runtime fixes." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Mistral tokenizer API additions and ExecuTorch dynamic-shape tests are unrelated." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Mistral tokenizer missing methods and MobileLLM model support are unrelated changes." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor robustness work for AudioFlamingo3/auto processors is unrelated to adding MobileLLM." + }, + { + "left": "pull_request:43424", + "right": "pull_request:47301", + "accept": false, + "reason": "ExecuTorch dynamic-shape testing and MTP docs are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "Adding MobileLLM and fixing config resolution order are unrelated code paths." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test fixes and layrnorm->layernorm renaming are different bugs and different code paths." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "ExecuTorch dynamic-shape tests and Mistral4 test fixes are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "MobileLLM model addition and Falcon output-collection refactor are unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "cross_entropy kwargs forwarding and MobileLLM model support are unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 tests and TDT loss kernel integration are unrelated." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "ExecuTorch dynamic-shape testing and MobileLLM addition are unrelated." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor and model-creation skill scaffolding are unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 test fixes and the model-creation skill are unrelated." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon refactor and Mistral4 tests do not address the same concrete bug." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "MobileLLM model addition and the model-creation skill are unrelated." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47090", + "accept": true, + "reason": "Same underlying CohereASR training-loss double-shift bug; both patch the same forward/loss path and could plausibly be merged into one fix." + }, + { + "left": "pull_request:46958", + "right": "pull_request:47090", + "accept": true, + "reason": "Same CohereASR double-shift bug and same loss computation path; 47090 is just a clearer variant of the same fix." + }, + { + "left": "pull_request:46895", + "right": "pull_request:46958", + "accept": true, + "reason": "Same CohereASR loss-shift issue; the two PRs differ in implementation detail but target the same concrete bug." + }, + { + "left": "pull_request:47173", + "right": "pull_request:47292", + "accept": false, + "reason": "Both are XPU test expectation updates, but they are model-specific failures on different code paths, not one shared bug." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47184", + "accept": false, + "reason": "Quantization test expansions and Qwen3Next/Videomt accelerator test updates are unrelated." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47175", + "accept": false, + "reason": "RoPE numerical test updates and OLMo3 CI expectation tweaks are different model/test changes." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47302", + "accept": false, + "reason": "MTP docs support and tokenizer fallback docs are unrelated documentation changes." + }, + { + "left": "pull_request:47184", + "right": "pull_request:47292", + "accept": false, + "reason": "Qwen3Next/Videomt multi-accelerator test changes and Tipsv2-DPT XPU fixes are unrelated." + }, + { + "left": "pull_request:47301", + "right": "pull_request:47304", + "accept": false, + "reason": "MTP docs and MPS graph cache docs are unrelated." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47304", + "accept": false, + "reason": "Japanese DistilBERT docs translation and MPS graph cache docs are unrelated." + }, + { + "left": "pull_request:46658", + "right": "pull_request:46836", + "accept": false, + "reason": "Step3p7 model build/conversion work and Unlimited OCR model addition are unrelated model-introduction PRs." + }, + { + "left": "pull_request:47210", + "right": "pull_request:47301", + "accept": false, + "reason": "Packed SDPA varlen attention routing and MTP docs are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47243", + "accept": false, + "reason": "OpenVINO exporter work and ExecuTorch exporter fixes are different exporter implementations and different bugs." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47143", + "accept": false, + "reason": "Nemotron Omni and NeuCodec are distinct model additions with different code and docs." + }, + { + "left": "pull_request:46376", + "right": "pull_request:46509", + "accept": false, + "reason": "Continuous batching multimodal support and Nemotron Omni model support are unrelated." + }, + { + "left": "pull_request:47291", + "right": "pull_request:47322", + "accept": false, + "reason": "Continuous-batching config deprecation and AMD Quark config changes are unrelated." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47166", + "accept": false, + "reason": "Paged-attention XPU tests and quantization test updates are unrelated." + }, + { + "left": "pull_request:47163", + "right": "pull_request:47184", + "accept": false, + "reason": "Paged-attention XPU tests and Qwen3Next/Videomt accelerator test updates are unrelated." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47181", + "accept": false, + "reason": "Nemotron Omni and Cosmos3 Edge are separate model additions, not the same underlying change." + }, + { + "left": "pull_request:46509", + "right": "pull_request:47163", + "accept": false, + "reason": "Model-addition PR and paged-attention XPU tests are unrelated." + }, + { + "left": "pull_request:46376", + "right": "pull_request:47163", + "accept": false, + "reason": "Continuous batching multimodal support and paged-attention XPU tests are unrelated." + }, + { + "left": "pull_request:47181", + "right": "pull_request:47301", + "accept": false, + "reason": "Cosmos3 Edge model support and MTP docs support are unrelated." + }, + { + "left": "pull_request:46376", + "right": "pull_request:47311", + "accept": false, + "reason": "Continuous batching multimodal support and Gemma4 H100 expectation updates are unrelated." + }, + { + "left": "pull_request:47143", + "right": "pull_request:47163", + "accept": false, + "reason": "NeuCodec model support and paged-attention XPU tests are unrelated." + }, + { + "left": "pull_request:47181", + "right": "pull_request:47311", + "accept": false, + "reason": "Cosmos3 Edge model support and Gemma4 H100 test expectations are unrelated." + }, + { + "left": "pull_request:46376", + "right": "pull_request:47322", + "accept": false, + "reason": "Continuous batching multimodal support and Quark config changes are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47243", + "accept": false, + "reason": "Molmo2 model addition and ExecuTorch exporter fixes 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": 1, + "component_count": 34, + "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-40546-76", + "nodes": [ + "pull_request:40546", + "pull_request:43451", + "pull_request:46509", + "pull_request:47003", + "pull_request:47096", + "pull_request:47143", + "pull_request:47163", + "pull_request:47166", + "pull_request:47175", + "pull_request:47181", + "pull_request:47184", + "pull_request:47253", + "pull_request:47291", + "pull_request:47298", + "pull_request:47306", + "pull_request:47311", + "pull_request:47322", + "pull_request:47323" + ], + "soft_pairs": [ + "pull_request:43451|pull_request:47291", + "pull_request:47096|pull_request:47291", + "pull_request:47253|pull_request:47306", + "pull_request:40546|pull_request:43451", + "pull_request:43451|pull_request:47175", + "pull_request:47096|pull_request:47175", + "pull_request:47166|pull_request:47311", + "pull_request:47184|pull_request:47311", + "pull_request:47181|pull_request:47306", + "pull_request:43451|pull_request:47143", + "pull_request:43451|pull_request:47163", + "pull_request:47298|pull_request:47323", + "pull_request:43451|pull_request:46509", + "pull_request:43451|pull_request:47311", + "pull_request:43451|pull_request:47096", + "pull_request:43451|pull_request:47003", + "pull_request:47003|pull_request:47096", + "pull_request:43451|pull_request:47323", + "pull_request:47096|pull_request:47323", + "pull_request:43451|pull_request:47298", + "pull_request:43451|pull_request:47322", + "pull_request:47096|pull_request:47298", + "pull_request:47003|pull_request:47323", + "pull_request:47003|pull_request:47298", + "pull_request:47096|pull_request:47143", + "pull_request:47096|pull_request:47163", + "pull_request:47143|pull_request:47298" + ], + "prepared_review_unit_hash": "b8d96a1e1fa5a8e86a7e1d7e9a79ef80e5e2332137426b15870addfaad0a7bd8", + "component_budget": { + "node_count": 76, + "item_count": 76, + "soft_pair_count": 215, + "serialized_chars": 265530, + "estimated_input_tokens": 66383, + "estimated_eval_tokens": 133022 + }, + "budget": { + "node_count": 18, + "item_count": 18, + "soft_pair_count": 27, + "serialized_chars": 50363, + "estimated_input_tokens": 12591, + "estimated_eval_tokens": 25438 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is dominated by unrelated PRs that only look superficially similar because they are all model/support/test/infra changes. None of the soft edges look like true duplicates of the same underlying change.", + "confidence": 0.57, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43451 is the most central hub in the cluster and the broadest, most representative support PR; several other items are only loosely similar to it by pattern, not by concrete code-path.", + "best_issue_reason": null, + "best_pr_reason": "PR 43451 is the best representative PR for the cluster because it is the hub with the strongest connectivity and the most complete implementation footprint among the items.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43451", + "right": "pull_request:47291", + "accept": false, + "reason": "Both involve framework/config changes, but one adds Molmo2 support and the other deprecates continuous batching config; different code paths and goals." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47291", + "accept": false, + "reason": "Exporter/Kimi changes are unrelated to generation config deprecation; same broad area, but not the same bug or fix." + }, + { + "left": "pull_request:47253", + "right": "pull_request:47306", + "accept": false, + "reason": "Tensor-parallel validation and Neuron experiment scripts are unrelated changes; no shared concrete failure mode." + }, + { + "left": "pull_request:40546", + "right": "pull_request:43451", + "accept": false, + "reason": "Both are model-support PRs, but for different models (VibeVoice vs Molmo2) and different implementation details." