diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 083c2a239650d18b1ade7e22f813e9f282f76391..9d3fcbdcda1d647a0c21e972357f5f03e5c41d9a 100644 --- a/analysis/current/analysis-report-hybrid.json +++ b/analysis/current/analysis-report-hybrid.json @@ -1,1037 +1,113 @@ { "schema_version": "1.0", "repo": "huggingface/transformers", - "snapshot_id": "20260707T000034Z", - "generated_at": "2026-07-07T00:11:37Z", + "snapshot_id": "20260707T060715Z", + "generated_at": "2026-07-07T06:19:22Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ { - "cluster_id": "cluster-43451-7", - "summary": "Cluster of 7 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-43979-11", + "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 43451, - "issue_numbers": [], + "confidence": 0.8, + "canonical_issue_number": 43979, + "canonical_pr_number": 44007, + "issue_numbers": [ + 43979 + ], "pr_numbers": [ - 43451, - 46196, - 46836, - 47003, - 47034, - 47092, - 47096 + 43996, + 44007, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 ], "evidence_types": [ + "closing_reference", + "shared_issue_target", "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 43451, - "right_pr_number": 46196, - "code_similarity": 0.171, - "size_similarity": 0.764, + "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": 0.123, + "patch_similarity": 0.429, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 43451, - "right_pr_number": 46836, - "code_similarity": 0.267, - "size_similarity": 0.816, - "file_overlap": 0.182, + "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": 0.331, - "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": [ - [ - 887, - 894 - ] - ] - }, - { - "filename": "src/transformers/conversion_mapping.py", - "left_ranges": [ - [ - 737, - 769 - ] - ], - "right_ranges": [ - [ - 1420, - 1486 - ] - ] - }, - { - "filename": "src/transformers/models/__init__.py", - "left_ranges": [ - [ - 290, - 296 - ] - ], - "right_ranges": [ - [ - 452, - 458 - ] - ] - }, - { - "filename": "src/transformers/models/auto/auto_mappings.py", - "left_ranges": [ - [ - 391, - 398 - ], - [ - 824, - 830 - ], - [ - 945, - 951 - ], - [ - 1068, - 1074 - ], - [ - 1202, - 1208 - ] - ], - "right_ranges": [ - [ - 628, - 638 - ], - [ - 905, - 914 - ], - [ - 1112, - 1118 - ], - [ - 1238, - 1244 - ] - ] - }, - { - "filename": "src/transformers/models/auto/modeling_auto.py", - "left_ranges": [ - [ - 331, - 338 - ], - [ - 1094, - 1100 - ], - [ - 1913, - 1919 - ] - ], - "right_ranges": [ - [ - 501, - 507 - ], - [ - 1107, - 1113 - ] - ] - }, - { - "filename": "utils/check_repo.py", - "left_ranges": [ - [ - 250, - 258 - ], - [ - 337, - 344 - ] - ], - "right_ranges": [ - [ - 267, - 274 - ], - [ - 489, - 496 - ] - ] - } - ] + "patch_similarity": 0.392, + "shared_filenames": [], + "shared_file_areas": [] }, { - "left_pr_number": 43451, - "right_pr_number": 47003, - "code_similarity": 0.179, - "size_similarity": 0.76, + "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": 0.178, + "patch_similarity": 0.479, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 43451, - "right_pr_number": 47034, - "code_similarity": 0.038, - "size_similarity": 0.112, + "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": 0.106, + "patch_similarity": 0.408, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 43451, - "right_pr_number": 47092, - "code_similarity": 0.001, - "size_similarity": 0.002, + "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": 0.003, + "patch_similarity": 0.528, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 43451, - "right_pr_number": 47096, - "code_similarity": 0.036, - "size_similarity": 0.099, + "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": 0.108, + "patch_similarity": 0.398, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 46196, - "right_pr_number": 46836, - "code_similarity": 0.205, - "size_similarity": 0.936, + "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": 0.12, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46196, - "right_pr_number": 47003, - "code_similarity": 0.231, - "size_similarity": 0.995, - "file_overlap": 0.055, - "area_overlap": 0.006, - "patch_similarity": 0.092, - "shared_filenames": [ - "src/transformers/models/unispeech/modeling_unispeech.py", - "src/transformers/models/unispeech/modular_unispeech.py", - "src/transformers/models/wavlm/modeling_wavlm.py", - "src/transformers/models/wavlm/modular_wavlm.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/unispeech/modeling_unispeech.py", - "left_ranges": [ - [ - 29, - 35 - ], - [ - 45, - 52 - ], - [ - 337, - 343 - ], - [ - 376, - 382 - ], - [ - 419, - 440 - ], - [ - 451, - 458 - ], - [ - 472, - 486 - ], - [ - 532, - 550 - ], - [ - 563, - 570 - ], - [ - 583, - 600 - ], - [ - 677, - 686 - ], - [ - 934, - 954 - ], - [ - 964, - 977 - ], - [ - 1029, - 1041 - ], - [ - 1048, - 1062 - ], - [ - 1080, - 1085 - ], - [ - 1091, - 1096 - ], - [ - 1165, - 1178 - ], - [ - 1181, - 1196 - ], - [ - 1223, - 1228 - ], - [ - 1267, - 1281 - ], - [ - 1289, - 1310 - ], - [ - 1322, - 1327 - ] - ], - "right_ranges": [ - [ - 1138, - 1150 - ] - ] - }, - { - "filename": "src/transformers/models/unispeech/modular_unispeech.py", - "left_ranges": [ - [ - 22, - 31 - ], - [ - 147, - 156 - ], - [ - 241, - 261 - ], - [ - 271, - 284 - ], - [ - 336, - 348 - ], - [ - 355, - 369 - ], - [ - 387, - 392 - ] - ], - "right_ranges": [ - [ - 383, - 395 - ] - ] - }, - { - "filename": "src/transformers/models/wavlm/modeling_wavlm.py", - "left_ranges": [ - [ - 15, - 21 - ], - [ - 28, - 37 - ], - [ - 153, - 160 - ], - [ - 184, - 190 - ], - [ - 194, - 199 - ], - [ - 222, - 229 - ], - [ - 314, - 334 - ], - [ - 338, - 344 - ], - [ - 357, - 382 - ], - [ - 394, - 402 - ], - [ - 411, - 431 - ], - [ - 446, - 454 - ], - [ - 462, - 484 - ], - [ - 561, - 570 - ], - [ - 998, - 1018 - ], - [ - 1030, - 1049 - ], - [ - 1118, - 1131 - ], - [ - 1134, - 1149 - ], - [ - 1176, - 1181 - ], - [ - 1220, - 1234 - ], - [ - 1242, - 1263 - ], - [ - 1275, - 1280 - ], - [ - 1316, - 1330 - ], - [ - 1338, - 1359 - ], - [ - 1362, - 1367 - ], - [ - 1480, - 1494 - ], - [ - 1502, - 1523 - ], - [ - 1547, - 1552 - ] - ], - "right_ranges": [ - [ - 241, - 251 - ] - ] - }, - { - "filename": "src/transformers/models/wavlm/modular_wavlm.py", - "left_ranges": [ - [ - 10, - 17 - ], - [ - 81, - 88 - ], - [ - 112, - 118 - ], - [ - 122, - 127 - ], - [ - 150, - 157 - ], - [ - 222, - 242 - ], - [ - 246, - 252 - ], - [ - 265, - 290 - ], - [ - 302, - 310 - ], - [ - 319, - 339 - ], - [ - 354, - 362 - ], - [ - 370, - 392 - ] - ], - "right_ranges": [ - [ - 172, - 182 - ] - ] - } - ] - }, - { - "left_pr_number": 46196, - "right_pr_number": 47034, - "code_similarity": 0.043, - "size_similarity": 0.146, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.094, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46196, - "right_pr_number": 47092, - "code_similarity": 0.002, - "size_similarity": 0.002, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.011, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46196, - "right_pr_number": 47096, - "code_similarity": 0.044, - "size_similarity": 0.129, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.124, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46836, - "right_pr_number": 47003, - "code_similarity": 0.214, - "size_similarity": 0.931, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.185, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46836, - "right_pr_number": 47034, - "code_similarity": 0.042, - "size_similarity": 0.137, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.097, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 46836, - "right_pr_number": 47092, - "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": 46836, - "right_pr_number": 47096, - "code_similarity": 0.041, - "size_similarity": 0.121, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.109, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47003, - "right_pr_number": 47034, - "code_similarity": 0.042, - "size_similarity": 0.147, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.084, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47003, - "right_pr_number": 47092, - "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": 47003, - "right_pr_number": 47096, - "code_similarity": 0.071, - "size_similarity": 0.13, - "file_overlap": 0.078, - "area_overlap": 0.015, - "patch_similarity": 0.108, - "shared_filenames": [ - "src/transformers/exporters/exporter_dynamo.py", - "src/transformers/exporters/exporter_executorch.py", - "src/transformers/exporters/utils.py", - "tests/exporters/test_export.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/exporters/exporter_dynamo.py", - "left_ranges": [ - [ - 43, - 49 - ], - [ - 449, - 470 - ], - [ - 487, - 507 - ], - [ - 548, - 559 - ], - [ - 594, - 606 - ], - [ - 675, - 682 - ] - ], - "right_ranges": [ - [ - 350, - 357 - ], - [ - 381, - 388 - ] - ] - }, - { - "filename": "src/transformers/exporters/exporter_executorch.py", - "left_ranges": [ - [ - 336, - 342 - ], - [ - 347, - 390 - ], - [ - 397, - 412 - ], - [ - 446, - 457 - ], - [ - 471, - 503 - ], - [ - 633, - 761 - ], - [ - 809, - 1120 - ], - [ - 1212, - 1241 - ], - [ - 1377, - 1393 - ] - ], - "right_ranges": [ - [ - 418, - 442 - ] - ] - }, - { - "filename": "src/transformers/exporters/utils.py", - "left_ranges": [ - [ - 232, - 238 - ], - [ - 444, - 461 - ], - [ - 621, - 644 - ], - [ - 647, - 663 - ], - [ - 761, - 768 - ] - ], - "right_ranges": [ - [ - 432, - 441 - ], - [ - 444, - 455 - ], - [ - 464, - 490 - ], - [ - 708, - 720 - ] - ] - }, - { - "filename": "tests/exporters/test_export.py", - "left_ranges": [ - [ - 24, - 30 - ], - [ - 35, - 41 - ], - [ - 46, - 92 - ], - [ - 98, - 104 - ], - [ - 109, - 149 - ], - [ - 153, - 185 - ], - [ - 226, - 301 - ], - [ - 333, - 349 - ], - [ - 442, - 449 - ], - [ - 455, - 487 - ], - [ - 588, - 595 - ], - [ - 602, - 634 - ] - ], - "right_ranges": [ - [ - 181, - 188 - ] - ] - } - ] - }, - { - "left_pr_number": 47034, - "right_pr_number": 47092, - "code_similarity": 0.004, - "size_similarity": 0.017, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.007, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47034, - "right_pr_number": 47096, - "code_similarity": 0.19, - "size_similarity": 0.883, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.09, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "left_pr_number": 47092, - "right_pr_number": 47096, - "code_similarity": 0.005, - "size_similarity": 0.019, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.011, - "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, + "patch_similarity": 0.229, "shared_filenames": [], "shared_file_areas": [] }, @@ -1445,626 +521,1302 @@ 611 ], [ - 633, - 639 + 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 ], [ - 645, - 652 + 658, + 663 ], [ - 657, - 673 + 666, + 674 ], [ - 716, - 728 + 717, + 722 ], [ - 737, - 743 + 738, + 744 ], [ - 748, - 771 + 749, + 772 ], [ - 789, - 794 + 790, + 795 ] ] } ] }, { - "left_pr_number": 44066, - "right_pr_number": 44129, - "code_similarity": 0.145, - "size_similarity": 0.526, + "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": 0.263, + "patch_similarity": 0.199, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 44066, - "right_pr_number": 44154, - "code_similarity": 0.174, - "size_similarity": 0.654, + "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": 0.286, + "patch_similarity": 0.247, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 44066, + "left_pr_number": 44154, "right_pr_number": 44722, - "code_similarity": 0.907, - "size_similarity": 0.964, - "file_overlap": 1.0, - "area_overlap": 0.808, - "patch_similarity": 0.874, + "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-44779-7", + "summary": "Cluster of 5 issues and 2 PRs centered on issue #45488.", + "status": "open", + "confidence": 0.6, + "canonical_issue_number": 45488, + "canonical_pr_number": 47017, + "issue_numbers": [ + 44779, + 45488, + 45812, + 45920, + 46489 + ], + "pr_numbers": [ + 46580, + 47017 + ], + "evidence_types": [ + "closing_reference", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46580, + "right_pr_number": 47017, + "code_similarity": 0.241, + "size_similarity": 0.352, + "file_overlap": 0.5, + "area_overlap": 0.0, + "patch_similarity": 0.136, "shared_filenames": [ - "src/transformers/models/codegen/modeling_codegen.py", - "src/transformers/models/gptj/modeling_gptj.py" + "src/transformers/models/auto/tokenization_auto.py" ], "shared_file_areas": [ { - "filename": "src/transformers/models/codegen/modeling_codegen.py", + "filename": "src/transformers/models/auto/tokenization_auto.py", "left_ranges": [ [ - 245, - 266 + 361, + 367 + ], + [ + 399, + 421 + ], + [ + 724, + 739 ] ], "right_ranges": [ [ - 228, - 234 + 428, + 475 + ], + [ + 807, + 825 + ] + ] + } + ] + } + ] + }, + { + "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": "src/transformers/models/gptj/modeling_gptj.py", + "filename": "tests/models/deimv2/test_modeling_deimv2.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 1734 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-46196-4", + "summary": "Cluster of 4 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 46196, + "issue_numbers": [], + "pr_numbers": [ + 46196, + 46658, + 47003, + 47096 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46196, + "right_pr_number": 46658, + "code_similarity": 0.163, + "size_similarity": 0.723, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.12, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46196, + "right_pr_number": 47003, + "code_similarity": 0.231, + "size_similarity": 0.995, + "file_overlap": 0.055, + "area_overlap": 0.006, + "patch_similarity": 0.092, + "shared_filenames": [ + "src/transformers/models/unispeech/modeling_unispeech.py", + "src/transformers/models/unispeech/modular_unispeech.py", + "src/transformers/models/wavlm/modeling_wavlm.py", + "src/transformers/models/wavlm/modular_wavlm.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/unispeech/modeling_unispeech.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 + 29, + 35 ], [ - 601, - 606 + 45, + 52 ], [ - 634, - 640 + 337, + 343 ], [ - 646, - 653 + 376, + 382 ], [ - 658, - 674 + 419, + 440 ], [ - 717, - 722 + 451, + 458 ], [ - 738, - 744 + 472, + 486 ], [ - 749, - 772 + 532, + 550 ], [ - 790, - 795 - ] - ], - "right_ranges": [ - [ - 33, - 42 + 563, + 570 ], [ - 175, - 182 + 583, + 600 ], [ - 251, - 258 + 677, + 686 ], [ - 399, - 420 + 934, + 954 ], [ - 426, - 435 + 964, + 977 ], [ - 460, - 467 + 1029, + 1041 ], [ - 472, - 489 + 1048, + 1062 ], [ - 519, - 540 + 1080, + 1085 ], [ - 554, - 560 + 1091, + 1096 ], [ - 566, - 575 + 1165, + 1178 ], [ - 580, - 598 + 1181, + 1196 ], [ - 601, - 612 + 1223, + 1228 ], [ - 634, - 640 + 1267, + 1281 ], [ - 646, - 653 + 1289, + 1310 ], [ - 658, - 663 - ], + 1322, + 1327 + ] + ], + "right_ranges": [ [ - 666, - 674 - ], + 1138, + 1150 + ] + ] + }, + { + "filename": "src/transformers/models/unispeech/modular_unispeech.py", + "left_ranges": [ [ - 717, - 722 + 22, + 31 ], [ - 738, - 744 + 147, + 156 ], [ - 749, - 772 + 241, + 261 ], - [ - 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": [ + [ + 271, + 284 + ], + [ + 336, + 348 + ], + [ + 355, + 369 + ], + [ + 387, + 392 + ] + ], + "right_ranges": [ + [ + 383, + 395 + ] + ] + }, { - "filename": "src/transformers/models/gptj/modeling_gptj.py", + "filename": "src/transformers/models/wavlm/modeling_wavlm.py", "left_ranges": [ [ - 33, - 40 + 15, + 21 ], [ - 173, - 180 + 28, + 37 ], [ - 249, - 256 + 153, + 160 ], [ - 397, - 405 + 184, + 190 ], [ - 408, - 419 + 194, + 199 ], [ - 425, - 434 + 222, + 229 ], [ - 459, - 465 + 314, + 334 ], [ - 470, - 489 + 338, + 344 ], [ - 517, - 539 + 357, + 382 ], [ - 553, - 559 + 394, + 402 ], [ - 565, - 574 + 411, + 431 ], [ - 579, - 597 + 446, + 454 ], [ - 600, - 611 + 462, + 484 ], [ - 633, - 639 + 561, + 570 ], [ - 645, - 652 + 998, + 1018 ], [ - 657, - 673 + 1030, + 1049 ], [ - 716, - 728 + 1118, + 1131 ], [ - 737, - 743 + 1134, + 1149 ], [ - 748, - 771 + 1176, + 1181 ], [ - 789, - 794 - ] - ], - "right_ranges": [ - [ - 33, - 42 + 1220, + 1234 ], [ - 175, - 182 + 1242, + 1263 ], [ - 251, - 258 + 1275, + 1280 ], [ - 399, - 420 + 1316, + 1330 ], [ - 426, - 435 + 1338, + 1359 ], [ - 460, - 467 + 1362, + 1367 ], [ - 472, - 489 + 1480, + 1494 ], [ - 519, - 540 + 1502, + 1523 ], [ - 554, - 560 + 1547, + 1552 + ] + ], + "right_ranges": [ + [ + 241, + 251 + ] + ] + }, + { + "filename": "src/transformers/models/wavlm/modular_wavlm.py", + "left_ranges": [ + [ + 10, + 17 ], [ - 566, - 575 + 81, + 88 ], [ - 580, - 598 + 112, + 118 ], [ - 601, - 612 + 122, + 127 ], [ - 634, - 640 + 150, + 157 ], [ - 646, - 653 + 222, + 242 ], [ - 658, - 663 + 246, + 252 ], [ - 666, - 674 + 265, + 290 ], [ - 717, - 722 + 302, + 310 ], [ - 738, - 744 + 319, + 339 ], [ - 749, - 772 + 354, + 362 ], [ - 790, - 795 + 370, + 392 + ] + ], + "right_ranges": [ + [ + 172, + 182 ] ] } ] }, { - "left_pr_number": 44129, - "right_pr_number": 44154, - "code_similarity": 0.099, - "size_similarity": 0.344, + "left_pr_number": 46196, + "right_pr_number": 47096, + "code_similarity": 0.044, + "size_similarity": 0.129, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.199, + "patch_similarity": 0.124, "shared_filenames": [], "shared_file_areas": [] }, { - "left_pr_number": 44129, - "right_pr_number": 44722, - "code_similarity": 0.146, - "size_similarity": 0.545, + "left_pr_number": 46658, + "right_pr_number": 47003, + "code_similarity": 0.174, + "size_similarity": 0.719, "file_overlap": 0.0, "area_overlap": 0.0, - "patch_similarity": 0.247, + "patch_similarity": 0.198, "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": 46658, + "right_pr_number": 47096, + "code_similarity": 0.035, + "size_similarity": 0.093, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.11, + "shared_filenames": [], + "shared_file_areas": [] + }, { - "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, + "left_pr_number": 47003, + "right_pr_number": 47096, + "code_similarity": 0.071, + "size_similarity": 0.13, + "file_overlap": 0.078, + "area_overlap": 0.015, + "patch_similarity": 0.108, "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" + "src/transformers/exporters/exporter_dynamo.py", + "src/transformers/exporters/exporter_executorch.py", + "src/transformers/exporters/utils.py", + "tests/exporters/test_export.py" ], "shared_file_areas": [ { - "filename": "docs/source/en/model_doc/deimv2.md", + "filename": "src/transformers/exporters/exporter_dynamo.py", "left_ranges": [ [ - 1, - 132 + 43, + 49 + ], + [ + 449, + 470 + ], + [ + 487, + 507 + ], + [ + 548, + 559 + ], + [ + 594, + 606 + ], + [ + 675, + 682 ] ], "right_ranges": [ [ - 1, - 65 + 350, + 357 + ], + [ + 381, + 388 ] ] }, { - "filename": "src/transformers/models/deimv2/__init__.py", + "filename": "src/transformers/exporters/exporter_executorch.py", "left_ranges": [ [ - 1, - 15 + 336, + 342 + ], + [ + 347, + 390 + ], + [ + 397, + 412 + ], + [ + 446, + 457 + ], + [ + 471, + 503 + ], + [ + 633, + 761 + ], + [ + 809, + 1120 + ], + [ + 1212, + 1241 + ], + [ + 1377, + 1393 ] ], "right_ranges": [ [ - 1, - 29 + 418, + 442 ] ] }, { - "filename": "src/transformers/models/deimv2/configuration_deimv2.py", + "filename": "src/transformers/exporters/utils.py", "left_ranges": [ [ - 1, - 74 + 232, + 238 + ], + [ + 444, + 461 + ], + [ + 621, + 644 + ], + [ + 647, + 663 + ], + [ + 761, + 768 ] ], "right_ranges": [ [ - 1, - 266 + 432, + 441 + ], + [ + 444, + 455 + ], + [ + 464, + 490 + ], + [ + 708, + 720 ] ] }, { - "filename": "tests/models/deimv2/test_modeling_deimv2.py", + "filename": "tests/exporters/test_export.py", "left_ranges": [ [ - 1, - 15 + 24, + 30 + ], + [ + 35, + 41 + ], + [ + 46, + 92 + ], + [ + 98, + 104 + ], + [ + 109, + 149 + ], + [ + 153, + 185 + ], + [ + 226, + 301 + ], + [ + 333, + 349 + ], + [ + 442, + 449 + ], + [ + 455, + 487 + ], + [ + 588, + 595 + ], + [ + 602, + 634 ] ], "right_ranges": [ [ - 1, - 1734 + 181, + 188 ] ] } @@ -2073,15 +1825,14 @@ ] }, { - "cluster_id": "cluster-46894-6", - "summary": "Cluster of 2 issues and 4 PRs centered on issue #46894.", + "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, - 46901 + 46894 ], "pr_numbers": [ 46895, @@ -2461,80 +2212,17 @@ "left_ranges": [ [ 15, - 21 - ], - [ - 180, - 221 - ] - ], - "right_ranges": [ - [ - 175, - 213 - ] - ] - } - ] - } - ] - }, - { - "cluster_id": "cluster-45488-4", - "summary": "Cluster of 2 issues and 2 PRs centered on issue #45488.", - "status": "open", - "confidence": 0.6, - "canonical_issue_number": 45488, - "canonical_pr_number": 47017, - "issue_numbers": [ - 45488, - 46489 - ], - "pr_numbers": [ - 46580, - 47017 - ], - "evidence_types": [ - "closing_reference", - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 46580, - "right_pr_number": 47017, - "code_similarity": 0.241, - "size_similarity": 0.352, - "file_overlap": 0.5, - "area_overlap": 0.0, - "patch_similarity": 0.136, - "shared_filenames": [ - "src/transformers/models/auto/tokenization_auto.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/auto/tokenization_auto.py", - "left_ranges": [ - [ - 361, - 367 - ], - [ - 399, - 421 + 21 ], [ - 724, - 739 + 180, + 221 ] ], "right_ranges": [ [ - 428, - 475 - ], - [ - 807, - 825 + 175, + 213 ] ] } @@ -2925,386 +2613,91 @@ "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-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-47013-3", - "summary": "Cluster of 1 issues and 2 PRs centered on issue #47013.", - "status": "open", - "confidence": 0.8, - "canonical_issue_number": 47013, - "canonical_pr_number": 47026, - "issue_numbers": [ - 47013 - ], - "pr_numbers": [ - 47026, - 47046 - ], - "evidence_types": [ - "closing_reference", - "shared_issue_target", - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 47026, - "right_pr_number": 47046, - "code_similarity": 0.308, - "size_similarity": 0.264, - "file_overlap": 0.5, - "area_overlap": 0.229, - "patch_similarity": 0.169, - "shared_filenames": [ - "src/transformers/models/xlstm/modeling_xlstm.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/xlstm/modeling_xlstm.py", - "left_ranges": [ - [ - 816, - 840 - ], - [ - 843, - 851 - ] - ], - "right_ranges": [ - [ - 25, - 39 - ], - [ - 789, - 795 - ], - [ - 808, - 817 - ], - [ - 823, - 846 + 983 ], [ - 849, - 857 + 986, + 996 ], [ - 1388, - 1394 - ], + 999, + 1007 + ] + ], + "right_ranges": [ [ - 1399, - 1405 + 976, + 982 ], [ - 1461, - 1473 + 985, + 995 ], [ - 1504, - 1522 + 998, + 1004 ] ] - } - ] - } - ] - }, - { - "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", + "filename": "src/transformers/models/vit_mae/modeling_vit_mae.py", "left_ranges": [ [ - 1360, - 1370 + 152, + 159 ], [ - 1374, - 1380 + 162, + 172 + ], + [ + 175, + 183 ] ], "right_ranges": [ [ - 1360, - 1370 + 152, + 158 ], [ - 1374, - 1380 + 161, + 171 + ], + [ + 174, + 180 ] ] - } - ] - } - ] - }, - { - "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", + "filename": "src/transformers/models/vit_mae/modular_vit_mae.py", "left_ranges": [ [ - 2227, - 2241 + 62, + 69 ], [ - 2303, - 2309 + 72, + 82 + ], + [ + 85, + 93 ] ], "right_ranges": [ [ - 2278, - 2291 + 62, + 68 ], [ - 2309, - 2322 + 71, + 81 + ], + [ + 84, + 90 ] ] } @@ -3313,184 +2706,89 @@ ] }, { - "cluster_id": "cluster-46950-3", - "summary": "Cluster of 3 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-45561-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45561.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 47058, - "issue_numbers": [], + "confidence": 0.8, + "canonical_issue_number": 45561, + "canonical_pr_number": 45645, + "issue_numbers": [ + 45561 + ], "pr_numbers": [ - 46950, - 46971, - 47058 + 45639, + 45645 ], "evidence_types": [ + "closing_reference", + "shared_issue_target", "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 46950, - "right_pr_number": 46971, - "code_similarity": 0.393, - "size_similarity": 0.759, - "file_overlap": 0.5, - "area_overlap": 0.182, - "patch_similarity": 0.184, + "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": [ - "tests/models/mamba2/test_modeling_mamba2.py" + "src/transformers/testing_utils.py", + "tests/utils/test_testing_utils.py", + "utils/notification_service.py" ], "shared_file_areas": [ { - "filename": "tests/models/mamba2/test_modeling_mamba2.py", + "filename": "src/transformers/testing_utils.py", "left_ranges": [ [ - 418, - 424 - ], - [ - 446, - 451 - ], - [ - 475, - 480 + 3611, + 3618 ], [ - 498, - 504 + 3835, + 3869 ] ], "right_ranges": [ [ - 418, - 424 - ], - [ - 450, - 456 - ], - [ - 480, - 497 + 3525, + 3558 ], [ - 508, - 514 + 3782, + 3788 ] ] - } - ] - }, - { - "left_pr_number": 46950, - "right_pr_number": 47058, - "code_similarity": 0.392, - "size_similarity": 0.69, - "file_overlap": 0.5, - "area_overlap": 0.127, - "patch_similarity": 0.395, - "shared_filenames": [ - "tests/models/mamba2/test_modeling_mamba2.py" - ], - "shared_file_areas": [ + }, { - "filename": "tests/models/mamba2/test_modeling_mamba2.py", + "filename": "tests/utils/test_testing_utils.py", "left_ranges": [ [ - 418, - 424 - ], - [ - 446, - 451 - ], - [ - 475, - 480 - ], - [ - 498, - 504 + 1, + 144 ] ], "right_ranges": [ [ - 404, - 411 - ], - [ - 420, - 426 - ], - [ - 452, - 458 - ], - [ - 482, - 488 - ], - [ - 503, - 509 + 1, + 114 ] ] - } - ] - }, - { - "left_pr_number": 46971, - "right_pr_number": 47058, - "code_similarity": 0.643, - "size_similarity": 0.909, - "file_overlap": 1.0, - "area_overlap": 0.339, - "patch_similarity": 0.286, - "shared_filenames": [ - "tests/models/mamba2/test_modeling_mamba2.py" - ], - "shared_file_areas": [ + }, { - "filename": "tests/models/mamba2/test_modeling_mamba2.py", - "left_ranges": [ - [ - 418, - 424 - ], - [ - 450, - 456 - ], - [ - 480, - 497 - ], + "filename": "utils/notification_service.py", + "left_ranges": [ [ - 508, - 514 + 935, + 967 ] ], "right_ranges": [ [ - 404, - 411 - ], - [ - 420, - 426 - ], - [ - 452, - 458 - ], - [ - 482, - 488 - ], - [ - 503, - 509 + 935, + 967 ] ] } @@ -3499,61 +2797,85 @@ ] }, { - "cluster_id": "cluster-46907-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-47013-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #47013.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 46907, - "issue_numbers": [], + "confidence": 0.8, + "canonical_issue_number": 47013, + "canonical_pr_number": 47026, + "issue_numbers": [ + 47013 + ], "pr_numbers": [ - 46907, - 47055 + 47026, + 47046 ], "evidence_types": [ + "closing_reference", + "shared_issue_target", "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 46907, - "right_pr_number": 47055, - "code_similarity": 0.925, - "size_similarity": 1.0, - "file_overlap": 1.0, - "area_overlap": 1.0, - "patch_similarity": 0.5, + "left_pr_number": 47026, + "right_pr_number": 47046, + "code_similarity": 0.308, + "size_similarity": 0.264, + "file_overlap": 0.5, + "area_overlap": 0.229, + "patch_similarity": 0.169, "shared_filenames": [ - "src/transformers/models/edgetam/modeling_edgetam.py", - "src/transformers/models/edgetam/modular_edgetam.py" + "src/transformers/models/xlstm/modeling_xlstm.py" ], "shared_file_areas": [ { - "filename": "src/transformers/models/edgetam/modeling_edgetam.py", + "filename": "src/transformers/models/xlstm/modeling_xlstm.py", "left_ranges": [ [ - 464, - 472 + 816, + 840 + ], + [ + 843, + 851 ] ], "right_ranges": [ [ - 464, - 472 - ] - ] - }, - { - "filename": "src/transformers/models/edgetam/modular_edgetam.py", - "left_ranges": [ + 25, + 39 + ], [ - 210, - 218 - ] - ], - "right_ranges": [ + 789, + 795 + ], + [ + 808, + 817 + ], + [ + 823, + 846 + ], + [ + 849, + 857 + ], + [ + 1388, + 1394 + ], + [ + 1399, + 1405 + ], + [ + 1461, + 1473 + ], [ - 210, - 218 + 1504, + 1522 ] ] } @@ -3562,77 +2884,57 @@ ] }, { - "cluster_id": "cluster-46908-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-45081-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45081.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 46908, - "issue_numbers": [], + "confidence": 0.8, + "canonical_issue_number": 45081, + "canonical_pr_number": 45317, + "issue_numbers": [ + 45081 + ], "pr_numbers": [ - 46908, - 47057 + 45086, + 45317 ], "evidence_types": [ + "closing_reference", + "shared_issue_target", "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 46908, - "right_pr_number": 47057, - "code_similarity": 0.565, - "size_similarity": 0.231, - "file_overlap": 1.0, - "area_overlap": 0.565, - "patch_similarity": 0.143, + "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": [ - "tests/models/deepseek_v32/test_modeling_deepseek_v32.py" + "src/transformers/tokenization_utils_tokenizers.py" ], "shared_file_areas": [ { - "filename": "tests/models/deepseek_v32/test_modeling_deepseek_v32.py", + "filename": "src/transformers/tokenization_utils_tokenizers.py", "left_ranges": [ [ - 13, - 19 - ], - [ - 251, - 257 - ], - [ - 270, - 276 - ], - [ - 291, - 297 + 1360, + 1370 ], [ - 332, - 338 + 1374, + 1380 ] ], "right_ranges": [ [ - 13, - 19 - ], - [ - 248, - 260 - ], - [ - 268, - 281 - ], - [ - 290, - 303 + 1360, + 1370 ], [ - 332, - 345 + 1374, + 1380 ] ] } @@ -3641,53 +2943,56 @@ ] }, { - "cluster_id": "cluster-46926-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "cluster_id": "cluster-43698-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43698.", "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 46926, - "issue_numbers": [], + "confidence": 0.75, + "canonical_issue_number": 43698, + "canonical_pr_number": 43779, + "issue_numbers": [ + 43698 + ], "pr_numbers": [ - 46926, - 47054 + 43779, + 43816 ], "evidence_types": [ - "soft_similarity" + "closing_reference", + "shared_issue_target" ], "pr_comparisons": [ { - "left_pr_number": 46926, - "right_pr_number": 47054, - "code_similarity": 1.0, - "size_similarity": 1.0, + "left_pr_number": 43779, + "right_pr_number": 43816, + "code_similarity": 0.418, + "size_similarity": 0.538, "file_overlap": 1.0, - "area_overlap": 1.0, - "patch_similarity": 1.0, + "area_overlap": 0.02, + "patch_similarity": 0.02, "shared_filenames": [ - "tests/models/minimax_m3_vl/test_modeling_minimax_m3_vl.py" + "src/transformers/integrations/integration_utils.py" ], "shared_file_areas": [ { - "filename": "tests/models/minimax_m3_vl/test_modeling_minimax_m3_vl.py", + "filename": "src/transformers/integrations/integration_utils.py", "left_ranges": [ [ - 14, - 20 + 2227, + 2241 ], [ - 487, - 493 + 2303, + 2309 ] ], "right_ranges": [ [ - 14, - 20 + 2278, + 2291 ], [ - 487, - 493 + 2309, + 2322 ] ] } @@ -3696,45 +3001,73 @@ ] }, { - "cluster_id": "cluster-46949-2", + "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": 47056, + "canonical_pr_number": 43656, "issue_numbers": [], "pr_numbers": [ - 46949, - 47056 + 43656, + 43842 ], "evidence_types": [ "soft_similarity" ], "pr_comparisons": [ { - "left_pr_number": 46949, - "right_pr_number": 47056, - "code_similarity": 0.54, + "left_pr_number": 43656, + "right_pr_number": 43842, + "code_similarity": 0.405, "size_similarity": 0.5, - "file_overlap": 0.5, - "area_overlap": 0.4, - "patch_similarity": 1.0, + "file_overlap": 1.0, + "area_overlap": 0.0, + "patch_similarity": 0.036, "shared_filenames": [ - "src/transformers/models/whisper/modeling_whisper.py" + "src/transformers/cli/serve.py" ], "shared_file_areas": [ { - "filename": "src/transformers/models/whisper/modeling_whisper.py", + "filename": "src/transformers/cli/serve.py", "left_ranges": [ [ - 309, - 315 + 11, + 18 + ], + [ + 30, + 36 + ], + [ + 315, + 323 + ], + [ + 665, + 671 + ], + [ + 931, + 937 + ], + [ + 1843, + 1849 + ], + [ + 1868, + 1874 ] ], "right_ranges": [ [ - 306, - 312 + 54, + 61 + ], + [ + 587, + 593 ] ] } @@ -4931,20 +4264,15 @@ ], "duplicate_issues": [ { - "cluster_id": "cluster-45488-4", + "cluster_id": "cluster-44779-7", "canonical_issue_number": 45488, "duplicate_issue_numbers": [ + 44779, + 45812, + 45920, 46489 ], - "reason": "Issues in cluster-45488-4 are treated as duplicates because they share closing_reference, soft_similarity evidence." - }, - { - "cluster_id": "cluster-46894-6", - "canonical_issue_number": 46894, - "duplicate_issue_numbers": [ - 46901 - ], - "reason": "Issues in cluster-46894-6 are treated as duplicates because they share closing_reference, shared_issue_target, soft_similarity evidence." + "reason": "Issues in cluster-44779-7 are treated as duplicates because they share closing_reference, soft_similarity evidence." } ], "duplicate_prs": [ @@ -4967,18 +4295,13 @@ "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-43451-7", - "canonical_pr_number": 43451, + "cluster_id": "cluster-43656-2", + "canonical_pr_number": 43656, "duplicate_pr_numbers": [ - 46196, - 46836, - 47003, - 47034, - 47092, - 47096 + 43842 ], "target_issue_number": null, - "reason": "PRs in cluster-43451-7 are treated as duplicates because they share soft_similarity evidence." + "reason": "PRs in cluster-43656-2 are treated as duplicates because they share soft_similarity evidence." }, { "cluster_id": "cluster-43698-3", @@ -5015,6 +4338,15 @@ "target_issue_number": null, "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." }, + { + "cluster_id": "cluster-44779-7", + "canonical_pr_number": 47017, + "duplicate_pr_numbers": [ + 46580 + ], + "target_issue_number": 45488, + "reason": "PRs in cluster-44779-7 are treated as duplicates because they converge on issue #45488 with closing_reference, soft_similarity evidence." + }, { "cluster_id": "cluster-45081-3", "canonical_pr_number": 45317, @@ -5024,15 +4356,6 @@ "target_issue_number": 45081, "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." }, - { - "cluster_id": "cluster-45488-4", - "canonical_pr_number": 47017, - "duplicate_pr_numbers": [ - 46580 - ], - "target_issue_number": 45488, - "reason": "PRs in cluster-45488-4 are treated as duplicates because they converge on issue #45488 with closing_reference, soft_similarity evidence." - }, { "cluster_id": "cluster-45561-3", "canonical_pr_number": 45645, @@ -5060,6 +4383,17 @@ "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-46196-4", + "canonical_pr_number": 46196, + "duplicate_pr_numbers": [ + 46658, + 47003, + 47096 + ], + "target_issue_number": null, + "reason": "PRs in cluster-46196-4 are treated as duplicates because they share soft_similarity evidence." + }, { "cluster_id": "cluster-46869-3", "canonical_pr_number": 46886, @@ -5070,7 +4404,7 @@ "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-46894-6", + "cluster_id": "cluster-46894-5", "canonical_pr_number": 46903, "duplicate_pr_numbers": [ 46895, @@ -5078,53 +4412,7 @@ 47090 ], "target_issue_number": 46894, - "reason": "PRs in cluster-46894-6 are treated as duplicates because they converge on issue #46894 with closing_reference, shared_issue_target, soft_similarity evidence." - }, - { - "cluster_id": "cluster-46907-2", - "canonical_pr_number": 46907, - "duplicate_pr_numbers": [ - 47055 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46907-2 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-46908-2", - "canonical_pr_number": 46908, - "duplicate_pr_numbers": [ - 47057 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46908-2 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-46926-2", - "canonical_pr_number": 46926, - "duplicate_pr_numbers": [ - 47054 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46926-2 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-46949-2", - "canonical_pr_number": 47056, - "duplicate_pr_numbers": [ - 46949 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46949-2 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-46950-3", - "canonical_pr_number": 47058, - "duplicate_pr_numbers": [ - 46950, - 46971 - ], - "target_issue_number": null, - "reason": "PRs in cluster-46950-3 are treated as duplicates because they share soft_similarity evidence." + "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-47013-3", @@ -5146,10 +4434,10 @@ } ], "best_issue": { - "cluster_id": "cluster-46894-6", - "issue_number": 46894, - "reason": "Issue #46894 is the strongest global issue candidate because it is open, belongs to a cluster with 6 artifacts, and carries 1 discussion comments plus 9 inbound references.", - "score": 171.0 + "cluster_id": "cluster-44779-7", + "issue_number": 45488, + "reason": "Issue #45488 is the strongest global issue candidate because it is open, belongs to a cluster with 7 artifacts, and carries 16 discussion comments plus 4 inbound references.", + "score": 206.0 }, "best_pr": { "cluster_id": "cluster-43979-11", diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index d942cfaeff73e063cd8a9209085f690770ee5757..f8eb7d18bfcc9b95febb2561c078b3e798c8dbc0 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": "20260707T000034Z", - "generated_at": "2026-07-07T00:11:37Z", + "snapshot_id": "20260707T060715Z", + "generated_at": "2026-07-07T06:19:22Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -132,20 +132,20 @@ "issue:45237|issue:46015", "issue:43039|issue:44805" ], - "prepared_review_unit_hash": "1ffdcf040065562dc090c27eb386c257002d87f8af83c4e3177e175a2176d79b", + "prepared_review_unit_hash": "026a428721709cf69668b57ab28acbb5a76fc55fcd8398c28cee16ca1572e265", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 63, - "serialized_chars": 46567, + "serialized_chars": 46565, "estimated_input_tokens": 11642, "estimated_eval_tokens": 23540 }, @@ -154,399 +154,495 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous noise: most items are different model/runtime regressions, docs issues, or hardware-specific failures that only look similar at a high level. The only clear duplicate-like pairs are the two ZeRO-3 BERT index-out-of-bounds reports and the two Qwen3_5Moe rotary-embedding/offload reports.", - "confidence": 0.34, - "canonical_issue_reason": "issue:43596 is the clearest true duplicate anchor in the cluster: it describes the same ZeRO-3 index-0 out-of-bounds failure as issue:43638, with slightly broader wording.", + "summary": "This cluster has three accepted soft-edge duplicate pairs, but they are separate families rather than one unified bug: the ZeRO-3/BERT zero-sized tensor IndexError (43596/43638), the Qwen3_5MoeTextRotaryEmbedding offload/materialization bug (45859/45902), and the torch 2.4.1 compatibility regression (45800/46515). Most other candidate links are correctly rejected because they only share a broad subsystem, model family, or version context, not the same concrete code path.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 43596 is the cleanest representative of the strongest duplicate family: the DeepSpeed ZeRO-3/BERT zero-sized tensor IndexError. It is broader and better-scoped than the narrower non-pretrained-BERT variant.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45237 is the broadest central representative of the cluster\u2019s recurring theme of model/device/runtime failures, but it is still not a true umbrella duplicate for most of the items.", + "best_issue_reason": "Issue 43596 is the best overall issue anchor because it corresponds to the clearest accepted duplicate pair and states the core failure mode succinctly.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Both are Qwen3.5-related failures, but one is a missing config attribute and the other is a streaming chat/completions bug; different code paths." + "reason": "Different failures: missing config attribute vs streaming chat/completions API error. Same model family is not enough." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Different hardware/backend failures: AMD GPU model support vs an MPS histogram op limitation." + "reason": "Different hardware/backend paths: AMD GPU failure vs MPS histogram Int limitation." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Both are device/dtype-related, but one is a multimodal dtype mismatch and the other is an AMD GPU execution failure on a different model." + "reason": "Both are model/device issues, but the concrete bugs differ: dtype mismatch in preprocessing vs AMD GPU runtime failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Different bugs: autocast dtype mismatch in Phi MoE versus a GPT-OSS AMD GPU issue." + "reason": "Different code paths: autocast dtype mismatch vs AMD GPU-specific failure." }, { "left": "issue:45237", "right": "issue:46468", "accept": false, - "reason": "Both mention model failures, but one is GPT-OSS on AMD GPUs and the other is a broken Nemotron example; no shared concrete bug." + "reason": "One is GPT-OSS on AMD GPUs, the other is a broken official example; no shared concrete bug." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different failures: AMD GPU compatibility versus an invalid method call in GraniteMoEHybrid." + "reason": "Different model classes and symptoms; too broad to be the same bug." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "One is a DeepSpeed collective mismatch, the other is an ONNX export error; unrelated code paths." + "reason": "Both involve runtime/model errors, but one is a collective mismatch and the other is ONNX export." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "ModernBERT ONNX export and Qwen3Moe FP8 issues are separate model/export/runtime problems." + "reason": "Different model/export failure vs FP8 configuration failure." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "ONNX export failure and Qwen3 auto device-map CUDA assert are unrelated." + "reason": "ONNX export and Qwen3 auto device map cudaErrorAssert are unrelated failures." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Table QA crash and outdated RAG examples are different features and different bugs." + "reason": "Both are documentation/example problems, but they affect different examples and different docs pages." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ONNX export failure and GenerationMixin import breakage are unrelated." + "reason": "ONNX export bug is not the same as the GenerationMixin import regression." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are stale docs after v5 removals, but they reference different removed items and are not the same concrete docs bug." + "reason": "Both are v5 docs cleanup items, but they concern different removed docs topics and not one identical change." }, { "left": "issue:44918", "right": "issue:46165", "accept": false, - "reason": "TRL SFT embedding unpacking failure and resume-training/trust-remote-code issue are different training flows." + "reason": "Training/trainer embedding unpacking is unrelated to resuming local model training with trust remote code." }, { "left": "issue:46165", "right": "issue:46869", "accept": false, - "reason": "Resume-training/remote-code and multi-GPU YOLO device-map failures are unrelated." + "reason": "Resume/trust-remote-code flow is unrelated to multi-GPU device-map failure." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Generic mask-shape mismatch and Gemma4 PLE device mismatch are distinct runtime errors." + "reason": "Mask shape mismatch and Gemma4 PLE device mismatch are different forward-path bugs." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both are IndexError-type reports, but one is ZeRO-3 BERT init and the other is a mask/tensor shape mismatch." + "reason": "Both are shape/index errors, but one is a zero-sized tensor under ZeRO-3 and the other is a mask/index mismatch." }, { "left": "issue:43638", "right": "issue:46165", "accept": false, - "reason": "ZeRO-3 BERT init error and resume-training remote-code issue are unrelated." + "reason": "DeepSpeed/BERT initialization bug is unrelated to resuming training with remote code." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Same exception class, different failure sites and model paths." + "reason": "The zero-sized tensor IndexError is specifically tied to ZeRO-3/BERT init, not a generic mask indexing failure." }, { "left": "issue:44805", "right": "issue:46294", "accept": false, - "reason": "Mask shape mismatch and Qwen3.5 video template mismatch are different multimodal bugs." + "reason": "Mask/index shape mismatch is unrelated to Qwen3.5 video token/template mismatch." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Mask indexing failure and GPT-OSS tensor-parallel issue are unrelated." + "reason": "Different failures: tensor mask mismatch vs GPT-OSS MoE only-TP problem." }, { "left": "issue:46165", "right": "issue:46468", "accept": false, - "reason": "Different problems: resume training with trust remote code versus a broken official example." + "reason": "Model resume/trust-remote-code issue is not the same as a broken official example." }, { "left": "issue:44560", "right": "issue:46165", "accept": false, - "reason": "Video chat template stop-iteration and remote-code resume issues do not share a concrete code path." + "reason": "Video chat-template failure is unrelated to local resume/trust-remote-code training." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice preset and ModernBERT ONNX export are unrelated." + "reason": "BarkProcessor voice preset handling and ModernBERT ONNX export are unrelated." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Different DeepSpeed failures: collective mismatch versus BERT ZeRO-3 index error." + "reason": "Both are distributed-training errors, but the concrete mechanisms differ: collective mismatch vs zero-sized BERT init." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Both describe the same ZeRO-3 index-0 out-of-bounds BERT initialization failure; only the model wording differs." + "reason": "Same underlying ZeRO-3/BERT zero-sized tensor IndexError; the second is just a narrower description of the same bug." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger Kernel cross-entropy being called and GPT-OSS tensor-parallel failure are different issues." + "reason": "Liger Kernel cross-entropy routing and GPT-OSS MoE TP-only failure are different code paths." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Incorrect detokenization and LoRA weight merge/save mismatch are unrelated." + "reason": "Different problems: detokenization regression vs LoRA weight merge/save/reload mismatch." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Qwen3-8B NVFP4 runtime failure and LoRA merge/save mismatch are different bugs." + "reason": "Different model/runtime issue vs LoRA merge/reload output mismatch." }, { "left": "issue:44841", "right": "issue:46661", "accept": false, - "reason": "Processor failure for Voxtral and remote-code model_class config_class error are different code paths." + "reason": "Processor failure for Voxtral and remote-code model_class config_class error are unrelated." }, { "left": "issue:43716", "right": "issue:46869", "accept": false, - "reason": "Both involve model/device placement, but they are different models and different failures." + "reason": "Different device-mismatch-style failures on different models and code paths." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "DeepSpeed weight-dimension error and collective mismatch are distinct training failures." + "reason": "Both are DeepSpeed-related, but one is a weight-dimension training error and the other is a collective mismatch on Jamba-v01." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Activation-checkpointing/FSDP tensor-size mismatch is not the same as a mask indexing error." + "reason": "Activation checkpointing/FSDP recompute mismatch is not the same as a mask shape indexing error." }, { "left": "issue:45685", "right": "issue:46869", "accept": false, - "reason": "MPS histogram limitation and multi-GPU YOLO device-map failure are unrelated backend issues." + "reason": "MPS histogram Int error is unrelated to YOLO multi-GPU device-map failure." }, { "left": "issue:45859", "right": "issue:45902", "accept": true, - "reason": "Both point to the same Qwen3_5MoeTextRotaryEmbedding state/materialization bug around offload or meta-to-empty handling." + "reason": "Same underlying Qwen3_5MoeTextRotaryEmbedding/offload-materialization bug around inv_freq handling and CPU/offload semantics." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "FSDP checkpointing size mismatch and CustomTrainer compute_loss signature breakage are unrelated training issues." + "reason": "FSDP checkpointing tensor-size mismatch is unrelated to CustomTrainer compute_loss signature breakage." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "ZeRO-3 BERT init and GPT-OSS tensor-parallel support are different problems." + "reason": "ZeRO-3 BERT init IndexError and GPT-OSS TP-only bug are unrelated." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "Same as above: different model families and failure mechanisms." + "reason": "Same rationale: different model families and failure modes." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Mistral dtype mismatch and GraniteMoE invalid method call do not share a concrete bug." + "reason": "Different bug classes: dtype mismatch vs invalid method call." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error and GPT-OSS AMD GPU failure are different model/backend issues." + "reason": "Qwen-3 NVFP4 runtime failure and GPT-OSS on AMD GPUs are different model/backend failures." }, { - "left": "issue:45874", - "right": "issue:46294", + "left": "issue:43828", + "right": "issue:45161", "accept": false, - "reason": "Gemma4 reverse-embedding lookup and Qwen3.5 video token mismatch are unrelated multimodal bugs." + "reason": "Autocast dtype mismatch and GPT-OSS MoE TP-only failure are unrelated." }, { - "left": "issue:39290", - "right": "issue:45081", + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Mixtral torch dynamo tracing failure is unrelated to Mistral image preprocessor dtype mismatch." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Different model families and failure mechanisms; only both are runtime errors." + }, + { + "left": "issue:43716", + "right": "issue:46468", + "accept": false, + "reason": "Mistral image preprocessor dtype mismatch and broken official Nemotron example are not the same bug." + }, + { + "left": "issue:45507", + "right": "issue:45685", + "accept": false, + "reason": "Invalid method call in GraniteMoEHybrid and MPS histogram Int error are different code paths." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime failure and Qwen3.5 embedding unpacking with TRL are different issues." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Mistral dtype mismatch and MPS Int histogram error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Different model-specific runtime failures; no shared code path." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT unpacking failure is unrelated to GPT-OSS MoE TP-only failure." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy routing and TRL embedding unpacking are different training-stack issues." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading IndexError and BERT ZeRO-3 loading IndexError are similar only at a high level, but not clearly the same model/code path." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs issues, but one is about removed pipeline tasks and the other about the removed transformers run command." + }, + { + "left": "issue:43828", + "right": "issue:45685", "accept": false, - "reason": "Gemma3TextConfig missing attribute and Mistral regex patch crash are different tokenizer/model config issues." + "reason": "Autocast dtype mismatch and MPS histogram Int error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Different code paths: Liger Kernel cross-entropy vs autocast dtype mismatch." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and CustomTrainer signature breakage are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation checkpointing/FSDP mismatch and GenerationMixin import regression are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "GPT-OSS on AMD GPUs and RAG docs are not the same kind of issue." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross-entropy dispatch and mask shape mismatch are unrelated." }, { "left": "issue:45800", "right": "issue:46515", - "accept": false, - "reason": "Both mention torch compatibility, but they concern different versions and likely different breakages." + "accept": true, + "reason": "Same torch 2.4.1 compatibility regression; one title is the specific transformers 5.8.0 pairing and the other is the generic version incompatibility." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Gemma4 device mismatch in forward and expensive reverse embedding lookup are different issues." + "reason": "Device-map forward mismatch and expensive reverse embedding lookup are different issues." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "FSDP checkpointing mismatch and Qwen3 auto device-map CUDA assert are unrelated." + "reason": "FSDP activation checkpointing mismatch is unrelated to Qwen3 auto device-map cudaErrorAssert." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Torch 2.10 incompatibility and Qwen-3 NVFP4 runtime failure are not the same bug." + "reason": "Torch-version incompatibility and Qwen-3 NVFP4 runtime failure are different problems." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Autocast dtype mismatch and GPT-OSS tensor-parallel support issue are different failures." + "reason": "Autocast dtype mismatch and GPT-OSS MoE TP-only failure are unrelated." }, { "left": "issue:43541", "right": "issue:43716", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm tracing error and Mistral-3 dtype mismatch are separate runtime problems." + "reason": "Mixtral torch dynamo tracing failure is unrelated to Mistral image preprocessor dtype mismatch." }, { "left": "issue:43828", "right": "issue:44292", "accept": false, - "reason": "Both are Qwen/MoE-adjacent runtime issues, but one is autocast dtype mismatch and the other is NVFP4 execution failure." + "reason": "Different model families and failure mechanisms; only both are runtime errors." }, { "left": "issue:43716", "right": "issue:46468", "accept": false, - "reason": "Different model families and different failure points." + "reason": "Mistral dtype mismatch and broken official Nemotron example are not the same bug." }, { "left": "issue:45507", "right": "issue:45685", "accept": false, - "reason": "Invalid method call in GraniteMoEHybrid versus an MPS operator limitation are unrelated." + "reason": "Invalid method call in GraniteMoEHybrid and MPS histogram Int error are different code paths." }, { "left": "issue:44292", "right": "issue:44918", "accept": false, - "reason": "Qwen-3 NVFP4 runtime failure and TRL SFT embedding unpacking failure are different training/inference bugs." + "reason": "Qwen-3 NVFP4 runtime failure and Qwen3.5 embedding unpacking with TRL are different issues." }, { "left": "issue:43716", "right": "issue:45685", "accept": false, - "reason": "Different backends and different error types." + "reason": "Mistral dtype mismatch and MPS Int histogram error are unrelated." }, { "left": "issue:43716", "right": "issue:44292", "accept": false, - "reason": "Different model-specific runtime failures." + "reason": "Different model-specific runtime failures; no shared code path." }, { "left": "issue:44918", "right": "issue:45161", "accept": false, - "reason": "TRL trainer embedding unpacking and GPT-OSS tensor parallelism are unrelated." + "reason": "TRL SFT unpacking failure is unrelated to GPT-OSS MoE TP-only failure." }, { "left": "issue:43039", "right": "issue:44918", "accept": false, - "reason": "Liger Kernel cross-entropy call and TRL SFT embedding unpacking are different issues." + "reason": "Liger Kernel cross-entropy routing and TRL embedding unpacking are different training-stack issues." }, { "left": "issue:41762", "right": "issue:43638", "accept": false, - "reason": "Gemma3 ZeRO-3 loading error and BERT ZeRO-3 loading error are similar in shape but not the same concrete bug." + "reason": "Gemma3 ZeRO-3 loading IndexError and BERT ZeRO-3 loading IndexError are similar only at a high level, but not clearly the same model/code path." }, { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are v5 docs cleanups, but they cover different obsolete references and are not the same issue." + "reason": "Both are docs issues, but one is about removed pipeline tasks and the other about the removed transformers run command." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Autocast dtype mismatch and MPS histogram backend limitation are unrelated." + "reason": "Autocast dtype mismatch and MPS histogram Int error are unrelated." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Liger Kernel cross-entropy invocation and Phi MoE autocast mismatch do not share a code path." + "reason": "Different code paths: Liger Kernel cross-entropy vs autocast dtype mismatch." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed collective mismatch and CustomTrainer compute_loss signature mismatch are unrelated." + "reason": "DeepSpeed collective mismatch and CustomTrainer signature breakage are unrelated." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "FSDP activation-checkpointing mismatch and GenerationMixin import error are different subsystems." + "reason": "Activation checkpointing/FSDP mismatch and GenerationMixin import regression are unrelated." }, { "left": "issue:45237", "right": "issue:46015", "accept": false, - "reason": "GPT-OSS AMD GPU failure and outdated RAG examples are not duplicates." + "reason": "GPT-OSS on AMD GPUs and RAG docs are not the same kind of issue." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Kernel selection/cross-entropy and mask shape mismatch are different bugs." + "reason": "Liger Kernel cross-entropy dispatch and mask shape mismatch are unrelated." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays mostly conservative. The two accepted duplicate-like pairs are well supported by the issue titles and wording: the ZeRO-3 BERT index-out-of-bounds reports, and the Qwen3_5Moe rotary-embedding/offload reports. The broader cluster characterization as heterogeneous is also reasonable. No clear overstatement of evidence stands out." + "accept": false, + "feedback": "The verdicts are mostly conservative, but the summary is not fully grounded in the packet: it says the ZeRO-3/BERT pair is the only clear duplicate and the rest are too broad, yet the analyst also accepted issue:45859 vs issue:45902 as a likely duplicate. That omission makes the aggregate summary imprecise. The accepted ZeRO-3/BERT pair and the torch 2.4.1 pair are well supported, but the summary should mention all accepted soft-edge pairs." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -657,20 +753,20 @@ "issue:43572|issue:45070", "issue:45800|issue:45823" ], - "prepared_review_unit_hash": "fb4f32f7d83214d53a6cc49af28f4ffce4b4efad59ef667efab7fa92f4d5607b", + "prepared_review_unit_hash": "75041e5e4c1bac764fc899a19738544d9a1428d220cae03c156d446c5a0fa482", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 42, - "serialized_chars": 40039, + "serialized_chars": 40040, "estimated_input_tokens": 10010, "estimated_eval_tokens": 20276 }, @@ -679,84 +775,84 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose set of version/regression reports that are not true duplicates. I found two likely duplicate families: AutoTokenizer token-ID regressions and DeepSpeed ZeRO-3 index-out-of-bounds loading failures. Everything else looks only superficially similar.", - "confidence": 0.6, - "canonical_issue_reason": "issue:45920 is the broadest concrete regression here: it describes a v5 AutoTokenizer token-ID bug across multiple model families, making it the best umbrella for the one clear duplicate family in this noisy set.", + "summary": "Mostly heterogeneous cluster. Keep only the AutoTokenizer v5 regression pair as a true duplicate; reject the ZeRO-3/BertModel IndexError pair as too generic and not sufficiently grounded. The rest are separate model-, version-, or docs-specific reports.", + "confidence": 0.68, + "canonical_issue_reason": "Issue 45812 is the clearest confirmed duplicate anchor: it states the broad AutoTokenizer v5 regression across multiple models, and 45920 is the same bug with an expanded model list.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best representative issue because it is broader than the Granite-specific tokenizer report and captures the shared tokenizer regression pattern more completely.", + "best_issue_reason": "Issue 45020 is the best global umbrella because it captures the broad recent-version / remote_code regression theme across otherwise heterogeneous reports, though it should not absorb unrelated runtime failures.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45800", "right": "issue:46304", "accept": false, - "reason": "Both are version-compatibility complaints, but one is about torch 2.4.1 and the other about USE_DISTRIBUTED=0 support; different failure modes and code paths." + "reason": "Both are compatibility regressions, but they concern different breakages and different code paths." }, { "left": "issue:43296", "right": "issue:46304", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and distributed-build support on PyTorch are unrelated bugs." + "reason": "PaddleOCR-VL loading in vLLM is unrelated to USE_DISTRIBUTED=0 support breaking in pytorch." }, { "left": "issue:43866", "right": "issue:46402", "accept": false, - "reason": "Checkpoint corruption and missing weight loading in Sam3Model are different model-specific loading problems." + "reason": "Different model-loading failures with different symptoms and no evidence of the same root cause." }, { "left": "issue:41762", "right": "issue:43596", - "accept": true, - "reason": "Both report DeepSpeed ZeRO-3 loading/indexing failures with an out-of-bounds index 0 on empty tensors/shards, which looks like the same underlying loading bug." + "accept": false, + "reason": "Both mention ZeRO-3 and IndexError, but the models and failure contexts differ enough that they do not look like the same bug." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Gemma3 sliding_window_pattern missing in vLLM and Qwen3 auto device mapping cuda asserts are unrelated." + "reason": "Gemma3 config attribute missing in vLLM and Qwen3 auto device-map CUDA assert are unrelated failures." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "Both are docs issues, but they refer to different outdated docs content and are not the same change." + "reason": "Both are docs-related, but they cover different documentation mismatches and should not be merged." }, { "left": "issue:44589", "right": "issue:46304", "accept": false, - "reason": "Float8 storage lookup failure and distributed-build support breakage are unrelated runtime/environment issues." + "reason": "Float8 storage lookup failure is a different runtime issue from the USE_DISTRIBUTED=0 compatibility break." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "AMD GPU support for GPT-OSS-20B and table-question-answering crashes are different subsystems and failure modes." + "reason": "AMD GPU model failure and table-question-answering crash are separate problems." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger Kernel cross-entropy usage and Qwen3-vl video StopIteration are unrelated bugs." + "reason": "Liger Kernel cross-entropy routing and Qwen3-VL video StopIteration are different code paths." }, { "left": "issue:43975", "right": "issue:46015", "accept": false, - "reason": "Incorrect detokenization in DeepSeek-Coder and outdated RAG examples are unrelated." + "reason": "Detokenization bug and outdated RAG docs are unrelated." }, { "left": "issue:45823", "right": "issue:46294", "accept": false, - "reason": "Gemma4 device-mapping mismatch and Qwen3.5 video token/feature mismatch are different model- and path-specific bugs." + "reason": "Gemma4 device-mismatch with auto mapping and Qwen3.5 video chat-template mismatch are different model/input issues." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Voxtral processor failure is unrelated to Gemma4 device placement issues." + "reason": "Processor failure for Voxtral and Gemma4 PLE device mismatch are not the same bug." }, { "left": "issue:43866", @@ -768,184 +864,166 @@ "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "NemotronH checkpoint loading and RAG docs updates have nothing in common beyond being issue reports." + "reason": "NemotronH checkpoint loading failure and RAG docs maintenance are unrelated." }, { "left": "issue:45020", "right": "issue:46304", "accept": false, - "reason": "Remote-code regressions and USE_DISTRIBUTED=0 breakage are both broad, but they are not the same bug or code path." + "reason": "One is a broad remote_code regression; the other is a specific PyTorch distributed build incompatibility." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Both describe the same v5 AutoTokenizer token-ID regression; 45920 is the broader multi-model report and 45812 is a narrower Granite-specific instance." + "reason": "Both report the same AutoTokenizer v5 regression producing wrong token IDs across multiple models." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Corrupted checkpoint versus checkpoint-loading implementation bug are different problems." + "reason": "Ovis2 checkpoint corruption and NemotronH load failure are distinct model issues." }, { "left": "issue:44292", "right": "issue:46015", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime failure and RAG docs issues are unrelated." + "reason": "Qwen-3-8B-NVFP4 runtime errors and RAG docs are unrelated." }, { "left": "issue:43872", "right": "issue:46304", "accept": false, - "reason": "bitsandbytes Int8Params API mismatch is unrelated to distributed-build support breakage." + "reason": "bitsandbytes Int8Params signature break and USE_DISTRIBUTED=0 support are different incompatibilities." }, { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Gemma3 config attribute missing and Qwen3-MoE sliding_window issues are separate model/config bugs." + "reason": "They both mention sliding-window-related terms, but affect different models and integration points." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "Both are IndexError reports, but one is a mask shape mismatch and the other is a zero-sized tensor during ZeRO-3; not clearly the same bug." + "reason": "This is only a weak ZeRO-3/BertModel/IndexError similarity; the evidence does not support calling them the same underlying bug." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper runtime failure and table-question-answering 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 test loading failure and Ovis2 checkpoint corruption are different model-specific issues." + "reason": "GLM-4.7 unit-test loading issues and Ovis2 checkpoint corruption are different artifacts." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch version incompatibility and Pydantic PretrainedConfig breakage are different incompatibilities." + "reason": "Torch 2.10 incompatibility and PretrainedConfig pydantic breakage are separate regressions." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "DeepSeek-Coder detokenization and table-question-answering crashes are unrelated." + "reason": "Detokenization bug and table-question-answering crash are unrelated." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper loading/running failure and Pydantic model-field breakage are unrelated." + "reason": "Whisper load failure and pydantic PretrainedConfig breakage are different bugs." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both involve Qwen3.5 and chat, but one is a chat crash and the other is a streaming inference API failure; too broad to treat as one bug." + "reason": "Qwen3.5 chat crash and streaming chat/completions failure are separate issues." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "ZeRO-3 zero-sized tensor loading and torch 2.4.1 compatibility are unrelated." + "reason": "ZeRO-3 BertModel index error and torch/transformers version incompatibility do not share the same root cause." }, { "left": "issue:43541", "right": "issue:46869", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm tracing error and YOLO multi-GPU device-map issue are unrelated." + "reason": "Mixtral grouped_mm tracing and YOLO multi-GPU device-map failure are unrelated." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "Qwen NVFP4 runtime failure and table QA crash are unrelated." + "reason": "Qwen NVFP4 runtime error and table-question-answering crash are unrelated." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Gemma3 vLLM config breakage and Gemma4 device-mismatch forward errors are separate model/version issues." + "reason": "Gemma3 sliding_window_pattern missing and Gemma4 device mismatch are different model-specific failures." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Qwen3.5 streaming API failure and MPS histogram kernel missing for MoE are unrelated." + "reason": "Qwen3.5 streaming inference failure and MPS histogram_mps Int error are separate code paths." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "ImportError for GenerationMixin and Gemma3 config attribute regression are unrelated." + "reason": "GenerationMixin import relocation and Gemma3 config attribute missing are different API breakages." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm and MPS histogram support are different bugs." + "reason": "Mixtral grouped_mm tracing and MPS histogram_mps errors are unrelated." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Grouped_mm tracing error and autocast dtype mismatch in Phi-tiny-MoE are different failure modes." + "reason": "Grouped_mm tracing failure and autocast dtype mismatch are different bugs." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Processor failure for Voxtral and non-template-node compile error are unrelated." + "reason": "Voxtral processor failure and non-template-node compile error are unrelated." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Mistral3 weight tying bug and corrupted Ovis2 checkpoint are different issues." + "reason": "Mistral3 weight-tying bug and Ovis2 checkpoint corruption are different issues." }, { "left": "issue:45507", "right": "issue:46869", "accept": false, - "reason": "GraniteMoEHybrid invalid method call and YOLO multi-GPU device-map failure are unrelated." + "reason": "GraniteMoEHybrid invalid method call and YOLO device-map auto failure are not the same bug." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are IndexError reports, but one is a mask-shape mismatch and the other is a zero-sized tensor loading failure; not the same bug." - }, - { - "left": "issue:45399", - "right": "issue:46661", - "accept": false, - "reason": "flash-attn2 fallback blockage and remote-code model_class config_class errors are different model-loading paths." - }, - { - "left": "issue:43572", - "right": "issue:45070", - "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and Pydantic PretrainedConfig breakage are unrelated." - }, - { - "left": "issue:45800", - "right": "issue:45823", - "accept": false, - "reason": "Torch/transformers version incompatibility and Gemma4 device-map mismatch are different problems." + "reason": "Both are ZeRO-3 IndexErrors, but the error shapes and affected models differ enough that they do not look like the same underlying bug." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The two accepted duplicate families are well supported by the titles: the AutoTokenizer token-ID regression pair (45812/45920) is clearly aligned, and the ZeRO-3 out-of-bounds loading pair (41762/43596) is plausibly the same failure mode. The rejected pairs do not overreach based on the available evidence, so I don\u2019t see unsupported conclusions here." + "accept": false, + "feedback": "The writeup is mostly careful, but one soft-edge verdict is too strong for the packet: `issue:41093` vs `issue:43638` is only a weak ZeRO-3/BertModel/IndexError similarity, and the evidence shown does not support calling them the same underlying bug. Because the instruction is to stay conservative, that accepted duplicate looks overstated. The AutoTokenizer pair is well grounded, but the overall result should be rejected on precision grounds." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1044,14 +1122,14 @@ "issue:44863|issue:45005", "issue:45183|issue:46463" ], - "prepared_review_unit_hash": "58265995341e221699b2a41d792c7c70281d84d41d1757d9d9fd3cc2620f10d2", + "prepared_review_unit_hash": "8dd474041cd1a0f0d54b97735e2b6fc962a1a09c8109095bb38df9f3f9077bbe", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, @@ -1066,208 +1144,205 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a mixed bag of unrelated issues. A few soft pairs do look like true duplicates, especially around remote-code custom-module loading, SigLIP2 output regressions, Qwen3.5 video-token mismatch, and decoder-input validation errors.", - "confidence": 0.63, - "canonical_issue_reason": "issue:45632 is the clearest root-cause report in one of the true duplicate pairs: it identifies the remote-code cache-path collision that can make local custom modules resolve incorrectly.", + "summary": "These issues are mostly separate bugs across different model families and failure modes. The soft matches share broad themes (remote code, MoE, validation, dtype/masking), but none are close enough to treat as the same underlying issue. Best representative issue is the concrete remote-code loading bug in #45698.", + "confidence": 0.83, + "canonical_issue_reason": "#45698 is the most concrete, reproducible, and broadly relevant issue here: it describes a specific wrong-module loading bug in `from_pretrained` after `save_pretrained`, with a clear code path and likely useful anchor for nearby remote-code bugs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45632 is the best canonical issue to keep because it describes the underlying bug directly, rather than a downstream symptom like loading the wrong module after save_pretrained.", + "best_issue_reason": "#45698 is the strongest standalone issue in the set because it is precise, actionable, and describes a specific loader regression rather than a vague symptom or model-specific anomaly.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "Different problems: PaddleOCR-VL loading failure versus GGUF gpt-oss architecture support." + "reason": "Both involve model loading for GPT-OSS/Nemotron-like architectures, but one is a PaddleOCR-VL load failure in transformers/vLLM and the other is GGUF architecture support for gpt-oss. Different bug surfaces and fix paths." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Both mention shape/runtime errors, but the failing code paths and model/trainer contexts are different." + "reason": "Both mention an IndexError shape mismatch, but the first is a generic mask/tensor length mismatch while the second is Qwen3.5 embedding unpacking with TRL SFT trainer. Too different in trigger and code path." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm tracing error is not the same bug as GPT-OSS tensor-parallel support." + "reason": "One is a torch dynamo tracing failure in Mixtral float32 grouped_mm, the other is TP-only behavior for GPT-OSS MoE. Same broad MoE area, but not the same concrete bug." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Different symptoms and paths: autocast dtype mismatch in Phi-tiny-MoE versus streaming chat/completions failure in Qwen3.5." + "reason": "These are different model families and error contexts: Phi-tiny-MoE dtype mismatch under autocast vs Qwen3.5 streaming chat/completions failure. Not the same underlying issue." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF architecture support and chat/completions streaming are unrelated." + "reason": "GPT-OSS GGUF architecture support and Qwen3.5 streaming inference are unrelated failures despite both involving recent model support." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Int4 OOM under v5.x and tied-weight translation-model issues are separate bugs." + "reason": "Both concern v5 quantization/tied-weights regressions, but one is CUDA reserved memory OOM under int4 and the other is tied weights on translation models. Different symptom and remedy." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "init_empty_weights argument handling and pydantic PretrainedConfig field regression are different code paths." + "reason": "Init-empty-weights TypeError from an unexpected `_is_hf_initialized` argument is a distinct initialization bug from the pydantic `PretrainedConfig` field regression." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding_window bug is unrelated to the pydantic model-field regression." + "reason": "Qwen3-MoE sliding_window behavior and a pydantic model field regression are unrelated problems." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Both are IndexErrors, but one is a mask-length mismatch and the other is a zero3/BertModel init failure." + "reason": "Both are IndexErrors, but one is a mask-vs-tensor shape mismatch during processing and the other is deepspeed zero3 init with BertModel and an empty-dimension access. Different causes." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code breakage and torch/transformers version incompatibility are not the same underlying bug." + "reason": "Both are version/regression complaints, but one is about remote_code models breaking and the other is torch 2.4.1 vs transformers 5.8.0 incompatibility. Too broad to merge." }, { "left": "issue:43716", "right": "issue:46294", "accept": false, - "reason": "Mistral-3 image-preprocessor dtype mismatch is a different modality and failure mode than Qwen3.5 video-token/template mismatch." + "reason": "Image preprocessor dtype mismatch in Mistral-3 is unrelated to Qwen3.5 video token/feature mismatch in chat template rendering." }, { "left": "issue:43054", "right": "issue:43994", - "accept": true, - "reason": "Both report broken SigLIP2 behavior for the same model family, with bad embeddings / nonsensical outputs pointing to the same underlying model implementation issue." + "accept": false, + "reason": "Both are SIGLIP2-related quality complaints, but one is about text embeddings being worse and the other is nonsensical AutoModel/pipeline outputs. Not the same failure mode." }, { "left": "issue:44479", "right": "issue:46294", - "accept": true, - "reason": "Both describe the same Qwen3.5-family video-input regression where video features/tokens are misaligned." + "accept": false, + "reason": "Video input regression in Qwen vl models and a video feature/token mismatch in Qwen3.5-9B are related by domain but not clearly the same bug or fix." }, { "left": "issue:46468", "right": "issue:46869", "accept": false, - "reason": "Nemotron example breakage and YOLO multi-GPU device-map failure are unrelated." + "reason": "Official Nemotron example broken vs YOLO multi-GPU device_map auto are completely different issues and subsystems." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite-config unwrapping and uninitialized inv_freq after materialization are different bugs." + "reason": "One is `from_config` not unwrapping `text_config` for composite Qwen multimodal configs; the other is uninitialized memory in Qwen3_5MoeTextRotaryEmbedding after materialization. Related model family, different defect." }, { "left": "issue:43596", "right": "issue:46165", "accept": false, - "reason": "Zero3/BertModel initialization and trust_remote_code resume behavior do not match the same root cause." + "reason": "Both touch model loading/training with remote/local models, but one is a zero3 init IndexError in BertModel and the other is resume-training with locally downloaded remote-code models. Not the same code path." }, { "left": "issue:46640", "right": "issue:46661", "accept": false, - "reason": "Flash-Attn2 availability/selection and remote-code model_class config handling are different issues." + "reason": "Flash Attention 2 integration problem and an AutoModel remote-code `config_class` AttributeError are unrelated." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Image-preprocessor dtype mismatch and SFT unpacking failure are separate model/trainer paths." + "reason": "Mistral-3 image preprocessing dtype mismatch is unrelated to Qwen3.5 input embedding unpacking with TRL." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP 2-D weight error is unrelated to the Qwen3.5/TR L unpacking problem." + "reason": "FSDP `'weight' must be 2-D` is a distributed training parameter-shape issue, not the same as Qwen3.5 embedding unpacking failure." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Jupyter custom-model initialization and flash-attn2 fallback gating are not the same bug." + "reason": "Notebook custom-model initialization in v5 and flash-attn2 fallback blocking are different regressions with different remediation." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "High perplexity on gpt-oss and GLM5 are unrelated model-quality reports." + "reason": "High perplexity on GPT-OSS-20B and the generic GLM5 issue are not the same underlying bug; one is a model-quality/regression report and the other is an umbrella model issue." }, { "left": "issue:45632", "right": "issue:45698", - "accept": true, - "reason": "Both concern remote-code custom modules being resolved incorrectly; one is the path-collision/root-cause report and the other is the wrong-module symptom after save_pretrained." + "accept": false, + "reason": "Both involve remote-code/local custom module loading, but one is a cache-path collision for local models sharing a leaf name and the other is loading the wrong custom module after save/load. Similar area, different root cause." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "FineGrainedFP8Config failure is unrelated to flash-attn2 fallback blocking." + "reason": "FineGrainedFP8Config failure for Qwen3Moe and flash-attn2 fallback gating are unrelated config/integration issues." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Tied lm_head weights in Mistral3 and translation-model tied-weight issues are not the same concrete bug." + "reason": "Untied lm_head in Mistral3 conditional generation is a model-weight tying bug, while the v5 translation-model issue is broader and different in scope." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Offloading/quantization behavior is separate from tied-weight handling." + "reason": "Quantization offloading behavior and translation-model tied weights are different features and failure modes; too broad to combine." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "A DSA indexer discussion and a docs typo are unrelated." + "reason": "A DSA indexer ReLU discussion and a docs typo about the removed run command are unrelated." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "Cache-strategy typo and removed run-command docs are separate documentation issues." + "reason": "Cache strategies typo and removed docs run command are both documentation/config text issues, but not the same issue." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight translation-model issues." + "reason": "PagedAttentionCache linear_attention crash on Qwen3.5 models is a runtime cache/kernel incompatibility, not the same as tied-weight regressions." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Qwen3.5 embedding unpacking and GraniteMoEHybrid invalid method calls are different failures." + "reason": "Qwen3.5 input embedding unpacking with TRL SFT and GraniteMoEHybrid calling an invalid method are different model-specific bugs." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint loading and translation-model tied weights are unrelated." + "reason": "NemotronH checkpoint loading failure and tied-weights regressions are separate model-integration problems." }, { "left": "issue:45183", "right": "issue:46463", - "accept": true, - "reason": "Both are the same input-validation bug class: incorrect XOR/error messaging for decoder input validation, with T5 being a specific instance." + "accept": false, + "reason": "These both concern input validation errors, but one is a broad XOR logic bug across models and the other is T5 decoder-input error messaging. Similar pattern, not the same bug." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet, and the accepted soft pairs are plausibly the same underlying bugs without overclaiming. The rejected pairs are phrased conservatively and do not overstate duplicate evidence." - }, - "evaluator_used": true, + "evaluator_result": null, + "evaluator_used": false, "retried": false, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -1364,223 +1439,223 @@ "issue:41553|issue:43881", "issue:43782|issue:44960" ], - "prepared_review_unit_hash": "9d25388d8f1a693eceb452e96b0cd5f9b461bfa7d915c9175a58591ad40ad7a6", + "prepared_review_unit_hash": "f851bfae60cd62c7c3f50b44a534b2210b5e9091df7b1ccb7fc0fa5a5db1cae7", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36283, - "estimated_input_tokens": 9071, - "estimated_eval_tokens": 18398 + "serialized_chars": 36287, + "estimated_input_tokens": 9072, + "estimated_eval_tokens": 18400 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous set of Transformers regression reports rather than one duplicate cluster. A few pairs do look like the same underlying bug (non-persistent buffer corruption, cache length contract, Gemma4 quantized embedder dtype casting), but most other pairs are only loosely related by subsystem/model family and should stay separate.", - "confidence": 0.81, - "canonical_issue_reason": "If a single representative is needed, issue 45020 is the broadest umbrella: it summarizes recent-version regressions that break model loading for remote-code models, which is the closest common thread across the set. That said, the cluster is still fairly heterogeneous.", + "summary": "Mostly unrelated issue families overall, with one tight duplicate subcluster around the Transformers v5 non-persistent-buffer regression (#43950, plus later symptom reports like #44534 and #45902). The #43643 pairing is not the same bug and should be rejected.", + "confidence": 0.91, + "canonical_issue_reason": "#43950 is the best anchor: it explicitly describes the v5 non-persistent-buffer corruption regression and is broad enough to cover later symptom reports like #44534 and #45902.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45020 is the best overall issue representative because it is the most general user-facing regression report in the group and captures the recurring theme of recent Transformers versions breaking previously working model loads.", + "best_issue_reason": "#43950 is the most representative issue overall because it names the concrete regression, identifies the affected code path, and matches the later buffer-corruption reports in the packet.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Both involve model/tokenizer loading failures, but one is a DeepSeek-Coder v1 detokenization regression and the other is a Qwen-3 NVFP4 loading/error case; different bugs and code paths." + "reason": "Different bugs: tokenizer detokenization regression vs Qwen NVFP4 load/runtime error." }, { "left": "issue:46489", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and symptoms: DeepSeek-Coder v1 incorrect output vs CodeLlama leading-space round-trip loss." + "reason": "Both are tokenizer regressions, but they affect different model families and symptoms; not clearly the same underlying bug." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "SigLIP2 docs example errors are not the same as a model-quality/embedding regression; one is an example mismatch/quantization failure, the other is output quality degradation." + "reason": "Docs/example mismatch and quantization failure vs a general text-embedding quality regression are not the same bug." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "These describe the same underlying v5 regression where non-persistent buffers (`register_buffer(persistent=False)`) are corrupted/filled with junk during loading." + "reason": "Same non-persistent-buffer regression in Transformers v5; #44534 reads like a later symptom of the same bug." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Unexpected compute_loss keyword argument handling is unrelated to Gemma3 ZeRO-3 loading failure; different APIs and failure modes." + "reason": "Different areas and failure modes: trainer API change vs DeepSpeed ZeRO-3 model loading." }, { - "left": "issue:43644", + "left": "issue:43643", "right": "issue:43950", "accept": false, - "reason": "One is a `trust_remote_code` config-field issue, the other is non-persistent buffer corruption; both touch loading but not the same bug." + "reason": "#43643 is about trust_remote_code=True returning missing fields from AutoConfig.from_pretrained, while #43950 is a Transformers 5.x non-persistent-buffer corruption regression; only superficial similarity." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "These are different quantization/back-end compatibility problems: bitsandbytes `Int8Params` signature mismatch vs missing float8 storage type." + "reason": "Unrelated incompatibilities: bitsandbytes parameter-constructor change vs missing Float8 storage." }, { "left": "issue:46899", "right": "issue:46932", "accept": true, - "reason": "Both are the same Gemma4 multimodal quantization bug: unsafe dtype casting in quantized vision/audio embedders." + "reason": "Same Gemma4 quantized vision/audio embedder dtype-casting bug; #46932 explicitly says it is the same issue." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "The Phi MoE autocast dtype mismatch and the Qwen3.5 RLHF NaN/SDPA fallback issue are distinct model-specific failures." + "reason": "Different model/code paths and symptoms: autocast dtype mismatch vs Qwen3.5 position-id/SDPA stability issues." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality regression is unrelated to the tokenizer behavior change in newer Transformers versions." + "reason": "Both mention tokenization/embeddings, but one is a quality regression and the other is a version-dependent tokenization change." }, { "left": "issue:45920", "right": "issue:46489", - "accept": false, - "reason": "Both are tokenizer regressions, but they concern different affected models and different symptoms; not the same concrete bug." + "accept": true, + "reason": "Same tokenizer-output regression in v5; #46489 is a model-specific manifestation of the broader wrong-token-IDs issue." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Both mention SigLIP2, but one is unexpected zero probabilities and the other is worse text embeddings; too different to treat as one bug without stronger evidence." + "reason": "Different underlying problems: zero probabilities on SigLIP2 vs a text-embedding quality complaint." }, { "left": "issue:45987", "right": "issue:46858", "accept": true, - "reason": "These are the same cache-length contract bug: `get_seq_length()` returns a tensor instead of an int, breaking downstream arithmetic and compilation paths." + "reason": "Same cache sequence-length contract bug; #46858 is the backend-specific manifestation of the same return-type problem." }, { "left": "issue:43950", "right": "issue:45902", - "accept": false, - "reason": "Both concern non-persistent buffers, but one is a generic from_pretrained corruption regression and the other is a specific uninitialized `inv_freq` materialization path; not clearly the same bug." + "accept": true, + "reason": "Same non-persistent-buffer initialization/corruption regression; #45902 is a specific buffer instance exposed by the same bug." }, { "left": "issue:43054", "right": "issue:43493", "accept": false, - "reason": "Both are about SigLIP2 behavior, but one is embedding quality and the other is HF-vs-JAX implementation discrepancy; related area, different underlying issue." + "reason": "Same model family, but different issues: quality regression vs HF/JAX implementation discrepancy." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Flash-Attention illegal memory access from 3D position_ids is unrelated to CPU-offload incompatibility in Qwen3_5Moe rotary embeddings." + "reason": "Different failures: flash-attention packed-sequence misinterpretation vs CPU offload incompatibility in rotary embedding." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing/LoRA and int4 quantization OOM are both memory-related, but they are different mechanisms and fixes." + "reason": "Both are memory-related, but one is LoRA checkpointing behavior and the other is int4 quantization reserve-memory growth." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Gemma4 FSDP2 per-layer shared_kv_states rebuilding is a different bug from non-persistent buffer initialization/corruption." + "reason": "Different code paths: Gemma4 FSDP2 per-layer keying issue vs non-persistent buffer materialization corruption." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Generic loading noise is not the same as a specific model failing to load; the latter is a concrete compatibility bug." + "reason": "Generic loading noise vs a specific model load failure; no shared underlying bug is evident." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to Molmo missing `all_tied_weights_keys`." + "reason": "Different model-loading bugs: bad AutoTokenizer error message for Voxtral vs missing all_tied_weights_keys on Molmo." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Tokenizer detokenization regression and Python 3.13 flash-attn import error are unrelated bugs." + "reason": "Different subsystems: tokenizer detokenization vs import_utils flash_attn mapping on Python 3.13." }, { "left": "issue:24643", "right": "issue:28530", "accept": false, - "reason": "DeepSpeed training weight-shape error and early-stopping metric configuration are unrelated training issues." + "reason": "Unrelated training/runtime error vs early-stopping metric selection issue." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Both mention loading, but one is generic loading noise and the other is a specific model load failure; not the same bug." + "reason": "Both are loading-related, but the reported models and failures differ too much to be the same bug." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "`AutoConfig.from_pretrained(..., trust_remote_code=True)` missing fields is a narrower config-construction issue, not the broader remote-code loading regressions in 45020." + "reason": "#45020 is a broad remote_code regression report; #43643 is a specific AutoConfig field-loss symptom, not the same concrete bug." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "LoRA gradient checkpointing behavior and Qwen3 MoE memory usage are different issues despite both involving memory." + "reason": "Different memory problems: ineffective gradient checkpointing vs MoE training memory inefficiency." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Gemma3 ZeRO-3 loading and Qwen3Moe FineGrainedFP8Config failure are unrelated model-specific errors." + "reason": "Distinct model/config errors with no shared failing code path." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading for a specific model and flash-attn fallback gating are unrelated subsystems." + "reason": "Tokenizer loading for a specific model vs flash-attn2 fallback being blocked; unrelated issues." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Missing Qwen2.5-VL export and bitsandbytes `Int8Params` incompatibility are separate compatibility bugs." + "reason": "Missing model import symbol vs bitsandbytes parameter-compatibility error." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Voxtral tokenizer error messaging and GLM-4V-9B loading failure are different model issues." + "reason": "Different load failures affecting different models and APIs." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL `weight_only=True` loading failure and the generic GLM5 issue are not the same code-path problem." + "reason": "Different problems: weight_only load failure for Qwen3VL vs a generic GLM5 issue without the same concrete failure mode." } ] }, "evaluator_result": { - "accept": true, - "feedback": "Grounded overall. The summary stays conservative about heterogeneity, and the accepted soft-edge matches are plausible same-bug pairs based on the packet (buffer corruption, cache-length contract, Gemma4 dtype casting). The rejected pairs are also phrased cautiously and do not overstate sameness." + "accept": false, + "feedback": "Mostly grounded, but one soft-edge acceptance is too strong: issue:43643 vs issue:43950 are not supported as the same bug by the packet. 43643 is about `trust_remote_code=True` returning missing fields from `AutoConfig.from_pretrained`, while 43950 is a Transformers 5.x non-persistent-buffer corruption regression. The similarity is only superficial/soft, so that verdict should be conservative and rejected." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1677,211 +1752,211 @@ "issue:41093|issue:44805", "issue:41553|issue:44991" ], - "prepared_review_unit_hash": "69cc3dc55badb0b5e1a08bf3b55c64511039ad9bee5da27bf7f66bacde0bdc69", + "prepared_review_unit_hash": "c6eceecac739ace0137042fc0d0dc187d2517d12f5097022ac547e898abc5498", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36748, - "estimated_input_tokens": 9187, - "estimated_eval_tokens": 18630 + "serialized_chars": 36754, + "estimated_input_tokens": 9189, + "estimated_eval_tokens": 18634 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The list is mostly heterogeneous, but there are a few clear duplicate pockets: docs about removed pipeline tasks, tokenizer regressions in DeepSeek-family models, video-input regressions in Qwen3-VL, remote-code model loading breakage, and the `_is_hf_initialized` loading incompatibility.", - "confidence": 0.74, - "canonical_issue_reason": "No single canonical for the entire mixed set; if one representative is needed for the strongest duplicate pocket, issue #45020 is the broadest umbrella for the remote-code loading regression and subsumes #46661.", + "summary": "The cluster is still heterogeneous overall. A few pairs are tight duplicates worth keeping together\u2014especially the `_is_hf_initialized` / `init_empty_weights` regression, the v5 docs cleanup, and the DeepSeek tokenizer/pre-tokenizer bug\u2014but the Qwen/DeepSpeed and video-related pairs are only same-family or same-subsystem similarities, not proven single-root-cause duplicates.", + "confidence": 0.72, + "canonical_issue_reason": "No single issue cleanly represents the whole cluster. If an anchor is needed, issue 43872/44291 is the strongest exact duplicate pair because it names a precise shared compatibility failure with a clear common root cause.", "canonical_pr_reason": null, - "best_issue_reason": "#45020 is the best global issue representative because it is the broadest, most central report among the accepted duplicates and has the clearest \u201cumbrella + specific symptom\u201d relationship with #46661.", + "best_issue_reason": "Issue 43872/44291 is the best representative issue pair in the cluster: it is a concrete, well-supported duplicate with a narrowly defined failure mode, unlike the broader same-family regressions elsewhere in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "Different failures and code paths: MPS integer histogram support vs Gemma4 device placement mismatch." + "reason": "Different bugs: an MPS `histogram_mps` int error vs a Gemma4 device-map mismatch." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Both mention Mistral/flash-attn-related areas, but one is a tokenizer regex crash and the other is a flash-attn2 fallback gating issue." + "reason": "Different code paths: a Mistral regex patch crash vs flash-attn fallback gating." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Remote-code loading regressions vs a Gemma4 device-map forward mismatch; not the same bug." + "reason": "Broad remote-code breakage is not the same concrete bug as the Gemma4 device mismatch." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Registration/mapping failure in add-new-model-like is distinct from the broader remote_code model-loading regression." + "reason": "Tokenizer-mapping failure is a separate loader bug from the generic remote-code regression." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Phi MoE autocast dtype mismatch is unrelated to Gemma4 device-map placement." + "reason": "Different runtime failures: autocast dtype mismatch vs device-placement mismatch." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Both involve Qwen3/LoRA warnings or compatibility, but they describe different symptoms and likely different fixes." + "reason": "Related Qwen/LoRA context, but one is a load-time incompatibility and the other is a warning about `tie_word_embeddings`." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Generic remote-code breakage is not the same as the Kimi-K2.5 tokenizer regression and misleading warning." + "reason": "The Kimi tokenizer regression is more specific than the broad remote-code breakage." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Different model families and different loading failures." + "reason": "Different model-loading failures for different architectures." }, { "left": "issue:45715", "right": "issue:46315", "accept": false, - "reason": "Performance regressions in inverse tokenizer conversions, but on different APIs (`convert_ids_to_tokens` vs `convert_tokens_to_ids`)." + "reason": "Both are tokenizer slowdowns, but they affect different conversion paths; not the same bug." }, { "left": "issue:43872", "right": "issue:46869", "accept": false, - "reason": "Bitsandbytes init/kwarg incompatibility is unrelated to YOLO multi-GPU device-map failure." + "reason": "Bitsandbytes constructor incompatibility is unrelated to the YOLO multi-GPU device-map failure." }, { "left": "issue:45759", "right": "issue:46661", "accept": false, - "reason": "Composite-config unwrapping vs remote-code `config_class` attribute error; different loading paths." + "reason": "Different failures in model construction/loading for different APIs and root causes." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same underlying `_is_hf_initialized`/`init_empty_weights` incompatibility during model initialization and loading." + "reason": "Same `_is_hf_initialized` / `init_empty_weights` compatibility failure causing the constructor TypeError." }, { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "Clear duplicate docs issue: both report obsolete pipeline-task references after v5 removals." + "reason": "Same docs regression: both still reference removed v5 pipeline tasks." }, { "left": "issue:45488", "right": "issue:46489", "accept": true, - "reason": "Both describe the same tokenizer regression around LlamaTokenizer overriding tokenizer.json behavior and breaking DeepSeek-family outputs." + "reason": "Same tokenizer/pre-tokenizer regression breaking DeepSeek-family tokenization on v5+." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Tied lm_head weights vs non-persistent buffer corruption are different regression types." + "reason": "Tied `lm_head` weights and non-persistent buffer corruption are different regression classes." }, { "left": "issue:45020", "right": "issue:46661", - "accept": true, - "reason": "#46661 is a concrete instance of the broader remote-code model breakage described by #45020." + "accept": false, + "reason": "Remote-code model breakage and missing `config_class` are different loader problems." }, { "left": "issue:46894", "right": "issue:46897", "accept": false, - "reason": "Similar training-loss double-shift pattern, but across different model families; not enough evidence of one shared bug." + "reason": "Same high-level loss-shift pattern, but the packet does not establish one shared code-path bug." }, { "left": "issue:43792", "right": "issue:44220", "accept": true, - "reason": "Likely the same audio feature-extraction bug path: Whisper failing because of `_torch_extract_fbank_features()`." + "reason": "The Whisper failure is plausibly the concrete symptom of the `_torch_extract_fbank_features()` bug." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Both touch DeepSpeed/Qwen, but one is weight conversion/missing params and the other is gradient averaging with GAS." + "reason": "Both involve DeepSpeed, but one is weight conversion/loading and the other is gradient averaging with GAS." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing ineffectiveness in PEFT LoRA is not the same as a missing trust_remote_code flag." + "reason": "Gradient-checkpointing ineffective under PEFT LoRA is not the same as a missing `supports_gradient_checkpointing` flag." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Different runtime errors and different tensor shapes/paths; only superficially similar Index/Type errors." + "reason": "Unexpected `num_items_in_batch` kwarg vs mask/tensor shape mismatch are unrelated trainer/runtime errors." }, { "left": "issue:43940", "right": "issue:45313", "accept": false, - "reason": "Both are ZeRO-3 load failures, but they affect different Qwen variants with different failure descriptions." + "reason": "Same model family and loader subsystem, but the packet only shows separate ZeRO-3 weight-loading reports; a shared root cause is not established." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Different model families and distinct loading problems." + "reason": "Different model checkpoint-loading failures for different model families." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "RLHF gradient explosion from 3D position ids is unrelated to flash-attention `s_aux=None` crashes." + "reason": "Different runtime failures: RLHF NaNs from position IDs vs flash-attention crash on `s_aux=None`." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed training weight-dimension error vs image processor void-segmentation-map failure." + "reason": "Training DeepSpeed weight shape error is unrelated to void segmentation-map preprocessing." }, { "left": "issue:44479", "right": "issue:44560", - "accept": true, - "reason": "Same video-input regression family affecting Qwen3-VL variants; #44560 looks like a specific symptom of the broader #44479 breakage." + "accept": false, + "reason": "Same general video area, but different model families and symptoms; the packet does not prove one shared bug." }, { "left": "issue:41093", "right": "issue:44805", "accept": false, - "reason": "Both are mask/tensor shape IndexErrors, but there is no evidence they come from the same code path." + "reason": "They share a shape-mismatch theme, but the actual failing paths and symptoms differ." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Tokenizer loading error for Voxtral is not the same as the est-roberta tokenizer failure." + "reason": "Bad Voxtral tokenizer messaging is unrelated to the EMBEDDIA tokenizer load failure." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative: it identifies only a few clearly related pockets and does not overclaim broad duplication. The accepted soft-edge matches are plausible from the issue titles, and the rejected pairs are appropriately cautious where the evidence is only superficial." + "accept": false, + "feedback": "The cluster-level conclusion is mostly grounded, but a few soft-edge accepts are a bit too strong for the evidence in the packet. The exact duplicate pairs 43872/44291, 43827/44509, and 45488/46489 look well supported. However, 43940/45313 and 44479/44560 are only shown to be same-model-family, same-general-area regressions; the packet does not establish a shared root cause strongly enough to call them clear duplicate/mergeable bugfix pairs. The summary also slightly overstates certainty by calling those subgroups \"clear duplicate subgroups.\"" }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1985,20 +2060,20 @@ "issue:43632|issue:45834", "issue:44661|issue:44805" ], - "prepared_review_unit_hash": "a9b4deeadcff14778a41d01acbab72afdb412ec40ddcab602c9061f9cab323e0", + "prepared_review_unit_hash": "ba8b4303a7e9318b4364536a4153b463f0ce5438a232f33872d0d1bd83e94898", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38354, + "serialized_chars": 38355, "estimated_input_tokens": 9589, "estimated_eval_tokens": 19434 }, @@ -2007,228 +2082,228 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated transformer/model-specific bug reports with a few clear duplicate pairs around the same underlying regression. The strongest duplicates are the custom model initialization breakage in Transformers 5, SigLIP2 bad outputs, and Gemma4 4-bit quantization issues.", - "confidence": 0.82, - "canonical_issue_reason": "issue:43646 is the best canonical issue: it names the broad regression directly ('Transformers 5.0.0 breaks custom model initialization') and can subsume the notebook-specific report as an example rather than a separate bug.", + "summary": "Most pairs are clearly unrelated bug reports. I only see a few likely duplicate/near-duplicate edges: the custom-model initialization regression, the SigLIP2 wrong-output regression, and the Gemma4 4-bit multimodal dtype-cast regression.", + "confidence": 0.58, + "canonical_issue_reason": "issue:43646 is the broadest, most canonical phrasing of the custom-model-initialization regression; issue:43645 is a narrower Jupyter-notebook manifestation of the same problem.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the most useful global issue anchor because it is the broadest and most general statement of the underlying regression among the issues in this cluster.", + "best_issue_reason": "issue:43646 is the best representative issue in this set because it describes the core regression without extra environment-specific constraints.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "Different model families and failure modes: Qwen3.5 PagedAttentionCache error vs NemotronH checkpoint loading." + "reason": "Different models and failures: Qwen3.5 PagedAttentionCache group parsing vs NemotronH checkpoint loading." }, { "left": "issue:44863", "right": "issue:46402", "accept": false, - "reason": "Both involve NemotronH, but one is a generic checkpoint-loading failure and the other is a specific text-encoder weight loading regression; not enough evidence they are the same bug." + "reason": "Both are loading bugs, but they affect different model families and different loading paths." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "Unrelated bugs: GGUF architecture support vs Phi-tiny-MoE dtype mismatch under autocast." + "reason": "gpt-oss GGUF support is unrelated to Phi-tiny-MoE autocast dtype mismatch." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "Different problems in different code paths: GGUF gpt-oss support vs Qwen3-MoE sliding_window behavior." + "reason": "GGUF architecture support and Qwen3-MoE sliding_window behavior are different issues." }, { "left": "issue:43596", "right": "issue:46515", "accept": false, - "reason": "BertModel/DeepSpeed ZeRO-3 index error is unrelated to a general torch 2.4.1 incompatibility report." + "reason": "A Bert/deepspeed zero3 index error is not the same as a broad torch-version incompatibility report." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both touch rotary/position-id logic, but they concern different models and distinct bugs." + "reason": "Different Qwen variants and different bugs: rope index scaling vs uninitialized rotary embedding memory after materialization." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Both are loading issues, but they refer to different models and different failure contexts; not the same underlying bug." + "reason": "Both are model-loading complaints, but they concern different models and different root causes." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "GGUF support for gpt-oss is unrelated to a Float8 storage deserialization error." + "reason": "GGUF support for gpt-oss is unrelated to Float8 storage lookup failures." }, { "left": "issue:43638", "right": "issue:46515", "accept": false, - "reason": "The ZeRO-3 BertModel index error is not the same as a broad torch-version incompatibility report." + "reason": "Specific deepspeed/Bert zero3 failure is not the same as a general torch 2.4.1 incompatibility." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "AudioFlamingo batching leak and Whisper empty-transcription failure are different bugs and code paths." + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription handling are different code paths." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4, but one is an FSDP2 shared_kv_states KeyError and the other is a device-map mismatch; separate issues." + "reason": "Same model family, but different failures: FSDP2 shared_kv_states rebuild vs device_map forward mismatch." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3MoE FP8Config failure and meta-device loading breakage are unrelated." + "reason": "Qwen3Moe FP8 loading and meta-device model loading are different regression classes." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Import error for a specific class and meta-device loading regressions are different failures." + "reason": "Missing Qwen2_5_VL symbol import is unrelated to meta-device initialization issues." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Mistral-3 image-preprocessor dtype mismatch is unrelated to Qwen3-vl video StopIteration." + "reason": "Different multimodal models and different failures: dtype mismatch vs StopIteration in video handling." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Non-persistent buffer corruption is a different regression from Kosmos2.5 long-OCR index errors." + "reason": "Non-persistent buffer corruption during from_pretrained is unrelated to Kosmos2.5 long-OCR index errors." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same underlying Transformers 5 custom-model initialization regression; 43645 is the notebook-specific manifestation and 43646 is the broader generic report." + "reason": "Both describe the same transformers v5 custom-model initialization regression; 43645 is the notebook-specific variant." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "3d_parallel.py runtime failure is unrelated to GGUF architecture support." + "reason": "3d_parallel.py runtime failure and gpt-oss GGUF support are not the same underlying bug." }, { "left": "issue:44910", "right": "issue:46858", "accept": false, - "reason": "Both involve attention/cache behavior, but the concrete bugs and affected paths differ substantially." + "reason": "Both involve caching/position arithmetic, but one is a Qwen3.5 Flash Attention crash and the other is a static-cache torch.compile backend issue." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Qwen3-vl video StopIteration and MPS histogram-on-Int are unrelated." + "reason": "Qwen3-vl video StopIteration and MPS histogram support are unrelated." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Bamba SDPA torch.compile failure is unrelated to Qwen3.5 flash-attention position_ids misinterpretation." + "reason": "Bamba SDPA torch.compile failure and Qwen3.5 Flash Attention position_ids crash are different concrete bugs." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "Importing GenerationMixin from transformers.generation and Qwen3MoE FP8 issues are unrelated." + "reason": "Importing GenerationMixin and Qwen3Moe FineGrainedFP8Config are unrelated failures." }, { "left": "issue:44683", "right": "issue:46515", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and a torch 2.4.1 incompatibility report are not the same bug." + "reason": "Torch>=2.9 flex_attention compilation breakage is not the same as a torch 2.4.1 incompatibility report." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Different Qwen3.5 bugs: video processing StopIteration vs embedding unpacking failure in TRL SFT." + "reason": "Video embedding StopIteration and TRL SFT unpacking failure are different Qwen3.5-related bugs." }, { "left": "issue:45812", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and likely different tokenizer code paths." + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes." }, { "left": "issue:45823", "right": "issue:46515", "accept": false, - "reason": "Gemma4 device mismatch under device_map=\"auto\" is unrelated to a general torch version incompatibility." + "reason": "Gemma4 device placement mismatch is unrelated to a general torch-version incompatibility." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "add-new-model-like/TOKENIZER_MAPPING_NAMES failure is unrelated to Gemma4 PLE device mismatch." + "reason": "Tokenizer-mapping-name validation and Gemma4 PLE device mismatch are unrelated." }, { "left": "issue:38175", "right": "issue:43994", "accept": true, - "reason": "Same SigLIP2 base model regression: one report shows zero probabilities, the other nonsensical outputs through AutoModel/pipeline." + "reason": "Both report the same siglip2-base-patch16-224 wrong-output regression, just phrased as zero probabilities vs nonsensical results." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Qwen3-vl video StopIteration and a generic mask/shape IndexError are different failures." + "reason": "Different failures: video StopIteration vs mask/tensor shape mismatch." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Both mention config validation, but they are distinct model/config-loading bugs." + "reason": "Both mention config loading, but they concern different models and different validation failures." }, { "left": "issue:45672", "right": "issue:46899", "accept": true, - "reason": "Same Gemma4 4-bit quantization family of bugs in multimodal/audio-vision embedders; 46899 is the broader statement of the same unsafe dtype-cast issue." + "reason": "These look like the same Gemma4 4-bit multimodal dtype-cast bug; 46899 is the broader restatement of 45672's failure." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "DeepSpeed ZeRO-3 BertModel index error is unrelated to add-new-model-like tokenizer mapping failure." + "reason": "Bert zero3 indexing and add-new-model-like tokenizer mapping validation are unrelated." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "BertModel/ZeRO-3 init failure is unrelated to tokenizer mapping registration." + "reason": "Deepspeed zero3/Bert failure is not the same as tokenizer mapping-name validation logic." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "Qwen3_moe LoRA compatibility on hf5.x and BitNet packed-weight loading are different issues." + "reason": "A qwen3_moe LoRA/hf4.x compatibility break is unrelated to BitNet packed-weight loading." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "_is_hf_initialized flag breakage and Kosmos2.5 long OCR index errors are unrelated." + "reason": "The _is_hf_initialized regression and Kosmos2.5 OCR index error are different issues." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer-mapping registration failure is unrelated to a mask/index shape mismatch." + "reason": "Tokenizer mapping validation and mask shape mismatch are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative: it frames the cluster as mostly unrelated reports with a few clear duplicate pairs, and the accepted pairs are plausible from the titles/body context. I didn\u2019t see any overstatement of evidence in the soft-edge verdicts." + "feedback": "The summary is grounded in the packet, and the soft-edge decisions are conservative. The three accepted pairs are plausibly the same underlying regressions based on their titles, while the rejected pairs are generally supported as different bug classes or different model/path combinations. The canonical/best-issue reasoning for 43646 vs 43645 is also consistent with the issue titles." }, "evaluator_used": true, "retried": false, @@ -2329,20 +2404,20 @@ "issue:44910|issue:45588", "issue:44402|issue:44488" ], - "prepared_review_unit_hash": "6678580ac2859891bcfffd61c5819bc7b797e1b8a04121d7daeab39b4dc43c40", + "prepared_review_unit_hash": "765a726ebad0f46243760839ed41e0cc2c7139bb32a6693fe407d1f1cb326a94", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36054, + "serialized_chars": 36053, "estimated_input_tokens": 9014, "estimated_eval_tokens": 18284 }, @@ -2351,198 +2426,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster contains several unrelated issue families. I only merge the clearly same-bug pairs: non-persistent buffer junk, tokenizer-load regressions on similar models, and the resize_token_embeddings propagation bug.", - "confidence": 0.77, - "canonical_issue_reason": "Issue 45276 is the broadest and clearest duplicate anchor among the accepted pairs; 45335 is a narrower T5Gemma-specific manifestation of the same resize_token_embeddings/output_embeddings bug.", + "summary": "The cluster is mostly heterogeneous, but it contains a few true duplicate pairs: the v5 non-persistent-buffer corruption reports, the resize_token_embeddings propagation bug, and the training-loss double-shift bug. Most other pairs are only thematically similar and should stay separate.", + "confidence": 0.83, + "canonical_issue_reason": "issue:43644 is the clearest broad regression report in the set and has an obvious near-duplicate (issue:44534); it best represents the non-persistent-buffer corruption bug.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45276 has the clearest duplicate evidence and the most general description of the resize_token_embeddings regression, making it the best representative issue.", + "best_issue_reason": "issue:43644 is the most representative standalone issue here: concise, regression-focused, and tied to the clearest duplicate pair in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "Different failures: dequantized-model offload/load behavior vs mmap-related OOM on Strix Halo." + "reason": "Different problems: device_map/offload loading failure vs mmap/OOM on Strix Halo." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "One is a model output regression; the other is a broken documentation example with model/processor mismatch and quantization issues." + "reason": "SigLIP2-related, but one is wrong probabilities and the other is a broken doc example; different root causes." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Same bug reported with minor wording differences: Transformers v5 filling non-persistent buffers with junk." + "reason": "Same v5 regression about non-persistent buffers being filled with junk; titles are paraphrases of the same bug." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "cudaErrorAssert on auto device mapping is a different problem from FineGrainedFP8Config failing on Qwen3MoE." + "reason": "Both touch Qwen3, but one is auto device mapping CUDA assert and the other is FineGrainedFP8Config; not the same bug." }, { "left": "issue:42915", "right": "issue:44756", "accept": false, - "reason": "Different subsystems and symptoms: FP8 config failure vs mmap/OOM workaround." + "reason": "Unrelated issues: Qwen3 MoE FP8 failure vs hardware-specific mmap OOM." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "LoRA checkpoint compatibility across hf4.x/hf5.x is unrelated to the resize_token_embeddings bug." + "reason": "Different code paths and symptoms: LoRA model incompatibility across versions vs token embedding resize propagation." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Different model families and different flash-attention failures; not the same underlying bug." + "reason": "Different models and failures; both mention generation/flash-attention, but not the same underlying bug." }, { "left": "issue:44451", "right": "issue:44488", - "accept": true, - "reason": "Both report the same current-version tokenizer/model loading regression affecting different BERT repos." + "accept": false, + "reason": "Both are loading failures for different BERT checkpoints, but the reports do not clearly point to the same root cause." }, { "left": "issue:44863", "right": "issue:46846", "accept": false, - "reason": "Checkpoint loading failure for NemotronH is not the same as the missing base_model_tp_plan entries causing TP>1 errors." + "reason": "Different model families and distinct failure modes: checkpoint loading vs TP plan missing entries." }, { "left": "issue:46459", "right": "issue:46512", "accept": false, - "reason": "Loading-time dtype handling vs stale config.dtype after from_config are related but distinct bugs." + "reason": "Both involve dtype handling, but one is checkpoint loading and the other is from_config initialization; not the same bug." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "Different code paths: DeepSpeed ZeRO-3 weight loading vs save_pretrained writing wrong visual encoder keys." + "reason": "Both are Qwen3.5 regressions, but one is ZeRO-3 weight loading and the other is save_pretrained key handling." }, { "left": "issue:45507", "right": "issue:46468", "accept": false, - "reason": "Invalid method call in GraniteMoEHybrid is unrelated to the broken Nemotron example." + "reason": "Different models and issues; no concrete shared bug path." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "Non-persistent buffer junk is unrelated to Jupyter notebook custom-model initialization breakage." + "reason": "Both are v5 regressions, but buffer corruption and Jupyter custom-model initialization are separate problems." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Documentation/example errors are not the same as the actual AutoModel/pipeline nonsensical-results bug." + "reason": "Related SigLIP2 area, but doc-example mistakes vs bad model outputs are not the same bug." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Qwen3 MoE accelerate+deepspeed weight conversion is a different issue from GLM-4.7-Flash unit-test loading." + "reason": "Different load/dispatch problems affecting different models and setups." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-loading problems, but the reported failures differ and there is no evidence they are the same bug." + "reason": "Both are tokenizer/model loading regressions, but they affect different checkpoints and do not clearly share one root cause." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Training-memory inefficiency in Qwen3 MoE is not the same as quantization offloading not working." + "reason": "Different memory issues: MoE training inefficiency vs quantization offloading behavior." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "Checkpoint compatibility for qwen3_moe LoRA is unrelated to resize_token_embeddings not updating output_embeddings." + "reason": "Unrelated bugs: LoRA model compatibility vs resize_token_embeddings not updating output embeddings." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "GLM-4.7-Flash test loading and Qwen3-VL-30B-A3B weight-shape mismatch are different model-loading failures." + "reason": "Different model-specific loading failures with different symptoms and likely causes." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "General MoE training memory usage and int4 quantization reserved-memory OOM are distinct regressions." + "reason": "Both are memory-related, but one is training inefficiency and the other is int4 reserved-memory/OOM regression." }, { "left": "issue:42491", "right": "issue:45335", "accept": false, - "reason": "LoRA/hf-version incompatibility is unrelated to the embedding-resize propagation bug." + "reason": "Different models and separate resize/load issues; not the same code-path bug." }, { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Trainer resume/max_steps calculation is unrelated to SGD arguments being ignored." + "reason": "Training resumption max_steps math is unrelated to optimizer argument handling." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "Flash-attention generation failure is unrelated to save/load breakage after sinq quantization." + "reason": "Different quantization/generation issues with no clear shared root cause." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "extra_state save/load handling is different from init_empty_weights passing an unexpected _is_hf_initialized argument." + "reason": "Both concern loading/saving internals, but extra_state handling and _is_hf_initialized are separate bugs." }, { "left": "issue:46897", "right": "issue:46901", - "accept": false, - "reason": "Similar training-loss symptom, but different model implementations and no evidence of a shared underlying loss-code bug." + "accept": true, + "reason": "Same training-loss double-shift bug: both report labels being shifted incorrectly instead of training against the intended targets." }, { "left": "issue:45276", "right": "issue:45335", "accept": true, - "reason": "Same resize_token_embeddings regression: embeddings/output_embeddings are not updated consistently; 45335 is the narrower T5Gemma report." + "reason": "Same resize_token_embeddings propagation bug; output embeddings/decoder embeddings are not updated across model variants." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Tokenizer pre-tokenizer override in LlamaTokenizer is a different bug from Granite AutoTokenizer producing wrong token IDs." + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Qwen3.5 chat crashing and Gemma4Processor missing _tokenizer are separate failures." + "reason": "Different issues: chat generation crash for Qwen3.5-35B vs Gemma4Processor missing _tokenizer in serve." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "FlashAttention illegal memory access from 3D position_ids is unrelated to the s_aux=None AttributeError." + "reason": "Both are flash-attention crashes, but one is packed-sequence misinterpretation and the other is s_aux=None handling." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Tokenizer vocab-size mismatch is a different issue from model loading failure." + "reason": "Token-vocab mismatch vs model loading failure; similar area, but not the same bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analyst\u2019s summary is grounded in the packet and stays conservative overall. The accepted pairs are the ones with the strongest title-level duplicate signals: the non-persistent buffer junk reports, the tokenizer-loading regressions on similar BERT repos, and the resize_token_embeddings/output_embeddings regression. The rejections are also appropriately cautious, with no obvious overreach into unrelated failures." + "feedback": "The summary is grounded in the packet, and the accepted soft-edge pairs are supported by very close title-level overlap. The rejections are generally conservative and do not overstate duplicate evidence." }, "evaluator_used": true, "retried": false, @@ -2644,226 +2719,226 @@ "issue:44589|issue:44756", "issue:44402|issue:44451" ], - "prepared_review_unit_hash": "e550ffd530ef60a1e36246189b15027dc965a84c2aa004266f875a2778de27f5", + "prepared_review_unit_hash": "1870a70a5a9ad5d2fc06a758b76aa172d80f194261f7d79df5d3eb6861e9ec52", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36445, - "estimated_input_tokens": 9112, - "estimated_eval_tokens": 18480 + "serialized_chars": 36443, + "estimated_input_tokens": 9111, + "estimated_eval_tokens": 18478 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are clearly unrelated. The only duplicate chain I\u2019d merge is the resize_token_embeddings regression across model variants: 45276 \u2194 45292 and 45292 \u2194 45335. No PRs are present.", + "summary": "This cluster is mostly a set of unrelated Transformers bug reports. The only clear duplicate-like pair is the resize_token_embeddings regression, where the Gemma4-specific report appears to be a narrower manifestation of the broader output_embeddings issue.", "confidence": 0.86, - "canonical_issue_reason": "issue:45292 is the broad, model-agnostic report of the resize_token_embeddings regression; the Gemma4 and t5gemma issues look like specific manifestations of the same underlying bug.", + "canonical_issue_reason": "issue:45292 is the broadest, model-agnostic report of resize_token_embeddings not updating output_embeddings, and it cleanly subsumes the Gemma4-specific variant in issue:45276.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 best represents the shared underlying problem and is the most general title to anchor the duplicate cluster.", + "best_issue_reason": "issue:45292 is the strongest global issue candidate because it describes a reusable API-level regression with broader impact than the model-specific duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Different failures: Torch version compatibility vs meta-device loading breakage." + "reason": "Both are loading/runtime failures, but one is a Torch 2.10 incompatibility and the other is a meta-device model-loading regression; different bugs." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Unrelated models and symptoms; docs/example errors vs a GLM5 loading issue." + "reason": "SigLIP2 doc/example errors and a GLM5 issue are unrelated model-specific problems." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Different subsystems and bugs: device-map CUDA assert vs AudioFlamingo batch leakage." + "reason": "Qwen3 auto device-map CUDA asserts and AudioFlamingo3 batched inference token leaks are different failure modes." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "init_empty_weights argument handling is unrelated to mmap/OOM on Strix Halo." + "reason": "init_empty_weights/_is_hf_initialized and Strix Halo mmap OOM are unrelated loading bugs." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Different code paths: Gemma4 FSDP attention state vs Qwen multimodal config unwrapping." + "reason": "Gemma4 FSDP shared_kv_states and Qwen multimodal text_config unwrapping are different code paths." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing/LoRA is unrelated to T5 apex RMSNorm memory leak." + "reason": "Gradient checkpointing with LoRA and a T5 apex RMSNorm memory leak are unrelated." }, { "left": "issue:43404", "right": "issue:43454", "accept": false, - "reason": "Both mention lm_head tying, but they target different model implementations and likely separate fixes." + "reason": "Both mention lm_head tying, but they affect different model implementations and don\u2019t look like the same concrete bug." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Both involve quantization/memory, but one is offloading behavior and the other is int4 reserved-memory OOM." + "reason": "Both involve quantization/OOM, but one is offloading behavior and the other is increased CUDA reserved memory under int4; too broad to merge." }, { "left": "issue:44079", "right": "issue:46829", "accept": false, - "reason": "General ModelOutput key assignment bug vs a VideoPrism output-field mixup." + "reason": "Generic ModelOutput key handling is not the same bug as VideoPrism returning the wrong tensor in hidden_states." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Different models and failures: Gemma4 FSDP KeyError vs Qwen3_5Moe CPU-offload incompatibility." + "reason": "Different model families and different runtime paths: FSDP shared_kv_states vs CPU offload rotary embedding." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization/loading problems with no clear same code path." + "reason": "init_empty_weights argument handling and MXFP4 quantization fallback are unrelated." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "Import path regression vs Whisper pipeline return_language regression." + "reason": "GenerationMixin import failure and Whisper return_language pipeline regression are distinct issues." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Both mention gradient checkpointing, but one is ineffective with LoRA and the other is eval-mode incompatibility." + "reason": "Both mention gradient checkpointing, but one is ineffective LoRA behavior and the other is eval-mode prohibition." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "AudioFlamingo batching bug and Gemma4 device-map mismatch are unrelated." + "reason": "AudioFlamingo3 batch leakage and Gemma4 device mismatch are different bugs." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading breakage is unrelated to disabling mmap for OOM." + "reason": "Meta-device loading regressions and Strix Halo mmap OOM are not the same change." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Quantization OOM and Strix Halo mmap OOM are different symptoms and fixes." + "reason": "Both can lead to OOM, but the causes are different: int4 reserved memory growth vs mmap on Strix Halo." }, { "left": "issue:43335", "right": "issue:45663", "accept": false, - "reason": "SwitchTransformers sparse-layer bug is unrelated to Gemma4 FSDP attention state." + "reason": "SwitchTransformers sparse-layer construction and Gemma4 FSDP attention state rebuilding are unrelated." }, { "left": "issue:44863", "right": "issue:46468", "accept": false, - "reason": "NemotronH checkpoint loading failure vs broken official Nemotron example are not the same bug." + "reason": "NemotronH checkpoint loading and a broken official Nemotron example are too different to treat as duplicates." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same resize_token_embeddings regression; 45276 is a Gemma4-specific manifestation of the broader output-embedding update bug." + "reason": "Same underlying resize_token_embeddings bug: the Gemma4 report is a model-specific instance of the broader output_embeddings failure." }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Video-input regression for Qwen VL models vs strict-config load failure for granite_speech." + "reason": "Video input regression for Qwen VL models and strict granite_speech config loading are unrelated." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Gradient checkpointing eval-mode bug is unrelated to the Qwen3.5 RLHF SDPA fallback/NaN issue." + "reason": "Eval-mode gradient checkpointing and Qwen3.5 NaN/SDPA fallback are different problems." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "save/from_pretrained extra_state handling is unrelated to mmap/OOM." + "reason": "save/from_pretrained extra_state handling is unrelated to the Strix Halo mmap OOM issue." }, { "left": "issue:42796", "right": "issue:46326", "accept": false, - "reason": "Both touch mistral_common tokenizers, but one is BPE-marker handling and the other is continue_final_message behavior." + "reason": "Both concern tokenizers, but one is BPE-marker handling and the other is continue_final_message behavior; not the same bug." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "flash_attention s_aux=None crash is unrelated to Gemma4 FSDP KeyError." + "reason": "Flash-attention sink-less crash and Gemma4 FSDP KeyError are unrelated." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Custom-model file permission propagation and wrong-module loading are different save/load bugs." + "reason": "Custom-model save permissions and wrong-module loading after save_pretrained are different issues in the same area, not the same bug." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading failure is not the same as Qwen3.5 save_pretrained regression." + "reason": "GLM-4.7-Flash loading in tests and Qwen3.5 save_pretrained regression are unrelated." }, { "left": "issue:45292", "right": "issue:45335", - "accept": true, - "reason": "Same underlying resize_token_embeddings bug: the embedding/output weights are not updated in the model-specific decoder path." + "accept": false, + "reason": "Very similar theme, but the titles suggest different affected embeddings and no evidence they\u2019re the same exact regression beyond resize_token_embeddings." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Jupyter custom-model initialization issue is separate from extra_state save/load handling." + "reason": "Custom model initialization in Jupyter and extra_state save/load handling are different failure modes." }, { "left": "issue:42175", "right": "issue:43957", "accept": false, - "reason": "Missing TensorFlow backend in a pip extra is unrelated to meta-device loading failures." + "reason": "Packaging/backend dependency omission and meta-device loading bugs are unrelated." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup error and Strix Halo mmap OOM are unrelated." + "reason": "Float8 storage lookup failure and Strix Halo mmap OOM are not the same code-path problem." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Tokenizer vocab-size mismatch is a different issue from loading failure for vesteinn/ScandiBERT." + "reason": "Tokenizer vocab mismatch and a model load failure are too general to call duplicates." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative. The only accepted duplicate chain is well supported by the titles: 45276 \u2194 45292 \u2194 45335 around resize_token_embeddings/output-embedding updates, and the summary correctly notes there are no PRs." + "feedback": "Grounded and conservative overall. The only accepted duplicate pair, issue:45276 vs issue:45292, is supported by the shared resize_token_embeddings/output_embeddings regression, and the rest of the soft-edge rejections avoid overclaiming duplicate status." }, "evaluator_used": true, "retried": false, @@ -2967,20 +3042,20 @@ "issue:43299|issue:43931", "issue:43596|issue:44683" ], - "prepared_review_unit_hash": "23cd4e891ab81a0cbf4a73468d43b9aecc04b6ea8f49aa35c530a02c1a093599", + "prepared_review_unit_hash": "f8d75a346dffa3b207fa8b5928735c84c7615e13b4eb583eb5437b18311eed71", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37541, + "serialized_chars": 37544, "estimated_input_tokens": 9386, "estimated_eval_tokens": 19028 }, @@ -2989,214 +3064,205 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous regression/reporting issues. Only one pair looks like the same underlying bug pattern: training-loss double-shift in two model heads.", - "confidence": 0.76, - "canonical_issue_reason": "issue:46894 is the best representative: it states the underlying training-loss double-shift pattern clearly and already frames itself as the same bug class seen elsewhere.", + "summary": "These items are mostly unrelated regression reports across tokenizer loading, multimodal inference, training, quantization, and docs. I don't see a single duplicate cluster or any soft edge that is strong enough to merge as the same underlying bug/change.", + "confidence": 0.94, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:46894 is the strongest cluster anchor because it describes the shared loss-label shifting bug in a generalizable way, not just one model instance.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Both are v5 regression reports, but one is custom model initialization in notebooks and the other is Trainer checkpoint resumption/max_steps calculation; different code paths and fixes." + "reason": "Different failure modes: custom model initialization in notebooks vs Trainer checkpoint-resume max_steps calculation." }, { "left": "issue:46894", "right": "issue:46901", - "accept": true, - "reason": "Both describe the same concrete bug pattern: training loss is shifted incorrectly/doubly shifted, causing models to learn against the wrong labels. These could plausibly be fixed by one shared loss-path change." + "accept": false, + "reason": "Same broad label of 'double-shift', but they affect different models and there is no evidence of one concrete shared code path." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch version incompatibility and autocast dtype mismatch are different failures with different triggers and fixes." + "reason": "Torch version incompatibility and autocast dtype mismatch are distinct problems with different triggers." }, { "left": "issue:46315", "right": "issue:46491", "accept": false, - "reason": "One is tokenizer conversion slowdown; the other is a CodeLlama encode/decode space regression. Different symptoms and code paths." + "reason": "One is a performance regression in token ID lookup; the other is a CodeLlama encode/decode spacing bug." }, { "left": "issue:45812", "right": "issue:46315", "accept": false, - "reason": "Wrong token IDs for Granite models and slower convert_tokens_to_ids are separate tokenizer regressions, not the same bug." + "reason": "Wrong Granite token IDs is a correctness regression, not the same issue as slower convert_tokens_to_ids." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper pipeline return_language breakage is unrelated to the broader tokenization changes reported in the other issue." + "reason": "Whisper return_language pipeline behavior and generic tokenization changes are not the same bug." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and template compilation failure are unrelated backend/compiler errors." + "reason": "Qwen3.5 cache group-type crash and template-node compilation error are unrelated." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "Both involve v5 model loading/custom models, but one is base_model_prefix handling and the other is notebook-defined model initialization. Not the same defect." + "reason": "Both are v5 loading problems, but one is base_model_prefix handling and the other is custom model initialization in notebooks." }, { "left": "issue:43425", "right": "issue:46869", "accept": false, - "reason": "Torch 2.10 incompatibility is not the same as YOLO device-map multi-GPU failure." + "reason": "Generic torch incompatibility is not the same as the multi-GPU device_map auto failure." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 PLE device mismatch are distinct issues." + "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 PLE device mismatch are different code paths." }, { "left": "issue:39404", "right": "issue:46665", "accept": false, - "reason": "Whisper pipeline return_language and from_pretrained inside Python classes do not share the same underlying bug." + "reason": "Whisper pipeline return_language and from_pretrained inside Python classes do not describe the same defect." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "Torch compatibility and Mistral-3 dtype mismatch are separate model/runtime problems." - }, - { - "left": "issue:43425", - "right": "issue:45823", - "accept": false, - "reason": "Generic torch incompatibility is not the same as a device_map auto forward mismatch." + "reason": "Torch 2.10 incompatibility is unrelated to the Mistral-3 image preprocessor/model dtype mismatch." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage lookup failure and template compilation errors are unrelated." + "reason": "Float8 storage lookup and non-template-node compilation are separate errors." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Video-input regression for several VL models and a mask shape IndexError are different concrete failures." + "reason": "One is a Qwen VL video-input regression; the other is a mask/tensor shape IndexError." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Both are documentation issues, but they cover different pipeline docs and different deprecations; not a single duplicate bug." + "reason": "Both are docs issues, but they cover different pipeline/task deprecations and should not be merged." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "Wrong outputs for SigLIP and ModelOutput key assignment are unrelated." + "reason": "Wrong SigLIP2 outputs and ModelOutput key assignment are different bugs." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "A reproduction of an earlier issue and a PagedAttentionCache crash are not the same bug." + "reason": "A reproduction of issue #38071 is not the same as the Qwen3.5 linear_attention cache crash." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export with torch_compilable_check and Perceiver non-default resolution failures are different model/export bugs." + "reason": "torch.export with torch_compilable_check and Perceiver non-default-resolution failure are unrelated." }, { "left": "issue:44164", "right": "issue:45759", "accept": false, - "reason": "extra_state save/load handling and composite Qwen config unwrapping are separate loader/saver issues." + "reason": "save/from_pretrained extra_state handling and AutoModelForCausalLM.from_config text_config unwrapping are different loading bugs." }, { "left": "issue:36032", "right": "issue:43452", "accept": false, - "reason": "T5 tokenizer special-tokens conflict and gguf AutoTokenizer/AutoModel loading breakage are unrelated." + "reason": "T5Tokenizer add_special_tokens conflict and gguf_file loading failure are distinct tokenizer/load paths." }, { "left": "issue:46050", "right": "issue:46326", "accept": false, - "reason": "Quantized save/load corruption and mistral_common continue_final_message behavior are different problems." + "reason": "Quantized save/load breakage and mistral_common continue_final_message behavior are unrelated." }, { "left": "issue:43906", "right": "issue:46326", "accept": false, - "reason": "An isolated reproduction of a prior issue is not the same as the mistral_common prompt-handling bug." + "reason": "A reproduction of an earlier issue does not match the mistral_common conversation bug." }, { "left": "issue:39404", "right": "issue:44912", "accept": false, - "reason": "Whisper return_language and MXFP4 quantization fallback are unrelated." + "reason": "Whisper return_language and MXFP4 quantization fallback are different subsystems." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "CamemBERT masked-LM prediction regression and tokenizer loading failure for est-roberta are different defects." + "reason": "Both involve tokenizer/model regressions, but CamemBERT masked LM predictions and est-roberta tokenizer loading are different concrete failures." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Two tokenizer-loading problems, but with different models and failure modes; not the same underlying bug." + "reason": "Tokenizer loading for est-roberta and Mistral regex patch crashes are not the same issue." }, { "left": "issue:45823", "right": "issue:46899", "accept": false, - "reason": "Device-map forward mismatch and 4-bit multimodal embedder dtype-cast issues are distinct." + "reason": "Gemma4 device placement mismatch and Gemma4 4-bit multimodal dtype-cast breakage are different bugs." }, { "left": "issue:45479", "right": "issue:46894", "accept": false, - "reason": "Degenerate zero-loss with num_labels=1 is a different training bug than the double-shift loss issue." + "reason": "Sequence-classification zero-loss behavior and ASR training-loss double-shift are different label-handling issues." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE memory inefficiency and Strix Halo mmap OOM avoidance are unrelated performance/memory issues." + "reason": "MoE training memory inefficiency and Strix Halo mmap OOM avoidance are unrelated." }, { "left": "issue:44164", "right": "issue:46821", "accept": false, - "reason": "extra_state save/load and TimesFM window_size attribute error are unrelated." + "reason": "save/from_pretrained extra_state handling and TimesFM window_size AttributeError are different problems." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Both concern Qwen3-VL loading, but one is a version-regression breaking Moe model loading and the other is a shape mismatch for a specific checkpoint. Not the same bug." + "reason": "Both are Qwen3-VL loading reports, but one is a dev0 breakage and the other is a weight-shape mismatch; not enough to treat as one bug." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "Deepspeed zero3 index error and compiled flex_attention torch>=2.9 failure are different runtime issues." + "reason": "Deepspeed ZeRO3/BertModel index error and compiled flex_attention torch>=2.9 failure are unrelated." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "Grounded overall. The summary stays conservative and matches the packet, and the only accepted soft-edge pair (issue:46894 / issue:46901) is reasonably supported by the titles as the same training-loss label-shift bug class. The rejected pairs also read 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 }, @@ -3294,223 +3360,217 @@ "issue:45084|issue:45507", "issue:43317|issue:43482" ], - "prepared_review_unit_hash": "579ab3f1a8ebf2e8b0bf119ccb7e2e025506ea739588266decc05f2bc5dd08b8", + "prepared_review_unit_hash": "0815f0a18d996f82438b34561c5c12dc7cfec27c3b2377b920d413f57889b2c8", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36743, - "estimated_input_tokens": 9186, - "estimated_eval_tokens": 18628 + "serialized_chars": 36746, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is mostly a set of adjacent Transformers v5 regression reports, but most pairs are only thematic, not true duplicates. The cluster still centers on custom-model initialization regressions, with one likely audio feature-extraction duplicate; the 43644/43646 pair should be treated as related but not merged as the same bug.", - "confidence": 0.85, - "canonical_issue_reason": "issue:43646 is still the broadest, most central formulation in the custom-model initialization family, so it works best as the cluster representative without overcommitting to narrower symptom overlap.", + "summary": "Mostly unrelated Transformers regression reports across tokenizer loading, model initialization, backend setup, device/offload, and export paths. Only the audio feature-extraction pair looks like the same underlying bug; the rest are too different to merge as duplicates.", + "confidence": 0.64, + "canonical_issue_reason": "Issue 44206 is the best canonical anchor: it describes a concrete, localized regression in the audio feature-extraction path and is the only pair with a plausible same-codepath duplicate (44220).", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best single representative because it describes the general custom-model initialization regression more abstractly than the nearby symptom reports.", + "best_issue_reason": "44206 is the clearest and most actionable report in the set, with a specific failure mode and code path; the other issues are broader or clearly about different subsystems.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43632", "right": "issue:47050", "accept": false, - "reason": "Different subsystems and failure modes: HF init flag regression vs ONNX attention dtype mismatch." + "reason": "Both are regressions, but one is about hf-initialization state and the other is an fp16 ONNX attention dtype mismatch; different bugs and code paths." }, { "left": "issue:45479", "right": "issue:46897", "accept": false, - "reason": "Both are training-loss bugs, but the model types and failure patterns are too different to treat as one concrete bug." + "reason": "Both involve training loss behavior, but the symptoms and model families differ; not enough evidence they share the same fix." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Both are load failures, but they point to different models and likely different loader/tokenizer regressions." + "reason": "Both are version-related loading failures, but they affect different models and likely different tokenizer/config paths." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Related area, but one is a generic model load failure and the other is a specific Mistral regex/backend_tokenizer crash." + "reason": "One is about ignoring tokenizer.json, the other is a Mistral regex patch crash; distinct failure modes." }, { "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; not the same bug." + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer state; not the same bug." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Completely different components: T5 tokenizer class conflict vs Qwen3Moe FP8 config failure." + "reason": "Tokenizer method-name conflict vs FineGrainedFP8Config failure; unrelated code paths." }, { "left": "issue:42796", "right": "issue:43404", "accept": false, - "reason": "Both involve Mistral-family models, but one is tokenizer BPE marker handling and the other is lm_head tying in a conditional generation model." + "reason": "Tokenizer BPE-marker handling and lm_head weight tying are separate issues in different model code." }, { "left": "issue:44462", "right": "issue:46489", "accept": false, - "reason": "Both are tokenizer-loading related, but the titles describe different breakages and no clear shared code-path." + "reason": "Both are tokenizer regressions, but one is repo tokenizer resolution and the other is DeepSeek-Coder tokenization behavior; different root causes." }, { "left": "issue:46285", "right": "issue:46402", "accept": false, - "reason": "Both concern model structure/weights, but they affect different architectures and different failure points." + "reason": "CLIP structure inconsistency and Sam3 text-encoder weight loading are unrelated model-architecture bugs." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Image preprocessor dtype mismatch is unrelated to the template-node compilation error." + "reason": "Image preprocessor dtype mismatch vs template compilation error; no shared code path." }, { "left": "issue:39404", "right": "issue:44671", "accept": false, - "reason": "Whisper pipeline language handling and CamemBERT MLM prediction issues are unrelated." + "reason": "Whisper pipeline return_language regression and CamemBERT MLM prediction error are unrelated." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Tokenizer mapping logic and torch 2.9 flex_attention compilation are unrelated." + "reason": "Tokenizer-mapping bookkeeping bug vs torch flex_attention compilation failure; different layers entirely." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "AutoConfig trust_remote_code field loss is unrelated to Qwen3.5 PagedAttentionCache group-type failures." + "reason": "Missing fields from trust_remote_code config vs PagedAttentionCache invalid group type; unrelated behavior." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "Both mention Qwen models, but FP8 config failure and PagedAttentionCache invalid group type are different code paths." + "reason": "FineGrainedFP8Config failure and Qwen3.5 PagedAttentionCache crash are different model/runtime issues." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Both are loading failures, but they target different models and likely different loader regressions." + "reason": "Both concern loading failures, but one is a specific model load regression and the other is tokenizer loading on a different model." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both involve device placement/offload, but one is a CPU arange placement bug and the other is a rotary embedding offload incompatibility." + "reason": "Both are device/offload related, but the models and failure mechanisms differ (CPU arange placement vs rotary embedding offload compatibility)." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Torch dynamo grouped_mm tracing failure is unrelated to the template-node compile error." + "reason": "Torch dynamo grouped_mm tracing failure is unrelated to the template-node compilation error." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity regression on a model is unrelated to tokenizer post-processor update behavior." + "reason": "High perplexity on a benchmark and runtime special-token post-processor updates are not the same bug." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "Trainer token accounting and multi-image iterable dataset failure are different issues." + "reason": "Trainer token counter not updating vs multimodal iterable-dataset finetuning failure; different subsystems." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "These are related v5 initialization-family reports, but 43644's non-persistent-buffer corruption is not clearly the same concrete bug as 43646's custom model initialization failure." + "reason": "Both are v5 breakages around model initialization, but one is buffer persistence/state handling and the other is custom-model init in notebooks; not the same concrete issue." }, { "left": "issue:42175", "right": "issue:46178", "accept": false, - "reason": "Backend installation/availability problems are similar in theme but not the same concrete bug." + "reason": "pip extra/backend packaging omission vs backend-mapping validation error; related theme but distinct bugs." }, { "left": "issue:40444", "right": "issue:46326", "accept": false, - "reason": "Multi-image VL training failure and mistral_common continue_final_message behavior are unrelated." + "reason": "IterableDataset multi-image finetuning failure and mistral_common continue_final_message behavior are unrelated." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError and _is_hf_initialized flag regression are unrelated." + "reason": "Apple Silicon allocator warmup TypeError and hf-initialization flag regression are separate problems." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "Tokenizer class method conflict and notebook custom model initialization are unrelated." + "reason": "Tokenizer add_special_tokens conflict vs custom model initialization in Jupyter; different failure modes." }, { "left": "issue:44206", "right": "issue:44220", "accept": true, - "reason": "These appear to describe the same audio feature-extraction/fbank regression, with one being the wrapper symptom and the other the underlying helper failure." - }, - { - "left": "issue:44811", - "right": "issue:45584", - "accept": false, - "reason": "Both are Whisper tokenizer/generation issues, but the reported failures are different enough to avoid merging as one bug." + "reason": "Both point to the same audio feature-extraction path; the titles suggest a shared `_torch_extract_fbank_features` regression, likely triggered by the unsupported `center` argument." }, { "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 model failure without clear proof of the same root cause." + "reason": "GGUF loading failures and v5 tokenizer-loading issues are separate model-format/code-path problems." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are v5 loading failures, but they affect different models and likely different tokenizer paths." + "reason": "Both are version-gated loading regressions, but they affect different models and likely different tokenizer/config logic." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Template compilation failure is unrelated to GraniteMoEHybrid calling an invalid method." + "reason": "Template compilation failure vs GraniteMoEHybrid invalid method call are unrelated." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "Dequantized gpu/cpu offload loading and Qwen2.5-GGUF load failure are not clearly the same bug." + "reason": "Dequantized model offload loading failure and Qwen2.5-GGUF v5 loading failure are distinct." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The overall clustering summary is plausible, but one soft-edge verdict is too strong for the evidence in the packet: issue:43644 vs issue:43646 is only supported as a broad v5-init family relation, not clearly the same concrete bug. The 43644 title (\u2018non-persistent buffers with junk\u2019) does not by itself justify \u2018same regression\u2019 / narrower-symptom duplicate treatment. The rest of the judgments are reasonably conservative, but this accepted pair makes the output insufficiently cautious." + "accept": true, + "feedback": "The assessment is grounded in the packet and stays conservative. The only accepted soft edge, 44206 \u2194 44220, is plausibly the same audio feature-extraction regression based on the titles and shared codepath. The rest of the pair verdicts are appropriately cautious and do not overclaim duplication across different subsystems." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -3608,208 +3668,208 @@ "issue:44479|issue:45325", "issue:43089|issue:45987" ], - "prepared_review_unit_hash": "735b4f46045111b486647a623abcdc57ccbe5973da7a388a28d5f8bca436bbdd", + "prepared_review_unit_hash": "f949ffcdf74203ba29d9a66ea152ac758735fce17b3f1b51596944ff43f0ed3c", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36612, - "estimated_input_tokens": 9153, - "estimated_eval_tokens": 18562 + "serialized_chars": 36613, + "estimated_input_tokens": 9154, + "estimated_eval_tokens": 18564 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose grouping of mostly unrelated Transformers bugs, with a few theme-adjacent v5/model-loading regressions but no true duplicate set. The closest recurring theme is custom model initialization/loading breakage in v5, but the individual reports still split across distinct code paths.", - "confidence": 0.38, - "canonical_issue_reason": "issue:43646 is the broadest representative of the repeated v5 custom-model initialization/load regressions in this set; it overlaps thematically with several nearby reports, though it is not a true duplicate canonical for the whole cluster.", + "summary": "This cluster is mostly heterogeneous: the soft pairs share broad subsystem keywords, but they do not appear to be the same concrete bug/change. No soft edge should be merged; the closest mini-cluster is the v5 custom-model initialization regressions, with issue:43646 as the best representative.", + "confidence": 0.72, + "canonical_issue_reason": "issue:43646 is the broadest representative of the only coherent mini-cluster here (Transformers v5 custom model initialization regressions), and it is more canonical than the narrower variants around `_is_hf_initialized` or `init_empty_weights`.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best thematic representative of the cluster because it captures the most general and reusable failure mode (v5 custom model initialization breakage) among otherwise heterogeneous issues.", + "best_issue_reason": "issue:43646 is the best overall issue representative in this set because it captures the broad v5 custom-model initialization breakage, while most other issues are model-specific or unrelated. It is still not a duplicate of the rest of the cluster, just the strongest anchor among the issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43746", "right": "issue:46326", "accept": false, - "reason": "Different subsystems: GraniteSpeech PEFT/local checkpoint loading vs Mistral-common chat templating." + "reason": "PEFT local-checkpoint loading and `continue_final_message` for mistral_common are different code paths and different failure modes." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Both are audio-related, but one is a sampling-rate default bug and the other is a padding-mask batching mismatch." + "reason": "Audio processor sampling-rate default vs MIMI batched-vs-single padding-mask mismatch are unrelated bugs." }, { "left": "issue:43208", "right": "issue:47013", "accept": false, - "reason": "Both mention xLSTM training, but one is a multi-bug umbrella report and the other is a specific unpacking crash." + "reason": "Both are xLSTM-related, but one is a bundle of training bugs for small models and the other is a specific unpacking crash in train mode; not the same bug." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Unnecessary loader noise is unrelated to the Mistral regex patch crash." + "reason": "Loading-noise behavior and the Mistral regex patch crash are unrelated issues." }, { "left": "issue:43792", "right": "issue:46869", "accept": false, - "reason": "Whisper runtime failure and YOLO multi-GPU device-map failure are unrelated model issues." + "reason": "Whisper inference failure and YOLO multi-GPU device-map failure are different models and different code paths." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Both are v5 regressions, but custom model init and non-persistent buffer serialization are different bugs." + "reason": "Custom model initialization regressions and non-persistent buffer junk are different v5 problems, not one concrete bug." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "Qwen3Moe FP8 loading failure and template compilation error are different code paths." + "reason": "Qwen3Moe FP8 config failure and template-node compilation failure are unrelated." }, { "left": "issue:43645", "right": "issue:46661", "accept": false, - "reason": "Both involve model initialization, but Jupyter custom-model init and remote-code config_class failure are distinct." + "reason": "Both concern custom models, but one is notebook re-initialization and the other is remote-code `config_class` handling; different underlying failures." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "MIMI padding-mask mismatch and DAC from_latents/STE mismatch are different audio-model bugs." + "reason": "MIMI padding-mask batching bug and DAC `from_latents`/forward mismatch are different audio-model bugs." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Different Qwen vision-language bugs: hidden_states output vs rope-index temporal position scaling." + "reason": "Hidden-states output bug in Qwen3.5 and Qwen2.5-VL rope/temporal scaling bug are unrelated." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Molmo tied-weights attribute error and loading cjvt/sleng-bert are unrelated." + "reason": "Missing `all_tied_weights_keys` on Molmo and loading `cjvt/sleng-bert` are unrelated model-loading problems." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Generic model load failure and Mistral tokenizer regex crash do not share the same root cause." + "reason": "General load failure for one model and Mistral tokenizer regex patch crash are different root causes." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Both concern lm_head/tied weights, but one is model-specific weight tying and the other is serialization inconsistency by device." + "reason": "Tied-weight absence in Mistral3 and device-dependent serialization inconsistency both mention `lm_head.weight`, but the failure modes are not the same concrete bug." }, { "left": "issue:28530", "right": "issue:34567", "accept": false, - "reason": "Trainer early-stopping metric handling and num_input_tokens_seen tracking are unrelated TrainerState issues." + "reason": "Early-stopping metric selection and `num_input_tokens_seen` not updating are unrelated Trainer issues." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "SigLIP2 AutoModel/pipeline behavior and OmDet-Turbo processor size mismatch are separate problems." + "reason": "Siglip2 nonsense outputs and OmDet-Turbo processor size mismatch are different model/processor bugs." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "Qwen3Moe FP8 config loading and init_empty_weights _is_hf_initialized argument handling are different initialization failures." + "reason": "FP8 config failure for Qwen3Moe and unexpected `_is_hf_initialized` arg in `init_empty_weights` are unrelated." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "T5 tokenizer special-token method conflict and MobileLLM key error are unrelated loading bugs." + "reason": "T5 tokenizer method-name conflict and MobileLLM key error are unrelated loading issues." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "ScandiBERT loading failure is not the same as the Mistral regex crash." + "reason": "ScandiBERT loading failure and Mistral regex patch crash are different models and different causes." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Compiled forward generation inconsistency and StaticCache length return-type bug are different code paths." + "reason": "Compiled chunked generation inconsistency and `StaticCache.get_seq_length()` return-type bug are related to generation APIs but not the same code-path defect." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "AddedToken keyword collision and generic model loading failure are unrelated." + "reason": "`AddedToken` argument collision and loading `physical-intelligence/fast` are unrelated." }, { "left": "issue:44008", "right": "issue:46899", "accept": false, - "reason": "Gemma3n variable-name collision and Gemma4 unsafe dtype casting are different multimodal bugs." + "reason": "Gemma 3n variable-name collision and Gemma4 quantization dtype-cast bug are different multimodal model issues." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "FP8 config loading for Qwen3Moe and Float8 storage lookup failure are unrelated." + "reason": "Qwen3Moe FP8 config failure and missing Float8 storage object are unrelated precision/storage issues." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export export failure and OmDet-Turbo processor resolution mismatch are not the same issue." + "reason": "`torch.export` failing on `torch_compilable_check` and a processor/model input-size mismatch are unrelated." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Llava Next segfault and Qwen3OmniMoe video-input ValueError are different multimodal loader/runtime bugs." + "reason": "Llava-Next segfault on load and Qwen3OmniMoe video-processing `ValueError` are different multimodal failures." }, { "left": "issue:43632", "right": "issue:46821", "accept": false, - "reason": "_is_hf_initialized regression and TimesFM window_size AttributeError are unrelated." + "reason": "`_is_hf_initialized` breakage and TimesFM `window_size` attribute error are unrelated regressions." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and granite_speech strict config loading are different bugs." + "reason": "Qwen3-MoE sliding-window behavior and granite_speech strict config loading are different bugs." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "Tokenizer special-token conflict and Qwen3 auto device-map cuda assert are unrelated." + "reason": "Tokenizer special-token conflict and Qwen3 auto device-map CUDA assert are unrelated." }, { "left": "issue:44479", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen VL temporal-position regressions, but one is a broad video-input regression and the other is a specific still-image tokens_per_second scaling bug; too different to merge confidently." + "reason": "Both touch Qwen2.5-VL/video-position logic, but one is a broad v5.3.0 video-input regression and the other is a specific still-image temporal-position scaling bug; not enough to treat as the same bug." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation sync/dispatch overhead and cache sequence-length return type are unrelated performance vs API bugs." + "reason": "Generation GPU-sync/dispatch overhead and `StaticCache.get_seq_length()` shape bug are different problems." } ] }, @@ -3916,233 +3976,236 @@ "issue:43317|issue:43856", "issue:42175|issue:43296" ], - "prepared_review_unit_hash": "6bf210ccabea92c9472b2c5c7d6c407ebaf54447dac6e30ad6194e9ec7dffefa", + "prepared_review_unit_hash": "d5f3a8ff6a1e0c2203870224ad0f7561ba32efbba8ed84deb6cdb60d93be379d", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37278, - "estimated_input_tokens": 9320, - "estimated_eval_tokens": 18896 + "serialized_chars": 37287, + "estimated_input_tokens": 9322, + "estimated_eval_tokens": 18900 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated issues that only share broad Transformers subdomains or model families. None of the soft pairs are strong enough to treat as duplicates or one mergeable PR-equivalent change.", - "confidence": 0.96, - "canonical_issue_reason": "No canonical issue stands out: the set spans many distinct bugs and features (tokenizers, model loading, multimodal training, quantization, Torch compatibility, etc.), so there is no single duplicate hub.", + "summary": "This cluster is mostly a grab bag of unrelated transformer/v5, tokenizer, model-loading, and training regressions. I only see one plausible duplicate pair: the two Qwen3.5 save_pretrained issues. No PRs are present.", + "confidence": 0.84, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No global best issue: the items do not form one coherent underlying bug/change, so picking a representative issue would be arbitrary.", + "best_issue_reason": "issue:45812 is the broadest representative of the tokenizer-regression subset and has clear impact across all Granite models.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Different problems: Torch 2.10 compatibility vs noisy transformer loading behavior." + "reason": "Torch-version incompatibility and generic loading noise are different failure modes." }, { "left": "issue:45216", "right": "issue:45357", - "accept": false, - "reason": "Both involve Qwen3.5 save_pretrained regressions, but one is a general checkpoint correctness issue and the other is specifically incorrect visual encoder keys in a later release." + "accept": true, + "reason": "Both describe the same Qwen3.5 save_pretrained checkpoint regression; the second is a narrower symptom of the same bug." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "Both mention model loading/regression patterns, but one is a vLLM PaddleOCR-VL load failure and the other is a sliding_window bug in Qwen3-MoE; different code paths." + "reason": "PaddleOCR-VL loading failure and Qwen3-MoE sliding_window behavior are unrelated." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen2.5-VL issues, but one is an IterableDataset multi-image finetuning failure and the other is a rope-index scaling bug in inference; not the same bug." + "reason": "Multi-image IterableDataset finetuning failure is not the same as the Qwen2.5-VL rope_index scaling bug." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Unrelated: Molmo tied-weights attribute error vs tokenizer loading regression for EMBEDDIA/est-roberta." + "reason": "Missing all_tied_weights_keys and tokenizer loading failure are unrelated." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "Different regressions: base_model_prefix loading failure vs non-persistent buffers being filled with junk." + "reason": "base_model_prefix loading breakage and non-persistent buffer junk are different code paths." }, { "left": "issue:44991", "right": "issue:46315", "accept": false, - "reason": "Both concern tokenizers, but one is a tokenizer load regression and the other is a performance regression in convert_tokens_to_ids for slow tokenizers with many added tokens." + "reason": "Tokenizer load failure and slow convert_tokens_to_ids regression are distinct bugs." }, { "left": "issue:43716", "right": "issue:46899", "accept": false, - "reason": "Both are dtype-related multimodal issues, but one is an image preprocessor/model dtype mismatch and the other is unsafe dtype casting in Gemma4 quantized embedders; different fixes." + "reason": "Both are multimodal dtype issues, but they involve different models and triggers, so they are not clearly the same bug." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Different failure modes: _torch_extract_fbank_features() bug vs compiled flex_attention breaking on newer torch." + "reason": "Audio feature extraction and compiled flex_attention on torch>=2.9 are unrelated." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Noise/logging issue vs tokenizer loading regression; not the same underlying bug." + "reason": "Generic loading noise is not the same as tokenizer repository-resolution failure." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing in eval mode vs missing supports_gradient_checkpointing flag on a trust_remote_code model; separate issues." + "reason": "Eval-mode gradient checkpointing and missing supports_gradient_checkpointing are different issues." }, { "left": "issue:42796", "right": "issue:43746", "accept": false, - "reason": "Both are tokenizer/model-loading issues, but one is BPE marker handling and the other is PEFT adapter loading from local checkpoints; different code paths." + "reason": "Tokenizer BPE-marker handling and GraniteSpeech PEFT checkpoint loading are unrelated." }, { "left": "issue:45812", "right": "issue:46491", "accept": false, - "reason": "Both are tokenizer regressions, but one is Granite token IDs being wrong and the other is CodeLlama leading-space round-trip behavior; not the same bug." + "reason": "Both are tokenizer regressions, but Granite token IDs and CodeLlama leading-space round-trip are different bugs." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "Transformers v5 _is_hf_initialized regression vs mmap OOM on Strix Halo; unrelated root causes." + "reason": "_is_hf_initialized breakage and mmap OOM on Strix Halo are unrelated." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Both are Qwen2-VL/Qwen2.5-VL training issues, but one is multi-GPU training and the other is multi-image IterableDataset finetuning failure; too broad to merge." + "reason": "Multi-GPU training and multi-image iterable finetuning are not the same failure." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Whisper run failure vs Float8 storage lookup TypeError; unrelated bugs." + "reason": "Whisper run failure and Float8 storage lookup error are different problems." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized argument mismatch vs missing Float8 storage object; different breakages." + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage object are unrelated." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "Perplexity regression on gpt-oss vs Whisper batch_decode skip_special_tokens handling; not the same issue." + "reason": "GPT-OSS perplexity regression and Whisper batch_decode skip_special_tokens bug are unrelated." }, { "left": "issue:42796", "right": "issue:46710", "accept": false, - "reason": "Tokenizer BPE-marker handling vs accuracy regression in DeepSeek-R1-Distill-Llama-8B; no shared code-path evidence." + "reason": "Tokenizer BPE-marker handling and DeepSeek accuracy regression are different issues." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Qwen2.5-GGUF loading failure and Llama4Config missing pad_token_id are different model/config issues." + "reason": "Qwen2.5-GGUF loading under v5 and Llama4Config missing pad_token_id are unrelated." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "inputs_embeds/past_key_values generation bug vs multi-image finetuning failure; unrelated." + "reason": "Progressive generation with inputs_embeds/past_key_values is not the same as the multi-image finetuning failure." }, { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "ModelOutput key assignment bug vs Gemma4 FSDP2 shared_kv_states KeyError; different layers of the stack." + "reason": "ModelOutput key assignment and Gemma4 shared_kv_states/FSDP2 are unrelated." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior vs init_empty_weights/_is_hf_initialized constructor mismatch; not the same bug." + "reason": "Qwen3-MoE sliding_window behavior and init_empty_weights argument handling are different bugs." }, { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Runtime special-token post-processor update feature vs mdeberta add_special_tokens BOS/EOS regression; different tokenizer subsystems." + "reason": "Runtime post-processor updates and add_special_tokens BOS/EOS omission are different tokenizer issues." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Qwen3 auto device map cudaErrorAssert vs Mistral regex patch crash; unrelated failures." + "reason": "Qwen3 auto device mapping cuda assert and Mistral regex patch crash are unrelated." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "TensorFlow backend packaging omission vs tensorboard-without-TensorFlow undefined backend error; related packaging area but different concrete bugs." + "reason": "These are both backend-packaging complaints, but they are not clearly the same underlying defect." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "Qwen2.5-GGUF loading failure vs CLIPTokenizer model_max_length constant; unrelated." + "reason": "GGUF loading failure and CLIPTokenizer model_max_length are unrelated." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "ModelOutput key bookkeeping vs buffer initialization junk; different subsystems." + "reason": "ModelOutput key assignment and non-persistent buffer initialization are different bugs." }, { "left": "issue:43296", "right": "issue:46821", "accept": false, - "reason": "PaddleOCR-VL load failure in vLLM vs TimesFM window_size AttributeError; unrelated model families and code paths." + "reason": "PaddleOCR-VL loading failure and TimesFM window_size AttributeError are unrelated." }, { "left": "issue:44462", "right": "issue:45812", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json vs Granite AutoTokenizer wrong token IDs; both tokenizer-related but different faults and fixes." + "reason": "Ignoring tokenizer.json and wrong Granite token IDs are both tokenizer regressions, but not the same code-path bug." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "device_map=auto dequantized load/offload failure vs Qwen3 MoE training memory inefficiency; different symptoms and mechanisms." + "reason": "device_map=auto offload failure and Qwen3 MoE memory usage are different issues." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "TensorFlow backend installation issue vs PaddleOCR-VL model load failure; not the same underlying bug." + "reason": "TensorFlow backend packaging and PaddleOCR-VL model loading are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, issue:45216 vs issue:45357, is plausibly the same Qwen3.5 save_pretrained regression with a narrower symptom. The other pairwise rejections are cautious and do not overclaim equivalence. No PR pairs are present." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -4209,8 +4272,8 @@ "issue:46909" ], "soft_pairs": [ - "issue:43611|issue:43646", "issue:44509|issue:46909", + "issue:43611|issue:43646", "issue:43377|issue:45491", "issue:43644|issue:44079", "issue:43818|issue:46402", @@ -4240,214 +4303,214 @@ "issue:44186|issue:44265", "issue:43606|issue:43819" ], - "prepared_review_unit_hash": "08b4ed089568bfbe5c9a9f3aad6df473f2473c8075621bffeba16965d2a92ba4", + "prepared_review_unit_hash": "428bbfe2a7334cea04387570a789843afc9545df5ceda275cd06861f2ef960a1", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36919, - "estimated_input_tokens": 9230, - "estimated_eval_tokens": 18716 + "serialized_chars": 36916, + "estimated_input_tokens": 9229, + "estimated_eval_tokens": 18714 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a broad mix of distinct issues: tokenizer regressions, multimodal position-id bugs, device-map/offload failures, docs problems, and unrelated model-loading breakages. I don\u2019t see any soft pair that clearly shares the same underlying bug/change, so all candidate merges are rejected. Issue 43873 is the best representative anchor because it is active, well-discussed, and phrased broadly enough for triage.", - "confidence": 0.74, - "canonical_issue_reason": "Issue 43873 is the most central representative: it is open, has the most discussion/inbound references, and describes a concrete runtime bug in a broad, reusable way.", + "summary": "The items are a heterogeneous mix of unrelated bugs and docs regressions across different models and subsystems. I do not see any soft pair that is clearly the same underlying change/bug, so all candidate duplicate edges should be rejected.", + "confidence": 0.89, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "43873 is the best issue to anchor this cluster for triage, but it should not be treated as a duplicate of the others.", + "best_issue_reason": "issue:43873 is the broadest and most active open bug in the set (offloading/quantization), so it is the best single representative issue, but it is not a true duplicate anchor for the other items.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43611", - "right": "issue:43646", + "left": "issue:44509", + "right": "issue:46909", "accept": false, - "reason": "Both concern Transformers 5.0 model initialization/loading regressions, but one is about base_model_prefix loading and the other about custom model init; not the same code-path bug." + "reason": "Both are docs-related, but one is about removed pipeline tasks in v5 while the other is a broken DIA example; different documentation problems." }, { - "left": "issue:44509", - "right": "issue:46909", + "left": "issue:43611", + "right": "issue:43646", "accept": false, - "reason": "Both are docs problems, but they target different documentation examples and different failures." + "reason": "Both mention Transformers 5.0.0 regressions, but one is about `base_model_prefix` loading and the other about custom model initialization; different failure modes/code paths." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Different models and different failure modes: missing padding-mask support versus NaNs from sliding-window/all-padding batches." + "reason": "Different models and bugs: missing padding mask support in MIMI versus NaNs from Gemma3 sliding-window attention on mixed-length batches." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Unrelated bugs: non-persistent buffer serialization junk versus ModelOutput key assignment behavior." + "reason": "One is about non-persistent buffer serialization in v5, the other about `ModelOutput` key assignment when a field was previously `None`; unrelated code paths." }, { "left": "issue:43818", "right": "issue:46402", "accept": false, - "reason": "Different model families and different defects: Video-LLaVA tower architecture versus Sam3 text-encoder loading." + "reason": "Different models and subsystems: Video-LLaVA tower architecture issues versus Sam3 text encoder weights not loading." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL issues, but one is iterable-dataset/multi-image fine-tuning and the other is video position-id computation." + "reason": "Both involve Qwen2.5-VL video/image handling, but one is finetuning with multi-image iterable datasets and the other is wrong video `vision_position_ids`; not the same concrete bug." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "One is a missing temporal-attention/video-tower architecture bug; the other is incorrect save_pretrained key naming for visual encoder weights." + "reason": "Video-LLaVA tower/temporal-attention problem is unrelated to Qwen3.5 `save_pretrained` saving wrong visual encoder keys." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "Different root causes: torch.export support for torch_compilable_check versus dtype mismatches in bfloat16 inference." + "reason": "`torch.export.export` failing on `torch_compilable_check` is a different issue from bf16 dtype mismatches in inference." }, { "left": "issue:43493", "right": "issue:43994", "accept": false, - "reason": "Same model family, but one is HF-vs-JAX parity and the other is bad AutoModel/pipeline outputs; too vague to treat as one concrete bug." + "reason": "Same model family, but one is a JAX-vs-HF implementation discrepancy and the other is a bad pipeline/AutoModel result; insufficient evidence they are the same code-path bug." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Both are Qwen2.5-VL, but the bugs differ: temporal scaling for still images versus OCR long-input index overflow." + "reason": "Both touch position/index handling, but they are different models and different inputs: Qwen2.5-VL still-image temporal positions versus Kosmos2.5 long OCR indexing." }, { "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Both involve Qwen2.5-VL position ids, but one is still-image temporal scaling and the other is video vision_position_ids." + "reason": "Both concern Qwen2.5-VL video/position IDs, but one is still-image temporal scaling and the other is video `vision_position_ids`; related area, not the same bug." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Both describe degenerate training/loss behavior, but the causes are different: FlashAttention-3 sequence classification versus num_labels=1 single-label classification." + "reason": "One is a training regression with `flash_attention_3`, the other is degenerate zero loss from `problem_type=\"single_label_classification\"` with `num_labels=1`; different causes." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "Both mention lm_head/tied weights, but one is a missing tie in model config and the other is inconsistent serialization by device." + "reason": "Both involve tied `lm_head` weights, but one is failing to tie weights at init and the other is inconsistent serialization depending on device; different problems." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Different loading failures: unexpected init_empty_weights argument versus strict config rejection for granite_speech." + "reason": "`init_empty_weights` rejecting `_is_hf_initialized` is a different compatibility issue from strict config blocking `granite_speech` config loading." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "One is a FineGrainedFP8Config failure and the other is a GGUF loading regression; different mechanisms." + "reason": "Qwen3Moe FP8 config failure and Qwen2.5-GGUF loading failure are unrelated." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL rope/position-id bugs." + "reason": "Checkpoint key mismatch during resume is unrelated to Qwen2.5-VL rope/position index scaling." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Quantization/offload behavior and tokenization changes are unrelated." + "reason": "Quantization/offloading behavior is unrelated to tokenizer output changes across versions." }, { "left": "issue:42796", "right": "issue:45381", "accept": false, - "reason": "Tokenizer BPE-marker handling is unrelated to Qwen2.5-VL video position-id computation." + "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL video position IDs are different bug classes." }, { "left": "issue:41553", "right": "issue:43582", "accept": false, - "reason": "Bad tokenizer error messaging and an Apple Silicon allocator TypeError are unrelated." + "reason": "Bad AutoTokenizer error reporting for Voxtral is unrelated to the Apple Silicon `caching_allocator_warmup` TypeError." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "Different models and symptoms: high perplexity on GPT-OSS versus Whisper empty-transcription generation failure." + "reason": "Extremely high perplexity on one model and Whisper empty-transcription failure are unrelated." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU offload device mismatch and torch.export export failure come from different code paths." + "reason": "CPU offload device mismatch and `torch.export` failure with `torch_compilable_check` are different runtime paths." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Both are device/runtime regressions, but they affect different models and different mechanisms." + "reason": "Different models and symptoms: video input regression versus Gemma4 PLE device mismatch." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio processor sample-rate defaulting and GraniteSpeech PEFT checkpoint loading are unrelated." + "reason": "Audio chat-template sampling-rate defaulting is unrelated to GraniteSpeech PEFT checkpoint loading." }, { "left": "issue:43828", "right": "issue:46821", "accept": false, - "reason": "Different bugs: autocast dtype mismatch versus TimesFM window_size AttributeError." + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to TimesFM `window_size` AttributeError." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to a Qwen3 auto-device-map CUDA assert." + "reason": "Progressive generation with `inputs_embeds`/`past_key_values` and Qwen3 auto-device-map CUDA asserts are different issues." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "SigLIP2 docs-example errors and Qwen2.5-VL multi-image training failures are different issues." + "reason": "SigLIP2 docs example mistakes and Qwen2.5-VL finetuning with multiple images are unrelated." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "BarkProcessor voice_preset failure and progressive generation cache behavior do not share the same bug." + "reason": "Bark `voice_preset` not working and progressive generation with embeddings/cache are unrelated model APIs." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Both mention device_map=\"auto\", but they are different models with different failures and code paths." + "reason": "Qwen3 auto device mapping CUDA assert and Gemma4 PLE device mismatch are different models and different failure modes." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "LayoutLMv2Tokenizer crashes and torch.export support issues are unrelated." + "reason": "LayoutLMv2 tokenizer crashes and `torch.export` failures are unrelated." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "CPU offload device mismatch is unrelated to DAC.from_latents versus forward-pass consistency." + "reason": "CPU offload device mismatch and DAC `from_latents` vs forward mismatch are different bugs." } ] }, @@ -4553,220 +4616,208 @@ "issue:43873|issue:44756", "issue:44186|issue:45072" ], - "prepared_review_unit_hash": "18a134356757817e61eaf2e897644bfcc3d46f326ccfb2764e9f73564e14babc", + "prepared_review_unit_hash": "e36926b1ce3216439fddceeef0641ec8c5d48efb846aa362c7c780307200cc0f", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36924, - "estimated_input_tokens": 9231, - "estimated_eval_tokens": 18718 + "serialized_chars": 36920, + "estimated_input_tokens": 9230, + "estimated_eval_tokens": 18716 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is largely heterogeneous. The soft pairs look like broad semantic similarities, not the same concrete bug/change, so I would keep them separate. If one issue must represent the set, 43873 is the broadest offloading/quantization/loading regression and has the strongest activity.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43873 is the most representative broad regression here: it targets offloading/quantization behavior, has the highest activity/inbound attention among the cluster, and is closer to several of the other loading/offloading-themed issues than the narrower model- or tokenizer-specific reports.", + "summary": "This cluster is heterogeneous: the soft pairs share broad themes like tokenization, loading, offloading, or quantization, but they do not look like the same concrete bug/change. No PRs are present.", + "confidence": 0.93, + "canonical_issue_reason": "No single true canonical duplicate stands out because the cluster spans many unrelated failure modes. If one issue must be used as an umbrella representative, #43873 is the broadest/highest-activity offloading/quantization report, but it is not a duplicate of the others.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the best single issue to anchor the cluster because it is broad, actively discussed, and sits in the main loading/offloading/quantization theme that appears repeatedly across the set.", + "best_issue_reason": "#43873 is the strongest representative only in the sense that it is broad, active, and central to the offloading/quantization subset; however, the cluster is too diverse for a real canonical duplicate issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "Different bugs: weight tying/garbage generation vs a torch.compile SDPA failure." + "reason": "Different models and failures: weight tying/text generation corruption vs torch.compile + SDPA failure." }, { "left": "issue:43257", "right": "issue:45923", "accept": false, - "reason": "Different components: MOE weight conversion/loading vs missing gradient-checkpointing flag." + "reason": "Qwen MoE weight-loading with accelerate/deepspeed is unrelated to a missing gradient_checkpointing flag on a trust_remote_code model." }, { "left": "issue:42796", "right": "issue:46209", "accept": false, - "reason": "Tokenizer BPE-marker handling is unrelated to FP8 inference on B200." + "reason": "Tokenizer BPE-marker handling is unrelated to FP8 inference failure on B200." }, { "left": "issue:44568", "right": "issue:45920", "accept": false, - "reason": "Both are tokenization regressions, but the concrete failures differ; not the same bug." + "reason": "Both are tokenizer regressions, but one is specific special-token insertion behavior while the other is wrong token IDs across multiple models." }, { "left": "issue:45823", "right": "issue:46821", "accept": false, - "reason": "Device-map forward mismatch is unrelated to the TimesFM window_size AttributeError." + "reason": "Device-mismatch during forward is unrelated to an AttributeError on window_size in TimesFM." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "Missing padding-mask support in an encoder is unrelated to a processor output-size mismatch." + "reason": "MIMI batching/padding-mask inconsistency is a different code path from OmDet processor output shape mismatch." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "Weight tying/serialization bug and dtype mismatch in bfloat16 inference are different issues." + "reason": "lm_head tying/serialization corruption is unrelated to bfloat16 dtype mismatches in CI." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "3d_parallel.py execution failure is unrelated to Qwen3Moe fine-grained FP8 loading." + "reason": "A script execution problem is unrelated to a Qwen3 MoE FineGrainedFP8Config failure." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "Perplexity on a model/dataset setup is unrelated to LoRA gradient-checkpointing ineffectiveness." + "reason": "High perplexity on a model/dataset is not the same bug as gradient checkpointing being ineffective with PEFT LoRA." }, { "left": "issue:46305", "right": "issue:46315", "accept": false, - "reason": "Sliding-window mask off-by-one and tokenizer lookup slowness are different code paths." + "reason": "Sliding-window masking off-by-one is unrelated to slow convert_tokens_to_ids performance." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_idx in a config is unrelated to unwrapping text_config in composite multimodal configs." + "reason": "A missing config field is unrelated to AutoModelForCausalLM.from_config failing to unwrap text_config." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "lm_head weight tying is unrelated to packed-weight unpacking during accelerate loading." + "reason": "Weight tying/text generation corruption is unrelated to BitNet packed-weight unpacking during accelerate loading." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "Missing padding-mask support and Perceiver interpolation failure are different problems." + "reason": "Batch-vs-single output mismatch in MIMI is unrelated to Perceiver image classification with non-default resolution." }, { "left": "issue:43749", "right": "issue:44534", "accept": false, - "reason": "FSDP CPU RAM-efficient loading and non-persistent-buffer junk are not the same bug." + "reason": "FSDP CPU RAM efficient loading is a different loading path from non-persistent buffer initialization junk." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Both mention offloading/quantization, but one is a specific dequantized-model load failure and the other is a broader offloading complaint; too broad to merge confidently." + "reason": "Both mention offloading/quantization, but one is a specific device_map auto load failure while the other is a broader offloading regression; not the same concrete bug." }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "Llava Next loading segfault and sinq save/load breakage are unrelated model issues." + "reason": "Llava Next loading segfault and sinq save/load corruption are different failure modes and code paths." }, { "left": "issue:36331", "right": "issue:42617", "accept": false, - "reason": "Unexpected compute_loss kwargs is unrelated to a 3d_parallel.py runtime failure." + "reason": "CustomTrainer compute_loss signature breakage is unrelated to not being able to run a 3d_parallel script." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Auto device-map cuda assert and Llava Next segfault are different failures." + "reason": "Qwen3 auto device mapping cudaErrorAssert and Llava Next segfault are unrelated load/runtime failures." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Int4 quantization memory regression and apex RMSNorm memory leak are unrelated." + "reason": "Int4 CUDA reserved-memory/OOM regression is unrelated to a T5 apex fused RMSNorm memory leak." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Both involve Qwen3.x generation/loading, but the concrete failure modes differ." - }, - { - "left": "issue:36331", - "right": "issue:42915", - "accept": false, - "reason": "CustomTrainer compute_loss signature mismatch is unrelated to Qwen3Moe FP8 loading." + "reason": "Both involve Qwen models, but one is a device-mapping assert and the other is flash-attention generation corruption." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Strict config loading and tokenizer loading regression are different issues." + "reason": "Strict config loading for granite_speech is unrelated to tokenizer loading failure for EMBEDDIA/est-roberta." }, { "left": "issue:43881", "right": "issue:44488", "accept": false, - "reason": "Two distinct model-loading failures with no shared concrete bug." + "reason": "Two separate model-loading failures with no evidence of the same root cause." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "High perplexity on one model is not the same as a general tokenization-version change." + "reason": "Perplexity regression on one model/dataset is not the same as a general tokenization-change report." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Batched video-shape bug and tokenization regression are unrelated." - }, - { - "left": "issue:41720", - "right": "issue:42617", - "accept": false, - "reason": "cudaErrorAssert under auto device mapping is unrelated to running 3d_parallel.py." + "reason": "Video processor batch-shape bug is unrelated to tokenization behavior changing across versions." }, { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both are save/serialization bugs, but they affect different artifacts and code paths." + "reason": "lm_head serialization inconsistency and wrong visual-encoder keys in Qwen3.5 are different save_pretrained issues." }, { "left": "issue:43065", "right": "issue:46402", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and missing Sam3Model text encoder weights are different issues." + "reason": "A dummy Conv2d in Sam3PixelDecoder is unrelated to missing text encoder weight loading in Sam3Model." }, { "left": "issue:26413", "right": "issue:28530", "accept": false, - "reason": "Checkpoint resumption epoch_iterator failure is unrelated to early-stopping metric lookup." + "reason": "resume_from_checkpoint epoch-iterator failure is unrelated to early stopping disabling from missing eval metric." }, { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Quantization/offloading trouble is not the same as Strix Halo mmap OOM." + "reason": "Quantization/offloading regression and disabling mmap on Strix Halo to avoid OOM are different problems." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "Tokenizer crash on NER/padding is unrelated to bfloat16 dtype mismatches." + "reason": "Tokenizer crashes on NER/padding inputs are unrelated to bfloat16 dtype mismatches." } ] }, @@ -4842,8 +4893,8 @@ "soft_pairs": [ "issue:46612|issue:46858", "issue:43295|issue:45701", - "issue:43577|issue:43708", "issue:45685|issue:46821", + "issue:43577|issue:43708", "issue:43550|issue:43720", "issue:43646|issue:46665", "issue:36010|issue:42617", @@ -4869,202 +4920,202 @@ "issue:44671|issue:44743", "issue:44186|issue:44361" ], - "prepared_review_unit_hash": "ddbf0de28d7ede5e6a89a1e90b259894eb47e1f0ef6342ba1c19834876f40aee", + "prepared_review_unit_hash": "2c33de9ecf4e2467b6865c1da0e36eb65939e03ec2843c488951b35837a88644", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36266, - "estimated_input_tokens": 9067, - "estimated_eval_tokens": 18390 + "serialized_chars": 36269, + "estimated_input_tokens": 9068, + "estimated_eval_tokens": 18392 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a loose set of Transformers bugs, mostly model- or subsystem-specific, not a duplicate cluster. I selected the broad tokenizer-regression report as the best canonical issue, but none of the soft pairs look like the same underlying bug/change.", - "confidence": 0.67, - "canonical_issue_reason": "issue:45701 is the broadest and most central report here: it describes a version-wide tokenization regression with multiple downstream references, making it the best umbrella issue among otherwise narrow, model-specific bugs.", + "summary": "No soft-edge pair looks like the same concrete bug/change; the set is a collection of unrelated Transformers regressions across loading, generation, tokenization, multimodal serving, and device-mapping paths.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43646 is the broadest umbrella-style issue in the set: a Transformers 5.0.0 regression breaking custom model initialization. It is more central and general than the more model-specific or symptom-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 best represents the cluster because it captures a general Transformers regression theme rather than a single model/backend failure.", + "best_issue_reason": "issue:43646 is the best single issue anchor for this cluster because it describes a foundational regression in model initialization/loading rather than a narrow model-specific failure.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46612", "right": "issue:46858", "accept": false, - "reason": "Both involve generation/cache behavior, but they are different bugs: beam-search reorder omission vs static-cache seq_length device arithmetic on non-CUDA backends." + "reason": "Both involve generation/cache behavior, but one is about beam-search cache reorder being skipped for several models and the other is a static-cache seq-length/device-tensor arithmetic bug on non-CUDA backends." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "One is a specific processor/tokenizer regression in v4.57.5; the other is a broad version-change tokenization complaint, not the same concrete failure." + "reason": "Both are version-related regressions, but one is about processor.tokenizer / image-to-tokenizer compatibility and the other is about tokenization changes; different code paths and symptoms." }, { - "left": "issue:43577", - "right": "issue:43708", + "left": "issue:45685", + "right": "issue:46821", "accept": false, - "reason": "Unrelated failures: dtype not propagating in Blip2 loading vs Trainer max_steps calculation after resume_from_checkpoint." + "reason": "Unrelated issues: an MPS MoE histogram/int dtype failure versus a TimesFM window_size AttributeError." }, { - "left": "issue:45685", - "right": "issue:46821", + "left": "issue:43577", + "right": "issue:43708", "accept": false, - "reason": "Different components and symptoms: an MPS int histogram backend error vs a TimesFM window_size AttributeError." + "reason": "Different subsystems: BLIP2 dtype propagation during load versus Trainer checkpoint-resume step calculation." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Different code paths: torch.compile+SDPA in Bamba vs packed-weight unpacking during accelerate loading for BitNet." + "reason": "Different failures in different model paths: torch.compile with SDPA on Bamba versus packed-weight handling during accelerate loading for BitNet." }, { "left": "issue:43646", "right": "issue:46665", "accept": false, - "reason": "Both touch custom loading, but the concrete problems differ: Transformers 5.0 custom model initialization breaks vs inconsistent from_pretrained behavior inside Python classes." + "reason": "Both are custom-model/loading related, but one is a broad initialization regression and the other is inconsistent from_pretrained behavior inside Python classes; not the same bug." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "No shared bug: GenerationMixin import failure vs inability to run 3d_parallel.py." + "reason": "Completely different problems: GenerationMixin import failure versus inability to run a specific 3d_parallel.py script." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Different failures: SDPA/torch.compile on Bamba vs CPU-offload device mismatch on suno/bark-small." + "reason": "Both are device/runtime bugs, but the concrete failures differ: SDPA+torch.compile on Bamba versus CPU offload device mismatch on suno/bark-small." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Both are processor-related, but one is LayoutLMv2Tokenizer NER/padding crash and the other is OmDet-Turbo image-size mismatch." + "reason": "Both touch vision preprocessing, but one is a tokenizer crash on NER/padding while the other is an image processor size mismatch for OmDet-Turbo." }, { "left": "issue:44451", "right": "issue:47020", "accept": false, - "reason": "Both concern tokenizer loading, but they are different concrete breakages on different models and different tokenizer constructors." + "reason": "Both are tokenizer-loading failures, but they concern different tokenizer implementations and distinct failure modes." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Different model-loading/runtime issues: FineGrainedFP8Config failure for Qwen3Moe vs dequantized model loading with gpu+cpu offload." + "reason": "Different quantization/device-map problems: FineGrainedFP8Config on Qwen3MoE versus dequantized model loading with gpu+cpu offload." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "Different model-specific state bugs: untied lm_head weights in Mistral3 vs incorrect visual-encoder keys on save_pretrained for Qwen3.5." + "reason": "Both concern model save/load correctness, but one is lm_head tying for Mistral3 while the other is incorrect visual encoder keys for Qwen3.5." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Different vision-model preprocessing bugs: wrong output resolution for OmDet-Turbo vs non-default-resolution failure in Perceiver." + "reason": "Both are image-size/preprocessing regressions, but they are not the same concrete bug: one is processor output size mismatch, the other is Perceiver interpolation failure." }, { "left": "issue:46821", "right": "issue:46869", "accept": false, - "reason": "No common underlying bug: TimesFM window_size AttributeError vs YOLO multi-GPU auto device-map failure." + "reason": "Different model bugs: TimesFM window_size attribute access versus YOLO multi-GPU device-map loading failure." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both affect serve with multimodal models, but the concrete causes differ: bad tensor-to-string handling vs missing _tokenizer on Gemma4Processor." + "reason": "Both affect transformers serve with multimodal processors, but one is a .to() call on a string during continuous batching and the other is a missing _tokenizer attribute on Gemma4Processor." }, { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "Separate tokenizer-loading regressions on different models; no evidence they stem from the same code path." + "reason": "Both are loading issues, but they are unrelated: glm-4v-9b load failure versus tokenizer loading failure for EMBEDDIA/est-roberta." }, { "left": "issue:44734", "right": "issue:46858", "accept": false, - "reason": "Both involve cache logic, but one is wrong tensor indexing in serve KV continuation and the other is static-cache seq_length device arithmetic during prefill." + "reason": "Both involve cache handling, but one is wrong tensor indexing in serve /v1/responses and the other is a device-tensor seq-length issue in _prefill with static cache and torch.compile." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "Completely different bugs: BeitImageProcessorFast label reduction vs Molmo missing all_tied_weights_keys." + "reason": "Different components entirely: BeitImageProcessorFast label reduction versus MolmoForCausalLM missing all_tied_weights_keys." }, { "left": "issue:28530", "right": "issue:33666", "accept": false, - "reason": "Unrelated training issues: early stopping metric selection vs Qwen2-VL multi-GPU training." + "reason": "No real overlap: early stopping metric selection versus Qwen2-VL multi-GPU training." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Both are distributed/device-map related, but one is a cudaErrorAssert on A800 and the other is a tensor-parallelism limitation in GPT-OSS MoE." + "reason": "Both are distributed/device-related, but they are different concrete bugs: cudaErrorAssert on A800 versus only-TP not working for GPT-OSS MoE." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Different multimodal forward bugs: variable-name collision causing Tensor attribute error vs device mismatch in Gemma4 PLE with auto device mapping." + "reason": "Both are Gemma multimodal forward issues, but the specific failures differ: variable-name collision in Gemma3n versus device mismatch in Gemma4 PLE." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Separate model/config regressions: Qwen3-MoE sliding_window handling vs missing pad_token_idx in StableLmConfig." + "reason": "Different model/config regressions: Qwen3-MoE sliding_window behavior versus StableLmConfig missing pad_token_idx." }, { "left": "issue:42915", "right": "issue:46291", "accept": false, - "reason": "Different failures: FineGrainedFP8Config model loading vs importing transformers.pipelines with kernels version/revision requirements." + "reason": "Different areas: MoE FP8 failure versus importing transformers.pipelines with a kernels dependency version/revision requirement." }, { "left": "issue:44464", "right": "issue:46858", "accept": false, - "reason": "Related to generation/compile, but not the same bug: inconsistent chunked generation outputs vs cache seq_length device-tensor arithmetic." + "reason": "Both involve generation/compile/cache, but one is inconsistent outputs from compiled forward and the other is a seq-length tensor arithmetic bug in prefill." }, { "left": "issue:36331", "right": "issue:41720", "accept": false, - "reason": "Trainer loss-signature mismatch is unrelated to Qwen3 auto device-mapping cuda asserts." + "reason": "Unrelated trainer API compatibility issue versus Qwen3 auto device-mapping crash." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Different save/load edge cases: cache-path collision for trust_remote_code vs read-only permission propagation in save_pretrained." + "reason": "Both concern custom-model save/load workflows, but one is a cache-path collision and the other is permissions propagation from source files." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "Both are model-specific regressions, but one is CamemBERT MLM predictions and the other is Qwen3.5 recurrent-state reset with cache." + "reason": "Different model-specific bugs: CamemBERT masked-LM predictions versus Qwen3.5 recurrent state reset with cache." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer crashes on different models/tasks; no shared underlying defect is evident." + "reason": "Both are tokenizer crashes, but they affect different tokenizers and failure points; not the same underlying bug." } ] }, @@ -5167,208 +5218,184 @@ "issue:44451|issue:44991", "issue:44077|issue:44625" ], - "prepared_review_unit_hash": "bb53dc8c2303706c61b74cedfba004cdd2413cb2c64000fc126f1c2bf8629246", + "prepared_review_unit_hash": "65c946df86779c240e3a0a52959b531ad396153a064e3d44a4fe5a98c51921dd", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35782, - "estimated_input_tokens": 8946, - "estimated_eval_tokens": 18148 + "serialized_chars": 35772, + "estimated_input_tokens": 8943, + "estimated_eval_tokens": 18142 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous; the soft-similarity pairs mostly share only broad subsystem vocabulary, not the same underlying bug. No PRs are present. I chose one concrete loading bug as the weakest representative anchor, but the grouping does not look like a real duplicate set.", - "confidence": 0.61, - "canonical_issue_reason": "issue:43452 is the clearest concrete code-path bug in the set: a shared `gguf_file` loading failure affecting both `AutoTokenizer.from_pretrained` and `AutoModelForCausalLM.from_pretrained`.", + "summary": "These are mostly superficially similar issues across Transformers, but the pairwise evidence points to different root causes and code paths. I do not see a true duplicate cluster here.", + "confidence": 0.88, + "canonical_issue_reason": "No single canonical issue: the linked items span unrelated bugs in tokenization, config loading, multimodal processing, serving, quantization, and docs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43452 is the best standalone representative because it describes a specific, actionable failure in a common loading API rather than a vague subsystem complaint.", + "best_issue_reason": "Issue 43452 is the broadest actionable loading regression in the set, but it still does not represent the others as a duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43425", "right": "issue:46821", "accept": false, - "reason": "Torch-version incompatibility and a TimesFM `window_size` AttributeError are different bugs with different failure paths." + "reason": "Different problems: Torch compatibility vs a TimesFM `window_size` attribute error." }, { "left": "issue:28530", "right": "issue:30064", "accept": false, - "reason": "Early stopping metric lookup and void segmentation-map processing are unrelated training vs. image-processing issues." + "reason": "Early-stopping metric lookup and segmentation-map processing are unrelated bugs." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Config API discrepancy in Gemma3 and GPTNeoX rotary_pct reload behavior are distinct config bugs, not one shared defect." + "reason": "Both are config-related, but one is a Gemma3 API mismatch and the other is GPTNeoX rotary_pct persistence." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "FineGrainedFP8Config failure for Qwen3Moe and Llava Next segfault are separate model-loading problems." + "reason": "Qwen3Moe FP8 config failure and Llava Next segfault are different models and failure modes." }, { "left": "issue:33357", "right": "issue:35707", "accept": false, - "reason": "MacOS CLIP bus error and generation with `inputs_embeds`/`past_key_values` are unrelated code paths." + "reason": "MacOS CLIP bus error and progressive generation with `inputs_embeds`/`past_key_values` are unrelated." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "Slow/fast tokenizer inconsistency and T5 `add_special_tokens` load conflict are different tokenizer bugs." + "reason": "Tokenizer inconsistency vs a T5 `add_special_tokens` name conflict; same area, different bug." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Segmentation-map processing and BarkProcessor `voice_preset` handling do not describe the same bug." + "reason": "Void segmentation-map handling is unrelated to BarkProcessor voice preset loading." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantization/offloading memory behavior and an apex RMSNorm memory leak are not the same underlying issue." + "reason": "Quantization offload behavior and an Apex RMSNorm memory leak are distinct issues." }, { "left": "issue:44387", "right": "issue:45701", "accept": false, - "reason": "Int4 quantization OOM and version-dependent tokenization changes are different regressions." + "reason": "CUDA OOM under int4 quantization is not the same as tokenization changing across versions." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Qwen3OmniMoe video-input ValueError and Qwen3VL `weight_only=True` load error affect different modalities and paths." + "reason": "One is video input processing in Qwen3OmniMoe; the other is a `from_pretrained` weight_only loading error in Qwen3VL." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Sliding-window handling in Qwen3-MoE and `text_config` unwrapping in composite Qwen configs are separate config/model bugs." + "reason": "Sliding-window behavior in Qwen3-MoE and unwrapping `text_config` for composite configs are different code paths." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "`torch.compile` + SDPA failure in Bamba and BigBird tokenizer special-token registration are unrelated." + "reason": "torch.compile/SDPA failure in Bamba is unrelated to BigBirdTokenizer mask token registration." }, { "left": "issue:44534", "right": "issue:46665", "accept": false, - "reason": "Non-persistent buffer initialization and `from_pretrained` behavior inside Python classes are not the same bug." + "reason": "Non-persistent buffer initialization and `from_pretrained` inside Python classes are separate regressions." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "A typo in cache strategies is unrelated to stale docs mentioning removed pipeline tasks." + "reason": "A cache-strategy typo does not match a docs update about removed pipeline tasks." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Qwen3 auto device mapping CUDA assert and GraniteSpeech PEFT local-checkpoint loading are separate model-specific failures." + "reason": "Qwen3 device-mapping CUDA assert and GraniteSpeech PEFT checkpoint loading are different failures." }, { "left": "issue:43452", "right": "issue:47020", "accept": false, - "reason": "`gguf_file` breaking `from_pretrained` loading and XLMRobertaTokenizer passing a dict to Unigram are different tokenizer/model-loading bugs." + "reason": "Both involve loading/tokenizers, but one is gguf/from_pretrained routing and the other is an XLMRobertaTokenizer/Unigram init bug." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "A trailing-slash import warning and undefined video-branch helpers in multimodal token counting are unrelated." + "reason": "Trailing-slash local model import warning is unrelated to the undefined video-branch helpers in multimodal token counting." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "CPU offload device mismatch for Bark and BitNet packed-weight unpacking during accelerate loading are different mechanisms." + "reason": "CPU-offload device mismatch in Bark-small and BitNet packed-weight unpacking are distinct loading/offload problems." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "Trainer token-count tracking and progressive generation cache usage are not the same defect." + "reason": "Trainer token-count tracking and progressive generation behavior are unrelated." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Gemma4Processor missing `_tokenizer` and Qwen3.5 streaming inference failures are different API issues." + "reason": "Gemma4Processor missing `_tokenizer` and Qwen3.5 streaming chat/completions failures are different serving bugs." }, { "left": "issue:43531", - "right": "issue:46821", + "right": "issue:44589", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and TimesFM `window_size` errors are unrelated despite the shared word 'window'." - }, - { - "left": "issue:43388", - "right": "issue:45059", - "accept": false, - "reason": "Last-batch label truncation and SAM3 text/bounding-box behavior are different evaluation vs. multimodal-input bugs." - }, - { - "left": "issue:34634", - "right": "issue:41720", - "accept": false, - "reason": "Bark voice-preset handling and Qwen3 device-mapping CUDA asserts are unrelated." - }, - { - "left": "issue:43606", - "right": "issue:43653", - "accept": false, - "reason": "CPU offload device mismatch and tokenizer special-token registration do not share the same code path." - }, - { - "left": "issue:43531", - "right": "issue:44589", - "accept": false, - "reason": "Sliding-window logic and missing `Float8_e4m3fnStorage` are different model/runtime failures." + "reason": "Qwen3-MoE sliding-window logic is unrelated to the Float8 storage lookup error." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Composite config unwrapping and CPU offload compatibility for Qwen3_5MoeTextRotaryEmbedding are distinct problems." + "reason": "Composite-config unwrapping and CPU-offload compatibility in Qwen3.5 MoE embedding are different issues." }, { "left": "issue:43540", "right": "issue:46858", "accept": false, - "reason": "Video-input handling in Qwen3OmniMoe and static-cache slice arithmetic on non-CUDA backends are unrelated." + "reason": "Video-input validation in Qwen3OmniMoe is unrelated to the static-cache slice arithmetic bug." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Loading a specific model failing and tokenizer loading regressions in v5 are different issues." + "reason": "Both are load failures, but the model-specific regression reports do not indicate the same tokenizer/root cause." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "`patchtsmixer` post-init policy and Qwen3.5 `num_labels` propagation are unrelated configuration issues." + "reason": "A `post_init` allowance policy for patchtsmixer is unrelated to Qwen3.5 `num_labels` propagation." } ] }, @@ -5471,20 +5498,20 @@ "issue:43782|issue:45381", "issue:46032|issue:46463" ], - "prepared_review_unit_hash": "0c88f7a358fbfbf7101a998ee555646a38fcf057a87d70d067f3bfe6f2cad0e8", + "prepared_review_unit_hash": "b345fda8be81ea60e77529db05256b7944d8cefbf4dbbbbffe1cc37a1057c679", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36370, + "serialized_chars": 36371, "estimated_input_tokens": 9093, "estimated_eval_tokens": 18442 }, @@ -5493,177 +5520,189 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous; only one true duplicate pair stands out: the Qwen2_5_VLProcessor.apply_chat_template crash with padding=False.", + "summary": "The cluster is mostly heterogeneous. The only clear duplicate issue pair is 44514/44545, which share the same Qwen2_5_VLProcessor batched-input crash with padding=False; 42886/44843 is thematically related around HF_HUB_OFFLINE tokenizer loading, but not strong enough to treat as the same bug from the visible evidence. No pull requests are present.", "confidence": 0.89, - "canonical_issue_reason": "issue 44514 is the clearest canonical report: it is an exact duplicate of 44545 and the earlier, more established entry for the actual duplicate bug.", + "canonical_issue_reason": "44514 is the best canonical issue anchor because it is the clearest and earliest report of the Qwen2_5_VLProcessor batched-input padding=False crash, with 44545 appearing to be a near-identical follow-up.", "canonical_pr_reason": null, - "best_issue_reason": "issue 44514 is the best representative because it captures the only clearly duplicated underlying bug in the cluster and is more canonical than the unrelated neighboring issues.", + "best_issue_reason": "44514 is the strongest representative issue in the set: it is specific, has the closest matching twin in 44545, and best captures the concrete batched-input crash path.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45357", "right": "issue:46402", "accept": false, - "reason": "Different multimodal loading bugs: Qwen3.5 saves wrong visual encoder keys, while Sam3Model fails to load text encoder weights." + "reason": "Different model families and failure modes: Qwen3.5 visual-encoder key saving vs Sam3 text-encoder weights not loading." }, { "left": "issue:43644", "right": "issue:46665", "accept": false, - "reason": "Unrelated failures: junk non-persistent buffers vs inconsistent from_pretrained behavior inside Python classes." + "reason": "Both involve loading behavior, but one is junk-filled non-persistent buffers and the other is inconsistent from_pretrained behavior inside Python classes." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "Different components and symptoms: BeitImageProcessorFast reduce_labels bug vs glm-4v-9b load failure." + "reason": "Unrelated bugs: BeitImageProcessorFast label reduction vs glm-4v-9b loading failure." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "A model loading failure is not the same as PIL image processors incorrectly requiring torchvision." + "reason": "Different concrete problems: loading a specific repo on newer transformers vs PIL image processors requiring torchvision." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver non-default resolution failure and bfloat16 dtype mismatch are distinct bugs." + "reason": "Different subsystems and symptoms: Perceiver image-size handling vs bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "AddedToken special-argument collision is unrelated to loading vesteinn/ScandiBERT." + "reason": "Both are tokenizer-related failures, but one is an AddedToken argument conflict and the other is a model loading failure for ScandiBERT." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Both touch sequence-classification config, but one is missing num_labels propagation and the other is a zero-loss problem_type bug." + "reason": "Both involve labels, but one is config propagation in Qwen3.5 and the other is a classification-loss degeneracy across multiple models." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Checkpointing-in-eval restriction and ZeRO-3 gradient growth are different training issues." + "reason": "Different issues: eval-mode gradient checkpointing restriction vs gradient growth with ZeRO-3 training." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both are vision-related, but one is a doc/example mismatch and the other is an OmDet-Turbo processor size mismatch." + "reason": "One is a SigLIP2 documentation/example problem; the other is an OmDet-Turbo processor image-size mismatch." }, { "left": "issue:41720", "right": "issue:45704", "accept": false, - "reason": "Qwen3 auto device mapping CUDA asserts and T5 RMSNorm memory leak are unrelated." + "reason": "Completely different failures: Qwen3 auto device mapping cuda assert vs a T5 RMSNorm memory leak." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "Both involve load/save flows, but PEFT local checkpoint loading and sinq quantized serialization are different bugs." + "reason": "Different code paths: PEFT adapter loading from local checkpoints vs quantized model save/load breakage." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "A tie_word_embeddings warning is not the same as incorrect visual encoder keys being saved." + "reason": "A warning about tie_word_embeddings in LoRA fine-tuning is not the same as incorrect visual encoder keys being saved." + }, + { + "left": "issue:43381", + "right": "issue:45704", + "accept": false, + "reason": "No shared underlying bug: gradient checkpointing eval-mode restriction vs apex RMSNorm memory leak." }, { "left": "issue:46655", "right": "issue:46693", "accept": false, - "reason": "Video processor padding calculation and FA2 long-context performance regression are different issues." + "reason": "Different problems: video padding calculation in SmolVLM vs performance regression with FA2 long-context generation." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "TRL SFT input embedding unpacking and trainer.evaluate failing after train are separate trainer problems." + "reason": "Different trainer failures; one is embedding unpacking with TRL SFT, the other is evaluate() failing after train()." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Both involve token handling, but one is runtime post-processor refresh and the other is lost custom token IDs after save/load." + "reason": "Both touch token/config handling, but one is runtime special-token post-processing and the other is saved custom token IDs causing IndexError." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "Apex RMSNorm memory leak and missing supports_gradient_checkpointing are unrelated." + "reason": "Unrelated: T5 FusedRMSNorm memory leak vs missing supports_gradient_checkpointing flag." }, { "left": "issue:45759", "right": "issue:45923", "accept": false, - "reason": "Composite Qwen config unwrapping and a missing gradient-checkpointing flag are different code paths." + "reason": "Different problems: unwrapping text_config in multimodal Qwen configs vs a trust_remote_code gradient-checkpointing flag omission." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "Local adapter checkpoint loading and offloading with quantization are not the same underlying bug." + "reason": "Different areas: local checkpoint loading with PEFT adapters vs offloading failure under quantization." }, { "left": "issue:42886", "right": "issue:44843", "accept": false, - "reason": "Both mention HF_HUB_OFFLINE, but one is a cache-loading regression and the other is a specific model_info call in _patch_mistral_regex." + "reason": "Both are about HF_HUB_OFFLINE tokenization paths, but one reports a cache-load failure and the other a specific unconditional model_info call; related, not clearly the same bug." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Exact duplicate reports of the same Qwen2_5_VLProcessor.apply_chat_template crash when padding=False." + "reason": "Same concrete Qwen2_5_VLProcessor apply_chat_template crash on batched input with padding=False; identical titles and matching failure path." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Different domains and failures: image processor segmentation-map handling vs GenerationMixin import error." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "SigLIP2 doc example errors and SigLIP2 implementation divergence from JAX are related only by model family, not the same bug." + "reason": "Same model family, but not the same bug: documentation example errors vs HF/JAX implementation discrepancy." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "SigLIP2 inference oddity and CamemBERT masked-LM regression are unrelated model-specific bugs." + "reason": "Different model families and symptoms: SigLIP2 nonsensical outputs vs CamemBERT masked-LM prediction regression." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Special-token post-processor updates and BigBirdTokenizer mask-token registration are different tokenization defects." + "reason": "Both concern special tokens, but one is about runtime post-processor updates and the other about a tokenizer-specific special-token registration bug." }, { "left": "issue:44464", "right": "issue:47050", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency and fp16 ONNX dtype mismatch are distinct issues." + "reason": "Different failures: compiled-forward chunked generation inconsistency vs ONNX export dtype mismatch in CanineModel attention." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Qwen3VL weight_only load error and qwen2.5-vl video vision_position_ids bug affect different paths." + "reason": "Different Qwen multimodal issues: weight_only loading error vs incorrect video vision_position_ids." }, { "left": "issue:46032", "right": "issue:46463", "accept": false, - "reason": "Mamba2 use_cache correctness and T5 decoder-input validation messaging are unrelated." + "reason": "Unrelated: Mamba2Mixer use_cache correctness bug vs T5 decoder validation error-message bug." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet: 44514/44545 are an exact duplicate pair, and the rest of the soft-edge decisions are conservative given only titles/body metadata. No pair is overstated as mergeable or identical without sufficient evidence." + "accept": false, + "feedback": "Mostly grounded, but the cluster summary overstates certainty. The packet supports 44514/44545 as a strong duplicate candidate because the titles are identical, yet it does not substantiate \"exact duplicate\" or \"identical reproduction conditions\" from the visible evidence. Likewise 42886/44843 looks plausibly related, but the summary should stay more conservative about claiming the same underlying bug. The soft-edge rejections are generally cautious, but the canonical/summary wording is a bit too strong for the packet." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -5764,220 +5803,226 @@ "issue:43618|issue:43697", "issue:43643|issue:44877" ], - "prepared_review_unit_hash": "dd3fe1477ce3fda28d8838320bd14d27d802799ca3b7476abceff6aa52f89aae", + "prepared_review_unit_hash": "d8f3c5e6060d0038c3e2817ff9c5056911a46f1a9c082e4e86283a8a9798d1b0", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37584, - "estimated_input_tokens": 9396, - "estimated_eval_tokens": 19048 + "serialized_chars": 37590, + "estimated_input_tokens": 9398, + "estimated_eval_tokens": 19052 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These issues are mostly unrelated one-off bugs across different models/subsystems. I would not merge any soft pair; at most, use the broad open quantization/offloading issue as a loose anchor if a canonical issue is required.", - "confidence": 0.85, - "canonical_issue_reason": "No strong duplicate cluster emerges here. If forced to pick one anchor, issue #43873 is the broadest and most active open issue, but it is still not a true canonical duplicate representative for the rest.", + "summary": "These items are mostly unrelated issues across different models, tokenizers, config loading, caching, and export/runtime paths. The soft links are mostly superficial topic overlap (especially around tokenizers/configs), but none look like the same underlying bug or change, so all soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue #43873 is the best generic anchor by breadth and activity, though the set does not actually form a single duplicate group.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44857", "right": "issue:47050", "accept": false, - "reason": "Different models and different failures: AMP/CUDA image loss crash vs fp16 ONNX attention dtype mismatch." + "reason": "Different failures and code paths: LwDetrImageLoss AMP crash vs CanineModel fp16 ONNX dtype mismatch." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both mention tokenizer regex handling, but one is an AttributeError in `_patch_mistral_regex` and the other is a Kimi tokenizer regression with codec handling; not clearly the same bug." + "reason": "Both involve tokenizer regex/loading, but the concrete bugs differ: missing backend_tokenizer vs codec handling regression." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Runtime special-token/post-processor update issue vs LayoutLMv2 NER padding/truncation crash; different code paths." + "reason": "Different tokenizer subsystems and symptoms; runtime post-processor updates are unrelated to LayoutLMv2 NER/padding crashes." }, { "left": "issue:44995", "right": "issue:46858", "accept": false, - "reason": "Both involve cache/sequence length, but one is a stale indexer cache in a model and the other is `get_seq_length()` returning the wrong type for static cache compile paths." + "reason": "Both mention cache behavior, but one is stale indexer state and the other is tensor-return shape in StaticCache." }, { "left": "issue:45356", "right": "issue:45907", "accept": false, - "reason": "Tokenizer regression vs Hub repo file listing forwarding bug; unrelated subsystems." + "reason": "Tokenizer regression vs repo-file listing API bug; no shared underlying issue." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "Doc-example model/processor mismatch and quantization failure vs a general offloading-with-quantization runtime issue." + "reason": "Both mention quantization, but one is a docs example mismatch and the other is offloading behavior." }, { "left": "issue:39692", "right": "issue:47050", "accept": false, - "reason": "Different failure modes and contexts: documentation example errors vs ONNX export dtype mismatch." + "reason": "Different model families and failures; docs/quantization example issues do not match ONNX export dtype errors." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "Packed-weight loading bug vs DAC latent/STE mismatch; different models and code paths." + "reason": "Separate model-specific bugs: packed-weight loading vs missing STE in DAC.from_latents." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Processor/tokenizer regression vs long OCR input index error; both multimodal but not the same bug." + "reason": "Both are multimodal regressions, but the concrete problems are different: processor/tokenizer API break vs long-OCR index error." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "BigBird special-token registration bug vs BitNet loading/unpacking bug." + "reason": "Tokenizer special-token registration bug is unrelated to BitNet packed-weight loading." }, { "left": "issue:46829", "right": "issue:46832", "accept": false, - "reason": "Both are RT/vision-model issues, but one is a wrong output field and the other is a feature-level mismatch TypeError." + "reason": "Different RT/vision model bugs: wrong output field vs feature-level mismatch TypeError." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "torch.compile/SDPA crash in Bamba vs config save/load losing token IDs in Dia; unrelated." + "reason": "Torch.compile SDPA failure and DiaConfig save/load token ID loss are unrelated." }, { "left": "issue:43550", "right": "issue:44361", "accept": false, - "reason": "Bamba compile/SDPA failure vs MLuke tokenizer AttributeError; different model families and bugs." + "reason": "No shared code path: Bamba compile/SDPA vs MLukeTokenizer task AttributeError." }, { "left": "issue:44625", "right": "issue:47050", "accept": false, - "reason": "Qwen3.5 label-propagation config bug vs Canine fp16 ONNX dtype mismatch; unrelated." + "reason": "Different components and failure modes: Qwen3.5 num_labels propagation vs Canine fp16 ONNX export." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "Deepspeed/Bert zero3 init IndexError vs fbank feature extraction issue; different subsystems." + "reason": "BertModel zero3 init IndexError and fbank feature extraction issue are unrelated." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior vs AutoConfig trust_remote_code field loss; not the same defect." + "reason": "Qwen3-MoE sliding_window bug is unrelated to AutoConfig missing fields with trust_remote_code." }, { "left": "issue:43540", "right": "issue:46294", "accept": false, - "reason": "Both are Qwen video-input issues, but one is Qwen3OmniMoe video processing ValueError and the other is Qwen3.5 chat-template feature/token mismatch." + "reason": "Both involve Qwen video inputs, but one is a direct video-processing ValueError and the other is a chat-template token/features mismatch." }, { "left": "issue:45381", "right": "issue:46693", "accept": false, - "reason": "Qwen2.5-VL video position IDs bug vs FA2 long-context performance regression; different symptoms and code paths." + "reason": "Qwen2.5-VL video position IDs bug is unrelated to FA2 long-context performance regression." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "AddedToken constructor error vs failure loading a specific model tokenizer; not the same bug." + "reason": "AddedToken constructor argument error is not the same as loading cjvt/sleng-bert failing." }, { "left": "issue:45987", "right": "issue:46829", "accept": false, - "reason": "StaticCache return-type bug vs VideoPrism output field misassignment; unrelated." + "reason": "StaticCache seq-length shape bug and VideoPrism hidden_states assignment are unrelated." }, { "left": "issue:43611", "right": "issue:46665", "accept": false, - "reason": "base_model_prefix loading regression vs inconsistent from_pretrained behavior in Python classes; too broad to be the same bug." + "reason": "Both are from_pretrained/loading issues, but the concrete failure modes and call paths differ." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert vs tokenizer load failure for est-roberta; unrelated." + "reason": "Different subsystems entirely: Qwen3 auto device mapping crash vs tokenizer loading failure for a specific model." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Audio feature extraction bug vs add-new-model-like tokenizer mapping failure; different areas." + "reason": "Audio feature extraction bug and tokenizer-mapping add-new-model-like failure are unrelated." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "ImageProcessor segmentation-map bug vs custom trainer loss signature mismatch; unrelated." + "reason": "Image segmentation-map processing and CustomTrainer compute_loss signature mismatch are unrelated." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Tokenizer crash on NER/padding vs Perceiver non-default resolution image classification bug; different models and paths." + "reason": "Tokenizer NER/padding crash is unrelated to Perceiver image classification interpolation failure." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Special-token registration bug vs DAC latent/forward mismatch; no shared concrete code path." - }, - { - "left": "issue:43550", - "right": "issue:43819", - "accept": false, - "reason": "Bamba compile/SDPA issue vs DAC STE mismatch; unrelated." + "reason": "Tokenizer special-token registration bug is unrelated to DAC.from_latents STE behavior." }, { "left": "issue:45987", "right": "issue:46932", "accept": false, - "reason": "StaticCache API/type bug vs Gemma4 quantized embedder dtype casting." + "reason": "StaticCache sequence-length typing and Gemma4 quantized embedder dtype casting are unrelated." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug vs Smollm3 RoPE-layer mismatch; different model configs." + "reason": "SwitchTransformers sparse-layer config bug is unrelated to Smollm3 RoPE-layer dropping." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Different models and mechanisms: SDPA/torch.compile failure vs DAC latent conversion bug." + }, + { + "left": "issue:44488", + "right": "issue:47020", + "accept": false, + "reason": "Both are tokenizer loading issues, but one is a specific model load failure and the other is an XLMRoberta Unigram init type mismatch." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "CLIPOutput attention field regression vs RTDetrV2 output difference; unrelated." + "reason": "CLIPOutput attentions regression and RTDetrV2 differing outputs are distinct vision model bugs." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "AutoConfig missing fields with trust_remote_code vs strict config blocking granite_speech load; different mechanisms." + "reason": "Both concern config loading strictness, but they are different code paths and symptoms, not the same bug." } ] }, @@ -6077,197 +6122,200 @@ "issue:44805|issue:44936", "issue:46294|issue:46463" ], - "prepared_review_unit_hash": "0a6956c7a704b1e287541e8f6df1b1a408d2fb744ab01fd0a078a63ca027f977", + "prepared_review_unit_hash": "43b58086ed355997b7538f919012529837503d6104b02a7217f7ebb0587e9325", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35343, - "estimated_input_tokens": 8836, - "estimated_eval_tokens": 17928 + "serialized_chars": 35345, + "estimated_input_tokens": 8837, + "estimated_eval_tokens": 17930 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a false-positive mix of largely unrelated issue reports across different models, tokenizers, runtime settings, and backends. None of the soft pairs look like the same underlying bug/change closely enough to merge.", - "confidence": 0.96, - "canonical_issue_reason": null, - "canonical_pr_reason": null, - "best_issue_reason": null, - "best_pr_reason": null, + "summary": "This cluster is mostly heterogeneous Transformers issues spanning unrelated models, tokenizers, multimodal/video paths, training, and runtime config. I found one plausible duplicate pair around the XLM-Roberta / est-roberta tokenizer-loading regression; the rest should stay separate.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue for the full cluster; issue:47020 is the clearest root-cause report for the only plausible duplicate pair, while the other items are unrelated.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "issue:47020 is the most specific and actionable report; it best represents the tokenizer-loading regression seen in issue:44991.", + "best_pr_reason": "No pull requests are present, so there is no best PR candidate.", "soft_edge_verdicts": [ { "left": "issue:44008", "right": "issue:46858", "accept": false, - "reason": "Different failures: Gemma3n forward name collision vs static-cache sequence length arithmetic on non-CUDA backends." + "reason": "Different failures in different code paths: Gemma3n forward variable collision vs static-cache slice arithmetic/device tensor issue." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "Unrelated areas: loading an AutoImageProcessor from URL vs a template compilation TypeError." + "reason": "Unrelated: loading AutoImageProcessor from URL vs template-node compilation error." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "Different bugs: BarkProcessor voice preset handling vs transformers GenerationMixin import path breakage." + "reason": "Different subsystems and symptoms: Bark voice_preset vs GenerationMixin import breakage." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Both involve tokenizer patching, but one is offline model-info access and the other is a backend_tokenizer attribute crash in Mistral regex fixing." + "reason": "Both touch mistral regex handling, but one is offline model_info access and the other is a backend_tokenizer attribute crash; distinct bugs." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "Doc-example SigLIP2 mismatches/quantization issues are not the same as Kosmos2.5 long OCR input indexing." + "reason": "SigLIP2 example issues and Kosmos2.5 long-ocr index error are unrelated model-specific bugs." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 NER/padding crash is unrelated to AutoTokenizer loading FastSpeech2ConformerTokenizer." + "reason": "Tokenizer crash on NER/padding vs AutoTokenizer failing to load a different tokenizer class." }, { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Different model paths and symptoms: Qwen3OmniMoe video processing vs Gemma3n audio_mel_mask attribute collision." + "reason": "Different multimodal paths: Qwen video processing ValueError vs Gemma3n audio-mask attribute collision." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Pegasus output drift and CamemBERT masked-LM predictions are separate model-specific regressions." + "reason": "Both are v5 output regressions, but in different models and code paths; not the same bug." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription handling after special-token alignment is unrelated to Kosmos2.5 OCR input length handling." + "reason": "Whisper empty-transcription generation failure is unrelated to Kosmos2.5 OCR indexing." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Input validation error messaging and single-label zero-loss classification are different code paths." + "reason": "Validation/error-message bug versus classification-loss bug; different underlying issues." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Llava Next loading segfault and MPS histogram integer support are entirely different backends/problems." + "reason": "Segfault on Llava Next loading is unrelated to MPS histogram type support." }, { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Non-persistent buffer initialization junk and FSDP CPU RAM efficient loading are not the same bug." + "reason": "Non-persistent buffer junk fill and FSDP CPU RAM-efficient loading are different loading/runtime problems." }, { "left": "issue:44991", "right": "issue:47020", - "accept": false, - "reason": "Both are tokenizer-loading failures, but the model families and concrete causes differ (est-roberta load failure vs Unigram expects Sequence)." + "accept": true, + "reason": "Likely the same tokenizer-loading regression: est-roberta fails to load because XLMRobertaTokenizer passes a dict to Unigram expecting a sequence." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "Device mismatch under CPU offload is not the same as bfloat16 dtype mismatch in different models." + "reason": "Both involve dtype/device mismatches, but for different models, backends, and concrete failures." }, { "left": "issue:44991", "right": "issue:46291", "accept": false, - "reason": "Tokenizer loading regression and kernels version/revision import failure are unrelated." + "reason": "Tokenizer load failure for one model is not the same as pipelines import breaking on kernels v0.15." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Both mention remote_code breakage, but one is specifically the removed is_torch_fx_available API and the other is a broader regression claim." + "reason": "Both concern trust-remote-code / recent-version breakage, but the specific removed helper and the broader remote-code regression are not the same bug." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Post-processor updates for special tokens and Whisper batch_decode skip_special_tokens behavior are different tokenizer/decoder issues." + "reason": "Special-token post-processor update bug is unrelated to Whisper batch_decode skip_special_tokens handling." }, { "left": "issue:43577", "right": "issue:43818", "accept": false, - "reason": "BLIP2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing defects are unrelated." + "reason": "Blip2 dtype propagation and Video-LLaVA temporal-attention/model-weight issues are unrelated." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "torch.compile with SDPA on Bamba and bfloat16 dtype mismatches in other models are not the same concrete bug." + "reason": "Both are compile/dtype-related, but the failing models and code paths differ." }, { "left": "issue:43582", "right": "issue:46512", "accept": false, - "reason": "Apple Silicon warmup TypeError and stale config.dtype after from_config(dtype=...) are separate dtype/config issues." + "reason": "AppleSilicon warmup TypeError and stale config.dtype after from_config are separate dtype/config bugs." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Qwen video-input regression and Kosmos2.5 OCR index error have no clear shared root cause." + "reason": "Qwen video regression and Kosmos2.5 OCR long-input indexing are unrelated." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Qwen3VL weight_only loading failure and Qwen3.5 flash-attention generation failure are distinct model/runtime problems." + "reason": "From_pretrained weight_only error in Qwen3VL is unrelated to Qwen3.5 flash-attention generation failure." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "An error-message wording regression in pipeline() is not the same as SigLIP2 producing nonsensical outputs." + "reason": "Pipeline translation-message regression and SigLIP2 output nonsense are different issues." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "gather_for_metrics batch-label truncation and Mamba2Mixer use_cache correctness are unrelated training/inference bugs." + "reason": "Metrics batch-label truncation and Mamba2 use_cache incorrect results are unrelated code paths." }, { "left": "issue:44805", "right": "issue:44936", "accept": false, - "reason": "Mask-shape IndexError and trainer.evaluate() failing after train are different symptoms with no shared concrete fix evident." + "reason": "Mask-shape IndexError and trainer.evaluate-after-train failure are not the same bug." }, { "left": "issue:46294", "right": "issue:46463", "accept": false, - "reason": "Qwen3.5 video chat-template token alignment and T5 decoder-input validation are unrelated issues." + "reason": "Qwen video chat-template mismatch and T5 decoder-input error-message bug are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: it correctly treats the set as mostly heterogeneous and only tentatively groups 44991 with 47020 as a plausible tokenizer-loading duplicate. The soft-edge verdicts are conservative, and there are no pull requests to evaluate. No obvious overstatement detected." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -6348,8 +6396,8 @@ "issue:42175|issue:45081", "issue:44671|issue:45701", "issue:45003|issue:46285", - "issue:46697|issue:46728", "issue:46175|issue:46932", + "issue:46697|issue:46728", "issue:43262|issue:43819", "issue:43582|issue:44857", "issue:43782|issue:45325", @@ -6362,203 +6410,206 @@ "issue:36010|issue:36331", "issue:43257|issue:43329" ], - "prepared_review_unit_hash": "0be1f359aff00def030914555dd3d22194d2e46ca10b6cf08f7b064c73a0873d", + "prepared_review_unit_hash": "1e8589e7632c878c9d1e6ad2348b350064bcf1ea02918ca7269101d5695930a8", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36244, - "estimated_input_tokens": 9061, - "estimated_eval_tokens": 18378 + "serialized_chars": 36248, + "estimated_input_tokens": 9062, + "estimated_eval_tokens": 18380 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a false-positive grouping of unrelated issues: tokenizer quirks, model-loading failures, multimodal processor bugs, dtype/quantization regressions, and training API incompatibilities. None of the soft pairs look like the same underlying bug or change, so I reject them all.", - "confidence": 0.84, - "canonical_issue_reason": "issue #45020 is the broadest umbrella-style report in the set, covering general version-regression / remote_code breakage; among these otherwise heterogeneous issues, it is the most representative central issue.", + "summary": "The set is mostly heterogeneous. The only clear duplicate-style pair is the tokenizer spacing bug versus the slower/faster tokenizer inconsistency report; the rest are distinct model, loading, dtype, or docs issues and should stay separate.", + "confidence": 0.74, + "canonical_issue_reason": "Issue #29159 is the broader formulation of the tokenizer spacing/inconsistency bug; #28218 reads like a narrower symptom of the same underlying behavior.", "canonical_pr_reason": null, - "best_issue_reason": "#45020 is the best single issue to anchor the cluster because it is the most general and cross-cutting of the reports, whereas the others are highly model- or subsystem-specific.", + "best_issue_reason": "Issue #29159 is the best duplicate target among the issues because it captures the shared tokenizer behavior more generally than #28218.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43931", "right": "issue:46846", "accept": false, - "reason": "Both are model-loading shape errors, but for different model families and different root causes (Qwen3-VL weight mismatch vs Qwen3.5 TP plan omission)." + "reason": "Different failures: Qwen3-VL weight-shape mismatch vs Qwen3.5 TP planning/linear_attn OOM-shape errors." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Different models and different failures: image preprocessor/model dtype mismatch vs a Gemma3n forward variable-name collision causing an AttributeError." + "reason": "Both are multimodal loading bugs, but the concrete code paths differ: image preprocessor/model dtype mismatch vs Gemma 3n audio tensor name collision." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both involve video inputs, but they describe different Qwen models and different bugs (processing ValueError vs incorrect vision_position_ids scaling)." + "reason": "Both mention video inputs, but they target different models and likely different bugs; not the same concrete change." }, { "left": "issue:28218", "right": "issue:29159", - "accept": false, - "reason": "Both are tokenizer issues, but one is about added-token spacing and the other is a broad slow-vs-fast inconsistency; not enough evidence they are the same defect." + "accept": true, + "reason": "Both describe the same tokenizer added-token spacing inconsistency; #29159 is the broader duplicate target." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Completely different subsystems: image size mismatch in OmDet-Turbo vs dtype mismatches in SwitchTransformers/TimmWrapperModel." + "reason": "Unrelated subsystems: OmDet image-size mismatch vs bf16 dtype mismatches in different models." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "Both concern loading, but the concrete failures differ: broken FSDP CPU RAM-efficient loading vs an unexpected _is_hf_initialized argument in init_empty_weights." + "reason": "Both involve loading infrastructure, but one is broken CPU RAM-efficient loading and the other is an unexpected init_empty_weights argument." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "Different bug classes: unpacking packed BitNet weights during accelerate loading vs inference dtype mismatches." + "reason": "Different issues: packed-weight unpacking during accelerate loading vs dtype mismatch inference." }, { "left": "issue:43645", "right": "issue:46665", "accept": false, - "reason": "Both mention from_pretrained / custom model initialization, but the reported behaviors and environments are different enough that they do not look like the same bug." + "reason": "Both concern loading custom models, but the reported environments and failure modes differ too much to treat as the same bug." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Related to v5 pipeline removals, but one is an error message suggesting unsupported translation tasks and the other is docs still referencing pipeline(); documentation/message issues, not the same bug." + "reason": "One is an error-message regression, the other is a documentation cleanup; related topic but not the same artifact or bug." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "Unrelated issues: BarkProcessor voice_preset failure vs CustomTrainer.compute_loss signature incompatibility." + "reason": "Completely different areas: BarkProcessor voice_preset vs Trainer compute_loss signature change." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "Different domains: Blip2 dtype propagation vs an Apple Silicon caching_allocator_warmup TypeError." + "reason": "Different failure modes and platforms: BLIP2 dtype loading vs Apple Silicon allocator warmup TypeError." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "No shared underlying change; one is backend dependency installation, the other is a tokenizer regex patch crash." + "reason": "No common code path: extras/install backend packaging vs Mistral regex patching during tokenizer load." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "Both mention tokenization/model behavior changes across versions, but one is CamemBERT masked-LM prediction drift and the other is a general tokenization change report." + "reason": "The titles are related to tokenization, but one is a specific CamemBERT MLM regression and the other is a broad version-to-version tokenization change report." }, { "left": "issue:45003", "right": "issue:46285", "accept": false, - "reason": "Different problems: unsafe sys.modules access in modeling_utils vs CLIP text-model structure inconsistency." + "reason": "Different problems: unsafe sys.modules access in modeling_utils vs CLIPTextModel structure divergence." }, { - "left": "issue:46697", - "right": "issue:46728", + "left": "issue:46175", + "right": "issue:46932", "accept": false, - "reason": "Both are image-processor sizing/padding bugs, but for different models and different calculations (over-padding vs wrong patch count)." + "reason": "Both touch Gemma4 vision/audio embedders, but one is a huge one-hot memory issue and the other is quantized dtype casting." }, { - "left": "issue:46175", - "right": "issue:46932", + "left": "issue:46697", + "right": "issue:46728", "accept": false, - "reason": "Same model family (Gemma4) but different bugs: one is a huge one-hot position-embedding materialization, the other is dtype casting for quantized embedders." + "reason": "Both are image-shape/count calculations, but for different processors and distinct bugs." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Different audio-related bugs: apply_chat_template default sampling rate vs DAC.from_latents/STE mismatch." + "reason": "Audio chat-template sampling-rate defaulting is unrelated to DAC latent/STE behavior." }, { "left": "issue:43582", "right": "issue:44857", "accept": false, - "reason": "Different failure modes and code paths: allocator warmup TypeError on Apple Silicon vs float16 AMP crash in LwDetrImageLoss." + "reason": "Apple Silicon allocator warmup and CUDA AMP image-loss crashes are unrelated." }, { "left": "issue:43782", "right": "issue:45325", "accept": false, - "reason": "Different Qwen-VL issues: a weight_only loading error vs an incorrect temporal position-id scaling algorithm." + "reason": "Different Qwen-VL issues: weight_only loading error vs rope-index / temporal position scaling bug." }, { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "No shared code path: BigBirdTokenizer special-token registration bug vs MXFP4 quantization fallback/loading issue." + "reason": "Tokenizer special-token registration is unrelated to MXFP4 quantization fallback." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Unrelated: missing Float8 storage object vs PIL backend image processors incorrectly requiring torchvision." + "reason": "Float8 storage lookup error vs PIL/torchvision dependency regression are unrelated." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "A specific MobileLLM key error is not the same underlying bug as the broad remote_code regression reported in #45020." + "reason": "A specific key error while loading one model is not the same as the broader remote_code breakage report." }, { "left": "issue:45372", "right": "issue:46326", "accept": false, - "reason": "Both involve mistral_common, but one is an import error during Gemma 4 processor loading and the other is continue_final_message behavior for tokenizers based on mistral_common." + "reason": "Both involve mistral_common, but one is an import error for Gemma 4 loading and the other is a tokenizer continue_final_message behavior bug." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Different issues: torch.compile + SDPA failure in Bamba-9B-v2 vs MXFP4 quantization loading fallback." + "reason": "torch.compile/SDPA failure is unrelated to quantization loading fallback." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "Different failure modes: missing GenerationMixin import vs notebook-defined custom model initialization." + "reason": "Import path regression for GenerationMixin vs Jupyter custom-model initialization failure are different bugs." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "No common bug: GenerationMixin import failure vs custom trainer loss-signature mismatch." + "reason": "One is an import regression, the other is a Trainer API signature change." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Different code paths and symptoms: Qwen3 MOE weight conversion under accelerate/deepspeed vs a multimodal-token counting video-branch bug." + "reason": "Qwen3 MoE loading/conversion and multimodal token counting are separate bugs." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is mostly conservative and grounded in the packet. The only accepted pair, #28218 and #29159, is a plausible tokenizer-duplicate match based on the titles, and the rest of the soft-edge pairs are reasonably rejected as distinct bugs/subsystems. No mergeability issues for PRs arise here since these are issue pairs only." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -6653,196 +6704,196 @@ "issue:42915|issue:43742", "issue:34567|issue:40990" ], - "prepared_review_unit_hash": "e89bb7e8e0d378d787020fed7ef1a73aec06a41e6b07ca0e9510b2cf2defe049", + "prepared_review_unit_hash": "f0b90a491c32a4721a35bc54e417bc00b5f9a5c46357ab44e05aaa56c7938b40", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36095, - "estimated_input_tokens": 9024, - "estimated_eval_tokens": 18304 + "serialized_chars": 36100, + "estimated_input_tokens": 9025, + "estimated_eval_tokens": 18306 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a duplicate set: it contains many unrelated Transformers bug reports spanning model loading, tokenizers, processors, quantization, training, and video/audio paths. The soft pairs mostly share a broad subsystem or symptom category, but not the same underlying bug or change.", - "confidence": 0.91, - "canonical_issue_reason": null, + "summary": "This cluster is heterogeneous: the issues cover unrelated models, loaders, processors, training, quantization, and backend/dependency bugs. None of the soft pairs look like the same underlying bug/change.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue: the items are not duplicates of one another, so there is no single issue to preserve as the cluster representative.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 44479 is the clearest self-contained regression report in the set, with a specific symptom and affected-model list, but it is still unrelated to the other items.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43906", "right": "issue:46209", "accept": false, - "reason": "Different problems: 43906 is an isolated reproduction of a prior issue, while 46209 is an FP8/B200 inference failure. No shared code path or fix target." + "reason": "Different problems: an isolated reproduction of an older issue versus an FP8 inference failure on B200." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BEiTImageProcessorFast reduce_labels bug vs RTDetrV2 output regression. Different components and symptoms." + "reason": "Unrelated models and symptoms: BEiT Fast label reduction versus RTDetrV2 output differences in v5." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to PIL backend requiring torchvision." + "reason": "Different code paths: Qwen3-MoE sliding-window behavior versus PIL image processors requiring torchvision." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "gguf_file breaks tokenizer/model loading; est-roberta tokenizer loading failure is a separate model-specific regression." + "reason": "Both are loading/tokenizer issues, but they affect different APIs, models, and failure modes." }, { "left": "issue:42175", "right": "issue:46291", "accept": false, - "reason": "One is an installation/backend dependency issue; the other is a pipelines import failure from kernels version/revision requirements." + "reason": "Different dependency regressions: missing TensorFlow backend on install versus kernels import requiring version/revision." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Zero3/non-pretrained Bert index error is a training setup bug; trainer.evaluate() after train() is a different lifecycle issue." + "reason": "Training-related bugs, but one is a DeepSpeed ZeRO3 Bert index error and the other is an evaluate-after-train failure." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "DAC.from_latents/STE mismatch is unrelated to DiaConfig losing custom token IDs on save/load." + "reason": "Different models and bugs: DAC forward/from_latents mismatch versus DiaConfig losing custom token IDs after save/load." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Both are attribute-related loading failures, but they affect different objects and code paths: Mistral regex patching vs Gemma4Processor in serve." + "reason": "Both involve tokenizer/processor loading, but one is a Mistral regex patch crash and the other is a Gemma4Processor attribute error." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Both mention tied-weights warnings in Qwen models, but they involve different model families and likely different config handling bugs." + "reason": "Related to tied-weights warnings, but the underlying model/config paths are different and not the same bug." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Video token counting undefined-variable bug is not the same as stale auto_mappings references to removed configs." + "reason": "Different subsystems: multimodal video token counting versus stale auto_mappings references breaking CI." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template compilation TypeError and CLIPTokenizer model_max_length are unrelated issues." + "reason": "Unrelated bugs: template compilation error versus CLIPTokenizer model_max_length handling." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "BEiT label reduction bug has no overlap with the MPS attention correctness issue." + "reason": "Different components entirely: label reduction in BEiT versus an MPS attention correctness issue." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Trailing-slash local model import warning is unrelated to Qwen3 MOE weight conversion with accelerate+deepspeed." + "reason": "Different loading failures: trailing-slash import warning versus Qwen3 MoE weight conversion under accelerate+deepspeed." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "PEFT local-checkpoint loading failure is unrelated to Whisper batch_decode ignoring skip_special_tokens." + "reason": "Different functionality: local PEFT checkpoint loading versus Whisper batch_decode skip_special_tokens handling." }, { "left": "issue:45859", "right": "issue:47050", "accept": false, - "reason": "CPU offload compatibility in Qwen3_5MoeTextRotaryEmbedding is unrelated to CanineModel ONNX dtype mismatch." + "reason": "Different backends and failures: CPU offload incompatibility versus fp16 ONNX export dtype mismatch." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "AutoConfig missing fields with trust_remote_code is a different config-loading bug than AutoModelForCausalLM.from_config not unwrapping text_config." + "reason": "Both are config/loading related, but one is trust_remote_code missing fields and the other is composite multimodal config unwrapping." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading failure and Qwen3.5 flash-attention generation bug are different models and different failures." + "reason": "Different models and symptoms: GLM-4.7-Flash unit-test loading failure versus Qwen3.5 flash-attention generation issues." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to OmDet-Turbo processor/model input-size mismatch." + "reason": "Both involve model/processor mismatch-like symptoms, but they are for different models and different concrete failures." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio apply_chat_template sample-rate default bug is separate from Whisper batch_decode skip_special_tokens handling." + "reason": "Different audio-processing bugs: apply_chat_template sample-rate default versus Whisper batch_decode token skipping." }, { "left": "issue:39692", "right": "issue:44912", "accept": false, - "reason": "SigLIP2 doc example errors and gpt-oss-20b MXFP4 quantization fallback are unrelated." + "reason": "Different quantization/documentation issues: SigLIP2 doc errors versus GPT-oss MXFP4 loading fallback behavior." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "Qwen3 MOE weight conversion with accelerate/deepspeed is unrelated to GPT-oss-20b CUDA OOM." + "reason": "Different failure classes: Qwen3 MoE weight conversion versus GPT-oss-20b CUDA out-of-memory." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Both involve video processing, but one is undefined variables in multimodal token counting and the other is batched output shape handling; not the same bug." + "reason": "Both are video-related, but one is multimodal token counting and the other is batched video processor shape handling." }, { "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "Qwen3 MOE conversion issue does not match Qwen3-VL weight-shape mismatch during loading." + "reason": "Different model-loading problems: MoE weight conversion under distributed loading versus Qwen3-VL weight shape mismatch." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "General video regression in v5.3.0 across Qwen models is broader and distinct from the specific undefined-variable bug in _get_num_multimodal_tokens." + "reason": "Related to video paths, but 43329 is a specific undefined-helper bug in token counting while 44479 is a broader video-input regression." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Int4 quantization reserved-memory OOM is a different problem from Sam3Video OOM." + "reason": "Different OOM causes: quantization-related CUDA reserved memory growth versus Sam3Video model memory exhaustion." }, { "left": "issue:42915", "right": "issue:43742", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8Config failure and MobileLLM key error are unrelated model-loading issues." + "reason": "Different loading errors: FineGrainedFP8Config failure in Qwen3Moe versus a key error loading MobileLLM-125M." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "TrainerState num_input_tokens_seen not updating is unrelated to high perplexity on gpt-oss-20b." + "reason": "Completely different areas: TrainerState token-count tracking versus an inference perplexity report on GPT-oss-20b." } ] }, @@ -6937,195 +6988,189 @@ "issue:44857|issue:46832", "issue:43697|issue:45357", "issue:45137|issue:45405", - "issue:28282|issue:29127", "issue:43404|issue:45127", + "issue:28282|issue:29127", "issue:45081|issue:47020", "issue:43742|issue:44291" ], - "prepared_review_unit_hash": "1c54988cbe742172f576cdb5c20b00e7e64ee5b95cf8cb0b2807c3bfafa77d99", + "prepared_review_unit_hash": "dcc228d062d03cf0fbbdd56f4bc9bce0218d67ad90ecb14e2e188bb336ceb5f8", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35108, - "estimated_input_tokens": 8777, - "estimated_eval_tokens": 17810 + "serialized_chars": 35117, + "estimated_input_tokens": 8780, + "estimated_eval_tokens": 17816 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of superficially similar but substantively different Transformers bug reports; I would not merge any of the soft pairs. The most reasonable anchor issue is the clear, user-facing import failure in 43824, but even that is not a duplicate of the others.", - "confidence": 0.69, - "canonical_issue_reason": "43824 is the strongest canonical anchor: it is a concrete, reproducible import failure with inbound references and broad user impact, whereas the other candidates are more model-, tokenizer-, or subsystem-specific.", + "summary": "This cluster is a broad semantic false-positive: the items span unrelated bugs in model loading, tokenization, cache behavior, vision/text models, training utilities, and dependency packaging. I found no pair that looks like the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue stands out; the set is heterogeneous and the apparent similarities are mostly broad subsystem overlap rather than the same defect.", "canonical_pr_reason": null, - "best_issue_reason": "43824 is the best overall representative of this mixed cluster because it is the most self-contained and actionable bug report; the rest are too diverse to serve as better duplicates.", + "best_issue_reason": "No issue is a strong global representative because the cluster does not share one concrete code path or user-facing failure mode.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "Different failure modes: init_empty_weights/_is_hf_initialized argument handling vs PIL image processor dependency requirements." + "reason": "Different failures: init_empty_weights argument handling vs PIL image processor dependency requirements." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Unrelated problems: FineGrainedFP8Config loading failure vs loading AutoImageProcessor from a URL." + "reason": "Unrelated areas: FineGrainedFP8Config failure vs loading AutoImageProcessor from URL." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Tokenizer fast/slow inconsistency and TrainerState token-count tracking are separate components with different bugs." + "reason": "Tokenizer inconsistency and TrainerState token counting are separate bugs." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve error handling, but one is protobuf decode handling in tokenizers and the other is a bare except in Fuyu tensor conversion; not the same bug." + "reason": "Both involve exception handling, but they are different code paths and different failure modes." }, { "left": "issue:46421", "right": "issue:46424", "accept": false, - "reason": "Both mention cache, but one is lazy-init recompilation and the other is max_cache_length behavior for Quantized Cache; different code paths." + "reason": "Both concern cache behavior, but one is recompilation with chunked prefill and the other is max_cache_length semantics." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "A vague data request is not the same as an isolated reproduction of a specific prior issue." + "reason": "A placeholder title vs an isolated reproduction of another issue; no shared bug." }, { "left": "issue:28282", "right": "issue:29870", "accept": false, - "reason": "Missing PyTorch import dependency and BART generate() decoder_attention_mask UnboundLocalError are unrelated." + "reason": "PyTorch import error and BART generate() UnboundLocalError are unrelated." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "Both involve memory/loading, but FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different root causes." + "reason": "FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different problems." }, { "left": "issue:43525", "right": "issue:46291", "accept": false, - "reason": "Llama4Config pad_token_id missing is unrelated to kernels v0.15 import-time version/revision validation." - }, - { - "left": "issue:28282", - "right": "issue:34689", - "accept": false, - "reason": "PyTorch-not-found import error and Llama 3.2 vision model loading regression are different issues." + "reason": "Missing pad_token_id on Llama4Config is unrelated to pipelines import failing on kernels version/revision." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "lm_head weight tying in AyaVision and BLIP2 dtype loading staying float32 are distinct bugs." + "reason": "Different model-loading bugs: lm_head tying vs dtype propagation in BLIP-2." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "Model load failure for physical-intelligence/fast and CLIPTokenizer model_max_length misuse are unrelated tokenizer/model issues." + "reason": "Loading a specific model and CLIPTokenizer max_length behavior are not the same defect." }, { "left": "issue:43540", "right": "issue:46701", "accept": false, - "reason": "Different multimodal pipelines and different errors: Qwen3OmniMoe video processing vs DiffusionGemma image-text-to-text output handling." + "reason": "Video input validation in Qwen3OmniMoe is unrelated to DiffusionGemma pipeline output handling." }, { "left": "issue:43425", "right": "issue:46291", "accept": false, - "reason": "Torch 2.10 incompatibility and kernels import failure involve different dependencies and failure mechanisms." + "reason": "Torch compatibility and kernels import requirements are separate dependency issues." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "DiaConfig save/load token-ID loss is not the same as MXFP4 quantization falling back to bf16." + "reason": "Save/load token ID loss in DiaConfig is unrelated to MXFP4 quantization fallback behavior." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and different tokenizer behaviors/codecs." + "reason": "Both are tokenizer regressions, but they affect different models and different underlying logic." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "MacOS bus error in CLIP and TrainerState.num_input_tokens_seen not updating are unrelated." + "reason": "CLIP bus error on macOS and TrainerState token counting are not the same issue." }, { "left": "issue:39379", "right": "issue:44554", "accept": false, - "reason": "Both touch attention/masking correctness, but they are model- and backend-specific with different underlying conditions." + "reason": "Different correctness bugs in different attention/multimodal paths." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Missing TensorFlow in a torch-only install and missing Qwen2.5-VL export are different packaging/import issues." + "reason": "Packaging backend composition issue vs missing exported model class import error." }, { "left": "issue:44857", "right": "issue:46832", "accept": false, - "reason": "LwDetrImageLoss AMP crash and RT-DETR forward TypeError are separate detection-model failures." + "reason": "Both are RT-DETR-family, but one is an AMP loss crash and the other is a forward-shape TypeError." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RTDetrV2 output drift in v5 and Qwen3.5 save_pretrained key regression are unrelated." + "reason": "RTDetrV2 output drift and Qwen3.5 visual encoder key save_pretrained regression are unrelated." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 deque error and a MIN_PEFT_VERSION bump are not the same bug." + "reason": "DeepSpeed ZeRO3 deque error and PEFT version pinning are unrelated." }, { - "left": "issue:28282", - "right": "issue:29127", + "left": "issue:43404", + "right": "issue:45127", "accept": false, - "reason": "A missing PyTorch dependency and LayoutLMv3 box-content error messaging are unrelated." + "reason": "Both mention tied weights, but one is a missing weight tying bug and the other is LoRA merge collapse with extended vocab." }, { - "left": "issue:43404", - "right": "issue:45127", + "left": "issue:28282", + "right": "issue:29127", "accept": false, - "reason": "Untied lm_head weights in Mistral3ForConditionalGeneration and LoRA merge collapse with extended vocabulary are different code-path problems." + "reason": "ImportError for PyTorch and a LayoutLMv3 error-message clarity issue are unrelated." }, { "left": "issue:45081", "right": "issue:47020", "accept": false, - "reason": "Mistral regex patch crashing on tokenizer internals and XLMRobertaTokenizer passing a dict to Unigram are separate tokenizer construction bugs." + "reason": "Mistral regex patch crash and XLMRobertaTokenizer/Unigram constructor misuse are different tokenizer code paths." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "KeyError loading MobileLLM-125M and unexpected _is_hf_initialized during init_empty_weights are different model-loading failures." + "reason": "Key error loading MobileLLM and init_empty_weights argument handling are separate model-loading failures." } ] }, @@ -7213,8 +7258,8 @@ "issue:42175|issue:44488", "issue:43540|issue:44560", "issue:44987|issue:45907", - "issue:38175|issue:44898", "issue:43540|issue:44479", + "issue:38175|issue:44898", "issue:44479|issue:45381", "issue:44291|issue:44821", "issue:34567|issue:36331", @@ -7229,20 +7274,20 @@ "issue:43756|issue:46399", "issue:43756|issue:45325" ], - "prepared_review_unit_hash": "b69b96f72202c4e04a6bb8dcb32e2f7d4ca23bb03fbd68e0c868e22f3d29044e", + "prepared_review_unit_hash": "0ba116de75251b9e2c97f0b582a01b1371991cd92389278f965e5a4d271ec9f4", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36059, + "serialized_chars": 36058, "estimated_input_tokens": 9015, "estimated_eval_tokens": 18286 }, @@ -7251,202 +7296,199 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated closed issues with a few weak thematic overlaps; the only soft edge that looks like a true duplicate-style match is the Qwen video-position-id regression pair 44479/45381.", - "confidence": 0.7, - "canonical_issue_reason": "issue:44479 is the broadest Qwen video-input regression report in the packet, covering multiple affected Qwen variants; issue:43540 predates it but appears narrower and model-specific, so 44479 is the better umbrella canonical issue.", + "summary": "The cluster is heterogeneous and does not collapse cleanly into a single bug/theme. All soft edges look like false positives: some are broadly related areas (Qwen video regressions, tokenizer regressions, trainer API changes), but they describe different concrete failures and fixes.", + "confidence": 0.93, + "canonical_issue_reason": "No canonical issue: the items span unrelated bugs across training, tokenization, multimodal models, packaging, and backend/runtime errors.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the best representative issue because it captures the shared Qwen video-input regression across several variants rather than a narrower single-model manifestation.", + "best_issue_reason": "issue:44479 is the broadest representative of the Qwen video-regression subset, but it still does not represent the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44936", "right": "issue:46821", "accept": false, - "reason": "Different subsystems and failures: trainer lifecycle bug vs TimesFM window_size AttributeError." + "reason": "Different failures and code paths: trainer evaluate-after-train behavior vs TimesFM window_size attribute handling." }, { "left": "issue:44479", "right": "issue:46693", "accept": false, - "reason": "Both mention Qwen/transformers regressions, but one is a video-input correctness bug and the other is a long-context FA2 performance regression." + "reason": "Both mention performance/regression, but one is a Qwen video-input bug and the other is an FA2 long-context slowdown." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "Unrelated: Qwen3-MoE FP8 loading failure vs CLIPTokenizer max_length constant bug." + "reason": "Unrelated: Qwen3 MoE FP8 loading vs CLIP tokenizer max_length." }, { "left": "issue:45538", "right": "issue:46869", "accept": false, - "reason": "Tokenizer constant bug vs YOLO multi-GPU device-map failure; no shared underlying code path." + "reason": "Different subsystems: tokenizer max_length vs YOLO multi-GPU device-map failure." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "AutoTokenizer/AutoModel gguf loading issue vs Apple Silicon allocator warmup TypeError." + "reason": "Unrelated runtime issues: gguf/AutoModel loading vs Apple Silicon allocator warmup TypeError." }, { "left": "issue:24643", "right": "issue:29159", "accept": false, - "reason": "DeepSpeed training RuntimeError and tokenizer slow/fast inconsistency are different bugs." + "reason": "Different areas: DeepSpeed weight shape error vs tokenizer slow/fast inconsistency." }, { "left": "issue:27957", "right": "issue:28282", "accept": false, - "reason": "Flash Attention 2 model-specific issue vs missing PyTorch ImportError; not the same problem." + "reason": "FA2/XLM-RoBERTa model behavior is unrelated to missing PyTorch ImportError." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-related, but one is an AddedToken constructor error and the other is a model-specific tokenizer load failure; too indirect to merge." + "reason": "Both are tokenizer-related, but one is an AddedToken argument conflict and the other is a model-specific tokenizer loading failure." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Different loading paths and symptoms: FP8 config failure vs FSDP CPU RAM efficient loading regression." + "reason": "Different bugs: Qwen3 MoE FP8 config vs FSDP CPU RAM efficient loading." }, { "left": "issue:43577", "right": "issue:46512", "accept": false, - "reason": "Both touch dtype handling, but one is a BLIP2 loading cast issue and the other is stale config metadata after from_config; not the same bug." + "reason": "Not the same bug: one leaves model/qformer weights in float32, the other leaves config.dtype stale after from_config(dtype=...)." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 hidden-states bug and a tokenizer-version regression are unrelated." + "reason": "Different topics: Qwen3.5 hidden_states behavior vs version-dependent tokenization changes." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Auto device-map CUDA assert vs Qwen3OmniMoe video processing failure; different code paths." + "reason": "Different failure modes and models: auto device-map CUDA assert vs Qwen3OmniMoe video input ValueError." }, { "left": "issue:42175", "right": "issue:44488", "accept": false, - "reason": "Packaging/backend dependency issue vs model loading failure for a specific checkpoint." + "reason": "Packaging/backend omission is unrelated to a model loading failure." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Both are video-input issues in Qwen variants, but the models and reported failures differ too much to treat as the same concrete bug." + "reason": "Both are video-input Qwen issues, but they affect different model families and report different concrete errors." }, { "left": "issue:44987", "right": "issue:45907", "accept": false, - "reason": "Model loading failure vs hub API revision-forwarding CI break; unrelated." + "reason": "Unrelated: model loading regression vs repo file listing not forwarding revision." }, { - "left": "issue:38175", - "right": "issue:44898", + "left": "issue:43540", + "right": "issue:44479", "accept": false, - "reason": "Different vision-model failures: zero probabilities in SigLIP2 vs Perceiver image classification on non-default resolution." + "reason": "Broadly similar Qwen video regressions, but not the same concrete bug or fix target." }, { - "left": "issue:43540", - "right": "issue:44479", + "left": "issue:38175", + "right": "issue:44898", "accept": false, - "reason": "Broad thematic overlap on Qwen video inputs, but the issue details and affected model families are distinct enough that this is not clearly one bug." + "reason": "Different vision-model bugs: zero probabilities vs Perceiver non-default resolution failure." }, { "left": "issue:44479", "right": "issue:45381", - "accept": true, - "reason": "Both describe the same Qwen2.5-VL / Qwen video-input position-id regression; 45381 is a narrower manifestation of the same underlying code-path bug." + "accept": false, + "reason": "Related Qwen2.5-VL video area, but one is a broad regression report and the other is a specific vision_position_ids symptom; not enough to treat as one bug." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "Unexpected _is_hf_initialized argument during init_empty_weights is unrelated to loading AutoImageProcessor from URL." + "reason": "Unrelated: init_empty_weights argument breakage vs AutoImageProcessor URL loading." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token-count tracking and CustomTrainer.compute_loss signature mismatch are separate trainer bugs." + "reason": "Both are trainer-related, but they are different API regressions with different impacted call sites." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Janus image-generation test failure and Phi-4 multimodal model loading issue are not the same concrete problem." + "reason": "Different multimodal model issues: Janus image-generation test failure vs Phi-4 multimodal loading failure." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Different tokenizer regressions with different mechanisms and models; too broad to merge." + "reason": "Both are tokenizer regressions, but they affect different tokenizer classes and different concrete behaviors." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed vs Smollm3 RoPE-layer dropping are unrelated model bugs." + "reason": "Unrelated model-specific bugs: Qwen3 MoE weight conversion vs Smollm3 RoPE layer dropping." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "AddedToken special-arg duplication vs Mistral regex patch crash; separate tokenizer internals." + "reason": "Different tokenizer bugs and different failure points." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image-processor dependency regression vs tokenizer regex patch crash; different layers." + "reason": "Image-processor torchvision dependency issue is unrelated to Mistral tokenizer regex patching." }, { "left": "issue:43582", "right": "issue:46821", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError is unrelated to TimesFM window_size AttributeError." + "reason": "Different runtime areas: allocator warmup on Apple Silicon vs TimesFM window_size handling." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "MobileLLM loading key error vs missing Float8 storage type; different failure modes." + "reason": "Different loading errors: MobileLLM key error vs missing Float8 storage object." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding_window bug is unrelated to loading AutoImageProcessor from URL." + "reason": "Qwen3-MoE sliding_window bug is unrelated to AutoImageProcessor URL loading." }, { "left": "issue:43756", "right": "issue:46399", "accept": false, - "reason": "Smollm3 RoPE-layer truncation and Qwen3.5 kernel registration failure are separate issues." + "reason": "Different model-specific regressions: Smollm3 RoPE layers vs Qwen3.5 kernel registration crash." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen-related but concern different model families and different rotary-position behavior; not enough to call them the same bug." + "reason": "Both touch rotary/position logic, but they are different models and different erroneous behaviors." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The analysis is mostly conservative, but the canonical issue rationale overstates the evidence: issue:44479 is not the earliest concrete regression report in the packet, since issue:43540 predates it. You can still argue 44479 is the broadest Qwen video-input report, but the \u2018earliest\u2019/\u2018subsumes follow-up reports\u2019 phrasing is too strong for the packet alone. The accepted soft edge 44479\u219445381 looks plausible; the rest are suitably cautious." - }, - "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 }, @@ -7534,26 +7576,26 @@ "issue:46463|issue:46899", "issue:43697|issue:45216", "issue:45357|issue:45920", - "issue:44466|issue:45005", "issue:44493|issue:46050", + "issue:44466|issue:45005", "issue:43720|issue:44898", "issue:45200|issue:46463", "issue:41762|issue:44936" ], - "prepared_review_unit_hash": "e00862def2aa2c021fa49f61754fa7e68a68f637a92ea45a55c5ac943ada1245", + "prepared_review_unit_hash": "ce7b32a528eab764b44894cbb7031fbb69c6b1c5910d6ac7a303dae19038055c", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35889, + "serialized_chars": 35891, "estimated_input_tokens": 8973, "estimated_eval_tokens": 18202 }, @@ -7562,174 +7604,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: it contains many unrelated bug reports across different models, tokenizers, training paths, quantization paths, and config/serialization issues. I do not see a single underlying duplicate topic, so no canonical issue/PR stands out.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bugs; the soft-similarity links do not look like true duplicates, so I would not merge any of them.", + "confidence": 0.87, + "canonical_issue_reason": "Issue 44466 is the best representative only because it is a concrete, well-scoped v5 save/load regression around tied weights; the rest of the cluster spans unrelated tokenizer, config, quantization, multimodal, and Trainer issues.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "44466 is the clearest standalone issue in the set: it has a specific symptom, an explicit v5 regression angle, and a likely actionable serialization bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "One is a PEFT version bump/release problem; the other is stale auto_mappings references. Different failure modes and fixes." + "reason": "Both are CI-related, but one is a missing PyPI release and the other is stale auto-mapping references; different failures and fixes." }, { "left": "issue:44737", "right": "issue:46463", "accept": false, - "reason": "XLNet CPU device placement bug vs T5 decoder-input validation error message. Unrelated code paths." + "reason": "XLNet CPU placement bug and T5 decoder-input error message are unrelated code paths and symptoms." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Both involve dtype mismatches, but they affect different models and different mechanisms." + "reason": "Both mention dtype issues, but they affect different models and different inference paths, so they are not the same bug." }, { "left": "issue:43716", "right": "issue:46701", "accept": false, - "reason": "Mistral-3 preprocessing/model dtype mismatch vs a broken diffusion generation output pipeline. Not the same bug." + "reason": "One is a dtype mismatch in Mistral-3 preprocessing, the other is a broken image-text-to-text pipeline for DiffusionGemmaGenerationOutput." }, { "left": "issue:45823", "right": "issue:46463", "accept": false, - "reason": "Gemma4 device_map forward mismatch vs T5 decoder validation messaging. Different models and causes." + "reason": "Gemma4 device-map forward mismatch and T5 decoder validation messaging are different problems." }, { "left": "issue:43906", "right": "issue:46693", "accept": false, - "reason": "An isolated reproduction of another issue vs a long-context FA2 performance regression. Not duplicates." + "reason": "The reproduction of issue 38071 and the FA2 long-context performance regression are not the same concrete bug." }, { "left": "issue:44568", "right": "issue:46491", "accept": false, - "reason": "Tokenizer special-token handling vs CodeLlama encode/decode leading-space regression. Different tokenizer bugs." + "reason": "Both are tokenizer regressions, but one is BOS/EOS insertion and the other is leading-space round-trip behavior; distinct issues." }, { "left": "issue:24643", "right": "issue:29870", "accept": false, - "reason": "DeepSpeed weight shape error vs Bart generate() unbound variable. Different training/generation paths." + "reason": "DeepSpeed training weight-shape error and BART decoder_attention_mask initialization error are unrelated failures." }, { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Both are quantization-related, but one is an unsafe uint8 dtype cast in Gemma4 audio modules; the other is BitsAndBytes skip-module exclusion handling." + "reason": "Both involve 4-bit inference, but one is a uint8/torch.finfo crash and the other is lm_head exclusion handling in BitsAndBytesConfig." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation vs strict config loading for granite_speech. Distinct config issues." + "reason": "Qwen3.5 num_labels propagation and granite_speech strict-config loading are different configuration bugs." }, { "left": "issue:43540", "right": "issue:46693", "accept": false, - "reason": "Video input validation in Qwen3OmniMoe vs FA2 long-context performance regression. Unrelated." + "reason": "Qwen3OmniMoe video processing failure and FA2 long-context performance regression do not share the same underlying code path." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "Generic gguf_file/from_pretrained breakage vs a specific model load failure for cjvt/sleng-bert. Not enough evidence of the same bug." + "reason": "Both mention loading failures, but one is gguf_file API handling and the other is a specific model not loading; no clear duplicate relation." }, { "left": "issue:43906", "right": "issue:46468", "accept": false, - "reason": "Reproduction of an earlier issue vs broken official Nemotron example. Different reported failures." + "reason": "Isolated reproduction of an older issue and an official Nemotron example breakage are not the same bug." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "Missing PyTorch import error vs Trainer broadcast NoneType error. Completely different problems." + "reason": "ImportError for missing PyTorch and Trainer/_gpu_broadcast_one NoneType error are entirely different issues." }, { "left": "issue:43697", "right": "issue:46829", "accept": false, - "reason": "RTDetrV2 output regression vs VideoPrism hidden_states field bug. Different models and outputs." + "reason": "RTDetrV2 output differences and VideoPrism hidden_states assignment are unrelated model-specific bugs." }, { "left": "issue:45663", "right": "issue:46846", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states bug vs Qwen3.5 tensor-parallel plan omission. Different underlying fixes." + "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Qwen3.5 TP plan omissions are different distributed-training problems." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "GPTNeoX config reload regression vs rope-validation utility TypeError. Related theme, but not the same defect." + "reason": "GPTNeoX rotary_pct reload regression and rope-parameter JSON list TypeError are both rope-related, but they are distinct failure modes." }, { "left": "issue:44737", "right": "issue:45588", "accept": false, - "reason": "Device placement bug in XLNet vs flash_attention handling of s_aux=None. Different code paths." + "reason": "XLNet CPU arange bug and flash_attention s_aux=None crash are separate low-level runtime errors." }, { "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug vs classification loss degeneracy for num_labels=1. Not the same issue." + "reason": "A BeitImageProcessor label-reduction bug is not the same as a sequence-classification zero-loss problem." }, { "left": "issue:46463", "right": "issue:46899", "accept": false, - "reason": "T5 decoder validation error message vs Gemma4 multimodal 4-bit dtype-cast bug. Different subsystems." + "reason": "T5 decoder validation messaging and Gemma4 4-bit multimodal embedding dtype casts are unrelated." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Model output regression vs Qwen3.5 save_pretrained checkpoint serialization regression. Unrelated." + "reason": "RTDetrV2 output regression and Qwen3.5 save_pretrained checkpoint corruption are different areas and symptoms." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 visual encoder key serialization bug vs tokenizer wrong-token-id regression across models. Different components." + "reason": "Incorrect Qwen3.5 visual-encoder keys on save and wrong token IDs across tokenizers are not the same bug." }, { - "left": "issue:44466", - "right": "issue:45005", + "left": "issue:44493", + "right": "issue:46050", "accept": false, - "reason": "Both mention tied weights, but one is device-dependent lm_head serialization while the other is translation-model tied-weight issues; too broad to merge." + "reason": "Unexpected position-id keys on load and sinq quantization save/load breakage are different serialization problems." }, { - "left": "issue:44493", - "right": "issue:46050", + "left": "issue:44466", + "right": "issue:45005", "accept": false, - "reason": "Unexpected position-id keys vs quantized-model save/load breakage. Different symptoms and roots." + "reason": "Both concern tied weights in v5, but one is device-dependent lm_head serialization and the other is translation-model tied-weight issues; not clearly the same fix." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "BitNet packed-weight loading bug vs Perceiver interpolation failure. Different models and code paths." + "reason": "Packed-weight loading for BitNet AutoBitLinear and Perceiver interpolate_pos_encoding failure are unrelated." }, { "left": "issue:45200", "right": "issue:46463", "accept": false, - "reason": "Gemma4 token-type defaults for text-only tuning vs T5 decoder-input validation text. Not duplicates." + "reason": "Gemma 4 mm_token_type_ids defaults and T5 decoder validation are different model-family bugs." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "DeepSpeed ZeRO-3 loading index error vs trainer.evaluate after train regression. Different trainer/model paths." + "reason": "Gemma3 DeepSpeed ZeRO-3 loading failure and trainer.evaluate-after-train failure are not the same underlying issue." } ] }, @@ -7803,8 +7845,8 @@ ], "soft_pairs": [ "issue:43643|issue:45375", - "issue:43688|issue:43749", "issue:38175|issue:43295", + "issue:43688|issue:43749", "issue:44361|issue:44442", "issue:43653|issue:44361", "issue:43452|issue:44843", @@ -7831,202 +7873,196 @@ "issue:42898|issue:46491", "issue:44843|issue:44991" ], - "prepared_review_unit_hash": "6d53be289983dd4b8b15769922da612a483bfebae8eb89c097a90cb76c53214c", + "prepared_review_unit_hash": "ce42da09ac859b7a1cb1420687b3ac27371344c086dbefad23c0a26cb9e0f6dd", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35722, - "estimated_input_tokens": 8931, - "estimated_eval_tokens": 18118 + "serialized_chars": 35725, + "estimated_input_tokens": 8932, + "estimated_eval_tokens": 18120 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a heterogeneous set of tokenizer/model regression reports rather than true duplicates. None of the soft pairs look like the same concrete bug. The broadest tokenizer consistency issue is the best representative, but only as an umbrella.", - "confidence": 0.62, - "canonical_issue_reason": "issue:29159 is the broadest, most general tokenizer-behavior report in the set and best matches the recurring slow/fast tokenizer inconsistency theme.", + "summary": "This cluster is a grab-bag of mostly unrelated Transformers bug reports across tokenizers, vision/audio models, config loading, and backend edge cases. The soft pairs share broad themes at best, but none look like the same concrete bug or change.", + "confidence": 0.47, + "canonical_issue_reason": "Issue #46491 is the clearest, most concrete tokenizer regression here: it has a specific model, a versioned regression vs v4, and a reproducible encode\u2192decode symptom.", "canonical_pr_reason": null, - "best_issue_reason": "It works as an umbrella representative for tokenizer-regression discussions, but it is not an exact duplicate of most model-specific items in the cluster.", + "best_issue_reason": "Issue #46491 is the best representative issue because it is open, narrowly scoped, and describes a user-facing tokenizer behavior regression with a clear comparison point.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both concern config fields getting lost, but they involve different mechanisms and models: remote-code config loading vs @strict dropping a custom field." + "reason": "Both involve config fields being dropped, but one is trust_remote_code returning incomplete configs and the other is a specific Qwen3_5MoeVisionConfig field being stripped; different root causes." }, { - "left": "issue:43688", - "right": "issue:43749", + "left": "issue:38175", + "right": "issue:43295", "accept": false, - "reason": "Unrelated bugs: auxiliary-loss normalization versus FSDP CPU-RAM-efficient loading." + "reason": "Siglip2 zero probabilities and processor.tokenizer/image-passing regression are unrelated model-path failures." }, { - "left": "issue:38175", - "right": "issue:43295", + "left": "issue:43688", + "right": "issue:43749", "accept": false, - "reason": "Different symptoms and code paths: zero probabilities in SigLIP2 versus processor.tokenizer/image-passing regression." + "reason": "Aux-loss normalization in MoE models is unrelated to the FSDP_CPU_RAM_EFFICIENT_LOADING breakage." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "Both are tokenizer failures, but they are different tokenizers and different failure modes (AttributeError on tasks vs AutoTokenizer load failure)." + "reason": "Different tokenizer load failures affecting different models and failure modes." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "Different concrete bugs in different tokenizers; special-token registration is not the same as the MLukeTokenizer AttributeError." + "reason": "BigBird special-token registration and MLukeTokenizer AttributeError are different tokenizer bugs." }, { "left": "issue:43452", "right": "issue:44843", "accept": false, - "reason": "Separate loading regressions: gguf_file handling versus an unconditional model_info() call in offline mode." + "reason": "Both touch AutoTokenizer loading, but one is gguf_file support and the other is an offline-mode failure in _patch_mistral_regex; not the same bug." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "Unrelated functionality: DAC latent decoding mismatch versus Whisper batch_decode skip_special_tokens handling." + "reason": "DAC.from_latents/STE mismatch is unrelated to Whisper batch_decode skip_special_tokens behavior." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Both are vision-model issues, but the affected models and failures differ substantially." + "reason": "Both are vision-model issues, but they are different models and different output/runtime problems." }, { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Different models and different errors: sliding_window handling in Qwen3-MoE versus a KeyError loading MobileLLM." + "reason": "Qwen3-MoE sliding_window and MobileLLM key error are distinct model-loading/runtime bugs." }, { "left": "issue:44857", "right": "issue:47000", "accept": false, - "reason": "Same broad detector family, but these are different loss crashes with different error conditions and likely different fixes." + "reason": "Both are detection-loss crashes, but they fail in different code paths and with different conditions." }, { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Both are dependency/import problems, but they are about different packages and platforms with different root causes." + "reason": "Gemma processor import breakage and Windows ROCm torch.distributed dependency issues are unrelated." }, { "left": "issue:43761", "right": "issue:46829", "accept": false, - "reason": "Both mention hidden_states, but they are different model implementations and different bugs in output plumbing." + "reason": "Both mention hidden_states, but they concern different models and different incorrect outputs." }, { "left": "issue:29159", "right": "issue:33357", "accept": false, - "reason": "Generic tokenizer inconsistency versus a MacOS bus error loading a CLIP community model; not the same bug." + "reason": "Tokenizer inconsistency vs MacOS bus error are not the same underlying bug." }, { "left": "issue:43525", "right": "issue:46701", "accept": false, - "reason": "Different issues: missing pad_token_id in a config versus a pipeline failure in DiffusionGemma generation output." + "reason": "Missing pad_token_id in Llama4Config is unrelated to DiffusionGemmaGenerationOutput pipeline failure." }, { "left": "issue:45542", "right": "issue:46178", "accept": false, - "reason": "Related backend-inference area, but different backend packages and distinct error paths; not one concrete bug." + "reason": "Different backend-registration problems with different offending packages and symptoms." }, { "left": "issue:46832", "right": "issue:46932", "accept": false, - "reason": "Completely different domains: RT-DETR feature-level mismatch versus Gemma4 dtype casting for quantized embedders." + "reason": "RT-DETR backbone/feature-level crash and Gemma4 dtype-casting fix are unrelated." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Different model components and failures: auxiliary-loss normalization versus an apex RMSNorm memory leak." + "reason": "MoE auxiliary-loss normalization and T5 apex RMSNorm memory leak are unrelated." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Both are regression reports, but they target different tokenizer/model paths and do not share the same underlying defect." + "reason": "Custom processor.tokenizer regression and Kimi-K2.5 codec/mistral-regex regression are different bugs." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Unrelated: auxiliary-loss normalization versus a broad tokenization-version-change report." + "reason": "A specific aux-loss normalization bug is not the same as a broad tokenization-change report." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both concern special tokens, but the concrete bug differs: mask token registration vs BOS/EOS insertion under add_special_tokens." + "reason": "Different tokenizers and different special-token failures; not one concrete defect." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "Custom model initialization regression versus FSDP loading break; different code paths and symptoms." - }, - { - "left": "issue:43756", - "right": "issue:46402", - "accept": false, - "reason": "Different model-specific weight/layout problems: RoPE layer dropping versus Sam3 text-encoder weights not loading." + "reason": "Custom model initialization breakage and FSDP CPU RAM efficient loading failure are unrelated." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "Tokenizer/loader breakage versus dtype casting on a different multimodal model; no shared code-path bug." + "reason": "gguf_file loading failure and BLIP2 dtype mismatch affect different code paths." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Both involve generation/cache behavior, but the models and failure mechanisms are different enough that they should stay separate." + "reason": "Both concern generation/cache behavior, but the models and mechanics differ; not enough to call them the same bug." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "Special-token registration in BigBirdTokenizer is not the same bug as DiaConfig losing custom token IDs after save/load." + "reason": "Special-token registration in a tokenizer and DiaConfig losing custom token IDs after save/load are different layers and symptoms." }, { "left": "issue:44821", "right": "issue:46821", "accept": false, - "reason": "Different subsystems: image processor loading from URL versus a TimesFM window_size attribute error." + "reason": "AutoImageProcessor URL loading and TimesFM window_size AttributeError are unrelated." }, { "left": "issue:42898", "right": "issue:46491", "accept": false, - "reason": "Both are spacing/tokenization regressions, but one is a global clean_up_tokenization_spaces change and the other is a CodeLlama round-trip issue." + "reason": "Both involve tokenizer behavior changes, but one is clean_up_tokenization_spaces semantics and the other is CodeLlama leading-space round-trip regression; different defects." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-loading regressions, but they stem from different models and different failure causes." + "reason": "Offline model_info() call in mistral regex patch and EMBEDDIA/est-roberta tokenizer loading failure do not appear to be the same bug." } ] }, @@ -8117,8 +8153,8 @@ "issue:43278|issue:43688", "issue:43504|issue:43720", "issue:43782|issue:46693", - "issue:43611|issue:46846", "issue:45491|issue:46002", + "issue:43611|issue:46846", "issue:43577|issue:44554", "issue:43720|issue:43927", "issue:43606|issue:43927", @@ -8131,214 +8167,214 @@ "issue:44625|issue:45584", "issue:43493|issue:43825" ], - "prepared_review_unit_hash": "57a92c9078b26be4b254203111b8223636e72b82623d7ffd3e1338fcaf01689f", + "prepared_review_unit_hash": "e6256b7cecee1fbdacd78036a3da22186db4a0af51a61c5b44deb7f5d4606bc2", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36964, - "estimated_input_tokens": 9241, - "estimated_eval_tokens": 18738 + "serialized_chars": 36968, + "estimated_input_tokens": 9242, + "estimated_eval_tokens": 18740 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: most items are distinct model-specific regressions, config bugs, or generation/loading issues rather than duplicates. I found no soft-edge pair that clearly shares the same concrete underlying bug or fix path.", - "confidence": 0.87, - "canonical_issue_reason": null, + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bug reports/regressions across different models and subsystems. The soft similarities are mostly superficial (shared keywords like Qwen, processor, save/load, FA2, or NaNs) rather than the same underlying defect.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 46693 is the most central active report in the set: it has the most discussion, inbound references, and describes a broad generation regression with clear ongoing impact.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46693 is the strongest anchor for the cluster because it has the highest activity, is still open, and is a clearly scoped regression report with broad impact (FA2 + long-context generation).", + "best_issue_reason": "Issue 46693 is the best representative issue here because it is active, well-discussed, and broadly scoped compared with the mostly isolated model-specific bugs in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Both are Qwen-related, but one is a video-input regression and the other is a num_labels propagation bug; different code paths and fixes." + "reason": "Both are Qwen3.5-related, but one is a video-input regression and the other is a num_labels propagation bug; different code paths and symptoms." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Multi-GPU training for Qwen2-VL is unrelated to the GenerationMixin import error; different subsystems and failure modes." + "reason": "Multi-GPU training for Qwen2-VL and a GenerationMixin import error are unrelated failures." }, { "left": "issue:43761", "right": "issue:47013", "accept": false, - "reason": "Both are runtime crashes, but one is CLIP vision hidden-states handling and the other is xLSTM return_last_states unpacking; not the same bug." + "reason": "CLIPVisionModel hidden_states regression and xLSTM train-mode unpacking crash are different model paths and bugs." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Both involve save/load behavior, but one is tied lm_head serialization and the other is quantized sinq model reload; not the same underlying defect." + "reason": "Both involve save/load behavior, but one is tied-weight serialization and the other is quantized sinq state restoration." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer GPU broadcast NoneType and Qwen2-VL multi-GPU training are separate distributed-training problems." + "reason": "Trainer GPU broadcast TypeError and Qwen2-VL multi-GPU training are not the same defect." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "One is about default shard size in save_pretrained, the other about AutoProcessor not forwarding kwargs to cached_file; unrelated." + "reason": "A max_shard_size question and an AutoProcessor kwarg-passing bug are unrelated." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Auto device mapping cuda assert and chunked generation compiled-forward inconsistency affect different execution paths." + "reason": "Auto device mapping cudaErrorAssert and compiled chunked generation inconsistency are different failures." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Both touch MoE models, but the first is auxiliary-loss normalization and the second is a DeepSeekV3MoE implementation divergence; different fixes." + "reason": "Both are model-fidelity bugs, but they affect different architectures and different internal computations." }, { "left": "issue:44977", "right": "issue:46693", "accept": false, - "reason": "Both mention generation and attention, but one is a Qwen3.5 flash-attention correctness issue while the other is a long-context performance regression." + "reason": "Qwen3.5 flash-attention correctness bug and FA2 long-context performance regression are not the same issue." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Compiled-forward chunked generation mismatch is not the same as Qwen3.5 flash-attention generation failure." + "reason": "Compiled-forward chunked generation mismatch is distinct from the Qwen3.5 flash-attention bug." }, { "left": "issue:44734", "right": "issue:46612", "accept": false, - "reason": "KV-cache continuation indexing in serve and beam-search cache reorder skipping are distinct generation bugs with different code paths." + "reason": "KV-cache continuation tensor indexing in serve and skipped beam-search cache reorder in generation are different code paths." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "Beit pretrained loading via legacy field and Bamba torch.compile SDPA failure are unrelated." + "reason": "Beit legacy-field loading failure and Bamba torch.compile SDPA failure are unrelated." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Qwen2-VL multi-GPU training is not the same as CustomTrainer.compute_loss signature mismatch." + "reason": "Qwen2-VL multi-GPU training and CustomTrainer compute_loss signature breakage are different training issues." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Both involve processor behavior, but one is audio sampling-rate defaults and the other is a regression around processor.tokenizer/image passing." + "reason": "Audio processor sample-rate defaulting and processor.tokenizer/image-passing regression are separate processor bugs." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "Embedding dtype drift on evaluate and MoE auxiliary-loss normalization are unrelated bugs." + "reason": "Embedding dtype mismatch during eval and auxiliary-loss normalization are unrelated." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Beit legacy-field loading and BitNet packed-weight unpacking during accelerate loading are different model-loading issues." + "reason": "Legacy field loading in Beit and packed-weight unpacking in BitNet are different loading bugs." }, { "left": "issue:43782", "right": "issue:46693", "accept": false, - "reason": "Qwen3VL weight_only loading failure is a model-load issue, not the FA2 long-context generation performance regression." + "reason": "Qwen3VL weight_only loading error and FA2 performance regression do not share the same underlying cause." }, { - "left": "issue:43611", - "right": "issue:46846", + "left": "issue:45491", + "right": "issue:46002", "accept": false, - "reason": "Base_model_prefix compatibility and missing linear_attn entries in tensor-parallel plans are separate v5 regressions." + "reason": "Both mention NaNs, but one is a Gemma3 sliding-window attention bug and the other is a classification-head init/checkpoint issue." }, { - "left": "issue:45491", - "right": "issue:46002", + "left": "issue:43611", + "right": "issue:46846", "accept": false, - "reason": "Gemma3 sliding-window NaNs on GPU and classification-head NaNs after checkout/init do not share the same failure mechanism." + "reason": "base_model_prefix loading regression and Qwen3.5 TP-plan OOM/shape errors are unrelated." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "Blip2 dtype mismatch and the MPS attention correctness issue are different problems." + "reason": "Blip2 dtype loading mismatch and MPS attention correctness with mismatched head dims are different bugs." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "BitNet accelerate-loading unpacking and DiaConfig token-id persistence are unrelated serialization/loading bugs." + "reason": "BitNet accelerate loading issue and DiaConfig token-ID loss after save/load are unrelated serialization/config bugs." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch for bark-small and DiaConfig save/load token-ID loss are not the same bug." + "reason": "CPU offload device mismatch and DiaConfig field loss after save/load are not the same failure." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "Beit legacy-field loading and OmDet-Turbo processor/model input-size mismatch are unrelated." + "reason": "Beit legacy field loading and OmDet-Turbo processor size mismatch are distinct model/processor problems." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both are chat-template crashes, but one is batched padding handling and the other is assistant tool-call/no-content handling; different edge cases." + "reason": "Both are apply_chat_template crashes, but one is batched padding=False handling and the other is assistant tool-call content handling." }, { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "Phi-4 multimodal loading failure and MIN_PEFT_VERSION bump are not the same underlying issue." + "reason": "Phi-4 multimodal loading failure and an unreleased PEFT version pin are unrelated." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Missing pad_token_id in Llama4Config and missing pad_token_idx in StableLmConfig are similar but model-specific config omissions, not one shared bug." + "reason": "Missing pad_token_id in Llama4Config and missing pad_token_idx in StableLmConfig are different model/config regressions." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "AutoConfig trust_remote_code missing fields and removal of is_torch_fx_available breaking trust_remote_code models are separate compatibility regressions." + "reason": "Both involve trust_remote_code, but one is missing config fields and the other is removal of is_torch_fx_available." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "BitNet packed-weight loading and MLukeTokenizer task AttributeError do not share a concrete fix path." + "reason": "BitNet packed-weight loading and MLukeTokenizer task AttributeError are unrelated." }, { "left": "issue:44625", "right": "issue:45584", "accept": false, - "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription generation after align_special_tokens are unrelated." + "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription generation are unrelated bugs." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 HF-vs-JAX discrepancy and pipeline translation-task error messaging are entirely different problems." + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task messaging are unrelated." } ] }, @@ -8441,208 +8477,208 @@ "issue:43334|issue:45759", "issue:45245|issue:46710" ], - "prepared_review_unit_hash": "6e56ef224491ab128ce2b0acf4d32ca3b8b078de16184cb4293a64dda4e3b036", + "prepared_review_unit_hash": "235acbfdb2590e4eb7b2fb81aa37eaff3b5c6a5acd590d1e32b35af6addf3061", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35926, - "estimated_input_tokens": 8982, - "estimated_eval_tokens": 18220 + "serialized_chars": 35930, + "estimated_input_tokens": 8983, + "estimated_eval_tokens": 18222 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All proposed soft edges look like false positives: they share only broad themes such as tokenizers, model loading, or version regressions, but not the same concrete bug. No PRs are present.", - "confidence": 0.94, - "canonical_issue_reason": "No true duplicate cluster is evident from the proposed pairs, so there is no safe canonical issue across this set. If a representative standalone issue must be named, issue:45812 is the clearest high-signal report, but it does not duplicate the others.", + "summary": "The set is mostly heterogeneous; a few pairs share broad areas like tokenization or model loading, but none are close enough to be true duplicates or the same fix scope.", + "confidence": 0.93, + "canonical_issue_reason": "No single canonical issue is justified across this cluster because the items span unrelated bugs (tokenizers, multimodal loading, training/runtime failures, and model-specific regressions).", "canonical_pr_reason": null, - "best_issue_reason": "issue:45812 is the strongest standalone issue in the batch: a concrete, reproducible tokenizer regression with clear impact and inbound references.", + "best_issue_reason": "If one issue must represent the most reusable triage anchor, issue 45812 is the strongest candidate because it is a broad, high-impact tokenizer regression with multiple references and nearby tokenizer-related reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44977", "right": "issue:46209", "accept": false, - "reason": "Different bugs: Qwen3.5 flash-attention generation failure vs FP8/B200 inference incompatibility." + "reason": "Both are inference regressions, but one is Qwen3.5 flash-attention behavior and the other is FP8/B200 support; different code paths and fixes." }, { "left": "issue:44554", "right": "issue:46365", "accept": false, - "reason": "MPS attention correctness issue and Trainer import failure from datasets are unrelated." + "reason": "MPS attention correctness and Trainer import failure are unrelated subsystems and failure modes." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Device-mismatch during forward and quantized save/load corruption are separate code paths." + "reason": "Device-map forward mismatch and quantized save/load breakage are different bugs with different triggers." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer spacing behavior and a MacOS bus error in a CLIP model are not the same bug." + "reason": "Tokenizer spacing bug and MacOS CLIP bus error are unrelated." }, { "left": "issue:43452", "right": "issue:44451", "accept": false, - "reason": "gguf_file loading API break and inability to load a specific BERT model have different causes." + "reason": "Both concern model loading, but one is gguf/from_pretrained plumbing and the other is a specific ScandiBERT load failure." }, { "left": "issue:43994", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 wrong outputs and return_tensors being ignored with text_kwargs are unrelated behaviors." + "reason": "SigLIP2 wrong outputs and ignored return_tensors are different API bugs." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "Documentation example errors and processor.tokenizer/image-passing regression are different failures." + "reason": "Documentation example errors and custom processor/tokenizer regression are not the same underlying issue." }, { "left": "issue:42796", "right": "issue:44977", "accept": false, - "reason": "Ministral BPE marker handling and Qwen3.5 flash-attention generation are distinct tokenizer/model issues." + "reason": "Tokenizer BPE-marker handling and Qwen3.5 flash-attention generation are unrelated." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation discrepancy and DeepSpeed gradient growth are unrelated problems." + "reason": "SigLIP2 implementation mismatch and DeepSpeed ZeRO-3 gradient growth are distinct model/training problems." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "Embedding dtype drift in evaluate does not match the ZeRO-3 gradient anomaly." + "reason": "Embedding dtype change and gradient explosion are different training behaviors with different likely fixes." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model init breakage and lm_head serialization inconsistency are different bugs." + "reason": "Custom model initialization breakage and tied-weight serialization inconsistency are separate regressions." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "Eval dtype change and resume checkpoint key mismatch do not share a concrete root cause." + "reason": "Dtype drift and resume_from_checkpoint key mismatch are unrelated." }, { "left": "issue:46209", "right": "issue:46710", "accept": false, - "reason": "FP8 inference failure on B200 and a DeepSeek accuracy regression after upgrade are separate issues." + "reason": "FP8 inference on B200 and DeepSeek accuracy regression after upgrade are not the same bug." }, { "left": "issue:44568", "right": "issue:46489", "accept": false, - "reason": "Both involve tokenizers, but they affect different models and different special-token/BPE paths." + "reason": "Both are tokenizer regressions, but they affect different models and different tokenization paths; not the same fix." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "MacOS bus error and Qwen2-VL multi-GPU training are unrelated." + "reason": "MacOS bus error with CLIP and Qwen2-VL multi-GPU training are unrelated." }, { "left": "issue:44849", "right": "issue:46209", "accept": false, - "reason": "Hidden-states bug in Qwen3.5 is unrelated to FP8/B200 inference failure." + "reason": "Hidden-state output bug and FP8 inference failure are unrelated." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "A failed janus test and a PEFT version bump are not the same underlying bug." + "reason": "A failing Janus image-generation test and an unreleased PEFT version bump do not share an underlying bug." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "DiaConfig save/load token-ID loss and MLukeTokenizer task AttributeError are different model-specific issues." + "reason": "DiaConfig save/load token ID loss and MLukeTokenizer AttributeError are different tokenizer/config issues." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "Tensor-parallel OOM and an unreleased PEFT dependency update are unrelated." + "reason": "Tensor-parallel OOM and dependency version bump are unrelated." }, { "left": "issue:44568", "right": "issue:45812", "accept": false, - "reason": "Different tokenizer regressions on different models; no shared concrete code-path bug." + "reason": "Both involve tokenizers, but one is missing special tokens for mdeberta and the other is wrong IDs for Granite; different concrete defects." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "ImportError for Qwen2_5_VLForConditionalGeneration and Float8 storage lookup failure are unrelated." + "reason": "ImportError for Qwen2.5-VL and missing Float8 storage are unrelated." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "Model loading failure for one model and offline mistral-regex network access are different breakages." + "reason": "Specific model load failure and offline-mode tokenizer patch bug are not the same root cause." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Qwen3.5 hidden-states behavior and incorrect visual encoder key serialization are separate regressions." + "reason": "Hidden-state behavior and incorrect visual-encoder key serialization are distinct Qwen3.5 issues." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and unexpected _is_hf_initialized arg handling are different 5.0 compatibility issues." + "reason": "Missing StableLmConfig field and unexpected init_empty_weights argument are different config/init regressions." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Float8 storage lookup and flash_attn import_utils KeyError are unrelated import/type failures." + "reason": "Float8 storage lookup failure and flash_attn import mapping error are unrelated." }, { "left": "issue:46821", "right": "issue:46832", "accept": false, - "reason": "TimesFM window_size AttributeError and RT-DETR forward TypeError are different model bugs." + "reason": "TimesFM window_size AttributeError and RT-DETR feature-level mismatch are different model-specific failures." }, { "left": "issue:43504", "right": "issue:43606", "accept": false, - "reason": "BEiT legacy-field loading and CPU offload device mismatch are unrelated." + "reason": "Legacy preset loading in Beit and CPU-offload device mismatch in bark-small are unrelated." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Both touch multimodal Qwen configs, but one is a missing pad_token_id field and the other is text_config unwrapping in from_config; different fixes." + "reason": "Missing pad_token_id in Qwen3-VL config and text_config unwrapping in composite configs are different loading bugs." }, { "left": "issue:45245", "right": "issue:46710", "accept": false, - "reason": "Category-count limit error and an accuracy regression after upgrade are unrelated." + "reason": "Category-limit runtime error and DeepSeek accuracy regression are not the same issue." } ] }, @@ -8728,11 +8764,11 @@ "issue:43844|issue:44964", "issue:29159|issue:32090", "issue:45588|issue:46050", - "issue:29942|issue:30333", "issue:44869|issue:46752", - "issue:42490|issue:43696", "issue:45229|issue:45823", + "issue:42490|issue:43696", "issue:45200|issue:45229", + "issue:29942|issue:30333", "issue:43937|issue:46701", "issue:45084|issue:45520", "issue:28218|issue:34567", @@ -8741,184 +8777,184 @@ "issue:43493|issue:43701", "issue:44554|issue:45446" ], - "prepared_review_unit_hash": "6b03375cf1bf495f07b81f58fd027a7611eaf8566c2a885382cf641893b9532e", + "prepared_review_unit_hash": "e1276c5024fc040ff737e45d9263d903ad07c515066eceef4246574156db2b07", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 25, - "serialized_chars": 34587, - "estimated_input_tokens": 8647, - "estimated_eval_tokens": 17550 + "serialized_chars": 34579, + "estimated_input_tokens": 8645, + "estimated_eval_tokens": 17546 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is a heterogeneous mix of unrelated model-loading, tokenizer, training, and CI issues. None of the proposed soft pairs look like true duplicates or the same underlying fix.", - "confidence": 0.94, - "canonical_issue_reason": "issue:43873 is the most active and broadly scoped report in the set, with the strongest discussion and inbound reference signal, so it is the best anchor among otherwise unrelated issues.", + "summary": "The proposed pairs are mostly near-matches only at a broad subsystem level (tokenizers, loading, attention, quantization, training), but they describe different concrete failures and fixes. I would not merge any of these as duplicates; the cluster is heterogeneous rather than one bug/change.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the most representative issue to keep as the cluster anchor, but only as a loose representative because the cluster does not form a coherent duplicate family.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44493", "right": "issue:46285", "accept": false, - "reason": "Different bugs: one is a positional-id key warning across many models, the other is a CLIP class structure regression." + "reason": "Both mention token-related behavior, but one is about unexpected position-id keys across models while the other is a CLIP text-model structural regression. Different symptoms and code paths." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Both involve Qwen3 MoE loading, but they describe different failure modes and code paths, not the same concrete bug." + "reason": "Both involve Qwen3 MOE loading, but one is accelerate+deepspeed weight conversion and the other is a transformers dev config breakage (`pad_token_id`). Same family, different bug." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Both are missing config fields after an update, but on different models and different attributes; not the same issue." + "reason": "These are separate missing-config-field issues in different models (`pad_token_id` vs `pad_token_idx`). Similar shape, not the same underlying defect." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "Unrelated subsystems: pretrained model loading vs tokenizer special-token registration." + "reason": "Beit pretrained load failure from a legacy field vs BigBirdTokenizer special-token registration causing empty decode. Different components and fixes." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "One is a real device-mismatch bug in Bark offload, the other is a GraniteSpeech CI test issue." + "reason": "CPU offload device mismatch in inference is not the same as a stale CI test failure in GraniteSpeech. Same general area, unrelated bugs." }, { "left": "issue:46399", "right": "issue:46846", "accept": false, - "reason": "Same model family, but different problems: kernel registration crash vs tensor-parallel plan gaps causing OOM/shape errors." + "reason": "One is a missing decorator for `apply_rotary_pos_emb` registration; the other is a missing tensor-parallel plan entry for `linear_attn`. Distinct issues." }, { "left": "issue:45072", "right": "issue:46832", "accept": false, - "reason": "Different models and different failure classes: dtype mismatch vs backbone-feature count mismatch." + "reason": "bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated to RTDetr feature-level shape/type errors." }, { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Both touch quantization/offloading, but one is about offload behavior and the other about save/load persistence after quantization." + "reason": "Both touch quantization/offloading, but one is offloading not working as expected and the other is save/load corruption after `sinq` quantization. Not the same code-path bug." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Different code paths and symptoms: compiled-forward chunked generation vs Whisper empty-transcription handling." + "reason": "Compiled-forward chunked-generation inconsistency is unrelated to Whisper empty-transcription failure after `align_special_tokens`." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "Different model-specific inference bugs; no shared underlying code-path problem is evident." + "reason": "Nonsensical SigLIP2 outputs and unexpected `position_id` keys are different model behaviors with different causes." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "Different topics: ZeRO-3 gradient behavior vs a Phi-4 multimodal loading regression." + "reason": "ZeRO-3 gradient growth during training is unrelated to a model-loading failure for Phi-4 multimodal." }, { "left": "issue:29159", "right": "issue:32090", "accept": false, - "reason": "Tokenizer slow/fast inconsistency and Trainer broadcast TypeError are unrelated." + "reason": "Tokenizer slow/fast inconsistency and `_gpu_broadcast_one` NoneType errors are entirely different subsystems and failures." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "Flash-attention aux-state crash and quantized save/load breakage are different bugs." + "reason": "Flash-attention `s_aux=None` crash and sinq quantized model save/load breakage are not the same bug." }, { - "left": "issue:29942", - "right": "issue:30333", + "left": "issue:44869", + "right": "issue:46752", "accept": false, - "reason": "CI flash-attention failures and MLflow reporting on job failure are unrelated." + "reason": "Whisper timestamp decode crash and empty chat-template input `IndexError` are unrelated edge-case crashes." }, { - "left": "issue:44869", - "right": "issue:46752", + "left": "issue:45229", + "right": "issue:45823", "accept": false, - "reason": "Both are empty/trailing-input edge cases, but they affect different APIs and code paths." + "reason": "Both are Gemma4, but one is multi-GPU OOM and the other is a device-map mismatch during forward. Different fixes." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "A path-parsing warning is not the same as a CUDA OOM issue." + "reason": "Local-path import warning with trailing slash is unrelated to GPT-oss-20b CUDA OOM." }, { - "left": "issue:45229", - "right": "issue:45823", + "left": "issue:45200", + "right": "issue:45229", "accept": false, - "reason": "Both are Gemma4 inference issues, but one is OOM from scale and the other is a device placement mismatch." + "reason": "Text-only `mm_token_type_ids` defaulting to zeros and Gemma4 multi-GPU OOM are distinct problems." }, { - "left": "issue:45200", - "right": "issue:45229", + "left": "issue:29942", + "right": "issue:30333", "accept": false, - "reason": "Text-only mm_token_type_ids handling is unrelated to multi-GPU CUDA OOM." + "reason": "Flash Attention 2 test failures and MLFlow job-status reporting after training failure are unrelated." }, { "left": "issue:43937", "right": "issue:46701", "accept": false, - "reason": "GenerationConfig validation and image-text pipeline failure are distinct bugs." + "reason": "Invalid `GenerationConfig` for GLM-5 and DiffusionGemma pipeline failure are different model/integration issues." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Compilation error for non-template nodes and Python 3.13 flash_attn import KeyError are unrelated." + "reason": "Compilation of non-template nodes and Python 3.13 `flash_attn` import handling are unrelated." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Tokenizer spacing behavior and TrainerState token counting are different features." + "reason": "Tokenizer spacing after added tokens and `num_input_tokens_seen` not updating in TrainerState are different bugs." }, { "left": "issue:43065", "right": "issue:46369", "accept": false, - "reason": "SAM3 dummy layer construction and SAM3 memory overflow are not the same concrete bug." + "reason": "A dummy Conv2d in Sam3PixelDecoder is not the same as memory overflow in sam3-litetext; one is architecture scaffolding, the other is runtime memory pressure." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript generator-expression export failure and SigLIP2 zero-probability outputs are unrelated." + "reason": "torch.jit.script GeneratorExp unsupported during export and SigLIP2 zero-probability outputs are unrelated." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 implementation discrepancy and resume-from-checkpoint key mismatch are different problems." + "reason": "SigLIP2 HF-vs-JAX discrepancy and checkpoint-resume key mismatch are different classes of problems." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness and a PyTorch version check bug in flex_attention are separate code paths." + "reason": "MPS attention correctness when value head dim differs from query is unrelated to the PyTorch version check bug for AuxRequest import." } ] }, @@ -9024,220 +9060,220 @@ "issue:44466|issue:45127", "issue:28218|issue:35141" ], - "prepared_review_unit_hash": "8aebb1528048f76da0c9508f031f8967d630982fac0169d70a1cc3ef2a57d3fb", + "prepared_review_unit_hash": "d9ea94e91d2f64327877bcbdf5f4f2ea116f3b2a68d28e31f338ca24c3415396", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37302, - "estimated_input_tokens": 9326, - "estimated_eval_tokens": 18908 + "serialized_chars": 37312, + "estimated_input_tokens": 9328, + "estimated_eval_tokens": 18912 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the issues span unrelated tokenizer, generation, serving, multimodal, loading, quantization, and backend-import bugs. I don't see any true duplicate pairs among the suggested soft edges, so there is no strong canonical issue or PR in this set.", - "confidence": 0.92, - "canonical_issue_reason": null, + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bug reports. The soft similarities are usually at the subsystem/model-family level, but the underlying failures differ, so none of the proposed edges should be merged.", + "confidence": 0.97, + "canonical_issue_reason": "issue:43295 is the best representative because it is the most connected soft hub and describes a broad regression in multimodal processor behavior, but the cluster is not a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:43295 is the strongest global representative: broad impact, clear regression framing, and multiple soft links to nearby multimodal-processing reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41720", "right": "issue:46710", "accept": false, - "reason": "Different problems: one is an auto device-mapping CUDA assert on A800, the other is an accuracy regression after a Transformers upgrade." + "reason": "Different failures: auto device-mapping CUDA assert vs an accuracy regression after upgrade; same broad model usage, not the same bug." }, { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Both involve serve crashes, but the failures are different code paths and errors: continuous batching multimodal AttributeError vs KV-cache continuation tensor indexing." + "reason": "Both touch transformers serve, but one is a multimodal continuous-batching crash and the other is a KV-cache continuation indexing bug." }, { "left": "issue:44977", "right": "issue:46701", "accept": false, - "reason": "Different model/features and failure modes: flash-attention generation bug for Qwen3.5 vs an image-text-to-text pipeline issue in DiffusionGemmaGenerationOutput." + "reason": "Both involve generation/output issues, but they are on different model families and different code paths." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Unrelated: missing flash_attention_utils import vs model loading with torch.device('meta') breaking some models." + "reason": "ImportError for a missing flash-attention helper is unrelated to meta-device model loading breakage." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Different regression surfaces: processor.tokenizer/image-tokenizer regression vs video processor output shape mismatch for batched inputs." + "reason": "One is a processor.tokenizer regression in custom multimodal code; the other is batched video processor shape handling." }, { "left": "issue:44977", "right": "issue:46710", "accept": false, - "reason": "Different bug classes: flash-attention generation failure vs model accuracy regression after version change." + "reason": "Same model family mention is not enough; one is flash-attention generation breakage, the other is an accuracy regression after version upgrade." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "Different subsystems and symptoms: dtype mismatch in bfloat16 inference vs Wav2Vec2 save_pretrained/tokenization failure." + "reason": "Completely different models and symptoms: bfloat16 dtype mismatch vs save/tokenization failure." }, { "left": "issue:45406", "right": "issue:46121", "accept": false, - "reason": "Completely different issues: Gemma4Processor _tokenizer AttributeError in serve vs rope-validation TypeError on JSON-loaded list." + "reason": "Processor missing _tokenizer is unrelated to rope-validation type handling for a JSON-loaded list." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Different bugs: flash_attention integration crashes on s_aux=None vs AutoModelForCausalLM.from_config not unwrapping text_config." + "reason": "Flash-attention sink handling and unwrapping text_config are separate code-path issues." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Different multimodal issues: video processing ValueError in Qwen3OmniMoe vs missing mm_token_type_ids default in Gemma 4 fine-tuning." + "reason": "Both are multimodal-adjacent, but one is a video-processing ValueError and the other is a missing default for mm_token_type_ids." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Different problem domains: processor/tokenizer regression vs offloading behavior with quantization." + "reason": "A processor/tokenizer regression and quantization offloading behavior are not the same underlying bug." }, { "left": "issue:43572", "right": "issue:46821", "accept": false, - "reason": "Different models and failures: StableLmConfig missing pad_token_idx vs TimesFM window_size AttributeError." + "reason": "Unrelated config/API issues in different model families: StableLm pad_token_idx vs TimesFM window_size." }, { "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "Different tokenizer bugs: LayoutLMv2 NER/padding crash vs return_tensors being ignored when text_kwargs is passed." + "reason": "Tokenizer crash on NER/batched padding is distinct from return_tensors being ignored when text_kwargs is explicitly passed." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Not the same bug: tokenizer/processor regression involving processor.tokenizer and images vs OmDet-Turbo image size mismatch." + "reason": "Processor/tokenizer regression and OmDet-Turbo image-size mismatch are different bugs in different models." }, { "left": "issue:45823", "right": "issue:46192", "accept": false, - "reason": "Unrelated: Gemma4 PLE device mismatch under device_map='auto' vs return_tensors handling in processor kwargs." + "reason": "Gemma4 device-map mismatch is unrelated to return_tensors being dropped in kwargs handling." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Different symptoms and components: video processor batched-shape bug vs tensor-parallel OOM when loading a large Qwen2 model." + "reason": "Video input shape handling and tensor-parallel OOM are different failure modes." }, { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "Different training/runtime issues: torch.compile SDPA failure on Bamba vs abnormal gradient growth with DeepSpeed ZeRO-3." + "reason": "torch.compile + SDPA failure is unrelated to ZeRO-3 gradient abnormality." }, { "left": "issue:29870", "right": "issue:29942", "accept": false, - "reason": "Unrelated failures: Bart generate() decoder_attention_mask assignment bug vs Flash Attention 2 tests failing." + "reason": "Bart generate mask bug and Flash Attention 2 test failures do not share the same code path." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "Different import/backend issues: flash_attn KeyError on Python 3.13 vs tensorflow_text backend mapping error." + "reason": "Both are import/backend errors, but they involve different missing backends and different failure mechanisms." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "Different loading/configuration failures: missing pad_token_idx in StableLmConfig vs failure loading physical-intelligence/fast." + "reason": "StableLm config regression and a transformers>=5.1.0 loading failure are not the same bug." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "Different import errors for different symbols and models; they do not describe the same broken code path." + "reason": "Different import problems: missing flash_attention helper vs missing Qwen2_5_VL class export." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Different Gemma 4 problems: defaulting mm_token_type_ids for text-only fine-tuning vs PLE device mismatch in forward." + "reason": "Both concern multimodal Gemma4, but one is a missing default tensor-type field and the other is a forward-device mismatch." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "No overlap in underlying bug: local-path import warning with trailing slash vs SigLIP2 implementation discrepancy." + "reason": "A path-trailing-slash import warning is unrelated to a SigLIP2 implementation discrepancy." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "Different issue scope: SigLIP2 documentation/model mismatch vs a processor return_tensors keyword bug." + "reason": "Documentation example errors are not the same as kwargs/return_tensors handling in preprocessing." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Different generation bugs: model_kwargs update after sync_gpus vs output_hidden_states behavior in Qwen3.5." + "reason": "Generation helper state update after sync_gpus is a different issue from a Qwen3.5 hidden-states bug." }, { "left": "issue:43295", "right": "issue:43493", "accept": false, - "reason": "Different models and failure modes: custom model code accessing processor.tokenizer vs SigLIP2 HF/JAX discrepancy." + "reason": "Processor/tokenizer regression and SigLIP2 HF-vs-JAX discrepancy are separate multimodal problems." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both involve chat templates, but one is all-zero assistant masks for multimodal inputs while the other crashes on tool-call assistant messages without content." + "reason": "Same apply_chat_template API, but different edge cases: multimodal assistant masks vs tool-call messages with no content." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Different problems: FineGrainedFP8Config failure in Qwen3Moe vs Qwen2_5_VL import error." + "reason": "Qwen3Moe FP8 failure and Qwen2.5-VL import error are unrelated." }, { "left": "issue:45381", "right": "issue:46209", "accept": false, - "reason": "Different hardware/model issues: Qwen2.5-VL video vision_position_ids bug vs FP8 inference failure on B200." + "reason": "Qwen2.5-VL video position ids and FP8 inference on B200 are different model/runtime problems." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Different serialization/training concerns: inconsistent lm_head.weight serialization in v5 vs LoRA merge collapse with extended vocabulary." + "reason": "Both mention tied weights, but one is inconsistent serialization and the other is post-LoRA model collapse after vocab extension." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Different tokenizer/model issues: extra space after added token vs output embedding reinitialized after resizing token embeddings." + "reason": "Tokenizer added-space behavior and post_init embedding reinitialization are distinct bugs." } ] }, @@ -9339,208 +9375,202 @@ "issue:44206|issue:45584", "issue:38175|issue:45412" ], - "prepared_review_unit_hash": "fae44aef7348ad27ac1a4e8a63bf7a3b5909767912506576b858eba986a9f4f1", + "prepared_review_unit_hash": "f3b3fae81bcf8e44948ee7e17505b443c5aa77afc2560d6bcc0b58703fc8a178", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35833, - "estimated_input_tokens": 8959, - "estimated_eval_tokens": 18174 + "serialized_chars": 35831, + "estimated_input_tokens": 8958, + "estimated_eval_tokens": 18172 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Nearly all candidate pairs are unrelated bugs or docs/test regressions. The only clear duplicate-like match is the Gemma4 quantized multimodal embedder dtype-casting issue (#46899/#46932).", - "confidence": 0.9, - "canonical_issue_reason": "#46932 is the best canonical issue: it is the clearest fix-oriented statement of the same Gemma4 quantized vision/audio embedder dtype-casting bug and directly overlaps with #46899.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues. The soft links are superficial theme matches (same model family, same subsystem, or both docs/runtime regressions) rather than true duplicates, so all soft edges should be rejected.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 45200 is the closest thing to a hub here: it is a concrete, reproducible Gemma4 multimodal bug and appears in multiple soft-similarity links. That said, it is not a true representative of the whole cluster because the rest of the items span unrelated models and code paths.", "canonical_pr_reason": null, - "best_issue_reason": "#46932 is the strongest issue to anchor the cluster because it names the concrete code-path and represents the only true duplicate pair in this set.", + "best_issue_reason": "45200 is the best single issue to anchor triage because it is specific and well-scoped, but the cluster is too heterogeneous for it to serve as a meaningful global canonical beyond its own Gemma4 area.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Different bugs: eval-time dtype drift vs Qwen3 MoE training memory usage." + "reason": "Both mention model internals, but one is a dtype regression in eval and the other is Qwen3 MoE training memory usage; different bugs and fixes." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "Different model failures: Qwen2.5-VL import error vs DeepSeek-Coder detokenization." + "reason": "Import/export symbol resolution for Qwen2.5-VL is unrelated to a tokenizer detokenization regression in deepseek-coder." }, { "left": "issue:44737", "right": "issue:46894", "accept": false, - "reason": "Unrelated code paths: XLNet CPU arange placement vs CohereASR loss shifting." + "reason": "XLNet CPU placement in relative positions and CohereASR double-shift training loss are unrelated code paths." }, { "left": "issue:43452", "right": "issue:46512", "accept": false, - "reason": "Different issues: gguf_file loading regression vs stale config.dtype after from_config()." + "reason": "gguf_file loading breaks model/tokenizer factories; the other issue is stale config.dtype after from_config. Not the same bug." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "Tokenizer special-token registration bug vs GraniteSpeech stale device override test." + "reason": "BigBirdTokenizer special-token registration and a stale CI device-override test in GraniteSpeech are different problems." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Different processors and failures: audio chat-template sample rate vs batched video shape handling." + "reason": "Both are processor bugs, but one is audio sampling-rate defaults and the other is batched video shape handling; separate fixes." }, { "left": "issue:43950", "right": "issue:46285", "accept": false, - "reason": "Different problems: non-persistent buffer loading corruption vs CLIP text model structural inconsistency." + "reason": "Non-persistent buffer loading corruption and CLIP text-model structural inconsistency are not the same underlying issue." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "Different tokenizer/model failures: FastSpeech2Conformer load vs Wav2Vec2 save/tokenization." + "reason": "Tokenizer loading for FastSpeech2Conformer is different from Wav2Vec2 save_pretrained/tokenization failure." }, { "left": "issue:43334", "right": "issue:43525", "accept": false, - "reason": "Both are pad_token_id AttributeErrors, but they involve different config classes and models, so not the same concrete bug." + "reason": "Both involve missing pad_token_id, but they are separate model/config classes and likely distinct fixes." }, { "left": "issue:45200", "right": "issue:46899", "accept": false, - "reason": "Different Gemma4 issues: missing mm_token_type_ids default vs quantization dtype-cast corruption." + "reason": "Both are Gemma4 multimodal issues, but one is token-type-id defaulting and the other is unsafe dtype casting under 4-bit quantization; different code paths." }, { "left": "issue:43526", "right": "issue:46192", "accept": false, - "reason": "Different processors and semantics: BEiT label reduction vs return_tensors being ignored with text_kwargs." + "reason": "reduce_labels returning one label and return_tensors being ignored are unrelated preprocessing bugs." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Unrelated failures: Perceiver non-default resolution vs RT-DETR memory not released." + "reason": "Perceiver resolution/interpolation failure and RT-DETR memory not being released are unrelated." }, { "left": "issue:45200", "right": "issue:46932", "accept": false, - "reason": "Different bugs in Gemma4: text-only mm_token_type_ids defaulting vs quantized embedder dtype casting." + "reason": "Both are Gemma4-related, but this is about mm_token_type_ids defaults while the other is about dtype casting in quantized embedders." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "Different generation test/regression issues with no shared concrete code path." + "reason": "A failed image-generation test and a stale num_return_sequences warning in paged generate are different symptoms and fixes." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "Different components: DAC latent reconstruction mismatch vs GraniteSpeech stale device override test." + "reason": "DAC forward/latent mismatch is unrelated to a GraniteSpeech CI stale test." }, { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Different multimodal bugs: Gemma4 token-type ids vs Qwen2.5-VL rope index scaling." + "reason": "Gemma4 token-type defaults and Qwen2.5-VL rope temporal position scaling are unrelated multimodal bugs." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "Different bugs in different models: SwitchTransformers sparse-layer creation vs Bamba torch.compile+SDPA failure." + "reason": "SwitchTransformers sparse-layer construction and Bamba torch.compile SDPA failure are unrelated." }, { "left": "issue:44561", "right": "issue:46234", "accept": false, - "reason": "Different missing utility regressions affecting different model families; not the same fix." + "reason": "Both concern utility availability, but one is a removed helper breaking trust_remote_code and the other is a missing import_utils symbol; not the same bug." }, { "left": "issue:41720", "right": "issue:46209", "accept": false, - "reason": "Both are hardware/runtime failures, but they are distinct device-specific issues on different models and accelerators." + "reason": "Qwen3 auto device mapping CUDA assert and FP8 inference on B200 are different hardware/runtime failures." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Different subsystems: OmDet-Turbo processor sizing vs LwDetrImageLoss AMP crash." + "reason": "Processor output resolution mismatch and LwDetr AMP/CUDA crash are unrelated." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs cleanup around old pipeline APIs, but they target different docs/pages and different API references." + "reason": "Both are docs regressions, but they affect different pages and different deprecations; not a duplicate change." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "Different helper/pathology: DAC from_latents mismatch vs qwen3_omni_moe feature-length helper behavior." + "reason": "DAC latent mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "Different loss/training bugs in unrelated models." + "reason": "AMP/CUDA crash in LwDetr loss and D-FINE auxiliary-loss omission are different bugs." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Different label-handling bugs in different processors and code paths." + "reason": "Metric gathering drops tuple labels, while BeitImageProcessorFast reduces labels incorrectly; similar domain, different functions and fixes." }, { "left": "issue:46512", "right": "issue:46821", "accept": false, - "reason": "Different config/runtime issues: stale dtype metadata vs TimesFM window_size AttributeError." + "reason": "Stale model config dtype and a missing window_size attribute in TimesFM are unrelated." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "Different failures: CPU offload device mismatch vs qwen3_omni_moe feature-length helper behavior." + "reason": "CPU offload device mismatch in suno/bark-small is unrelated to qwen3_omni_moe helper behavior." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Different audio/model problems: unsupported feature-extractor arg vs Whisper empty-transcription generation failure." + "reason": "Unsupported center argument crash in LasrFeatureExtractor and empty-transcription Whisper generation failure are different issues." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "Different issues: SigLIP2 zero probabilities vs RT-DETR memory cleanup." - }, - { - "left": "issue:46899", - "right": "issue:46932", - "accept": true, - "reason": "Both describe the same Gemma4 quantized vision/audio embedder dtype-casting bug; one is the symptom report and the other is the fix-oriented duplicate." + "reason": "Siglip2 zero probabilities and RT-DETR memory release are unrelated." } ] }, @@ -9645,20 +9675,20 @@ "issue:39692|issue:44493", "issue:28218|issue:32090" ], - "prepared_review_unit_hash": "8a30211c1102599c0fbfdbcf76708f00cb75dd16c5a418b3e214f1fc1aa9a1f5", + "prepared_review_unit_hash": "1faecc408d582c50e0134c3539de917fa374018341508743406c7d34f5abd7a4", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 37138, + "serialized_chars": 37137, "estimated_input_tokens": 9285, "estimated_eval_tokens": 18826 }, @@ -9667,192 +9697,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous issue cluster with a few loose topical overlaps (especially chat-template and loading/compatibility bugs), but no pair looks like the same underlying bug/change. I would keep one high-signal user-facing issue as the representative, but treat the soft edges as separate reports.", - "confidence": 0.31, - "canonical_issue_reason": "issue:44521 is the strongest representative candidate: it has clear user impact, active discussion, and a concrete reproducible API bug in a high-traffic area (chat templating for multimodal inputs).", + "summary": "This looks like a noisy semantic cluster rather than a duplicate set: the items span many unrelated Transformers subsystems, so there is no single underlying bug/change to collapse onto.", + "confidence": 0.97, + "canonical_issue_reason": "No clear canonical duplicate: the issues are only loosely related by being bug reports, but they cover different code paths (tokenization, model loading, cache, chat templates, CI, etc.).", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the best standalone issue to retain because it is specific, well-scoped, and has the most engagement among the candidates.", + "best_issue_reason": "No strong best issue for deduping; none of these is a reliable representative of the others because the cluster is not cohesive.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "Both are compatibility/import failures, but one is a missing flash-attention symbol and the other is a Torch 2.10 incompatibility; different root causes and code paths." + "reason": "Different failures: missing flash-attention helper import vs general Torch 2.10 incompatibility." }, { "left": "issue:45759", "right": "issue:46285", "accept": false, - "reason": "Both involve model/config structure, but one is Qwen multimodal config unwrapping and the other is CLIP class structure inconsistency; not the same bug." + "reason": "Different model paths: Qwen multimodal config unwrapping vs CLIP text-model structure inconsistency." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and BigBirdTokenizer special-token registration are unrelated subsystems and failures." + "reason": "Unrelated bugs: SwitchTransformers sparse-layer config vs BigBird mask-token registration." }, { "left": "issue:46424", "right": "issue:46858", "accept": false, - "reason": "Cache max-length handling for quantized cache and device-tensor seq-length arithmetic in static cache are distinct cache bugs." + "reason": "Both involve cache, but one is max_cache_length handling and the other is device-tensor seq_length arithmetic under static cache/compile." }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "PyTorch-missing ImportError vs MacOS bus error on a community CLIP model; different environments and failure modes." + "reason": "ImportError for missing PyTorch vs MacOS bus error loading a community CLIP model." }, { "left": "issue:29942", "right": "issue:32090", "accept": false, - "reason": "Flash Attention 2 test failures and Trainer _gpu_broadcast_one NoneType handling are different problems." + "reason": "Flash Attention 2 test failures vs Trainer GPU broadcast NoneType; not the same code path." }, { "left": "issue:43611", "right": "issue:44466", "accept": false, - "reason": "Both relate to v5 loading/weights behavior, but one is base_model_prefix loading breakage and the other is device-dependent lm_head serialization; not mergeable as one fix." + "reason": "Different loading/serialization regressions: base_model_prefix handling vs device-dependent lm_head.weight serialization." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "Sparse encoder-layer config bug in SwitchTransformers vs BEiT legacy-field load failure; different models and causes." + "reason": "Different causes: sparse layer creation with zero layers vs legacy field blocking BEiT pretrained loading." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba and a stale CI device-override test in GraniteSpeech are unrelated." + "reason": "torch.compile SDPA failure in Bamba vs stale device-override CI test in GraniteSpeech." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 torch.jit parsing/import failure vs qwen3_omni_moe feature-length helper behavior; different code paths." + "reason": "TorchScript parsing/import error vs helper returning unexpected feature lengths in qwen3_omni_moe." }, { "left": "issue:45520", "right": "issue:46170", "accept": false, - "reason": "flash_attn import key error on Python 3.13 and Windows ROCm torch.distributed dependency issues are separate environment/import problems." + "reason": "Different environment issues: flash_attn import key error on Python 3.13 vs Windows ROCm/torch.distributed dependency problems." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Pegasus v4/v5 output difference and Qwen3.5 modular recurrent-state reset are unrelated model-behavior bugs." + "reason": "Pegasus v4/v5 output difference vs recurrent states resetting in modular_qwen3_5; not the same bug." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Missing flash-support helper and cannot import HybridCache are both import errors, but for different symbols and modules." + "reason": "Different missing imports: _flash_supports_window_size vs HybridCache." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "SwitchTransformers config bug and qwen3_omni_moe length helper bug do not share an underlying implementation issue." + "reason": "Sparse-layer config bug and qwen3_omni_moe length helper bug are unrelated." }, { "left": "issue:43335", "right": "issue:44112", "accept": false, - "reason": "Config creation bug vs CI test flake; no shared code-path problem." + "reason": "No shared concrete failure: config construction bug vs CI test flake." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "SwitchTransformers sparse-layer misconfiguration and D-FINE auxiliary-loss omission are different model bugs." + "reason": "SwitchTransformers config bug vs D-FINE auxiliary-loss behavior; different subsystems." }, { "left": "issue:45290", "right": "issue:46326", "accept": false, - "reason": "Both are chat-template related, but one is a crash on assistant messages with tool calls and no content, while the other is continue_final_message behavior for mistral_common tokenizers; distinct bugs." + "reason": "Both are chat-template related, but one crashes on tool-call messages and the other is continue_final_message on mistral_common tokenizers." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype drift during evaluate and a pipeline error message for unsupported translation tasks are unrelated." + "reason": "Embedding dtype drift during eval vs pipeline error message about unsupported translation tasks in v5." }, { "left": "issue:44514", "right": "issue:44521", "accept": false, - "reason": "Both involve apply_chat_template, but one crashes on batched input with padding=False and the other returns zero assistant masks for multimodal inputs; separate defects." + "reason": "Both touch apply_chat_template, but one is a batched padding=False crash and the other is all-zero assistant masks for multimodal inputs." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Model loading regression for Llama 3.2 Vision and post_init reinitializing output embeddings after resize are different weight-loading/initialization issues." + "reason": "Model loading regression for Llama 3.2 vision vs embedding reinitialization after resize; different mechanisms." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "flash_attention integration crash on sink-less models vs CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding; different failure points." + "reason": "Flash-attention sink-less AttributeError vs Qwen3 MoE rotary embedding CPU-offload incompatibility." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Local-path slash import warning and GLM-4.7-Flash unit-test loading failure are not the same issue." + "reason": "Local-path trailing-slash warning vs inability to load a specific model in unit tests; not the same bug." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 error messaging for insufficient boxes and Trainer GPU broadcast NoneType error are unrelated." + "reason": "LayoutLMv3 error-message clarity issue vs Trainer broadcast NoneType failure." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC latent reconstruction mismatch and unexpected position-id key warnings are distinct model/runtime behaviors." + "reason": "DAC from_latents / STE mismatch vs unexpected position-id keys; distinct code paths." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "SigLIP2 docs/example errors and BeitImageProcessorFast label reduction bug are unrelated." + "reason": "SigLIP2 documentation example problems vs BeitImageProcessorFast reduce_labels bug." }, { "left": "issue:46431", "right": "issue:46566", "accept": false, - "reason": "Missing chat template on a processor and DiffusionGemma staying on meta device with device_map=auto are different loading/configuration issues." + "reason": "Processor missing chat template vs DiffusionGemma staying on meta with device_map=auto." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "CPU offload device mismatch for bark-small and float16 AMP crash in LwDetrImageLoss are separate device/dtype bugs." + "reason": "CPU-offload device mismatch in bark-small vs float16 AMP crash in LwDetrImageLoss." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "BitNet packed-weight unpacking during accelerate loading and a GraniteSpeech stale CI device-override test do not match." + "reason": "BitNet packed-weights loading issue vs GraniteSpeech CI stale override test." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "Doc/example mistakes for SigLIP2 and unexpected position-id key issues are different categories of bugs." + "reason": "Documentation/example failures vs position-id key mismatch; unrelated." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer extra-space behavior after added token and Trainer GPU broadcast NoneType error are unrelated." + "reason": "Tokenizer extra-space bug vs GPU broadcast NoneType error in Trainer." } ] }, @@ -9939,8 +9969,8 @@ "issue:44857|issue:45072", "issue:45593|issue:47000", "issue:38175|issue:44493", - "issue:27957|issue:34689", "issue:43550|issue:45083", + "issue:27957|issue:34689", "issue:44464|issue:46424", "issue:29942|issue:35141", "issue:29870|issue:32090", @@ -9954,202 +9984,202 @@ "issue:43278|issue:43381", "issue:44360|issue:44561" ], - "prepared_review_unit_hash": "77405b4711b4036eaff5e3e203b255ab3915fc832de53ce4373b1c3f28a1e06b", + "prepared_review_unit_hash": "03e336c7da0804860e56454f6f87302d208a65ca0388c3a63b861ab3431001d0", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35706, - "estimated_input_tokens": 8927, - "estimated_eval_tokens": 18110 + "serialized_chars": 35692, + "estimated_input_tokens": 8923, + "estimated_eval_tokens": 18102 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly heterogeneous, with issues spanning unrelated models, training, generation, and backend bugs. Only one soft pair looks like a true duplicate: the Qwen3.5 MoE `from_pretrained` regression reported twice.", - "confidence": 0.88, - "canonical_issue_reason": "issue:45310 is the best canonical issue here because it has an almost identical follow-up report (issue:45478) about the same Qwen3.5 MoE `from_pretrained` regression on `transformers>=5.4.0`.", + "summary": "This set is mostly a grab-bag of unrelated Transformers issues; only the two Qwen3.5 MoE `from_pretrained` reports look like true duplicates. All other soft pairs are false positives based on broad subsystem overlap or similar error wording.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45310 is the earlier and more complete report of the Qwen3.5 MoE `from_pretrained` failure; issue:45478 is effectively the same bug with a near-identical title.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest representative issue in this set: it is concrete, specific, and has the clearest duplicate signal. The rest of the items do not form a single coherent duplicate cluster.", + "best_issue_reason": "issue:45310 is the best issue anchor in this set because it most clearly identifies the concrete failing model and version-specific regression.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Same title, same model family, and same `transformers>=5.4.0` `from_pretrained` failure; this looks like a duplicate report of the same regression." + "reason": "Same model, same `transformers>=5.4.0` `from_pretrained` failure, near-identical titles; these look like duplicates." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Both mention v5 regressions, but one is Pegasus output drift and the other is a Deepseek tokenizer bug; different code paths and symptoms." + "reason": "Both mention output changes across versions, but they concern different models and different failure modes." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Different subsystems: audio feature extractor argument handling vs Gemma4 device placement with `device_map=auto`." + "reason": "One is a `LasrFeatureExtractor` crash from an unsupported arg; the other is a Gemma4 device-map mismatch." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "`resume_from_checkpoint` key mismatch is a training-state restore bug; `_get_feat_extract_output_lengths` in qwen3_omni_moe is a model-specific helper issue." + "reason": "Both are bug reports, but one is about checkpoint key mismatch and the other about a Qwen3 Omni feature-extraction helper." }, { "left": "issue:44617", "right": "issue:46697", "accept": false, - "reason": "CUDA OOM in Sam3Video and over-padding in Swin2SR are unrelated problems with different root causes." + "reason": "CUDA OOM for Sam3Video is unrelated to Swin2SR image over-padding." }, { "left": "issue:26413", "right": "issue:31501", "accept": false, - "reason": "Both affect training startup, but one is a resume/epoch iterator failure and the other is a FineWeb SLM startup issue; not the same bug." + "reason": "Both affect training startup/resume, but the concrete errors and code paths differ." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "LoRA/vocabulary tying collapse and a category-count limit error are unrelated failures." + "reason": "LoRA/vocab-collapse is unrelated to the 2^24 category limit runtime error." }, { "left": "issue:45459", "right": "issue:46234", "accept": false, - "reason": "One is tokenizer error masking when protobuf is absent; the other is a missing import utility symbol. Different fixes." + "reason": "Tokenizer protobuf error handling and a missing `import_utils` symbol are different bugs." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "Trailing-slash local model import warning is unrelated to tensor-parallel OOM while loading Qwen2-57B." + "reason": "Local-path import warning and tensor-parallel OOM on Qwen2 are unrelated." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Unsupported `NoneType` in GPU broadcast and a MacOS bus error with CLIP are different backend/platform issues." + "reason": "A `NoneType` broadcast failure in Trainer is not the same as a MacOS bus error loading CLIP." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "Both involve mixed precision, but one is LwDetr loss under float16 AMP and the other is dtype mismatches in inference for different models." + "reason": "Both involve dtype/precision, but they hit different models and different execution paths." }, { "left": "issue:45593", "right": "issue:47000", "accept": false, - "reason": "D-FINE missing auxiliary losses and RTDetr loss infeasible cost matrix are distinct detection-training bugs." + "reason": "Different DETR-family losses and different concrete crashes; not the same bug." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "Unexpected zero probabilities in siglip2-base-patch16-224 is unrelated to unexpected key warnings about position ids." + "reason": "Zero probabilities in SigLIP2 is unrelated to unexpected position-id keys across models." }, { - "left": "issue:27957", - "right": "issue:34689", + "left": "issue:43550", + "right": "issue:45083", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta and Llama 3.2 model loading breakage are different issues." + "reason": "torch.compile/SDPA failure in Bamba is unrelated to the Qwen3 Omni feature-length helper bug." }, { - "left": "issue:43550", - "right": "issue:45083", + "left": "issue:27957", + "right": "issue:34689", "accept": false, - "reason": "torch.compile + SDPA failure in Bamba and a multimodal feature-length helper bug are unrelated." + "reason": "Flash Attention 2 support in XLMRoberta and Llama 3.2 vision model loading are different problems." }, { "left": "issue:44464", "right": "issue:46424", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is not the same as cache-length handling limited to Quantized Cache." + "reason": "Compiled chunked generation inconsistency and cache-length handling are only broadly cache/generation related, not the same defect." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Failing Flash Attention 2 tests and embedding reinitialization after resizing are unrelated code paths." + "reason": "Flash Attention test failures and embedding reinitialization on resize are unrelated." }, { "left": "issue:29870", "right": "issue:32090", "accept": false, - "reason": "Bart `generate()` unbound local error and `_gpu_broadcast_one` unsupported type error are different failures." + "reason": "Bart generate mask initialization and Trainer GPU broadcast type handling are separate code paths." }, { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Chat-template tool-call crash and Qwen2.5-VL video `vision_position_ids` bug are separate multimodal issues." + "reason": "Chat-template tool-call crash and Qwen2.5-VL video position ids are different multimodal bugs." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "BF16-to-FP32 evaluation dtype change and undefined video-branch helpers in multimodal token counting are unrelated." + "reason": "Eval dtype drift and undefined video-branch variables are unrelated." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "German DistilBertModel error and TorchScript generator-expression export failure are unrelated legacy issues." + "reason": "German DistilBertModel issue and GeneratorExp/JIT export failure are completely different." }, { "left": "issue:29870", "right": "issue:33290", "accept": false, - "reason": "Bart generation crash and Adafactor/deepspeed OOM are different problems." + "reason": "Bart generation mask bug and DeepSpeed Adafactor OOM do not share the same underlying defect." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Pegasus output differences and Qwen3.5 flash-attention generation issues are not the same bug." + "reason": "Pegasus v4/v5 output drift and Qwen3.5 flash-attention generation failure are different issues." }, { "left": "issue:45290", "right": "issue:46655", "accept": false, - "reason": "Chat-template tool-call handling and SmolVLM video padding are unrelated preprocessing bugs." + "reason": "Chat template tool-call handling and SmolVLM video padding are unrelated." }, { "left": "issue:37428", "right": "issue:42175", "accept": false, - "reason": "Missing `_flash_supports_window_size` import and TensorFlow absent from `.[torch]` installs are separate packaging issues." + "reason": "Missing flash-attention utility import and TensorFlow not being installed with `.[torch]` are unrelated packaging/import problems." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "Evaluation dtype change and gradient checkpointing disallowed in eval mode are different runtime behaviors." + "reason": "Eval-mode checkpointing restriction is not the same as BF16-to-FP32 embedding dtype drift." }, { "left": "issue:44360", "right": "issue:44561", "accept": false, - "reason": "DSA indexer activation bug and removal of `is_torch_fx_available` breaking trust-remote-code models are unrelated." + "reason": "A missing ReLU in the DSA indexer and removal of `is_torch_fx_available` are unrelated changes." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The duplicate call on issue:45310 / issue:45478 is well supported by the identical title, same model family, and same transformers>=5.4.0 from_pretrained regression. The rest of the pairs are reasonably treated as unrelated, and there is no overstatement of evidence." + "feedback": "The duplicate call on issue:45310/issue:45478 is well supported by the packet: they have the same title and very similar issue bodies, with 45310 earlier in time. The other soft-edge rejections are conservative and consistent with the titles. No clear overstatement beyond a mild inference on completeness." }, "evaluator_used": true, "retried": false, @@ -10221,8 +10251,8 @@ ], "soft_pairs": [ "issue:45920|issue:46192", - "issue:44855|issue:46399", "issue:40444|issue:43526", + "issue:44855|issue:46399", "issue:32090|issue:33290", "issue:44977|issue:45229", "issue:43927|issue:45698", @@ -10244,207 +10274,207 @@ "issue:44186|issue:45198", "issue:44361|issue:45198", "issue:45198|issue:45803", - "issue:43611|issue:46121", "issue:44977|issue:45563", + "issue:43611|issue:46121", "issue:29870|issue:34689", "issue:29942|issue:30064" ], - "prepared_review_unit_hash": "cfaa260ab22f6180a7b63841e7cf45a1e621ae28d2ac338a6b06a7e91e92c7e4", + "prepared_review_unit_hash": "303e627fb920a70631847f12120ede073148ed9fee7f3400f0f2abc9b109da95", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36384, - "estimated_input_tokens": 9096, - "estimated_eval_tokens": 18448 + "serialized_chars": 36377, + "estimated_input_tokens": 9095, + "estimated_eval_tokens": 18446 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very heterogeneous: most items are distinct bug reports spanning tokenizers, cache/attention, training, loading, and docs. I do not see any strong duplicate pairs, so the safest canonical anchor is the broad v5 tokenizer regression issue.", - "confidence": 0.41, - "canonical_issue_reason": "issue:45920 is the broadest, most representative bug report in the set: it describes a v5 AutoTokenizer regression affecting multiple models, which makes it the best cluster anchor among otherwise unrelated issues.", + "summary": "This cluster is heterogeneous: the soft pairs share broad themes like tokenizers, save/load, cache, or device mapping, but they describe different concrete bugs and code paths. None look like safe duplicates to merge.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best overall issue candidate because it is high-level, user-impacting, and broad enough to summarize the most plausible shared regression theme in this cluster.", + "best_issue_reason": "issue:43873 is the strongest representative here: it is active, has multiple inbound references, and describes a concrete quantization/offloading bug. The rest of the items are mostly unrelated regressions or model-specific failures.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "Tokenizer ID regression across several models vs return_tensors being ignored when text_kwargs is passed; different behavior and code path." + "reason": "Both involve tokenizer/processor behavior, but one is wrong token IDs across models and the other is a return_tensors handling bug; different code paths." }, { - "left": "issue:44855", - "right": "issue:46399", + "left": "issue:40444", + "right": "issue:43526", "accept": false, - "reason": "Python import/IndentationError from decorator/comment parsing vs missing decorator registration for rotary embedding; unrelated failures." + "reason": "Both are vision/image-processing related, but one is dataset collation for multi-image prompts and the other is Beit label reduction; not the same bug." }, { - "left": "issue:40444", - "right": "issue:43526", + "left": "issue:44855", + "right": "issue:46399", "accept": false, - "reason": "Multi-image IterableDataset finetuning failure vs BeitImageProcessorFast reduce_labels bug; different subsystems and symptoms." + "reason": "Both are import/runtime failures, but one is a Python 3.13 IndentationError in DebertaV2 and the other is a missing decorator for Qwen3.5 kernels; unrelated fixes." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer NoneType broadcast error vs deepspeed/adafactor OOM; not the same underlying bug." + "reason": "Trainer broadcast NoneType error vs Deepspeed Adafactor OOM are distinct failures with different root causes." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Flash-attention generation failure vs multi-GPU CUDA OOM; only broad model-family overlap." + "reason": "Flash-attention generation failure and multi-GPU CUDA OOM are different problems, not one underlying bug." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Lost custom token IDs after save/load vs wrong custom module loaded after save_pretrained; both serialization-related but distinct defects." + "reason": "Both are save/load related, but one loses custom token IDs and the other loads the wrong custom module; different persistence bugs." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Trailing-slash import warning vs quantization offloading not working; unrelated." + "reason": "Trailing-slash import warning and quantization offloading failure are unrelated." }, { "left": "issue:46421", "right": "issue:46858", "accept": false, - "reason": "Lazy cache init causing recompilation vs static-cache seq_length returning a device tensor; different cache bugs." + "reason": "Both concern cache behavior, but one is spurious recompilation and the other is a device-tensor slice arithmetic issue in static cache; not mergeable as one fix." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Quantization/offloading failure vs Gemma4 multi-GPU OOM; memory-related but not the same code-path problem." + "reason": "Both can surface as memory-related, but offloading/quantization misbehavior is not the same as a model-specific CUDA OOM." }, { "left": "issue:25251", "right": "issue:32090", "accept": false, - "reason": "Pipeline top_k changes output shape vs trainer GPU broadcast NoneType; no meaningful duplication." + "reason": "Pipeline top_k nesting bug and Trainer GPU broadcast NoneType error are entirely separate subsystems." }, { "left": "issue:44448", "right": "issue:44466", "accept": false, - "reason": "Pegasus output mismatch across v4/v5 vs inconsistent lm_head.weight serialization depending on device; different model behaviors." + "reason": "Both are v5 regressions, but one is output-difference on a specific model and the other is serialization of tied weights; different issues." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Gradient checkpointing not usable in eval mode vs resume_from_checkpoint key mismatch; separate training-state issues." + "reason": "Eval-mode gradient checkpointing and resume_from_checkpoint key mismatch are unrelated trainer behaviors." }, { "left": "issue:43976", "right": "issue:45440", "accept": false, - "reason": "Python version support regression vs native DeepseekV3MoE divergence from remote implementation; unrelated." + "reason": "Python version compatibility and DeepseekV3MoE divergence are different bugs with no shared code path." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "Inconsistent eos_token_id config vs Qwen3.5 flash-attention generation problem; different bug classes." + "reason": "EOS token configuration inconsistency and flash-attention generation failure are not the same defect." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "Clarifying an error message for LayoutLMv3 vs TrainerState num_input_tokens_seen not updating; different components." + "reason": "One is a clearer error message for LayoutLMv3 box inputs; the other is TrainerState token counting not updating." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "MacOS bus error loading a CLIP model vs documentation example errors for SigLIP2; not the same bug." + "reason": "A MacOS bus error in a CLIP model and documentation example errors for SigLIP2 are separate problems." }, { "left": "issue:43278", "right": "issue:43575", "accept": false, - "reason": "BF16-to-FP32 dtype drift in eval vs tensor-parallel OOM when loading a large model; unrelated." + "reason": "BF16-to-FP32 dtype drift during eval and tensor-parallel load OOM are not the same bug." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "flash_attention.py crash on s_aux=None vs Gemma4 PLE device mismatch under device_map=auto; different failure points." + "reason": "One is an AttributeError on s_aux=None in flash_attention integration; the other is a PLE device mismatch in Gemma4 forward pass." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Removal of is_torch_fx_available breaking trust_remote_code vs DeepSeek tokenizer producing wrong results; unrelated regressions." + "reason": "A removed helper breaking trust_remote_code and a tokenizer regression in DeepSeek are unrelated." }, { "left": "issue:44568", "right": "issue:45491", "accept": false, - "reason": "add_special_tokens not adding BOS/EOS vs NaN embeddings from sliding-window attention with mixed-length batches; different layers of the stack." + "reason": "Tokenizer special-token insertion and Gemma3 NaN embeddings from sliding-window attention are distinct code paths." }, { "left": "issue:41720", "right": "issue:45229", "accept": false, - "reason": "Qwen3 auto device mapping assert vs Gemma4 multi-GPU OOM; both model-scale issues but not the same code-path." + "reason": "Auto device mapping CUDA assert and multi-GPU OOM are different model/runtime failures." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash vs Wav2Vec2 save_pretrained/tokenization failure; different tokenizer families and behaviors." + "reason": "LayoutLMv2 tokenizer padding/truncation crash and Wav2Vec2 save/tokenization failure are unrelated tokenizer bugs." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "MLukeTokenizer AttributeError vs Wav2Vec2 save_pretrained/tokenization failure; unrelated bugs." + "reason": "MLuke tokenizer AttributeError and Wav2Vec2 save/tokenization failure do not share the same underlying fix." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 tokenization/save failure vs FuyuBatchFeature bare except swallowing errors; different modules and symptoms." + "reason": "Saving/tokenization failure in Wav2Vec2 and a bare-except swallowing errors in Fuyu are different problems." }, { - "left": "issue:43611", - "right": "issue:46121", + "left": "issue:44977", + "right": "issue:45563", "accept": false, - "reason": "base_model_prefix loading break in v5 vs TypeError in convert_rope_params_to_dict with JSON-loaded lists; unrelated." + "reason": "Flash-attention generation failure and a stale warning in paged generate are not the same bug." }, { - "left": "issue:44977", - "right": "issue:45563", + "left": "issue:43611", + "right": "issue:46121", "accept": false, - "reason": "Flash-attention generation bug vs stale warning for num_return_sequences in paged generate(); not the same underlying defect." + "reason": "Base-model-prefix loading breakage and rope-parameter list handling are unrelated." }, { "left": "issue:29870", "right": "issue:34689", "accept": false, - "reason": "BART generate() decoder_attention_mask UnboundLocalError vs Llama 3.2 Vision model loading regression; unrelated." + "reason": "BART generate UnboundLocalError and Llama 3.2 vision model loading breakage are distinct regressions." }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Failing Flash Attention 2 tests vs image processor failing on void segmentation maps; distinct issues in different areas." + "reason": "Flash Attention 2 test failures and void segmentation map processing are unrelated." } ] }, @@ -10539,8 +10569,8 @@ "issue:45440|issue:45593", "issue:45372|issue:45406", "issue:43575|issue:44387", - "issue:43421|issue:45491", "issue:45406|issue:46701", + "issue:43421|issue:45491", "issue:45412|issue:46369", "issue:43066|issue:44462", "issue:45372|issue:45459", @@ -10556,256 +10586,256 @@ "issue:43994|issue:46369", "issue:42915|issue:43931" ], - "prepared_review_unit_hash": "f3ead12dc169c59f9049d5372ae1b6cf2139ebdbbe71d56be8188bcf9514bde9", + "prepared_review_unit_hash": "5d8b240c4c33638534a2834cbb602b5c0d2d6880be176f532a8b02422e25cfd1", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 37, - "serialized_chars": 38693, - "estimated_input_tokens": 9674, - "estimated_eval_tokens": 19604 + "serialized_chars": 38699, + "estimated_input_tokens": 9675, + "estimated_eval_tokens": 19606 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft links are all false positives: the issues cover unrelated models, APIs, regressions, and error messages. There is no true duplicate cluster here.", - "confidence": 0.94, - "canonical_issue_reason": "No canonical issue: the set is heterogeneous, so there is no single duplicate anchor.", - "canonical_pr_reason": "No PRs are present in this candidate set.", - "best_issue_reason": "No best issue: none of the issues cleanly represents the whole set as a duplicate target.", - "best_pr_reason": "No PRs are present in this candidate set.", + "summary": "The candidate pairs are mostly unrelated issues that only share broad subsystem names or vague keyword similarity. None should be merged as duplicates/aliases. No PRs are present. The cleanest representative issue in this set is 44462 because it describes a concrete, reproducible AutoTokenizer loading bug with a narrow fix surface.", + "confidence": 0.9, + "canonical_issue_reason": "No single canonical issue emerges from the set overall, but issue:44462 is the strongest representative because it is a concrete AutoTokenizer regression with a specific failure mode and clear code-path.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44462 is the most self-contained and actionable bug report here: precise repro, specific subsystem, and a focused fix path.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "Different bugs: tokenizer special-token registration vs recurrent state reset in Qwen3.5 cache handling." + "reason": "Different tokenization/output-decoding bug vs Qwen recurrent-state/cache behavior; only superficial similarity in model/inference context." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "Unrelated regressions: unsupported `center` arg in an audio extractor vs video input regression in Qwen VL models." + "reason": "LasrFeatureExtractor center-arg crash and Qwen VL video regression are different failure modes in different components." }, { "left": "issue:42915", "right": "issue:42994", "accept": false, - "reason": "One is FP8 config failure on Qwen3-MoE; the other is quantized model saving." + "reason": "FP8 config failure during model use is not the same as quantized model saving failure; separate code paths." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Different model implementations and failure modes: Smollm3 RoPE-layer mismatch vs DeepSeekV3MoE divergence from remote code." + "reason": "Different models and different defects: RoPE layer selection mismatch vs native/remote implementation divergence." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "Tokenizer special-token decode bug is not the same as AutoTokenizer failing to load a specific tokenizer class." + "reason": "BigBirdTokenizer special-token decode bug is unrelated to FastSpeech2Conformer tokenizer loading failure." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Sam3 pixel decoder dummy Conv2d is unrelated to SigLIP2 producing bad outputs." + "reason": "Dummy Conv2d in Sam3PixelDecoder and Siglip2 nonsensical outputs are unrelated model-specific problems." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "Dummy pixel-decoder layer issue is unrelated to RT-DETR memory not being released." + "reason": "A dummy layer implementation bug is not the same as an RT-DETR memory-release leak." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "Tensor-parallel OOM when loading a model is not the same as a resume-from-checkpoint key mismatch." + "reason": "Tensor-parallel OOM on Qwen2 load vs resume checkpoint key mismatch are different bugs and code paths." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "Torch export failure with `torch_compilable_check` is unrelated to Wav2Vec2 save/tokenization failures." + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 save/tokenization failure." }, { "left": "issue:29127", "right": "issue:33357", "accept": false, - "reason": "Clarity of an error message in LayoutLMv3 is unrelated to a MacOS bus error loading CLIP." + "reason": "LayoutLMv3 error-message clarity and MacOS CLIP bus error are not the same underlying bug." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Gemma 4 processor import/loading dependency failure is unrelated to a flash-attention crash on `s_aux=None`." + "reason": "Gemma 4 processor import failure is unrelated to flash_attention.py crashing on s_aux=None." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Executorch export guard/support issues are a different code path from flash-attention sink-less model crashes." + "reason": "Executable export failure and flash-attention attribute error are different concrete failures." }, { "left": "issue:45491", "right": "issue:46693", "accept": false, - "reason": "Gemma3 NaN embeddings from sliding-window padding is not the same as FA2 long-context performance regression." + "reason": "Gemma3 NaNs from sliding-window padding are not the same as a performance regression in FA2 long-context generation." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Trailing-slash local model import warning is unrelated to gradient checkpointing in eval mode." + "reason": "Local-path import warning with trailing slash is unrelated to gradient-checkpointing eval-mode restriction." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "Error-message clarity for boxes vs embedding reinitialization during `post_init` are different problems." + "reason": "Clarifying an error message is not the same as embedding reinitialization during post_init." }, { "left": "issue:45245", "right": "issue:45563", "accept": false, - "reason": "Category-count limit runtime error is unrelated to a stale warning in paged `generate()`." + "reason": "Category-count runtime error and stale num_return_sequences warning are unrelated." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Tokenizer decode-error masking due to missing protobuf is unrelated to leftover dummy classes in `dummy_pt_objects.py`." + "reason": "Tokenizer error masking from protobuf handling is unrelated to leftover dummy-class cleanup." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 inference nonsense and RT-DETR memory retention are unrelated issues." + "reason": "Siglip2 inference output corruption and RT-DETR memory not releasing are different bugs." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "DeepSeekV3MoE implementation divergence is unrelated to D-FINE auxiliary-loss behavior." + "reason": "DeepSeekV3MoE divergence from remote impl is unrelated to D-FINE auxiliary-loss behavior." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Both concern Gemma 4 processor loading, but one is an import/dependency failure and the other is a missing `_tokenizer` attribute in serve." + "reason": "Gemma 4 processor loading import error and Gemma4Processor missing _tokenizer in serve are related by model family but not the same concrete failure." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Tensor-parallel OOM on a specific model is not the same as int4 quantization reserved-memory regression." + "reason": "Tensor-parallel OOM on one model is not the same as increased CUDA reserved memory under int4 quantization." }, { - "left": "issue:43421", - "right": "issue:45491", + "left": "issue:45406", + "right": "issue:46701", "accept": false, - "reason": "Runtime special-token post-processor update bug is unrelated to Gemma3 NaN embeddings under mixed-length batching." + "reason": "Gemma4Processor serve crash is unrelated to DiffusionGemmaGenerationOutput pipeline failure." }, { - "left": "issue:45406", - "right": "issue:46701", + "left": "issue:43421", + "right": "issue:45491", "accept": false, - "reason": "Gemma 4 serve `_tokenizer` crash is unrelated to DiffusionGemmaGenerationOutput pipeline failure." + "reason": "Runtime special-token post-processor update issue is unrelated to Gemma3 NaN embeddings." }, { "left": "issue:45412", "right": "issue:46369", "accept": false, - "reason": "RT-DETR memory leak on deletion is not the same as sam3-litetext memory overflow during use." + "reason": "RT-DETR memory leak and sam3-litetext memory overflow are different model-specific memory issues." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 is unrelated to AutoTokenizer ignoring repository `tokenizer.json`." + "reason": "Wrong tokenizer decoder type in v5 is not the same as AutoTokenizer ignoring tokenizer.json from the repo." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma 4 loading fails due to a missing `ReasoningEffort` import; this is not the same as protobuf-related tokenizer error handling." + "reason": "Gemma 4 import failure and protobuf-related tokenizer exception masking are unrelated." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Different tokenizer/model families and failure modes: AutoTokenizer load failure vs bfloat16 dtype mismatch." + "reason": "Tokenizer loading failure and dtype mismatch in bfloat16 inference are unrelated." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to Kosmos2.5 OCR index errors." + "reason": "Incorrect label reduction in BeitImageProcessorFast is unrelated to Kosmos2.5 long-ocr index errors." }, { "left": "issue:27957", "right": "issue:33290", "accept": false, - "reason": "XLM-RoBERTa Flash Attention 2 support and Adafactor+DeepSpeed OOM are different issues." + "reason": "Flash Attention 2 support for XLMRoberta and Adafactor/deepspeed OOM are unrelated." }, { "left": "issue:27957", "right": "issue:29127", "accept": false, - "reason": "Flash Attention 2 support for XLM-RoBERTa is unrelated to a LayoutLMv3 error-message clarity issue." + "reason": "FA2 model support and LayoutLMv3 error-message clarity are different issues." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "ImportError for a missing Qwen2.5-VL class is not the same as a generic dependency issue with transformers." + "reason": "Missing Qwen2_5_VL import and a generic dependency issue are not the same underlying bug." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "Adafactor/DeepSpeed OOM is unrelated to Llama 3.2 Vision model loading breaking in 4.46.2." + "reason": "Optimizer-related OOM and model-loading shape mismatch are unrelated." }, { "left": "issue:43526", "right": "issue:46693", "accept": false, - "reason": "BeitImageProcessorFast label handling is unrelated to FA2 long-context generation performance regression." + "reason": "BeitImageProcessorFast label handling and FA2 long-context performance regression are unrelated." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "DeepSpeed Adafactor OOM is unrelated to token-embedding resize reinitialization in `post_init`." + "reason": "Deepspeed Adafactor OOM and embedding reinit during post_init are different defects." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "AMP/CUDA crash in LwDetrImageLoss is unrelated to flash-attention sink-less model attribute errors." + "reason": "AMP/CUDA crash in LwDetrImageLoss is unrelated to flash_attention.py s_aux=None handling." }, { "left": "issue:43994", "right": "issue:46369", "accept": false, - "reason": "SigLIP2 bad outputs and sam3-litetext memory overflow are different model/runtime problems." + "reason": "Siglip2 inference corruption and sam3-litetext memory overflow are different bugs." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "FineGrainedFP8Config failure on Qwen3-MoE is unrelated to Qwen3-VL weight-shape mismatch during loading." + "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are different model-loading/training issues." } ] }, @@ -10906,200 +10936,197 @@ "issue:43643|issue:44913", "issue:44855|issue:45137" ], - "prepared_review_unit_hash": "67980c62772e133b32afa06cce1a78bf10b0a6bf56d0163592c4983bf69eec0b", + "prepared_review_unit_hash": "4411fd35e1b7ad1988d6016668b95a539a605c079affa7e7120b21e753de0487", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35628, - "estimated_input_tokens": 8907, - "estimated_eval_tokens": 18070 + "serialized_chars": 35630, + "estimated_input_tokens": 8908, + "estimated_eval_tokens": 18072 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a collection of unrelated transformers bug reports. The only plausible duplicate pair is the two generic Flash Attention 2 issues; the rest are too model- or symptom-specific to safely merge.", - "confidence": 0.57, - "canonical_issue_reason": "issue:46640 is the broadest and most representative of the only likely duplicate pair, with a generic Flash Attention 2 title and recent discussion.", - "canonical_pr_reason": null, - "best_issue_reason": "issue:46640 is the best global issue candidate because it is the most generic Flash Attention 2 report and the clearest duplicate anchor in this set.", - "best_pr_reason": null, + "summary": "The cluster is a mixed bag of unrelated issues across model configs, tokenizers, vision/audio processors, flash-attention, and runtime regressions. None of the soft pairs look like the same concrete bug/change, so all proposed merges should be rejected.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue is appropriate: the set spans many unrelated bugs and feature regressions, so there is no clear cluster anchor.", + "canonical_pr_reason": "No PRs are present in this cluster.", + "best_issue_reason": "If one issue must be highlighted, issue 43873 is the strongest representative by activity and inbound references, but it is not a duplicate of the others.", + "best_pr_reason": "No PRs are present, so there is no best PR candidate.", "soft_edge_verdicts": [ { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Both are tokenizer/processor bugs, but they affect different components and failure modes; not the same underlying bug." + "reason": "Different processor/tokenizer bugs: one is BeitImageProcessorFast label reduction, the other is LayoutLMv2 NER/padding crashes." }, { "left": "issue:45250", "right": "issue:46640", - "accept": true, - "reason": "Both are generic Flash Attention 2 issues and appear to describe the same underlying topic/bug." + "accept": false, + "reason": "Both mention Flash Attention 2, but the reports are too generic and not the same concrete failure mode." }, { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Different Qwen model families and different symptoms; no clear shared code-path bug." + "reason": "Qwen3-VL pad_token_id loading failure is unrelated to Qwen3-MoE sliding_window behavior." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "One is about eval/gradient-checkpointing behavior, the other about offloading with quantization." + "reason": "Gradient checkpointing in eval mode is unrelated to quantization/offloading behavior." }, { "left": "issue:45588", "right": "issue:45600", "accept": false, - "reason": "Different code areas and failures: flash-attention sink handling versus auto-mapping references to removed configs." + "reason": "Different code paths: flash_attention sink-less crash vs removed Sam3LiteText auto_mapping references." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "The themes overlap around initialization, but the concrete bugs are different and not safely mergeable." + "reason": "Both concern model initialization changes, but they describe different breakages and model-specific behavior." }, { "left": "issue:46121", "right": "issue:46399", "accept": false, - "reason": "Rope-parameter validation and Qwen3.5 kernel registration are unrelated bugs." + "reason": "Rope-validation TypeError and Qwen3.5 kernel registration ValueError are distinct failures." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding and the other is video position-id math." + "reason": "Both involve Qwen2.5-VL, but one is chat-template batching and the other is video position-id scaling." }, { "left": "issue:44977", "right": "issue:45245", "accept": false, - "reason": "Completely different failure modes and subsystems; only share being user-facing runtime errors." + "reason": "Flash-attention generation issues are unrelated to the category-count RuntimeError." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "Different models and code paths; one is hidden-states output, the other CPU offload compatibility." + "reason": "CLIPVisionModel hidden_states regression is unrelated to Qwen3_5Moe rotary embedding CPU offload compatibility." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Unrelated functionality: recurrent cache state handling versus processor batch decoding behavior." + "reason": "Recurrent cache-state reset and Whisper batch_decode skip_special_tokens are different bugs in different components." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Both touch audio processors, but the bugs are distinct and in different APIs." + "reason": "Audio chat-template sampling-rate defaulting is unrelated to LasrFeatureExtractor passing an unsupported center argument." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "Different modalities and different failure mechanisms; not a duplicate." + "reason": "FastSpeech2Conformer tokenizer loading and Perceiver image-size handling are unrelated issues." }, { "left": "issue:42490", "right": "issue:44743", "accept": false, - "reason": "Local model-path import warnings and Qwen modular cache-state reset are unrelated." + "reason": "Local-path import warnings and Qwen3.5 recurrent-state resets do not share the same underlying bug." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet accelerate-loading bug versus tokenizer loading failure; no shared underlying issue." + "reason": "BitNet accelerate-loading unpacking bug is unrelated to AutoTokenizer loading FastSpeech2ConformerTokenizer." }, { "left": "issue:27957", "right": "issue:29870", "accept": false, - "reason": "Different models and different errors; both are generation-related but not the same bug." + "reason": "XLMRoberta Flash Attention 2 support and BART generate() decoder_attention_mask error are different problems." }, { "left": "issue:42175", "right": "issue:46234", "accept": false, - "reason": "Packaging/backend selection and missing import_utils helper are distinct issues." + "reason": "TensorFlow extra packaging and a missing import_utils helper are unrelated packaging/API issues." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "PatchTSMixer constructor policy and Qwen video-input regression are unrelated." + "reason": "PatchTSMixer post_init policy and Qwen video-input regression are separate model-specific regressions." }, { "left": "issue:45003", "right": "issue:45440", "accept": false, - "reason": "Generic modeling_utils safety issue versus a DeepSeekV3MoE implementation divergence." + "reason": "Unsafe sys.modules access in modeling_utils is unrelated to DeepseekV3MoE divergence from remote implementation." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Both are config API discrepancies, but they concern different models and different missing fields." + "reason": "Gemma3TextConfig API mismatch and Llama4Config missing pad_token_id are different config-schema issues." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 implementation mismatch and docs referencing removed pipeline APIs are unrelated." + "reason": "SigLIP2 implementation discrepancy and pipeline() documentation drift are unrelated." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization/offloading behavior and eos_token_id config inconsistency are not the same bug." + "reason": "Quantization/offloading behavior is unrelated to Gemma-3 eos_token_id configuration inconsistency." }, { "left": "issue:45325", "right": "issue:46697", "accept": false, - "reason": "Different models, different image/video preprocessing paths, and different symptoms." + "reason": "Qwen2.5-VL rope index scaling and Swin2SR over-padding are unrelated image/video preprocessing bugs." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "Tokenizer AddedToken duplication and list_repo_files revision forwarding are unrelated." + "reason": "AddedToken special-argument duplication and list_repo_files revision forwarding are unrelated API bugs." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Both involve config loading, but the specific bugs differ: missing fields under trust_remote_code versus rotary_pct not persisting on reload." + "reason": "Missing fields from trust_remote_code config loading and GPTNeoX rotary_pct reload loss are different config deserialization bugs." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python import parsing issue and DeepSpeed ZeRO3 deque error are unrelated." + "reason": "Python 3.13 TorchScript parsing failure is unrelated to the DeepSpeed ZeRO3 deque underflow." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft pair, issue:45250 and issue:46640, is plausibly the same generic Flash Attention 2 bug, while the other pairwise rejections are appropriately cautious and supported by the titles/body signals." - }, - "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 }, @@ -11170,8 +11197,8 @@ "issue:43316|issue:45759", "issue:44987|issue:46178", "issue:44190|issue:44496", - "issue:42617|issue:43575", "issue:43931|issue:45216", + "issue:42617|issue:43575", "issue:43701|issue:46506", "issue:43232|issue:43388", "issue:44442|issue:45803", @@ -11196,208 +11223,208 @@ "issue:43646|issue:44913", "issue:43452|issue:44496" ], - "prepared_review_unit_hash": "845806ff6e50326e36ca818360349d3c914b58b53303474eb494329e0ca9ef7a", + "prepared_review_unit_hash": "171d405fe182bf3143cf265faf4cde767881ac2da860116c6578ab36b410d5d3", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36311, - "estimated_input_tokens": 9078, - "estimated_eval_tokens": 18412 + "serialized_chars": 36315, + "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 not a duplicate set; it\u2019s a grab-bag of unrelated Transformers bugs/regressions across loading, generation, multimodal, training, and config handling. None of the soft pairs look like the same underlying change or defect.", - "confidence": 0.89, - "canonical_issue_reason": "No single canonical issue stands out: the items cover many unrelated code paths and model families, with no explicit linkage or shared filenames/targets.", + "summary": "This cluster is a loose collection of unrelated Transformers issues. The soft pairs mostly share broad keywords like loading, config, generation, or compilation, but they do not appear to describe the same concrete bug or fix.", + "confidence": 0.96, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43761 is the most concrete and self-contained bug report: a specific regression, a specific model, clear expected-vs-actual behavior, and an actionable code path.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "Both are model-loading failures, but one is an FSDP CPU RAM-efficient loading issue and the other is a Qwen3-VL weight-shape mismatch; different failure modes and likely different fixes." + "reason": "Both are model-loading failures, but one is about FSDP CPU RAM efficient loading and the other is a Qwen3-VL weight-shape mismatch; different code paths and symptoms." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Both mention composite config/text_config behavior, but one is an API discrepancy in Gemma3TextConfig while the other is missing text_config unwrapping in AutoModelForCausalLM.from_config for Qwen multimodal configs; not the same bug." + "reason": "Both involve config APIs for multimodal models, but they target different model families and different failures; not the same bug." }, { "left": "issue:44987", "right": "issue:46178", "accept": false, - "reason": "These are unrelated import/load errors: one is a specific model load failure on transformers>=5.1.0, the other is a backend mapping error for tensorflow_text." + "reason": "One is a specific model load regression, the other is an import-utils backend registration error; unrelated causes." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in an example script is unrelated to an unrecognized model/config error when loading a specific model." + "reason": "Different failure modes: dataset loading in a training script vs unrecognized model/config loading." }, { - "left": "issue:42617", - "right": "issue:43575", + "left": "issue:43931", + "right": "issue:45216", "accept": false, - "reason": "Both involve parallel/model loading, but one is failure to run a 3D parallel example and the other is TP loading OOM for a specific Qwen model; not the same underlying problem." + "reason": "Both mention loading problems, but one is a checkpoint weight-shape mismatch and the other is a save_pretrained regression for a different model family." }, { - "left": "issue:43931", - "right": "issue:45216", + "left": "issue:42617", + "right": "issue:43575", "accept": false, - "reason": "Both are Qwen-related loading/save issues, but one is a weight-shape mismatch on load and the other is an incorrect checkpoint saved by save_pretrained; distinct code paths." + "reason": "Both are distributed/model loading issues, but one is a script execution problem and the other is an OOM during tensor-parallel loading; not the same defect." }, { "left": "issue:43701", "right": "issue:46506", "accept": false, - "reason": "Resume-from-checkpoint key mismatch is unrelated to CPU/CUDA output divergence in PPDocLayoutV3." + "reason": "Resume-from-checkpoint key mismatch is unrelated to CPU/CUDA output divergence in a vision model." }, { "left": "issue:43232", "right": "issue:43388", "accept": false, - "reason": "A generation kwargs update bug after sync_gpus is unrelated to gather_for_metrics dropping tuple labels in the last batch." + "reason": "Generation state-update logic and gather_for_metrics label truncation are separate training/runtime bugs." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "Tokenizer loading failure for one model is unrelated to Fuyu batch feature tensor conversion swallowing exceptions." + "reason": "Tokenizer loading failure and a bare-except in tensor conversion are different subsystems and symptoms." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba is unrelated to checkpoint resume key mismatches." + "reason": "torch.compile/SDPA failure is unrelated to checkpoint key mismatch." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "TP OOM while loading one model is not the same as a key error loading another model." + "reason": "OOM when loading a large Qwen model is not the same as a key error for a small MobileLLM model." }, { "left": "issue:27957", "right": "issue:32090", "accept": false, - "reason": "Flash Attention 2 support for XLMRoberta is unrelated to a Trainer GPU broadcast NoneType error." + "reason": "Flash Attention 2 support and a Trainer broadcast type error are unrelated." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "These are both about model outputs, but one is a specific CLIPVisionModel regression and the other is a generic ModelOutput key-assignment bug; insufficient evidence of the same root cause." + "reason": "One is missing hidden_states from CLIPVisionModel.forward; the other is generic ModelOutput key assignment behavior." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "CLIP hidden_states regression is unrelated to attention-kernel divergence under torch.compile." + "reason": "Different issues in model outputs vs attention kernel branching under torch.compile." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "CPU offload device mismatch for bark-small is unrelated to resume-from-checkpoint key mismatches." + "reason": "CPU offload device mismatch is unrelated to resume_from_checkpoint key mismatches." }, { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Pipeline top_k output shape change is unrelated to AutoModel missing PyTorch import errors." + "reason": "Pipeline output nesting and missing PyTorch import are unrelated." }, { "left": "issue:46829", "right": "issue:46932", "accept": false, - "reason": "These are both vision/multimodal issues, but one is a VideoPrism output-field bug and the other is a Gemma4 quantized embedder dtype-casting fix; different components and fixes." + "reason": "Video classification output field misassignment is not the same as Gemma4 quantized-embedder dtype casting." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Saving dequantized Ministral/Devstral models is a different problem from from_pretrained loading the wrong custom module after save_pretrained." + "reason": "Both touch save/load flows, but one is saving dequantized models and the other is loading the wrong custom module after save_pretrained; different bugs." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "A regression in custom processor/tokenizer behavior is unrelated to an incorrect pipeline error message about translation tasks." + "reason": "A regression in processor/tokenizer image handling is unrelated to pipeline error-message wording." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "A model_max_length question for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." + "reason": "Model_max_length behavior for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Processor/tokenizer API regression is unrelated to Gemma-3 eos_token_id inconsistency." + "reason": "Both involve tokenizer/config regressions, but one is processor.tokenizer/image passing and the other is eos_token_id mismatch; not the same fix." }, { "left": "issue:44871", "right": "issue:45584", "accept": false, - "reason": "Both involve special tokens, but Gemma-3 eos_token_id mismatch and Whisper empty-transcription failure are different bugs." + "reason": "Both mention special-token handling, but one is an eos_token_id config inconsistency and the other is empty-transcription generation failure after align_special_tokens." }, { "left": "issue:46234", "right": "issue:46291", "accept": false, - "reason": "Missing is_falcon_mamba_ssm_available and pipelines import failing on kernels version/revision requirements are separate import-time issues." + "reason": "Missing availability helper and kernel dependency validation are distinct import-time issues." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "A patchtsmixer post_init API constraint issue is unrelated to non-persistent buffers being filled with junk." + "reason": "Optional post_init handling in one model class is unrelated to non-persistent buffer corruption in v5." }, { "left": "issue:43316", "right": "issue:43716", "accept": false, - "reason": "Gemma3TextConfig API discrepancy is unrelated to Mistral-3 image preprocessor dtype mismatch." + "reason": "Config API discrepancy in Gemma3TextConfig is unrelated to a vision dtype mismatch in Mistral-3." }, { "left": "issue:44464", "right": "issue:46421", "accept": false, - "reason": "Both mention generation/compilation, but inconsistent chunked generation outputs and chunked prefill recompilation are different behaviors and likely different fixes." + "reason": "Both relate to chunked/compiled generation, but one is incorrect outputs and the other is spurious recompilation from cache lazy init." }, { "left": "issue:43316", "right": "issue:46821", "accept": false, - "reason": "Gemma3 config API discrepancy is unrelated to TimesFM window_size AttributeError." + "reason": "Gemma3 config API mismatch and TimesFM window_size AttributeError are unrelated." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Custom model initialization breakage in Transformers 5.0.0 is unrelated to GPTNeoXConfig rotary_pct not persisting on reload." + "reason": "Custom model initialization breakage is unrelated to GPTNeoX rotary_pct not persisting on reload." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "GGUF support for AutoTokenizer/AutoModel loading is unrelated to an unrecognized model config error." + "reason": "gguf_file handling for tokenizer/model loading is a different problem from unrecognized model config loading." } ] }, @@ -11496,14 +11523,14 @@ "issue:43388|issue:43688", "issue:44190|issue:44964" ], - "prepared_review_unit_hash": "cac736cfd8dd58fc16bf517f863d9fbfc48294f198543fbbe984fd3f038ea2b6", + "prepared_review_unit_hash": "ef50b5e5ad29e3779882d1b77b56205e4a08b31d4995f61ee55737d38c7dfa85", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, @@ -11518,162 +11545,162 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a true duplicate set; the soft edges mostly connect unrelated issues that only share broad vocabulary like loading, cache, flash-attention, or model family names. I would keep the cluster anchored on the high-activity model-loading bug, but reject every soft pair.", - "confidence": 0.95, - "canonical_issue_reason": "issue:45698 is the strongest representative anchor: it has the highest discussion/inbound-reference signal and is a broad, user-facing `from_pretrained` / `save_pretrained` loading bug.", + "summary": "This cluster is mostly a loose similarity bucket of unrelated issues; none of the soft pairs look like true duplicates or the same concrete fix. The strongest standalone issue is the save/load custom-module regression.", + "confidence": 0.34, + "canonical_issue_reason": "issue:45698 is the strongest representative: it has broad user impact, clear save_pretrained/from_pretrained reproduction, and the highest discussion/reference activity among the set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45698 is the best cluster representative because it is the most prominent and generally scoped issue in the set, even though the surrounding items are not true duplicates.", + "best_issue_reason": "issue:45698 is the best overall issue candidate because it is concrete, widely impactful, and more central than the other mostly model-specific edge cases.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Both involve model/config persistence, but one is an `eos_token_id` mismatch and the other is a quantized-model reload failure; different bugs and code paths." + "reason": "Both involve config/state roundtrips, but one is eos_token_id inconsistency and the other is quantized model reload failure; different bugs and fixes." }, { "left": "issue:45803", "right": "issue:46365", "accept": false, - "reason": "Completely different failures: a broad exception in tensor conversion vs an import failure caused by `datasets`; no shared underlying bug." + "reason": "One is a bare-except bug in tensor conversion, the other is an import failure tied to datasets/Trainer loading; not the same code path." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 breakage and a Janus image-generation test failure are unrelated model areas and symptoms." + "reason": "ESM2 model breakage and a janus image-generation test failure are unrelated model-specific issues." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "Same general model family, but one is tied-weight initialization and the other is FlashAttention support; not the same concrete issue." + "reason": "Both mention newer model generation problems, but they target different models and likely different FlashAttention/config paths." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "Both mention SAM3, but dummy `Conv2d` in the pixel decoder and video OOM are distinct problems." + "reason": "Both are SAM 3 video-related, but one is a dummy Conv2d placeholder and the other is an OOM symptom; not the same fix." }, { "left": "issue:44242", "right": "issue:46894", "accept": false, - "reason": "Both are training-loss-related, but they affect different models and different loss computation bugs." + "reason": "Both concern training loss, but the models and mechanics differ; no evidence they share the same underlying bug." }, { "left": "issue:44610", "right": "issue:46369", "accept": false, - "reason": "OmDet-Turbo input-size mismatch and SAM lite text memory overflow are unrelated." + "reason": "Processor/model input-size mismatch is unrelated to a memory overflow when calling sam3-litetext." }, { "left": "issue:45632", "right": "issue:46424", "accept": false, - "reason": "Both mention cache behavior, but one is a path-collision bug and the other is a cache-length configuration limitation." + "reason": "Both touch cache behavior, but one is a path collision for trust_remote_code and the other is max_cache_length semantics; different bugs." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Both are multimodal input-processing issues, but they hit different processors and different failure modes." + "reason": "Both are Qwen multimodal input issues, but one is video processing and the other is batched chat-template padding behavior." }, { "left": "issue:45542", "right": "issue:46234", "accept": false, - "reason": "Optional-dependency/import plumbing is the only overlap; the actual failures are unrelated." + "reason": "Both involve backend availability, but tensorboard/TensorFlow setup is not the same as missing falcon_mamba backend export." }, { "left": "issue:43572", "right": "issue:46178", "accept": false, - "reason": "A missing config attribute and an invalid backend mapping are different API breakages." + "reason": "Missing pad_token_idx and a tensorflow_text backend mapping error are unrelated configuration/import failures." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "One is a test failure for Janus generation, the other is a stale auto-mapping reference; not the same bug." + "reason": "A janus test failure and stale auto_mappings references are different problems in different code paths." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "Bare `except` swallowing errors and a Windows/ROCm import dependency problem are unrelated." + "reason": "Bare except in Fuyu tensor conversion is unrelated to a Windows/ROCm torch.distributed import dependency issue." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Both touch generation, but Janus image tests and Qwen3.5 FlashAttention behavior are separate issues." + "reason": "Both are generation-related, but janus image generation and Qwen3.5 flash-attention behavior are not the same bug." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "Both involve model loading, but one is a legacy field in BEiT preset loading and the other is wrong custom-module restoration after save/load." + "reason": "Both concern loading/saving models, but BEiT legacy preset loading is not the same as custom-module resolution after save_pretrained." }, { "left": "issue:43650", "right": "issue:46202", "accept": false, - "reason": "No meaningful overlap beyond being issue titles; the data request and Gemma pipeline roundtrip bug are unrelated." + "reason": "The titles are too generic to imply the same bug; no shared code path or concrete change is evident." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "LoRA merge collapse with extended vocab and Gemma 4 token-type handling are different bugs." + "reason": "One is LoRA merge collapse with tied embeddings, the other is missing mm_token_type_ids defaults; different failure modes and fixes." }, { "left": "issue:43352", "right": "issue:46640", "accept": false, - "reason": "One is a specific model not supporting FlashAttention 2 yet; the other is a broader FlashAttention 2 issue." + "reason": "Both mention Flash Attention 2, but model/trigger details are missing and there is no evidence they are the same concrete compatibility fix." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Both are vision/video-adjacent, but one is a processor-selection bug and the other is a missing attribute on a SAM3 output object." + "reason": "A pipeline using image processors and a missing Sam3VisionEncoderOutput attribute are different SAM/video bugs." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "A generic v5 buffer-initialization regression and a GPTNeoX config reload bug are not the same failure." + "reason": "Non-persistent buffer junk filling and GPTNeoX rotary_pct reloading are unrelated state/config regressions." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "A missing config attribute and a sliding-window behavior issue are unrelated." + "reason": "Missing pad_token_id in Llama4Config and a Qwen3-MoE sliding_window issue are distinct model/config bugs." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Token-embedding resize initialization and trailing-slash local-model warning are unrelated." + "reason": "Embedding reinitialization after resize and a local-path trailing-slash import warning are unrelated." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both involve cache/state handling, but one is missing GenerationMixin cache in chunked prefill and the other resets recurrent states with `seq_len>1`." + "reason": "Both involve cache semantics, but chunked_prefill cache handling and recurrent-state reset in modular_qwen3_5.py are different issues." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Evaluation label truncation and auxiliary-loss normalization are different training/eval bugs." + "reason": "Label gathering in metrics and auxiliary-loss normalization are unrelated training/runtime bugs." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Local dataset loading in a script and a multimodal model load failure are unrelated." + "reason": "A local dataset loading failure and inability to load Phi-4 multimodal instruct are separate loading problems." } ] }, @@ -11772,25 +11799,25 @@ "issue:43065|issue:44610", "issue:43335|issue:44190", "issue:44617|issue:46369", - "issue:45588|issue:47013", "issue:43854|issue:44792", "issue:44479|issue:44734", + "issue:45588|issue:47013", "issue:45381|issue:46655" ], - "prepared_review_unit_hash": "fabb22b03f8b0fe6607154003a8c541f4d381f6d1cbe352d1749e6eb4ede63f1", + "prepared_review_unit_hash": "5871efaee22c50a914093502cdddd4c65234f62ed6608b8b2be8790e740c8f83", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36534, + "serialized_chars": 36535, "estimated_input_tokens": 9134, "estimated_eval_tokens": 18524 }, @@ -11799,186 +11826,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated issue reports across different models and subsystems. I do not see a true duplicate cluster here; at most a few pairs share a broad area like cache or video processing, but not the same concrete bug.", - "confidence": 0.96, - "canonical_issue_reason": "No single issue is a good canonical duplicate target because the set spans many distinct failures (generation, loading, configs, video processing, caching, docs, and model-specific bugs).", + "summary": "This set is a heterogeneous mix of unrelated issues across different models, processors, generation paths, and config behaviors. The soft edges mostly reflect superficial wording similarity (e.g. cache, padding, dtype, generation) rather than the same concrete bug, so none should be merged as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No global best issue is suitable; the cluster is heterogeneous and does not represent one underlying bug.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "Different bugs: KV-cache continuation tensor indexing in serve vs Gemma4Processor missing _tokenizer." + "reason": "Both are runtime crashes in serving/generation, but the concrete failures differ: KV-cache continuation tensor indexing vs missing _tokenizer on Gemma4Processor." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Unrelated failures: Python 3.13 import parsing/indentation vs flash_attention crash on s_aux=None." + "reason": "Different failure modes: Python 3.13 IndentationError in a scripted model file vs flash_attention crash on s_aux=None for sink-less models." }, { "left": "issue:29870", "right": "issue:35141", "accept": false, - "reason": "One is a generate() decoder_attention_mask bug; the other is output embedding reinitialization after resize. Different code paths." + "reason": "Both involve generation/model internals, but one is a BART decoder_attention_mask bug at generate() and the other is embedding reinitialization after resize_token_embeddings." }, { "left": "issue:45072", "right": "issue:46932", "accept": false, - "reason": "Both mention dtype issues, but they are model-specific and arise in different paths; not the same concrete bug." + "reason": "Both mention dtype/precision issues, but one is a CI bfloat16 mismatch in SwitchTransformers/TimmWrapperModel and the other is Gemma4 quantized vision/audio embedder casting." }, { "left": "issue:44869", "right": "issue:46832", "accept": false, - "reason": "Whisper timestamp decoding crash vs RT-DETR feature-level mismatch; different subsystems and symptoms." + "reason": "Whisper timestamp decoding crashes on a trailing replacement character, while RT-DETR fails because num_feature_levels exceeds available backbone outputs." }, { "left": "issue:45632", "right": "issue:46617", "accept": false, - "reason": "Both involve trust_remote_code cache behavior, but one is a leaf-directory collision and the other is a cache FileNotFoundError; not the same bug." + "reason": "Both concern trust_remote_code cache/loading, but one is a cache path collision for same leaf directory names and the other is FileNotFoundError when loading from cache." }, { "left": "issue:45245", "right": "issue:46858", "accept": false, - "reason": "Number-of-categories limit error is unrelated to static-cache seq-length tensor arithmetic under torch.compile." + "reason": "Both are numeric/runtime edge cases, but one is a categories cardinality limit and the other is a cache.get_seq_length() device-tensor bug affecting slice arithmetic." }, { "left": "issue:46082", "right": "issue:46697", "accept": false, - "reason": "LlamaConfig head_dim validation and Swin2SR over-padding are different model/config bugs." + "reason": "Both are shape/config constraints, but LlamaConfig head_dim validation is unrelated to Swin2SR over-padding when size_divisor is already satisfied." }, { "left": "issue:43827", "right": "issue:43994", "accept": false, - "reason": "Docs still referencing pipeline() after removals is unrelated to SigLIP2 producing bad results in pipeline/AutoModel." + "reason": "The first is stale docs referencing removed pipeline() APIs; the second is a model behavior bug where AutoModel/pipeline gives nonsensical SigLIP2 results." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "Sam3Video OOM and Whisper empty-transcription handling are unrelated failures." + "reason": "Sam3Video OOM is a memory-pressure issue, while Whisper empty transcription failure after align_special_tokens is a generation edge-case crash." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk fill is a different issue from GPTNeoX rotary_pct not persisting on reload." + "reason": "Both involve configuration/state persistence, but one is transformers v5 non-persistent buffer initialization junk and the other is GPTNeoXConfig rotary_pct not surviving reload." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Weight-shape mismatch in Qwen3-VL loading is not the same as a wrong tied-weights warning in Qwen3-Next." + "reason": "Both are Qwen model loading issues, but one is Qwen3-VL weight shape mismatch and the other is Qwen3-Next incorrect tied-weights warning across dt_bias." }, { "left": "issue:25251", "right": "issue:27957", "accept": false, - "reason": "Top_k pipeline nesting regression and XLMRoberta Flash Attention 2 support are unrelated." + "reason": "Pipeline top_k nested-list output bug is unrelated to XLMRoberta Flash Attention 2 support." }, { "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug is unrelated to DebertaV2 import parsing on Python 3.13." + "reason": "SwitchTransformers sparse-layer creation with zero sparse layers is unrelated to the DebertaV2 Python 3.13 IndentationError." }, { "left": "issue:44077", "right": "issue:44877", "accept": false, - "reason": "Optional post_init validation and strict-config loading failure are different configuration issues." + "reason": "PatchTSMixer optional post_init policy and strict config blocking granite_speech are different config-validation concerns with no shared code-path bug." }, { "left": "issue:44242", "right": "issue:46897", "accept": false, - "reason": "Both concern losses, but the load-balancing loss omission and Florence2 double-shift bug are not the same underlying bug." + "reason": "Load-balancing loss not added when output_router_logits=False is an MoE loss-computation bug; Florence2 double-shift is a training-target alignment bug." }, { "left": "issue:43673", "right": "issue:46421", "accept": false, - "reason": "Both are cache/chunked-prefill related, but one reports missing cache in v5.0.0 and the other spurious recompilation from lazy init; too different to merge." + "reason": "Both mention cache, but one is missing GenerationMixin cache in v5.0.0 during chunked_prefill and the other is lazy cache init causing spurious recompilation." }, { "left": "issue:45381", "right": "issue:46369", "accept": false, - "reason": "Qwen2.5-VL vision_position_ids regression and Sam3-litetext memory overflow are unrelated." + "reason": "Both mention model regressions and video/image-like workloads, but qwen2.5-vl vision_position_ids is unrelated to Sam3-litetext memory overflow." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "Paged-generate warning behavior is unrelated to Whisper empty transcription after align_special_tokens." + "reason": "Paged generate stale warning for num_return_sequences is not the same bug as Whisper empty transcription failure after align_special_tokens." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "Missing import from image_utils and Qwen3.5 chat crash are different problems." + "reason": "Nonexistent image_utils import is a broken reference; Qwen3.5-35B crashes with transformers chat is a model/chat integration runtime failure." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Auto device mapping cudaErrorAssert in Qwen3 is unrelated to the 2^24 category limit error." + "reason": "Auto device mapping cudaErrorAssert on Qwen3 is unrelated to the max-categories RuntimeError." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "torch.compile+SDPA failure in Bamba is unrelated to loading a local dataset in the no-trainer script." + "reason": "Bamba-9B-v2 torch.compile+SDPA crash and local dataset loading in a training script are different subsystems and failure modes." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and OmDet-Turbo processor size mismatch are different vision-model issues." + "reason": "Dummy Conv2d in Sam3PixelDecoder and OmDet-Turbo processor producing 640px instead of 224px are different model/processor bugs." }, { "left": "issue:43335", "right": "issue:44190", "accept": false, - "reason": "Sparse layer creation in SwitchTransformersConfig is unrelated to local dataset loading failure." + "reason": "SwitchTransformers sparse-layer config bug and local dataset loading script failure are unrelated." }, { "left": "issue:44617", "right": "issue:46369", "accept": false, - "reason": "Both mention SAM3 memory problems, but they affect different models and appear to have different causes." - }, - { - "left": "issue:45588", - "right": "issue:47013", - "accept": false, - "reason": "Flash-attention s_aux=None crash and xLSTM return_last_states unpacking bug are unrelated." + "reason": "Both are Sam3-related, but one is OOM in Sam3Video and the other is memory overflow for sam3-litetext; different models and likely different causes." }, { "left": "issue:43854", "right": "issue:44792", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading failure is unrelated to janus image-generation test failure." + "reason": "GLM-4.7-Flash unit-test loading failure is unrelated to janus test_model_generate_images failure." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Video-input regression in Qwen VL models is not the same as serve KV-cache tensor indexing failure." + "reason": "Video input regression across Qwen VL variants is unrelated to KV-cache continuation tensor indexing in serve responses." + }, + { + "left": "issue:45588", + "right": "issue:47013", + "accept": false, + "reason": "Both are runtime crashes, but one is a flash_attention s_aux=None AttributeError and the other is xLSTM train-mode unpacking mismatch with bare tensors." }, { "left": "issue:45381", "right": "issue:46655", "accept": false, - "reason": "Both are video-processing issues, but they concern different models and different bugs (position IDs vs padding calculation)." + "reason": "Both involve video/image preprocessing, but qwen2.5-vl vision_position_ids and SmolVLM padding calculation are distinct bugs." } ] }, @@ -12081,14 +12108,14 @@ "issue:36246|issue:41628", "issue:36246|issue:38617" ], - "prepared_review_unit_hash": "430ba1dfecccafd8fa38b7b5927839b30b7190f950afe8bbe2c71f3d46ee2e70", + "prepared_review_unit_hash": "6a4a5bf9356b9e5aca947823bbe1ce0c05be70d1c046083f49a3cd5a7e5adc98", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, @@ -12103,187 +12130,190 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated, model-specific bug reports and API regressions that only share broad Transformers themes. None of the suggested soft pairs look like the same underlying issue/change, so I reject all of them.", - "confidence": 0.89, - "canonical_issue_reason": "issue:41628 is the broadest, clearest top-level API regression in the set (a direct import failure), so it is the best single representative of the cluster even though the cluster is heterogeneous.", + "summary": "This cluster is mostly a loose semantic grouping of unrelated Transformer issues; only one pair looks plausibly like the same underlying tokenizer regression. No PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 44521 is the best anchor if a single issue must be chosen: it is open, concrete, and has the strongest discussion/inbound activity. That said, the overall cluster is heterogeneous, so it is not a true umbrella duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41628 is the most generally recognizable user-facing failure and the least niche among the issues here; several others are narrower model- or environment-specific bugs.", + "best_issue_reason": "44521 is the most actionable and representative issue in the set, with the clearest bug report and strongest follow-on interest.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "Different problems: SigLIP2 docs/example errors vs loading a local dataset in the no-trainer image classification script." + "reason": "Different problems: a SigLIP2 doc example/model-processor mismatch versus a local dataset loading failure in `run_image_classification_no_trainer.py`." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "Different code paths: LoRA merge with extended vocab causes collapse, while the other is a Qwen3.5 save_pretrained visual-key regression." + "reason": "Both touch model saving/resizing, but the bugs differ: LoRA merge with extended vocab collapse vs Qwen3.5 saving wrong visual encoder keys." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "One is a regression in processor/tokenizer access for custom multimodal code; the other is resize_token_embeddings not updating Gemma4 embeddings." + "reason": "Different code paths: processor/tokenizer image-handling regression versus `resize_token_embeddings` not updating Gemma4 embedding/output layers." }, { "left": "issue:44464", "right": "issue:45563", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is a different bug from a stale warning in paged generate()." + "reason": "Different symptoms: inconsistent outputs from compiled chunked generation vs a stale warning about `num_return_sequences` in paged generate." }, { "left": "issue:44779", "right": "issue:46489", - "accept": false, - "reason": "Both involve DeepSeek tokenizers, but they target different models/versions and different concrete tokenizer failures." + "accept": true, + "reason": "Both report the same DeepSeek tokenizer/BPE-marker regression in v5; 46489 reads like a model-specific manifestation or missing edge case from the same fix path." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "GraniteSpeech CI stale device override is unrelated to auto_mappings referencing removed Sam3LiteText configs." + "reason": "Unrelated CI failures: GraniteSpeech device-override test flakiness vs auto-mapping references to removed Sam3LiteText configs." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Resize-token-embeddings/output-embedding handling is unrelated to Whisper empty transcription after align_special_tokens." + "reason": "Different bugs in different models: Gemma4 embedding resize state propagation vs Whisper failing on empty transcription after special-token alignment." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "apply_chat_template assistant_mask bugs and Qwen2.5-VL video vision_position_ids are different multimodal issues." + "reason": "Both are multimodal template issues, but one is assistant-mask generation and the other is video vision-position IDs; not the same bug." }, { "left": "issue:45290", "right": "issue:46752", "accept": false, - "reason": "Both hit apply_chat_template, but one is tool-call assistant messages with no content and the other is an empty-list IndexError." + "reason": "Same function, but different edge cases: tool-call assistant messages with no content versus an entirely empty message list." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-input validation failure is unrelated to transformers serve KV-cache continuation indexing." + "reason": "Different subsystems and failures: Qwen3OmniMoe video-input validation versus server-side KV-cache continuation tensor indexing." }, { "left": "issue:25251", "right": "issue:29870", "accept": false, - "reason": "Pipeline top_k nesting behavior and BART generate() decoder_attention_mask initialization are different bugs." + "reason": "Different areas: pipeline `top_k` output nesting regression versus BART generate() hitting `decoder_attention_mask` initialization error." }, { "left": "issue:43526", "right": "issue:46697", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels and Swin2SR over-padding are separate image-processor issues." + "reason": "Both are image-processor bugs, but one is `reduce_labels` returning too little data and the other is Swin2SR over-padding images." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "AutoConfig missing fields with trust_remote_code is not the same as an unrecognized model/config missing model_type." + "reason": "Different loading/config issues: missing fields when `trust_remote_code=True` versus an unrecognized model due to missing `model_type`." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to list_repo_files not forwarding revision." + "reason": "Different concerns: AutoTokenizer ignoring `tokenizer.json` versus `list_repo_files` not forwarding `revision` for CI." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 import failure and missing TensorFlow backend from a torch-only install are different dependency problems." + "reason": "Different environment/backend problems: Python 3.14 import failure versus TensorBoard installed without TensorFlow causing backend errors." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "They are both model-loading failures, but for different models and distinct root causes." + "reason": "Both are model-loading complaints, but they concern different models and likely different loader paths." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Slow import behavior is a performance issue; unsafe sys.modules access is a correctness/safety bug." + "reason": "One is slow import performance; the other is an unsafe `sys.modules` access bug in `modeling_utils`." }, { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "XLNet CPU tensor creation in positional encoding is unrelated to FuyuBatchFeature swallowing KeyboardInterrupt and errors." + "reason": "Different bugs: XLNet device placement regression versus Fuyu swallowing exceptions with a bare `except:`." }, { "left": "issue:43673", "right": "issue:46858", "accept": false, - "reason": "Both touch cache/generation internals, but one is a missing cache in chunked_prefill and the other is a device-tensor arithmetic bug in static cache with torch.compile." + "reason": "Both involve cache/prefill, but one is missing GenerationMixin cache in v5.0.0 and the other is a device-tensor `seq_length` arithmetic issue on non-CUDA backends." }, { "left": "issue:27957", "right": "issue:35141", "accept": false, - "reason": "XLMRoberta Flash Attention 2 support and token-embedding resize reinitialization are unrelated." + "reason": "Different topics: XLMRoberta Flash Attention 2 support versus token embedding resize reinitialization with untied output embeddings." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "A model load failure for Phi-4 multimodal is unrelated to stale autogenerated mappings in CI." + "reason": "Different models and failures: Phi-4 multimodal loading versus stale auto-mapping references to removed configs." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "GraniteSpeech CI override behavior and Phi-4 multimodal loading are different issues with no shared code path." + "reason": "GraniteSpeech CI test failure and Phi-4 multimodal model loading are unrelated." }, { "left": "issue:45632", "right": "issue:46234", "accept": false, - "reason": "Cache-path collision for trust_remote_code local models is unrelated to a missing import-utils availability helper." + "reason": "Different hub/import issues: local cache-path collision for `trust_remote_code` versus a missing utility export in `import_utils`." }, { "left": "issue:42175", "right": "issue:44279", "accept": false, - "reason": "A packaging extra missing TensorFlow is not the same as a vague dependency issue with transformers." + "reason": "Both are dependency-related, but one is a specific packaging extra omission and the other is a generic transformers dependency complaint." }, { "left": "issue:42796", "right": "issue:44521", "accept": false, - "reason": "Tokenizer BPE-marker handling and multimodal assistant_mask generation are different bugs." + "reason": "Different features: DeepSeek tokenizer BPE-marker handling versus multimodal assistant-mask generation in `apply_chat_template`." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer _gpu_broadcast_one NoneType failure is unrelated to embedding reinitialization after resizing tokens." + "reason": "Different runtime errors: trainer GPU broadcast rejecting `NoneType` versus embedding reinitialization after vocab resize." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they concern different symbols and different underlying packaging/export failures." + "reason": "Different import failures: a missing model-specific processor export versus missing top-level `AutoImageProcessor`." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Missing Qwen2_5_VLImageProcessor and missing layer_type_validation are distinct import/export problems." + "reason": "Unrelated missing imports in different modules; no shared code path." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative about cluster cohesion. The chosen anchor issue, 44521, is supported by its open state, discussion activity, and inbound references. The note that the cluster is heterogeneous is also consistent with the mostly dissimilar issue titles and the soft pairs. The only accepted soft edge, 44779/46489, is plausibly the same DeepSeek tokenizer regression path and is not overstated. No PR-pair mergeability concerns apply here since the packet contains only issues." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -12321,7 +12351,6 @@ "issue:44220", "issue:44279", "issue:44295", - "issue:44327", "issue:44336", "issue:44496", "issue:44521", @@ -12342,11 +12371,12 @@ "issue:45759", "issue:46082", "issue:46121", + "issue:46123", + "issue:46369", "issue:46424", "issue:46506", "issue:46515", "issue:46693", - "issue:46752", "issue:46932" ], "soft_pairs": [ @@ -12372,230 +12402,230 @@ "issue:43650|issue:44220", "issue:44016|issue:44062", "issue:43976|issue:44279", - "issue:43441|issue:45588", "issue:44792|issue:46424", + "issue:43441|issue:45588", "issue:43824|issue:44933", "issue:44521|issue:46693", "issue:44279|issue:46515", "issue:43479|issue:45375", "issue:45335|issue:45584", "issue:43299|issue:45216", - "issue:44327|issue:46752" + "issue:46123|issue:46369" ], - "prepared_review_unit_hash": "b2e8e75aea40f0f1298d001c127d50d9c2657bd0d36b263dc263e065f6c22c61", + "prepared_review_unit_hash": "b3b16ef7aa0d8bcbae1ac0fce1a48cf41cf9346cc49795e0f389d1c19c65917c", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36871, - "estimated_input_tokens": 9218, - "estimated_eval_tokens": 18692 + "serialized_chars": 36847, + "estimated_input_tokens": 9212, + "estimated_eval_tokens": 18680 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a loose collection of unrelated Transformers issues, many in multimodal/generation/config code paths. None of the soft pairs look like the same underlying bug closely enough to merge as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "issue:44521 is the strongest representative: it is a current, well-scoped multimodal pipeline bug with active discussion and references, and it best reflects the recurring theme in this set better than the narrower or one-off reports.", + "summary": "This is a loose cluster of mostly unrelated Transformers issues, with repeated themes around multimodal model/config loading and processor regressions but no true duplicate pair. The best anchor is a Qwen multimodal loading regression; the soft edges are mostly same-subsystem, not same-bug matches.", + "confidence": 0.43, + "canonical_issue_reason": "issue:43299 is the clearest central bug in the set: a concrete multimodal model-loading regression with a well-defined failure mode and release window, and it best represents the Qwen/config-loading slice of the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the best overall issue because it is open, concrete, and broadly relevant to multimodal chat template behavior; it is also the clearest candidate for ongoing triage.", + "best_issue_reason": "issue:43299 is the most suitable representative issue overall because it is high-signal, concrete, and closest to the recurring multimodal loading/config failures seen across the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Both are multimodal checkpoint/loading regressions, but one is lm_head tying and the other is save_pretrained checkpoint correctness for a different model family." + "reason": "Related multimodal family, but one is untied lm_head weights and the other is save_pretrained checkpoint correctness; different bugs/code paths." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Different bugs: Qwen3-MoE sliding window behavior vs stale auto_mappings references breaking CI." + "reason": "Qwen3-MoE sliding_window behavior vs stale auto_mappings references to removed configs; unrelated root causes." }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "Unrelated config-validation failures: explicit head_dim handling vs JSON-loaded rope ignore-keys type error." + "reason": "Both are config validation issues, but head_dim handling and rope ignore-key parsing are distinct failures." }, { "left": "issue:44295", "right": "issue:46506", "accept": false, - "reason": "Different subsystems and symptoms: position_ids buffer access vs CPU/CUDA result divergence." + "reason": "Position_ids buffer read error vs CPU/CUDA output divergence; not the same defect." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Video-classification processor lookup is unrelated to a notebook syntax error." + "reason": "Video-classification processor lookup bug vs a notebook syntax error; no shared underlying issue." }, { "left": "issue:43673", "right": "issue:46424", "accept": false, - "reason": "Both involve cache terms, but one is chunked_prefill cache access and the other is max_cache_length semantics for quantized cache." + "reason": "Both mention cache, but chunked_prefill cache missing and max_cache_length semantics for quantized cache are different." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Different code paths and models: processor.tokenizer regression vs t5gemma embedding resize not propagating." + "reason": "Processor.tokenizer/image regression vs resize_token_embeddings not updating decoder.embed_tokens; different code paths." }, { "left": "issue:43582", "right": "issue:46932", "accept": false, - "reason": "Apple Silicon allocator TypeError is unrelated to Gemma4 quantized embedder dtype casting." + "reason": "Apple Silicon TypeError in warmup vs Gemma4 quantized-embedder dtype casting; unrelated." }, { "left": "issue:45083", "right": "issue:46506", "accept": false, - "reason": "Different model-specific helper bug vs cross-device output mismatch." + "reason": "Helper length computation bug vs CPU/CUDA result mismatch; not the same failure." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "Model auto-loading failure is unrelated to PIL image-processor torchvision dependency checks." + "reason": "Unrecognized model config vs PIL image processors needing torchvision; different problems." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "Python version compatibility issue is not the same as loading a specific model repo failing." + "reason": "Python version compatibility issue vs a specific model-loading failure; different root cause." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug and docs mismatch are distinct issues." + "reason": "BeitImageProcessorFast label reduction bug vs documentation mismatch in TextClassificationPipeline." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Documentation stale-API issue is unrelated to unrecognized-model config loading." + "reason": "Docs mentioning return_all_scores vs model config import failure; unrelated." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Cache missing during generation is unrelated to ANSI code emission in loading_report." + "reason": "Cache-missing regression vs ANSI codes in loading_report; no overlap beyond general utility code." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "Unrecognized model/config loading failure is different from composite multimodal text_config unwrapping." + "reason": "Unrecognized model loading vs failing to unwrap text_config in composite multimodal configs; different code paths." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Both touch flex_attention/versioning, but one is a compiled flex_attention failure on torch>=2.9 and the other is an import version check bug." + "reason": "Compiled flex_attention failure on torch>=2.9 vs incorrect version gate for AuxRequest import; different issues." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss handling are unrelated." + "reason": "Gradient checkpointing in eval mode vs D-FINE auxiliary-loss behavior; unrelated." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error is unrelated to the fbank feature extraction bug." + "reason": "Notebook syntax error vs audio feature extraction helper bug; no shared underlying defect." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Both mention Ministral, but one is save-dequantized serialization and the other is FlashAttention loading/runtime failure." + "reason": "Dequantized model saving issue vs flash-attention runtime failure on Ministral-3; same model family but different bug." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "Generic placeholder issue is unrelated to audio feature extraction." + "reason": "Placeholder/invalid issue vs audio feature extraction bug; no relation." }, { "left": "issue:44016", "right": "issue:44062", "accept": false, - "reason": "Notebook syntax error is unrelated to the AddedToken keyword-argument TypeError." + "reason": "Notebook syntax error vs AddedToken special-argument TypeError; unrelated." }, { "left": "issue:43976", "right": "issue:44279", "accept": false, - "reason": "Python version support issue is unrelated to a vague dependency problem." + "reason": "Version-support problem vs generic dependency issue; not the same bug." }, { - "left": "issue:43441", - "right": "issue:45588", + "left": "issue:44792", + "right": "issue:46424", "accept": false, - "reason": "Both are FlashAttention-related, but they target different failure modes and code paths." + "reason": "Janus image-generation test failure vs cache max_length semantics; different areas." }, { - "left": "issue:44792", - "right": "issue:46424", + "left": "issue:43441", + "right": "issue:45588", "accept": false, - "reason": "Janus image-generation test failure is unrelated to cache max length handling." + "reason": "Both are flash-attention themed, but one is Ministral-3-specific and the other is an s_aux=None AttributeError for sink-less models; not mergeable as one fix." }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Missing Qwen2.5-VL import is unrelated to a nonexistent image_utils import." + "reason": "Missing Qwen2_5_VL export vs nonexistent image_utils import; both import errors, but different symbols and causes." }, { "left": "issue:44521", "right": "issue:46693", "accept": false, - "reason": "apply_chat_template assistant mask bug is not the same as a FA2 long-context performance regression." + "reason": "All-zero assistant masks in multimodal chat templates vs FA2 long-context performance regression; not the same issue." }, { "left": "issue:44279", "right": "issue:46515", "accept": false, - "reason": "General dependency issue is not the same as torch 2.4.1 incompatibility." + "reason": "Dependency issue vs torch 2.4.1 incompatibility; environment problems, but different ones." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both are config issues, but they affect different models and different mechanisms: default init vs strict field dropping." + "reason": "Phi4MultimodalConfig default initialization bug vs missing Qwen3_5MoeVisionConfig field; different config objects and failures." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Token embedding resize propagation bug is unrelated to Whisper empty-transcription handling after token alignment." + "reason": "Token embedding resize bug vs Whisper empty-transcription failure after align_special_tokens; unrelated." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Qwen multimodal loading regression and checkpoint save correctness are different lifecycle bugs." + "reason": "Both are Qwen multimodal regressions, but one is model loading and the other is save_pretrained checkpoint corruption; different fixes." }, { - "left": "issue:44327", - "right": "issue:46752", + "left": "issue:46123", + "right": "issue:46369", "accept": false, - "reason": "QA decode_spans top-k crash is different from apply_chat_template on an empty list." + "reason": "Final partial batch dropping vs sam3-litetext memory overflow; unrelated performance/behavior issues." } ] }, @@ -12638,6 +12668,7 @@ "issue:44242", "issue:44279", "issue:44291", + "issue:44327", "issue:44351", "issue:44368", "issue:44462", @@ -12658,17 +12689,16 @@ "issue:45812", "issue:45823", "issue:46082", - "issue:46123", "issue:46209", - "issue:46369", "issue:46399", "issue:46512", "issue:46614", + "issue:46752", "issue:47000", "issue:47085" ], "soft_pairs": [ - "issue:46123|issue:46369", + "issue:44327|issue:46752", "issue:46614|issue:47085", "issue:44279|issue:45405", "issue:43295|issue:44779", @@ -12696,112 +12726,112 @@ "issue:44683|issue:45542", "issue:43479|issue:43525" ], - "prepared_review_unit_hash": "9add2fd764aa5ad7675b6fb4f15ca1a72a1efd5ff66c1a2d71ea79350b9d1773", + "prepared_review_unit_hash": "a95247e3a6657f9a110aaaa8022592e4e54ff08a44d28bf1dabedf7b920feb16", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35396, - "estimated_input_tokens": 8849, - "estimated_eval_tokens": 17954 + "serialized_chars": 35414, + "estimated_input_tokens": 8854, + "estimated_eval_tokens": 17964 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly superficially similar regression/config/tokenizer reports across different models and subsystems, but they do not look like true duplicates of the same underlying bug/change. No PRs are present.", - "confidence": 0.37, - "canonical_issue_reason": "issue:45020 is the broadest umbrella-style report in the set (recent transformers versions breaking remote_code-based model loading). It is the closest thing to a duplicate target, though most other items are still too specific or unrelated to be merged into it.", + "summary": "This cluster is heterogeneous: the soft pairs mostly share only broad themes (tokenizers, multimodal models, docs, regressions) but not the same underlying bug or change, so none should be merged as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out; the items span unrelated failures across different models, pipelines, configs, and documentation.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the best single issue representative because it is the most general regression report and can plausibly subsume some of the model-loading/version breakages. That said, there is no strong cluster-wide canonical duplicate here.", + "best_issue_reason": "No strong global best issue: 45020 is the broadest recent regression report, but it does not meaningfully represent the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:46123", - "right": "issue:46369", + "left": "issue:44327", + "right": "issue:46752", "accept": false, - "reason": "Different failures: batch-output dropping in a mask pipeline vs memory overflow in a model call." + "reason": "Different code paths and failure modes: QA decode_spans top-k bounds vs apply_chat_template on an empty list." }, { "left": "issue:46614", "right": "issue:47085", "accept": false, - "reason": "Both mention tokenizers, but one is a stale phonemizer backend bug and the other is a thread-safety regression in fast tokenizers." + "reason": "Tokenizer backend state bug vs Rust tokenizer thread-safety regression; unrelated underlying issues." }, { "left": "issue:44279", "right": "issue:45405", "accept": false, - "reason": "Both are dependency/version complaints, but one is a generic transformers dependency issue and the other is an unreleased PEFT minimum version bump." + "reason": "A generic dependency problem is not the same as bumping MIN_PEFT_VERSION to an unreleased release." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different APIs/models and describe different code paths." + "reason": "Both involve tokenizers, but one is processor/tokenizer API regression in multimodal code and the other is DeepSeek tokenization output corruption." }, { "left": "issue:42796", "right": "issue:43650", "accept": false, - "reason": "Unrelated issues: tokenizer BPE marker handling vs a vague request to add data." + "reason": "Ministral BPE handling bug is unrelated to the data request issue." }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "Both are documentation-related, but one is a SigLIP2 example mismatch/quantization failure and the other is a pipeline docs mismatch for return_all_scores." + "reason": "Both are docs-related, but they concern different examples and different behavior, not the same mistake." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Different subjects: optimizer arguments not used vs a configuration warning about tie_word_embeddings." + "reason": "SGD args being ignored is unrelated to the tie_word_embeddings warning." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "Both are regressions around model saving/loading, but they concern different models and distinct serialization/dtype behaviors." + "reason": "Different regressions: loading dtype stays float32 vs save_pretrained serialization of Qwen3.5 checkpoints." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "Both are import errors, but one is a specific missing symbol and the other is a broad post-upgrade import breakage report." + "reason": "A specific missing import/export is not the same as broad import errors after upgrade." }, { "left": "issue:41950", "right": "issue:45020", "accept": false, - "reason": "Different pipeline/model areas: video-classification processor lookup vs remote_code model loading regressions." + "reason": "Video-classification processor lookup bug and remote_code breakage are different problems." }, { "left": "issue:43976", "right": "issue:45405", "accept": false, - "reason": "Python support/version-compatibility issue vs a PEFT version requirement problem; not the same bug." + "reason": "Python version compatibility issue vs dependency pin/version-release issue." }, { "left": "issue:43381", "right": "issue:47000", "accept": false, - "reason": "Gradient checkpointing eval-mode behavior is unrelated to RTDetr loss infeasibility." + "reason": "Gradient-checkpointing eval-mode guard is unrelated to RTDetrLoss cost-matrix infeasibility." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Both involve Gemma/tied-weight style behavior, but one is serialization of lm_head.weight and the other is resize_token_embeddings not updating specific Gemma4 embeddings." + "reason": "Serialization of tied lm_head weights is different from resize_token_embeddings not propagating to embeddings." }, { "left": "issue:44779", @@ -12813,79 +12843,79 @@ "left": "issue:46082", "right": "issue:46399", "accept": false, - "reason": "Different Qwen3.5 issues: config validation for head_dim vs a kernel-registration crash." + "reason": "Config validation for head_dim is unrelated to a missing kernel decorator causing rotary-pos-emb registration failure." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen3.5-related regressions, but one is model loading and the other is incorrect visual-encoder key saving." + "reason": "Model loading failure and save_pretrained visual-key serialization bug are different code paths." }, { "left": "issue:46082", "right": "issue:46512", "accept": false, - "reason": "Config validation on head_dim vs stale config.dtype after from_config; different underlying bug." + "reason": "Hidden-size/head-dim validation is unrelated to from_config leaving config.dtype stale." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "Tokenizer special-token registration bug vs missing auxiliary losses when denoising is disabled." + "reason": "BigBirdTokenizer special-token registration is unrelated to D-FINE auxiliary-loss behavior." }, { "left": "issue:43650", "right": "issue:46209", "accept": false, - "reason": "No shared bug path: vague data request vs FP8 inference failure on B200." + "reason": "The titles are unrelated; 'ADD THE DATA' is not the same as FP8 inference failing on B200." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "Both are pipeline/model integration problems, but they involve different pipelines and a different device-map forward mismatch." + "reason": "Processor lookup in a video pipeline is unrelated to Gemma4 device-placement mismatch." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Both touch training losses, but one is about router load-balancing loss gating and the other about Gemma4 multimodal token type IDs." + "reason": "Load-balancing loss gating and mm_token_type_ids defaults are different multimodal training issues." }, { "left": "issue:43299", "right": "issue:43854", "accept": false, - "reason": "Both are model-loading failures, but for different models and with different failure modes." + "reason": "Qwen3VL loading regression and GLM unit-test loading issue are separate model-specific failures." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "Tokenizer loading issue vs a specific Mistral regex patch crash; same area, not the same code-path bug." + "reason": "Ignoring tokenizer.json and crashing in _patch_mistral_regex are distinct tokenizer-loading bugs." }, { "left": "issue:45588", "right": "issue:47000", "accept": false, - "reason": "Different crash sites: flash-attention auxiliary state handling vs RTDetr assignment/cost-matrix infeasibility." + "reason": "Flash-attention s_aux=None crash and RTDetrLoss cost-matrix infeasibility are unrelated." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error is unrelated to init_empty_weights argument handling." + "reason": "A notebook syntax error is not the same as an init_empty_weights TypeError." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "Compiled flex_attention on torch 2.9 vs tensorboard/TensorFlow backend installation issue; unrelated." + "reason": "Compiled flex_attention on new torch versions is unrelated to a TensorFlow backend install/import issue." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Both are config-default issues, but they are for different models and different missing/incorrect fields." + "reason": "Multimodal default config initialization and missing pad_token_id are separate config bugs." } ] }, @@ -12964,8 +12994,8 @@ "issue:42994|issue:43317", "issue:43441|issue:45834", "issue:44336|issue:46851", - "issue:45103|issue:45372", "issue:44623|issue:44749", + "issue:45103|issue:45372", "issue:45859|issue:46851", "issue:44261|issue:44263", "issue:44623|issue:45397", @@ -12988,202 +13018,196 @@ "issue:46655|issue:46726", "issue:44246|issue:44496" ], - "prepared_review_unit_hash": "e714f0c9a984748cf897cb5bfbc19627e2f62ce763f01e66bb07584bbeca8030", + "prepared_review_unit_hash": "c79ca437ad2fe0c901a479f28b15bcb0921f0903965251547e6819915da2cd3e", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36015, - "estimated_input_tokens": 9004, - "estimated_eval_tokens": 18264 + "serialized_chars": 36017, + "estimated_input_tokens": 9005, + "estimated_eval_tokens": 18266 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is heterogeneous: the proposed soft edges mostly connect broadly related transformer bugs, but not the same underlying defect. I would not merge any of the candidate pairs as duplicates.", + "summary": "The items are largely heterogeneous bug reports across unrelated models, loading paths, processors, and docs/performance regressions. None of the soft pairs look like the same underlying bug or change.", "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:47000 is the most actionable standalone issue here: it is open, recent, has discussion/inbound references, and describes a concrete failing code path. It is not a duplicate anchor for the rest of the set.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Both involve DeepSpeed/optimizer behavior, but one is OOM with Adafactor and the other is double optimizer stepping. Different symptoms and likely different bugs." + "reason": "Both mention Deepspeed/optimizer behavior, but one is Adafactor OOM and the other is double-stepping; different failure modes and fixes." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Mistral tokenizer regex crash vs Hugging Face Hub revision forwarding in CI are unrelated code paths." + "reason": "Mistral tokenizer regex crash vs hub API revision forwarding; unrelated code paths." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Both touch quantized/dequantized model handling, but one is save failure and the other is load/offload failure. Not the same bug." + "reason": "One is quantized model saving, the other is loading a dequantized model with device offload; different lifecycle stages." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "Tokenizer BPE marker handling and Kosmos OCR long-input indexing are distinct failures in different model pipelines." + "reason": "Different models and bugs: Mistral tokenizer/BPE markers vs Kosmos2.5 OCR index error." }, { "left": "issue:44336", "right": "issue:46851", "accept": false, - "reason": "ANSI-code terminal logging and Qwen3.5 position_ids issues are unrelated." + "reason": "ANSI color logging issue vs Qwen3.5 variable-length position_ids bug; unrelated." }, { - "left": "issue:45103", - "right": "issue:45372", + "left": "issue:44623", + "right": "issue:44749", "accept": false, - "reason": "Auto docstring annotation handling and Gemma 4 processor import failure are different subsystems and symptoms." + "reason": "Processor save_pretrained missing files vs a reported slowdown after upgrading Transformers; not the same problem." }, { - "left": "issue:44623", - "right": "issue:44749", + "left": "issue:45103", + "right": "issue:45372", "accept": false, - "reason": "Processor save-pretrained file loss and Chinese performance regression during filtering are unrelated." + "reason": "Auto-docstring annotation crash vs Gemma 4 processor import failure; distinct subsystems." }, { "left": "issue:45859", "right": "issue:46851", "accept": false, - "reason": "CPU-offload rotary embedding incompatibility is not the same as variable-length position_ids handling." + "reason": "Both are Qwen3.5-related, but one is CPU offload incompatibility and the other is variable-length input handling." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "Different model internals and different failure modes: missing rms_norm_eps precision issue vs torch.split return value issue." + "reason": "Different model internals and symptoms: MLA layer norm precision/config issue vs GLM-MoE split-return behavior." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Processor serialization missing files and Gemma-4 zero3 from_pretrained loading are separate bugs." + "reason": "Processor file-saving bug vs Gemma-4 ZeRO-3 loading failure; unrelated." }, { "left": "issue:42175", "right": "issue:43976", "accept": false, - "reason": "Packaging backend inclusion and Python version compatibility are unrelated release/configuration problems." + "reason": "Package backend inclusion problem vs Python version compatibility; different environment issues." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Docs mismatch for return_all_scores is not the same as intermittent import slowness." + "reason": "Docs mismatch for pipeline arguments vs intermittent import slowdown; unrelated." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Both are repo-maintenance/CI issues, but they concern different files and defects: stale auto_mappings entries vs leftover dummy classes leaking into exports." + "reason": "Both are repo maintenance/CI issues, but one is stale auto_mappings references and the other is leftover dummy classes; different fixes." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "CamemBERT masked-LM regression and DeepSeek tokenizer regression are different models and different failure surfaces." + "reason": "CamemBERT masked-LM regression vs DeepSeek tokenizer regression; different models and code paths." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Missing pad_token_id on Llama4Config and glm-4v-9b loading failure do not point to the same underlying bug." + "reason": "Llama4Config missing pad_token_id vs glm-4v-9b loading failure; different model-specific errors." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache missing during chunked prefill and CLIPVisionModel hidden_states regression are unrelated." + "reason": "Generation cache regression in chunked_prefill vs CLIPVision hidden_states regression; not the same bug." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error and Gemma4 device mismatch are completely different issues." + "reason": "Notebook syntax error vs Gemma4 device-map mismatch during forward; completely different." }, { "left": "issue:44186", "right": "issue:47000", "accept": false, - "reason": "LayoutLMv2 tokenizer NER/padding crash and RTDetrLoss infeasible cost matrix are unrelated model-specific failures." + "reason": "Tokenizer padding/truncation crash vs RTDetr loss infeasible cost matrix; unrelated model failures." }, { "left": "issue:43824", "right": "issue:43976", "accept": false, - "reason": "Missing Qwen2_5_VL import and Python version support are different problems." + "reason": "Missing Qwen2_5_VL export vs Python support/version issue; unrelated." }, { "left": "issue:44987", "right": "issue:45542", "accept": false, - "reason": "Model loading regression and missing TensorFlow backend due to install extras are unrelated." + "reason": "Model loading failure vs missing TensorFlow backend installation; different problems." }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "torch.compile attention-kernel divergence and ANSI code generation in loading_report are unrelated." + "reason": "Attention-kernel divergence under torch.compile vs terminal ANSI output issue; unrelated." }, { "left": "issue:42796", "right": "issue:43441", "accept": false, - "reason": "Both mention Ministral, but one is tokenizer BPE markers and the other is FlashAttention failure; not the same bug." + "reason": "Same model family, but tokenizer BPE-marker handling and FlashAttention failure are different concrete bugs." }, { "left": "issue:43525", "right": "issue:43643", "accept": false, - "reason": "Llama4Config missing pad_token_id and trust_remote_code AutoConfig field loss are different config-loading issues." + "reason": "Both involve missing config fields, but one is a model config attribute bug and the other is trust_remote_code returning incomplete config objects." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "AutoTokenizer loading regression and ESM2 broken are unrelated model/config problems." + "reason": "Generic tokenizer loading issue vs ESM2 breakage; different models and failure modes." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Float8 storage lookup failure and Qwen3.5 chat crash are distinct errors." - }, - { - "left": "issue:44246", - "right": "issue:44496", - "accept": false, - "reason": "Slow import and unrecognized model config are different failure modes." + "reason": "Float8 storage type lookup failure vs Qwen chat crash; unrelated." }, { "left": "issue:46655", "right": "issue:46726", "accept": false, - "reason": "Video processor padding miscalculation and AutoProcessor missing-vision-backend error are different processor-level bugs." + "reason": "Video processor padding bug vs missing vision backend/image processor error; different causes." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import latency and model-type recognition are not the same underlying issue." + "reason": "Import slowdown vs unrecognized model/config.json failure; not the same underlying issue." } ] }, @@ -13287,20 +13311,20 @@ "issue:43257|issue:45137", "issue:44496|issue:45003" ], - "prepared_review_unit_hash": "866b886d8f530bc054ad1f805f99c0748cde9bf6f9272130f77d0aaeb7338eac", + "prepared_review_unit_hash": "8bb6f3e3217beb590a54f105001bdca9504540e44efd228ce1f107e0101c17c5", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36470, + "serialized_chars": 36472, "estimated_input_tokens": 9118, "estimated_eval_tokens": 18492 }, @@ -13309,192 +13333,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is highly heterogeneous: it mixes unrelated import errors, model-loading regressions, tokenizer bugs, docs issues, and runtime/device problems. None of the suggested soft pairs look like the same underlying bug, so all soft edges should be rejected.", - "confidence": 0.93, - "canonical_issue_reason": null, + "summary": "These items are mostly unrelated single-issue reports. The soft pairs share broad keywords or model families, but none look like the same underlying bug or change, so I would not merge any of them as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 46851 is the strongest standalone bug report in this set: it describes a concrete, reproducible runtime failure in a specific code path (Qwen3.5 variable-length `position_ids`) with a clear model surface and likely fix location.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45003 is the broadest generic runtime bug in the set, but there is no strong true canonical because the issues are not a single duplicate cluster.", + "best_issue_reason": "Issue 46851 is the best issue candidate overall because it is specific, actionable, and centered on a clear runtime regression rather than a vague import/docs/config mismatch.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45325", "right": "issue:46851", "accept": false, - "reason": "Different Qwen bugs: rope-index temporal scaling vs variable-length position_ids handling." + "reason": "Both mention Qwen and `position_ids`, but one is a Qwen2.5-VL temporal rope-scaling bug and the other is a Qwen3.5 variable-length input crash; different model families and code paths." }, { "left": "issue:46291", "right": "issue:46399", "accept": false, - "reason": "Both mention kernels, but one is pipeline import-time dependency resolution and the other is a Qwen3.5 registration/value error." + "reason": "Both involve kernels-related failures, but one is a `transformers.pipelines` import error from kernel version requirements and the other is a Qwen3.5 kernel registration/decorator crash." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer save-pretrained metadata mismatch is unrelated to Gemma-4 Zero3 from_pretrained loading." + "reason": "Tokenizer save metadata mismatch and a Gemma-4 ZeRO3 `from_pretrained` failure are unrelated bugs with different surfaces and fixes." }, { "left": "issue:45083", "right": "issue:46851", "accept": false, - "reason": "Different code paths in different Qwen models: helper length calculation vs position_ids behavior." + "reason": "Both are Qwen3-related, but one is an omni-MoE feature-length helper issue and the other is a text/chat `position_ids` regression; not the same bug." }, { "left": "issue:43784", "right": "issue:45542", "accept": false, - "reason": "Both are import/backend issues, but one is missing nn in sentence-transformers and the other is TensorFlow backend setup with tensorboard." + "reason": "Both are import/dependency problems, but the affected packages and failure modes differ (`nn` NameError in sentence-transformers vs missing TensorFlow backend handling for tensorboard)." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "An import symbol error is not the same as a Python-version compatibility regression." + "reason": "Missing `AutoImageProcessor` export and a Python-version compatibility regression are different issues, even though both are import-time failures." }, { "left": "issue:46399", "right": "issue:46851", "accept": false, - "reason": "Same model family, but the failures are distinct: kernels registration decorator vs variable-length position_ids." + "reason": "One is a kernel registration decorator error for Qwen3.5, the other is a variable-length `position_ids` failure; same model family but not the same concrete defect." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to modeling_utils sys.modules access." + "reason": "BEiT Fast `reduce_labels` behavior and unsafe `sys.modules` access are unrelated code-path bugs." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "Pytest-xdist file race and stale generation warning are different subsystems and symptoms." + "reason": "A pytest-xdist file race in testing utils is unrelated to a stale `num_return_sequences` warning in generation." }, { "left": "issue:43540", "right": "issue:46752", "accept": false, - "reason": "Video input ValueError in Qwen3OmniMoe is unrelated to empty-list apply_chat_template IndexError." + "reason": "A Qwen3OmniMoe video-input ValueError and an empty-list chat-template IndexError are different runtime failures." }, { "left": "issue:45103", "right": "issue:46399", "accept": false, - "reason": "Auto-docstring annotation parsing crash is unrelated to Qwen3.5 kernels registration." + "reason": "Auto-docstring future-annotations crash and Qwen3.5 kernel registration failure are unrelated." }, { "left": "issue:45072", "right": "issue:46829", "accept": false, - "reason": "Different model bugs: dtype mismatch under torch.compile vs hidden_states/output packaging." + "reason": "One is a dtype mismatch under `torch.compile`/SDPA, the other is a wrong field assignment in `ImageClassifierOutput`; not the same bug." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import-time slowdown and eos_token_id config inconsistency are unrelated." + "reason": "Import slowness and Gemma-3 `eos_token_id` inconsistency are unrelated issues." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Position_ids buffer access bug is unrelated to Gemma-4 Zero3 loading." + "reason": "Reading `position_ids` after buffer registration and a model-loading failure in Gemma-4 are different defects." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Both involve Qwen, but one is sliding_window behavior in MoE and the other is a chat crash." + "reason": "Both are Qwen-related, but one is a `sliding_window` issue in Qwen3-MoE and the other is a chat crash in Qwen3.5; different symptoms and likely fixes." }, { "left": "issue:46614", "right": "issue:46665", "accept": false, - "reason": "Tokenizer backend state leakage is unrelated to from_pretrained behavior inside Python classes." + "reason": "A