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47175", + "accept": false, + "reason": "Molmo2 support and Olmo3 CI expectation updates are different artifacts, with no shared underlying bug." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47175", + "accept": false, + "reason": "Kimi exportability and Olmo3 test expectation tweaks are unrelated; they do not fix the same code path." + }, + { + "left": "pull_request:47166", + "right": "pull_request:47311", + "accept": false, + "reason": "Both adjust test expectations, but for different models and hardware targets; not a duplicate fix." + }, + { + "left": "pull_request:47184", + "right": "pull_request:47311", + "accept": false, + "reason": "Qwen3 Next test-annotation cleanup and Gemma4 H100 expectation updates are separate test-only changes." + }, + { + "left": "pull_request:47181", + "right": "pull_request:47306", + "accept": false, + "reason": "Cosmos3 Edge support and Neuron experiment scripts are unrelated model/infrastructure additions." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47143", + "accept": false, + "reason": "Both add model support, but for different models (Molmo2 vs NeuCodec) with distinct files and runtime behavior." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47163", + "accept": false, + "reason": "Molmo2 model support and XPU paged-attention test coverage are unrelated changes." + }, + { + "left": "pull_request:47298", + "right": "pull_request:47323", + "accept": false, + "reason": "Both mention hardware/runtime issues, but one patches image loading fallbacks for MI450 while the other changes DeepGEMM multi-device behavior; different concrete failures." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46509", + "accept": false, + "reason": "Molmo2 and Nemotron Omni are separate model integrations, not the same underlying feature." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47311", + "accept": false, + "reason": "Molmo2 support and Gemma4 H100 test expectations are unrelated model-specific updates." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47096", + "accept": false, + "reason": "Molmo2 support and Kimi exportability are different feature areas and code paths." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47003", + "accept": false, + "reason": "Molmo2 model support and OpenVINO exporter support are different subsystems and changes." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47096", + "accept": false, + "reason": "OpenVINO exporter support and Kimi exportability are both exporter-related, but they target different models and runtime paths." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47323", + "accept": false, + "reason": "Model support for Molmo2 and DeepGEMM multi-device fixes are unrelated changes." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47323", + "accept": false, + "reason": "Kimi exportability and DeepGEMM multi-device fixes address different parts of the stack." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47298", + "accept": false, + "reason": "Molmo2 support and MI450 image-loading/model-failure fixes are unrelated code paths." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47322", + "accept": false, + "reason": "Molmo2 support and AMD Quark config class changes are distinct features with no shared fix." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47298", + "accept": false, + "reason": "Exporter support for Kimi and MI450 runtime fallback fixes are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47323", + "accept": false, + "reason": "OpenVINO exporter support and DeepGEMM multi-device fixes are separate runtime changes." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47298", + "accept": false, + "reason": "OpenVINO exporter work and MI450 image fallback fixes do not share a concrete bug or code path." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47143", + "accept": false, + "reason": "Kimi exportability and NeuCodec support are different model integrations." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47163", + "accept": false, + "reason": "Kimi exportability and XPU paged-attention tests are unrelated." + }, + { + "left": "pull_request:47143", + "right": "pull_request:47298", + "accept": false, + "reason": "NeuCodec support and MI450 image-loading fixes are unrelated model/runtime 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": 2, + "component_count": 34, + "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": "6fc7c337b5add6b7a211a4fdc8d2030ca261496352d4f5214f50a2462c22394d", + "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 feature additions in different model areas, so this cluster should be split rather than deduplicated.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: PR 40755 adds TimesFM covariate forecasting, while PR 41224 adds DINOv3 image-classification support; they touch different models, code paths, and user-facing features.", + "best_issue_reason": null, + "best_pr_reason": "No single PR is a valid global representative for this cluster because the changes are unrelated. If forced, 41224 is the more narrowly scoped PR, but it is not a duplicate of 40755.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Reject: these PRs implement different features for different architectures (TimesFM forecasting with covariates vs DINOv3 image classification) and do not fix the same underlying bug or change the same code path." + } + ] + }, + "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": 34, + "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": 34, + "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": "a2255e5043b2da60e886c9cafaf826baf6f06af807e60e9dfaf1d7365f4a8408", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate workflow-maintenance PRs, not duplicates: one upgrades several GitHub Actions for Node 24 compatibility, while the other bumps docker/build-push-action in multiple Docker image workflows. They overlap in workflow files but change different actions and address different upgrade tracks.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR 42975 is the broader and more descriptive maintenance change, covering multiple core GitHub Actions upgrades tied to Node 24 compatibility.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Different underlying changes: 42975 updates checkout/cache/upload-artifact and related workflow actions for Node 24 compatibility, while 42976 only upgrades docker/build-push-action. Shared workflow files alone are not enough to treat them 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": 5, + "component_count": 34, + "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": "5fa43a3276a9dc573c55a56a33117bed1c16b5c1d944be8157812760b58a3dd4", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 13045, + "estimated_input_tokens": 3262, + "estimated_eval_tokens": 6780 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 13045, + "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 (ViT NEPA, Molmo, CTSM, Step3p7). They share only generic Transformers scaffolding such as docs TOC entries and auto-registration files; no pair looks like the same underlying change or bug fix.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true canonical PR: the PRs target different models and different code paths. If one must be used as a representative, #46562 is the most complete and review-heavy, but it is not a duplicate of the others.", + "best_issue_reason": null, + "best_pr_reason": "#46562 is the strongest standalone PR in this set: it has the broadest model-specific implementation surface and the most review activity. It is still not a duplicate of the other PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo). Shared files are only generic docs/auto-registry scaffolding, not the same underlying change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "Different model families and implementations. Overlap is limited to shared Transformers plumbing, which is not enough to treat them as one PR." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are unrelated new-model PRs; shared filenames are broad integration points only, not a common code-path fix." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46562", + "accept": false, + "reason": "CTSM and Step3p7 are distinct model additions with different architectures and files; the overlap is generic auto/docs infrastructure, 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": 6, + "component_count": 34, + "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-43291-7", + "nodes": [ + "pull_request:43291", + "pull_request:43665", + "pull_request:44815", + "pull_request:46380", + "pull_request:46539", + "pull_request:46886", + "pull_request:46922" + ], + "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: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": "e8c6033e70abff46d5a751de5173c9c174356f44a095b749a8820d3e2f159b57", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19601, + "estimated_input_tokens": 4901, + "estimated_eval_tokens": 10058 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19601, + "estimated_input_tokens": 4901, + "estimated_eval_tokens": 10058 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly unrelated PR noise. The only true duplicate-like pair is the two YOLOS fixes, which target the same device-mismatch bug in the same file and share the same issue target; 46922 is the more complete representative.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46922 is the best canonical PR: it directly fixes the YOLOS device-mismatch path in `forward()` and also includes the `_no_split_modules` adjustment, making it the more complete version of the same underlying fix than 46886.", + "best_issue_reason": null, + "best_pr_reason": "PR 46922 is the strongest standalone candidate because it addresses the concrete YOLOS runtime bug at the actual failing code path and appears to subsume the narrower change in 46886.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both are test-focused, but they touch different models and different failures; the similarity is only superficial." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper test cleanup is unrelated to the FP8 dequant/loading fix; no shared code path or concrete bug." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46380", + "accept": false, + "reason": "Whisper tokenization/tests and generation watermark validation are different subsystems and different bugs." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46922", + "accept": false, + "reason": "Whisper decoding/tests do not overlap with the YOLOS device-mismatch fix." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46539", + "accept": false, + "reason": "These are unrelated test/workflow changes in different model areas and do not appear to fix the same issue." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "One is ad hoc test changes; the other is a genuine FP8 dequantization/loading fix. Not the same bug." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46380", + "accept": false, + "reason": "Different models and different failure modes; only broad 'fix tests' similarity." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46922", + "accept": false, + "reason": "The CLIP/ViT test edits are unrelated to YOLOS inference/device placement behavior." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46539", + "accept": false, + "reason": "FP8 quantization/loading changes and Phimoe CI/device-memory test infrastructure are unrelated." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46380", + "accept": false, + "reason": "They address different issues: FP8 dequantization vs generation config validation." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46922", + "accept": false, + "reason": "Different model families and different bug classes; no plausible single merged fix." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46539", + "accept": false, + "reason": "Generation config watermark validation and Phimoe integration-test CI setup are separate problems." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46922", + "accept": false, + "reason": "These PRs touch different subsystems and fix different code paths." + }, + { + "left": "pull_request:46539", + "right": "pull_request:46922", + "accept": false, + "reason": "CI memory limiting for Phimoe tests is unrelated to YOLOS device placement." + }, + { + "left": "pull_request:46886", + "right": "pull_request:46922", + "accept": true, + "reason": "Both PRs target the same YOLOS device-mismatch bug in `src/transformers/models/yolos/modeling_yolos.py` and share the same issue target; 46922 effectively subsumes the narrower fix in 46886." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft-edge pair, 46886/46922, is clearly mergeable: both target the same YOLOS device-mismatch issue in the same file and share the same issue target, with 46922 subsuming the narrower `_no_split_modules` change. The remaining rejections are reasonably supported by the diffs and titles as unrelated or at most superficially similar." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 7, + "component_count": 34, + "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": "bfb47689d01057127d3093d60b2eeaca6408ae9a09c0758c51281759fe6805bd", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 6348, + "estimated_input_tokens": 1587, + "estimated_eval_tokens": 3430 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 6348, + "estimated_input_tokens": 1587, + "estimated_eval_tokens": 3430 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate '[don't merge]' test PRs with different concrete changes: packaging/dependency setup, assorted CI/comment workflow changes, and a PR security-gate workflow change. They are not duplicates of the same underlying bug or change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46747 is the cleanest single-purpose change: a small, focused workflow edit in one file. It is the best representative only in the sense of being the most self-contained artifact here, not because it duplicates the others.", + "best_issue_reason": null, + "best_pr_reason": "PR 46747 is the most focused and least noisy of the three, with one workflow file and a single code-path change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:46747", + "accept": false, + "reason": "Different changes: 43488 edits setup/dependency metadata and a test file, while 46747 changes the PR security-gate workflow trigger. Same broad 'test' theme, but not the same underlying fix." + }, + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "43488 is about package/version/dependency setup plus a Vit test tweak; 43532 is about CI/comment workflows and related test/support files. The overlap is incidental, not a shared concrete bug or change." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46747", + "accept": false, + "reason": "Both touch GitHub Actions, but they modify different workflows for different purposes: self-comment CI vs security gate. They are not plausibly the same mergeable 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": 8, + "component_count": 34, + "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": "6aefa12ef70689b4feec190b7038e3d305fde925897c7de55bd71ac6aa38f10e", + "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": "Two related but distinct PRs adding different segmentation pipelines. They share infrastructure/doc updates, but target different tasks, model integrations, and code paths.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43613 is the better representative of the cluster because it is the broader, more complete promptable-visual-segmentation implementation and includes multiple model/test integrations; 43612 is a separate promptable-concept-segmentation pipeline.", + "best_issue_reason": null, + "best_pr_reason": "PR 43613 is the strongest standalone change: it adds a full visual segmentation pipeline, docs, auto-mapping, and tests across several relevant models. It is more substantial and coherent than 43612.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Both are segmentation-related and share boilerplate docs/registry updates, but they implement different pipelines for different tasks and model sets. They are not the same underlying change and should not be merged 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": 9, + "component_count": 34, + "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": "1184515d962f1fc5dff8050a36988813ec59a608197996093e76c996a431c557", + "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 PRs both touch `trainer.py` and add override hooks for custom trainers, but they address different code paths: gradient-accumulation loss scaling vs. dataloader column/signature filtering. They are not the same underlying change and should not be deduplicated.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the two pull requests fix different trainer behaviors and are not mergeable into one coherent change.", + "best_issue_reason": null, + "best_pr_reason": "No single best PR for a duplicate cluster here; each PR is a separate, focused improvement to a different trainer subsystem.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different bugs/changes: 43651 adjusts loss scaling during training_step, while 43663 refactors signature-column filtering for dataset column pruning. Shared file alone is not enough." + } + ] + }, + "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": 34, + "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": "6702be00d97ef3b6aa42c9cee0dae55c1eb745916f771019ac1e88893fb3296d", + "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 touch the same file and both address annotation-related NameErrors in the CLI, but they target different fixes and different issues: one is broader future-annotations/type-hint cleanup, the other is a narrow optional-pydantic TypeAdapter fix.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43842 is the cleaner canonical PR: it directly fixes the TypeAdapter NameError when pydantic is absent, with a focused change that matches its issue and title.", + "best_issue_reason": null, + "best_pr_reason": "PR #43842 is the most representative of the cluster because it is a precise, self-contained fix for the concrete runtime failure in `src/transformers/cli/serve.py`.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Reject: although both are in `src/transformers/cli/serve.py` and concern annotation errors, #43656 is a broader `from __future__ import annotations` / type-hint cleanup tied to issue 43576, while #43842 specifically fixes a `TypeAdapter` NameError when pydantic is not installed for issue 43824. They are not the same concrete 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": 11, + "component_count": 34, + "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": 34, + "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:44018|pull_request:44068", + "pull_request:44066|pull_request:44071", + "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:44161", + "pull_request:44073|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44018|pull_request:44059", + "pull_request:44056|pull_request:44101", + "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": "8fda3c0773e46d177e288178f7bd4b14b788758cb56e1319c11765828f215589", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84505, + "estimated_input_tokens": 21127, + "estimated_eval_tokens": 42510 + }, + "budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 72, + "serialized_chars": 76938, + "estimated_input_tokens": 19235, + "estimated_eval_tokens": 38726 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a broad wave of model-specific output-tracing refactors, so most pairs are not true duplicates. The only clear duplicate-style overlaps are the GPT-J/CodeGen refactor set around 44066/44085/44722 and the GPT-Neo pair 44018/44068.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "44722 is the best representative of the duplicate subset: it targets the same GPT-J/CodeGen output-tracing code path as 44066/44085, touches the same files, and looks like the most complete version of that concrete change.", + "best_issue_reason": null, + "best_pr_reason": "Best PR for the cluster's main duplicate set. It most cleanly represents the shared GPT-J/CodeGen output-tracing refactor, while the rest of the cluster is mostly separate model-by-model work.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families and code paths; same tracking theme only." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "CVT/FNet vs DeBERTa V2; not the same underlying change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Both are GPT-J output-tracing refactors on the same core code path; 44066/44085 overlap materially." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "ResNet/RegNet vs EfficientNet; separate model implementations." + }, + { + "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": "Different models and implementation paths; only the refactor theme is shared." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 vs DeBERTa V2; no shared concrete bug or change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen vs MGP-STR; same general refactor pattern but not the same code path." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both target GPT-Neo output tracing in the same model file/test area; plausible duplicate or direct revision of the same change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen and MPT are unrelated model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen vs GPT-Neo; similar refactor pattern, different code path." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen; not the same underlying model change." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet and GPT-J/CodeGen are separate model families." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen do not share a code path." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet vs GPT-J; same theme only, not the same fix." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are unrelated implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet/RT-DETR backbone work vs MPT; no concrete overlap." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are different model code paths." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen; related pattern, but not the same model change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT vs GPT-J/CodeGen are different implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo vs GPT-J; same refactor theme, different code path." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated model implementations." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo vs MGP-STR; not the same underlying change." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR touch different models and logic." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are unrelated code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 vs MGP-STR; shared refactor theme only." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT vs Flaubert/XLM; separate model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are different model implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44071", + "accept": false, + "reason": "SegFormer and MPT are unrelated model paths." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet and Flaubert/XLM are different code paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 are unrelated model implementations." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet and GPT-Neo are unrelated." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 are separate model code paths." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet and LongT5 are unrelated models." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo vs GPT-2; similar refactor pattern, not the same change." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "Flaubert/XLM and LongT5 are different model families." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet vs MPNet; no concrete overlap." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are unrelated model code paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBert and MPT are different implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-Neo and SpeechT5 are unrelated model paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBert and VisualBert are different code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa V2 and TextNet are unrelated model implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44018", + "accept": false, + "reason": "SqueezeBert and GPT-Neo are not the same change." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are unrelated." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44161", + "accept": false, + "reason": "VisualBert and LongT5 are different model paths." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44101", + "accept": false, + "reason": "VisualBert vs Flaubert/XLM; only broad output-tracing theme matches." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are unrelated models." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert/XLM are different code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are separate refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are unrelated model implementations." + }, + { + "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." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", + "accept": false, + "reason": "CVT/FNet vs SegFormer; separate model code paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBert and DeBERTa V2 are unrelated implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa V2 and SpeechT5 are different model paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa V2 and VITS do not share a concrete fix." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet and ImageGPT are unrelated model code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: it correctly centers the GPT-J/CodeGen duplicate set and treats the rest as mostly separate model-specific refactors. The accepted soft-edge pairs are plausible and mergeable as single bugfix PRs, and the rejected pairs are stated conservatively without overclaiming duplication." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 34, + "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": "a46aabcb8bd3c72f7769b93be898d654c3e24bbff3ca4dcc7eb76e295d5426f0", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84505, + "estimated_input_tokens": 21127, + "estimated_eval_tokens": 42510 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 5, + "serialized_chars": 13067, + "estimated_input_tokens": 3267, + "estimated_eval_tokens": 6790 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These PRs are all model-specific migrations toward standardized output tracing, but they affect different architectures and code paths. They are related in theme, not duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44073 is the best representative of the cluster: it shows the core standardized output-tracing refactor clearly, adds the model-specific output-capture mapping, and uses the new capture_outputs/can_return_tuple pattern in a canonical way.", + "best_issue_reason": null, + "best_pr_reason": "PR #44073 is the strongest single exemplar of the shared refactor theme, with explicit output-capture plumbing and a clean model migration. The others are model-specific variants rather than the same change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Same broad migration pattern, but different models and concrete code paths (VisualBert vs TextNet). This is not one fix that could be merged into a single PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "Both are output-tracing refactors, but for different architectures (MPNet vs VisualBert) with different layer mappings and implementation details; not the same underlying change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "Shared refactor theme only. MPNet and TextNet have distinct model internals and output-capture wiring, so these are not duplicates." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "Both migrate to the new tracing helpers, but RegNet and SqueezeBert are unrelated model implementations and the diffs do not indicate the same concrete bug or patch." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBert\u2019s decorator migration and SpeechT5\u2019s output/caching signature changes are different code-path changes in different models; they should 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": 13, + "component_count": 34, + "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:44028", + "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": "45a6cdceef7fe949a867ae2aa714569765ee692e50e5a524d5549ebdb0a99b02", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40182, + "estimated_input_tokens": 10046, + "estimated_eval_tokens": 20348 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40182, + "estimated_input_tokens": 10046, + "estimated_eval_tokens": 20348 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a set of model-specific pull requests that all refactor output tracing / tuple-return behavior, but they target different architectures and code paths. I found no true duplicate PR pairs among the soft candidates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44003 is the best representative of the cluster: it covers a broader, more central output-tracing refactor across the related Mamba/FalconMamba implementations, making it more representative than the single-model wrapper changes elsewhere.", + "best_issue_reason": null, + "best_pr_reason": "PR 44003 is the strongest single example of the common theme because it changes the core tracing pattern in closely related model code, rather than only adjusting one isolated model wrapper.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families and files; both are output-tracing refactors, but not the same concrete change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different architectures (`timm_backbone` vs `vision_text_dual_encoder`); same theme, not the same bug or patch." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "`timm_backbone` and `univnet` are unrelated code paths; too broad to be duplicates." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Both are tracing refactors, but they affect different model families and distinct implementations." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Separate models and separate forward paths; no shared concrete fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "`mobilenet_v1` and `univnet` are unrelated; similarity is only at the refactor theme level." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different models and different output types; not mergeable as one PR." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families (`mobilenet_v1` vs `depth_anything`); same broad refactor pattern, not the same change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "`mobilenet_v1` and `vision_encoder_decoder` touch unrelated code paths." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "Different model implementations; no shared bug or patch." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "`vision_text_dual_encoder` and `univnet` are unrelated models, so this is only thematic similarity." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different architectures and files; not the same concrete output-tracing change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "`vision_text_dual_encoder` and `falcon_mamba/mamba` are distinct code paths; reject as duplicate." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families; both refactor outputs, but not the same fix." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "`univnet` and `upernet` are unrelated implementations." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and different forward signatures; not one underlying change." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "Separate model code paths; similarity is too coarse for duplicate triage." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "`upernet` and `rwkv` are unrelated models with different output-capture logic." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Different architectures and modules; no plausible single merged PR." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "`falcon_mamba/mamba` and `superpoint` are unrelated models; only the refactor style matches." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "`mamba` family versus `codegen`; different code paths and different API changes." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Different models and different refactor scopes; not the same underlying change." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "`univnet` and `codegen` are unrelated; this is not a duplicate pair." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model code paths and different output contracts." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "`depth_anything` and `vision_encoder_decoder` are separate model families; same meta-pattern only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "`depth_anything` and `speech_encoder_decoder` are not the same concrete bug or change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Different models; one is depth estimation, the other keypoint detection." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model implementations; no shared code path." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "`depth_anything` and `dpr` are unrelated models." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Two different encoder-decoder wrappers; similar refactor, but not the same PR." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "`vision_encoder_decoder` and `superpoint` address different subsystems." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Different models and different output-tracing code paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "`vision_encoder_decoder` and `dpr` are separate model families." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "`speech_encoder_decoder` and `superpoint` are unrelated changes." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures; no concrete overlap beyond generic refactoring." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "`speech_encoder_decoder` and `dpr` are different code paths." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "`superpoint` and `rwkv` are unrelated models." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model families and different tracing implementations." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "`rwkv` and `dpr` are unrelated; not mergeable 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": 14, + "component_count": 34, + "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": "e2dceb60cb6e22461f95e1294acdf940a708e4822bccdd46f9ae2f8d4aeaaa4c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6552, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6552, + "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: one adds PenguinVL, the other adds Voxtral TTS. They share some boilerplate files for model registration/docs, but they do not fix the same code path or implement the same feature.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No duplicate PR pair here; the closest representative is #44662 because it is the broader, older model implementation PR, but it is not a duplicate of #45401.", + "best_issue_reason": null, + "best_pr_reason": "#44662 is the better cluster representative because it is the more comprehensive and established implementation PR, while #45401 is a separate model integration.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different models and different underlying implementations (PenguinVL vs Voxtral TTS). Shared auto/docs files are boilerplate, not evidence of the same bug/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": 34, + "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": "82b6e899cf7b709a4eacd4f442bcb77ec55baa6d7727c618df955805837a812d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related by the general theme of model/kernel fusion, but they are not the same change: one introduces a standalone Module Fusion API, while the other implements n-to-1 kernel fusion through KernelConfig and integration hooks.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR for a duplicate set: PR #44979 adds a new Module Fusion API surface, while PR #45363 adds a separate KernelConfig-based kernel fusion feature.", + "best_issue_reason": null, + "best_pr_reason": "PR #45363 is the better representative of the cluster topic because it is the broader, more integrated user-facing fusion feature; PR #44979 is a separate foundational API addition.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979 (Module Fusion API)", + "right": "pull_request:45363 (n-to-1 kernel fusion via `KernelConfig`)", + "accept": false, + "reason": "They share a fusion theme, but the code changes implement different abstractions and use cases. PR #44979 adds a generic module-fusion API, while PR #45363 wires kernel fusion into model loading/integration paths; they would not plausibly be merged as one 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": 16, + "component_count": 34, + "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": "b4b574542e26fa1ed2da2bd59f2d7f82d5c94264f69d65fa3d35e5061b5c12ef", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4413, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4413, + "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 address the same Mistral tokenizer AttributeError in `_patch_mistral_regex` by switching from `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; the later PR adds tests and is the stronger candidate.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45317 is the better canonical PR because it implements the same fix as #45086 but also includes test coverage, making it the more complete change set.", + "best_issue_reason": null, + "best_pr_reason": "PR #45317 is the best overall PR candidate: same concrete bug fix, clearer title, and added regression tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying code-path bug in `_patch_mistral_regex` causing an AttributeError; the second PR is a test-augmented version of the same fix and could plausibly be merged as one 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": 17, + "component_count": 34, + "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": "9826515c08658a68eb70c52442b47f3ec7eabe9084b25426fc6ea597ded8cc19", + "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 PRs are related NemotronH architecture cleanup work, but they are not the same change: one refactors NemotronH to inherit GraniteMoeHybrid and adds conversion mappings, while the other splits NemotronH into dense/sparse variants and updates docs/auto-mapping accordingly.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45569 is the broader, more representative NemotronH rework: it introduces the dense/sparse model split, updates docs, and touches the model auto-mapping path across the package.", + "best_issue_reason": null, + "best_pr_reason": "PR #45569 is the best single representative of the cluster because it covers the more complete product-level change and subsumes the visible model-organization work, whereas #45115 is a narrower implementation refactor around inheritance/conversion mapping.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "They share NemotronH files and a general architectural theme, but the concrete changes differ: #45115 is a refactor to inherit GraniteMoeHybrid and add conversion renames, while #45569 introduces separate dense/sparse variants plus docs and auto-mapping updates. They are related, not the same underlying 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": 18, + "component_count": 34, + "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": "199898ce010fc5fc892ac7a6f2b7d34dccc5a575effd432a9afc8050cc72a891", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5513, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5513, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests are typing-focused, but they are not the same change: #45415 is a broad type-checking sweep across many `src/transformers/*.py` files, while #45425 adds specific typing helpers/protocols and touches `modeling_utils`-related code. They overlap in a few filenames, but the concrete code-path changes differ enough that they should not be merged as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45425 is the more specific, self-contained typing change with a clearer scope and smaller blast radius. PR #45415 is a broad repository-wide type-checking sweep, so it is less suitable as a canonical representative for this cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #45425 is the best representative because it targets a concrete typing refactor around `_typing.py`, `modeling_utils`, and related utilities, whereas #45415 is an expansive annotation pass that is too broad to serve as the best single exemplar.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Do not treat as duplicates: both are typing-related, but #45415 is a wide type-checking sweep across many files, while #45425 adds specific typing protocols/helpers and related `modeling_utils` updates. They do not appear to fix the same concrete problem or be safely merged into 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": 19, + "component_count": 34, + "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": "2de982b868b57affdd9b7865289ccab1a05ce0eb16ccd2a9dd283e3a8da6dfdc", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "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 concern expert parallelism in Transformers, but they fix different failure modes: one handles serialized distributed config / ZeRO-3 loading, the other fixes experts being overwritten during FSDP2-style sharded loading. They overlap in subsystem, not in the underlying bug.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45662 is the stronger representative because it targets a concrete loading corruption path for expert parameters and includes the most directly relevant code changes around sharded parameter assignment.", + "best_issue_reason": null, + "best_pr_reason": "PR #45662 is the best overall PR here: it addresses a specific, user-visible expert-parallel loading bug with clear code-path coverage across loading and tensor-parallel integrations.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Do not merge as duplicates: #45548 fixes EP + DeepSpeed ZeRO-3/accelerate launch serialization and adds has_ep support, while #45662 fixes experts being silently overwritten by rank-0 broadcast during sharded loading. They touch adjacent EP machinery but not the same bug or mergeable change set." + } + ] + }, + "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": 34, + "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": "c53d48fec60e4187c3907553cfc9d7293087751881810550aebfaa59f51e931b", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4673, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4673, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same xdist collision problem around patched testing debug artifacts, with PR 45645 effectively subsuming the core fix from 45639 and adding CI log collection support.", + "confidence": 0.83, + "canonical_issue_reason": "Issue 45561 is the shared tracker for the xdist collision in captured/patched testing debug output.", + "canonical_pr_reason": "PR 45645 is the stronger canonical PR because it includes the core xdist-safe artifact fix plus the CI workflow update to collect multiple captured_info files.", + "best_issue_reason": "Issue 45561 is the only explicit issue target and matches the underlying bug described by both PRs.", + "best_pr_reason": "PR 45645 is the best representative because it covers the same core code-path fix as 45639 and also addresses downstream CI log preservation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "They share the same underlying xdist-safe captured-info fix in testing_utils; 45645 is a broader follow-on that appears to subsume 45639 rather than a separate 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": 21, + "component_count": 34, + "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": 34, + "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": "c79cf511183156b9e58724aa55ef41be49ecfb68fcc1d77de2c4ca1fbdf3aa2c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5525, + "estimated_input_tokens": 1382, + "estimated_eval_tokens": 3020 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5525, + "estimated_input_tokens": 1382, + "estimated_eval_tokens": 3020 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related distributed-training PRs but not duplicates: one is a focused EP + Trainer integration, the other is a broad distributed branch base spanning FSDP, MoE, generation, and CI plumbing.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46269 is the broader umbrella change and the better representative of the distributed refactor/integration work, while #46126 is a narrower feature slice.", + "best_issue_reason": null, + "best_pr_reason": "PR #46269 is the strongest single PR to keep in a cluster because it covers the widest set of distributed changes and subsystems.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46126", + "right": "pull_request:46269", + "accept": false, + "reason": "Reject: they share some distributed files, but the concrete changes differ. #46126 is specifically EP + Trainer integration on top of DistributedConfig, while #46269 is a much larger branch-base PR adding FSDP CI and broader distributed/model integration. They do not look like the same underlying change or something that could plausibly be merged into 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": 23, + "component_count": 34, + "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": "1e3202457ce4f4cff1e6efa583b868ea52fe8e73bbcd5bfcb67cafb41a5deb34", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 31, + "serialized_chars": 26675, + "estimated_input_tokens": 6669, + "estimated_eval_tokens": 13594 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 31, + "serialized_chars": 26675, + "estimated_input_tokens": 6669, + "estimated_eval_tokens": 13594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly a bundle of unrelated CI/workflow maintenance PRs. The only tight subgroup is the pr-ci-security-gate.yml changes, but those PRs make different security/permission/reusable-workflow edits and are not duplicates of the same underlying fix.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46554 is the best single anchor if one must choose: it is the most substantive and central change within the only closely related subgroup (pr-ci-security-gate.yml), but it still represents a distinct workflow update rather than a duplicate.", + "best_issue_reason": null, + "best_pr_reason": "PR 46554 is the most representative PR in the cluster because it has the strongest connectivity and the clearest concrete workflow/security-gate logic change among the related items.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46138", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflows and different problems: self-comment CI argument handling vs build-ci-docker-images commit-message parsing." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows; one updates self-comment CI stdin handling, the other adds PR security-gate checks." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflows and unrelated changes; not the same bug or fix path." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow update and different security-gate behavior; shared CI theme is too broad." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflows and different code paths; not a duplicate change." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46501", + "accept": false, + "reason": "SSH runner hardening is unrelated to anti-slop workflow edits." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46502", + "accept": false, + "reason": "SSH runner workflow changes are unrelated to assign-reviewers checkout behavior." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflows and different runtime behavior; not the same underlying fix." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46554", + "accept": false, + "reason": "SSH runner hardening vs PR security-gate logic; related only at a very broad CI level." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflows and distinct security concerns; not mergeable as one fix." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46563", + "accept": false, + "reason": "SSH runner changes are unrelated to the reusable PR CI security-gate update." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflows and different security-hardening contexts; not a duplicate." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46502", + "accept": false, + "reason": "Anti-slop workflow vs assign-reviewers workflow; different functionality and no shared bug." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46503", + "accept": false, + "reason": "Distinct CI workflows with unrelated changes." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflows and different fixes; same broad CI area is not enough." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46554", + "accept": false, + "reason": "Anti-slop workflow tweak is unrelated to PR security-gate logic changes." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46503", + "accept": false, + "reason": "Assign-reviewers vs benchmark workflow; different code paths and purposes." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46504", + "accept": false, + "reason": "Unrelated workflow files and distinct behavior changes." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46554", + "accept": false, + "reason": "Assign-reviewers checkout ref change is unrelated to PR security-gate hardening." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46504", + "accept": false, + "reason": "Benchmark gating change vs build-ci-docker-images commit-message parsing; different fixes." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows and no shared concrete bug." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46554", + "accept": false, + "reason": "Build-ci-docker-images change is unrelated to PR security-gate workflow changes." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow files and different operational concerns." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow files; not the same underlying change." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow files and different fix scope." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46558", + "accept": false, + "reason": "Same file, but different changes: one adds checkout/diff-based security checks, the other changes permissions and approval flow." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46563", + "accept": false, + "reason": "Same workflow file, but one switches to a reusable workflow and adds comments while the other adds concrete gate logic; not the same fix." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46716", + "accept": false, + "reason": "Same workflow file, but one changes security checks and the other changes secrets handling; distinct patches." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46563", + "accept": false, + "reason": "Both touch the same workflow file, but they address different aspects of the gate and are not the same code-path problem." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46716", + "accept": false, + "reason": "Different edits to the same workflow file: permissions/approval flow vs secrets inheritance; not duplicates." + }, + { + "left": "pull_request:46563", + "right": "pull_request:46716", + "accept": false, + "reason": "Reusable-workflow commentary/invocation change vs secrets configuration change; distinct fixes." + } + ] + }, + "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": 34, + "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": 34, + "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": "7807af6dc58bf033998dbc0e99c240c285965d8df0e3ac8c30956a16399d1b83", + "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 address an off-by-one in the right side of a sliding window mask, but PR 47162 also expands the fix to Gemma4 and adds tests, so it is broader than PR 46310 rather than a clean duplicate.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47162 is the stronger canonical PR: it fixes the same MoonshineStreaming boundary bug as 46310, extends the correction to Gemma4, and includes test coverage.", + "best_issue_reason": null, + "best_pr_reason": "PR 47162 is the best overall PR because it covers the broader code-path impact and includes tests, while 46310 is a narrower subset fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46310", + "right": "pull_request:47162", + "accept": false, + "reason": "Related off-by-one fixes, but not the same single change: 47162 also modifies Gemma4 and test coverage, so these are not a clean one-to-one duplicate pair." + } + ] + }, + "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": 34, + "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": "205e93df042b5bd7c3bd7d4127a5ef680af611b34fd08e1341901e06ffae286c", + "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": "The two PRs are unrelated: #46574 fixes CodeLlama decode stripping/leading-space handling, while #46580 fixes DeepSeek-Coder/DeepSeek-LLM tokenizer-class selection in auto tokenization.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the diffs touch different code paths and solve different bugs, so neither PR is a duplicate of the other.", + "best_issue_reason": null, + "best_pr_reason": "#46574 is the more self-contained bugfix with a direct issue target, but it is not a representative match for #46580 because the underlying problems differ.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46574", + "right": "pull_request:46580", + "accept": false, + "reason": "Reject: one changes CodeLlama decoder initialization to preserve leading spaces; the other adds checkpoint-prefix overrides so DeepSeek models load the correct tokenizer class. Different bugs, different fix paths." + } + ] + }, + "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": 34, + "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-46696-2", + "nodes": [ + "pull_request:46696", + "pull_request:47215" + ], + "soft_pairs": [ + "pull_request:46696|pull_request:47215" + ], + "prepared_review_unit_hash": "7990a36faed7c891ab9e64200610b8f66f1afa3faad1a7d69b252a84a3924b2c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4579, + "estimated_input_tokens": 1145, + "estimated_eval_tokens": 2546 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4579, + "estimated_input_tokens": 1145, + "estimated_eval_tokens": 2546 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same FA2 long-context performance regression in `modeling_flash_attention_utils.py`, centered on eager-vs-compile handling of scalar lengths to avoid syncs. PR #47215 is the broader, later fix and adds related adjustments/tests, so it is the better canonical PR.", + "confidence": 0.84, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #47215 is the more complete fix for the shared FA2 long-context perf regression: it keeps scalar lengths as tensors only when compiling and otherwise converts eagerly, while also covering related helper paths and tests. PR #46696 is a narrower variant of the same underlying change.", + "best_issue_reason": null, + "best_pr_reason": "PR #47215 is the best representative because it most fully addresses the regression and includes the supporting test/benchmark updates, making it the most useful single artifact for triage.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46696", + "right": "pull_request:47215", + "accept": true, + "reason": "Same underlying FA2 long-context decode/perf regression in the flash-attention helper path; both fix eager-vs-compile scalar handling to avoid costly syncs and could plausibly be merged into 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": 28, + "component_count": 34, + "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": "4330197a5f33cf9ad16398d4507f205088c119c96cb8a7aed55d5ad8cba9f856", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5432, + "estimated_input_tokens": 1358, + "estimated_eval_tokens": 2972 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5432, + "estimated_input_tokens": 1358, + "estimated_eval_tokens": 2972 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs both touch padding-mask logic, but they address different bugs in different execution paths: one fixes batch-size-1 masking in linear-attention model forwards, the other fixes recurrent-layer masking being skipped on continued forwards/cache continuation.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #47087 is the cleaner canonical representative because it has an explicit issue link, a narrower and well-scoped fix, and the diff targets a single shared masking utility rather than many model-specific forward paths.", + "best_issue_reason": null, + "best_pr_reason": "PR #47087 is the strongest standalone candidate: it clearly states the bug, links to an issue, and fixes a concrete recurrent-mask code path with focused tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46773", + "right": "pull_request:47087", + "accept": false, + "reason": "Reject: both concern padding masks, but #46773 fixes batch-size-1 masking being skipped in linear-attention model forwards, while #47087 fixes recurrent masking being skipped on continued forwards/chunked prefill/cache continuation. Different underlying bugs and not plausibly one merged 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": 29, + "component_count": 34, + "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": "3c1a3355a7069df85c403c705e3de90f4ed93d39bc487abb2f069c5ce5d5686f", + "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": "These are three unrelated pull requests: one fixes a key-name bug in AnyToAnyPipeline, one changes Gemma4 processor placeholder validation, and one adjusts candidate-generation length capping. None appear to be duplicates or mergeable into a single PR.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47074 is the clearest and most self-contained change, with the strongest discussion activity, but it is not actually a duplicate of the other two PRs.", + "best_issue_reason": null, + "best_pr_reason": "PR 47074 is the best representative only in the sense that it is the most focused and well-scoped of the three; it should not be treated as the same change as 47076 or 47195.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47074", + "right": "pull_request:47076", + "accept": false, + "reason": "Different code paths and bugs: 47074 normalizes AnyToAnyPipeline input keys, while 47076 validates Gemma4 audio/image placeholder batching. They touch different processors and fix different behaviors." + }, + { + "left": "pull_request:47074", + "right": "pull_request:47195", + "accept": false, + "reason": "Unrelated functionality: 47074 fixes multimodal pipeline input keys, while 47195 changes generation candidate length limiting. No shared concrete bug or mergeable combined patch is evident." + } + ] + }, + "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": 34, + "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-47105-2", + "nodes": [ + "pull_request:47105", + "pull_request:47129" + ], + "soft_pairs": [ + "pull_request:47105|pull_request:47129" + ], + "prepared_review_unit_hash": "12ae9a914892d4ccb1dce9188cf03e8b5d4c7227140baae64df08b97acd996ee", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4181, + "estimated_input_tokens": 1046, + "estimated_eval_tokens": 2348 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4181, + "estimated_input_tokens": 1046, + "estimated_eval_tokens": 2348 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are not duplicates: one re-enables a VITS model-forward test, while the other fixes RT-DETR v2 weight-conversion key mappings. They touch different code paths and should remain separate.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47129 is the more substantive code fix, but there is no shared underlying change that would make these PRs a duplicate cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR 47129 is the better representative of the cluster because it changes production conversion logic, whereas PR 47105 is only a test re-enable.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47105", + "right": "pull_request:47129", + "accept": false, + "reason": "Different subsystems and different defects: VITS test expectations versus RT-DETR v2 converter key mapping. They do not fix the same concrete bug and would not plausibly merge into 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": 31, + "component_count": 34, + "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:47150", + "pull_request:47224", + "pull_request:47266", + "pull_request:47280", + "pull_request:47281", + "pull_request:47282", + "pull_request:47310" + ], + "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:47150", + "pull_request:47124|pull_request:47282", + "pull_request:47124|pull_request:47281", + "pull_request:47150|pull_request:47310", + "pull_request:47150|pull_request:47280", + "pull_request:47150|pull_request:47282", + "pull_request:47150|pull_request:47281", + "pull_request:47150|pull_request:47224", + "pull_request:47266|pull_request:47280", + "pull_request:47266|pull_request:47281", + "pull_request:47124|pull_request:47266", + "pull_request:47266|pull_request:47310", + "pull_request:47280|pull_request:47281", + "pull_request:47280|pull_request:47310", + "pull_request:47281|pull_request:47282", + "pull_request:47282|pull_request:47310", + "pull_request:47280|pull_request:47282" + ], + "prepared_review_unit_hash": "d90ba3baa839bc0afa94f4ef4823f7aedf4effc3b9f4ecaa2c4a24c4047c34b6", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 27, + "serialized_chars": 29375, + "estimated_input_tokens": 7344, + "estimated_eval_tokens": 14944 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 27, + "serialized_chars": 29375, + "estimated_input_tokens": 7344, + "estimated_eval_tokens": 14944 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: one real duplicate pair stands out for MiniMax-M3-VL load_error/offload_folder handling, while the rest are unrelated model-specific fixes. The broad umbrella PRs are not duplicates of the focused ones because they bundle additional unrelated changes.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47224 is the broad umbrella fix with the widest scope and strongest centrality in the cluster, but it is not a clean duplicate of the focused model-specific PRs; it best represents the cluster\u2019s mixed serge-failure batch.", + "best_issue_reason": null, + "best_pr_reason": "PR 47224 is the best representative because it is the largest umbrella change and links to multiple files/code paths seen in the cluster, even though the cluster does not form a single true duplicate set.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47122", + "right": "pull_request:47281", + "accept": false, + "reason": "Different models and code paths: Edgetam import_or_config/backbone kwargs bug versus GLM46V integration changes." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47123", + "accept": false, + "reason": "Both are test-failure fixes, but for different models and different failure causes; not the same underlying bug." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47280", + "accept": false, + "reason": "Same symptom class, but different models and different artifacts; this is not the same concrete change." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47224", + "accept": false, + "reason": "47224 bundles the Edgetam fix with several unrelated fixes, so the PRs are not duplicates as a whole." + }, + { + "left": "pull_request:47122", + "right": "pull_request:47310", + "accept": false, + "reason": "Edgetam backbone kwarg cleanup versus Gemma3 cache_implementation handling are unrelated." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47280", + "accept": true, + "reason": "Same MiniMax-M3-VL load_error in the same test file; both add offload-folder handling, and 47280 is a fuller version of the same fix." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47266", + "accept": false, + "reason": "Same load_error label, but different models and loading scenarios; not the same underlying bug." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47281", + "accept": false, + "reason": "MiniMax-M3-VL test loading versus GLM46V model changes are unrelated." + }, + { + "left": "pull_request:47123", + "right": "pull_request:47124", + "accept": false, + "reason": "Different model and failure mode: MiniMax load_error versus Falcon-Mamba output mismatch." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47310", + "accept": false, + "reason": "Falcon-Mamba output mismatch and Gemma3 cache_implementation behavior are distinct code paths." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47150", + "accept": false, + "reason": "Falcon-Mamba generation output mismatch versus Whisper max_length adjustment are unrelated bugs." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47282", + "accept": false, + "reason": "Falcon-Mamba output mismatch versus Mamba2 OOM is not the same issue." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47281", + "accept": false, + "reason": "Falcon-Mamba generation fix versus GLM46V changes are unrelated." + }, + { + "left": "pull_request:47150", + "right": "pull_request:47310", + "accept": false, + "reason": "Both involve generation behavior, but Whisper max_length setting and Gemma3 cache_implementation defaulting are different concrete problems." + }, + { + "left": "pull_request:47150", + "right": "pull_request:47280", + "accept": false, + "reason": "Whisper generation fix versus MiniMax-M3-VL load_error are unrelated." + }, + { + "left": "pull_request:47150", + "right": "pull_request:47282", + "accept": false, + "reason": "Whisper generation logic and Mamba2 OOM test adjustments are unrelated." + }, + { + "left": "pull_request:47150", + "right": "pull_request:47281", + "accept": false, + "reason": "Whisper generation behavior versus GLM46V model changes are unrelated." + }, + { + "left": "pull_request:47150", + "right": "pull_request:47224", + "accept": false, + "reason": "47224 includes the Whisper fix but also unrelated tokenizer/Edgetam changes, so it is not the same standalone PR/change." + }, + { + "left": "pull_request:47266", + "right": "pull_request:47280", + "accept": false, + "reason": "Different models and different load_error causes; not duplicates." + }, + { + "left": "pull_request:47266", + "right": "pull_request:47281", + "accept": false, + "reason": "DeepSeekV3.2 loading/offload handling versus GLM46V changes are unrelated." + }, + { + "left": "pull_request:47124", + "right": "pull_request:47266", + "accept": false, + "reason": "Falcon-Mamba output mismatch versus DeepSeekV3.2 load_error are unrelated." + }, + { + "left": "pull_request:47266", + "right": "pull_request:47310", + "accept": false, + "reason": "DeepSeek load_error and Gemma3 generation cache behavior are different bugs." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47281", + "accept": false, + "reason": "MiniMax-M3-VL load_error fix versus GLM46V changes are unrelated." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47310", + "accept": false, + "reason": "MiniMax-M3-VL load_error fix versus Gemma3 generation cache behavior are unrelated." + }, + { + "left": "pull_request:47281", + "right": "pull_request:47282", + "accept": false, + "reason": "GLM46V changes versus Mamba2 OOM test fix are unrelated." + }, + { + "left": "pull_request:47282", + "right": "pull_request:47310", + "accept": false, + "reason": "Mamba2 OOM-related test adjustment versus Gemma3 cache_implementation fix are unrelated." + }, + { + "left": "pull_request:47280", + "right": "pull_request:47282", + "accept": false, + "reason": "MiniMax-M3-VL load_error fix versus Mamba2 OOM fix 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": 32, + "component_count": 34, + "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": "db90b54083490ae5cda54a85df6d7ddde35819d1f8f4c2ee23267ee88f64b8fd", + "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 address the chat CLI management URL bug when the server is mounted under `/v1` versus the service root; 47303 is the broader root-routing variant, while 47140 is a narrower normalization approach.", + "confidence": 0.74, + "canonical_issue_reason": null, + "canonical_pr_reason": "47303 is the cleaner representative: it explicitly routes management requests to the service root and packages the fix in a helper rather than only normalizing the constructor input.", + "best_issue_reason": null, + "best_pr_reason": "47303 most directly captures the user-facing fix for chat management calls and is the broader, more representative implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47140", + "right": "pull_request:47303", + "accept": true, + "reason": "Same underlying chat CLI management URL problem (`/v1` base vs service root); the patches are alternative implementations of the same fix and could plausibly be combined into one PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The shared chat CLI management URL/service-root theme is supported by the titles and diff previews, and the overlapping files make the soft-pairing plausible. The summary stays reasonably conservative, though the differing issue targets mean the \u2018same bugfix\u2019 link should be treated as suggestive rather than certain." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 33, + "component_count": 34, + "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-47165-2", + "nodes": [ + "pull_request:47165", + "pull_request:47289" + ], + "soft_pairs": [ + "pull_request:47165|pull_request:47289" + ], + "prepared_review_unit_hash": "22fa01ba2f28cba1147651ce11b352263040833ade250bf8edab99c0b7906795", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5904, + "estimated_input_tokens": 1476, + "estimated_eval_tokens": 3208 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5904, + "estimated_input_tokens": 1476, + "estimated_eval_tokens": 3208 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two PRs are only loosely related through distributed-training/inference infrastructure, but they change different code paths: one adds FSDP plans to many model classes, while the other adds pipeline-parallel naive inference support and test/marker plumbing. They should not be treated as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47165 is the more complete, non-draft distributed-model change and is the better representative of the cluster than the pipeline-parallel feature PR.", + "best_issue_reason": null, + "best_pr_reason": "PR 47165 is the strongest overall representative here because it is broader, more mature, and focused on concrete model-level FSDP plan additions rather than auxiliary pipeline-parallel scaffolding.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47165", + "right": "pull_request:47289", + "accept": false, + "reason": "Different features and code paths: FSDP plan annotations across model implementations vs pipeline-parallel naive inference and test markers. 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": 34, + "component_count": 34, + "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": "49046d1cea32082a35ba982144ad15477d7fd5ebb8fecba9fcc6799261dda54c", + "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 pull requests are not duplicates: one fixes label smoothing + gradient accumulation in training, the other changes TopHLogitsWarper to honor min_tokens_to_keep in generation. Different code paths, different bugs, different subsystems.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47261 is the better canonical representative because it has a clear, explicit issue target and a narrowly defined training bug fix; PR 47262 is a separate generation change.", + "best_issue_reason": null, + "best_pr_reason": "PR 47261 is the strongest representative of the cluster because it is issue-backed and addresses a concrete, reproducible correctness problem. PR 47262 is unrelated and should not be grouped with it as a duplicate.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47261", + "right": "pull_request:47262", + "accept": false, + "reason": "These fix different problems in different areas of the library: trainer loss scaling vs. generation token filtering. They do not appear to be the same underlying change and should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/manifest.json b/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..64403d25175e8bd9a15e1cfc21713aa689e4ca92 --- /dev/null +++ b/snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260714t180038z", + "artifacts": { + "hybrid": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-07-14T18:15:00Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260714T180038Z", + "variant": "hybrid" +} diff --git a/snapshots/20260714T180038Z/manifest.json b/snapshots/20260714T180038Z/manifest.json index f8b43a440f2043e87dec27f21af3297420a70473..63523e4292fc52df2aa1f6ba42d4260ce34a1c4b 100644 --- a/snapshots/20260714T180038Z/manifest.json +++ b/snapshots/20260714T180038Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 632 }, "extracted_at": "2026-07-14T18:00:38Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260714t180038z", + "runs": { + "hybrid-model-20260714t180038z": { + "analysis_id": "hybrid-model-20260714t180038z", + "artifacts": { + "hybrid": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260714T180038Z/analysis-runs/hybrid-model-20260714t180038z/manifest.json", + "model": null, + "published_at": "2026-07-14T18:15:00Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260714T180038Z", "watermark": {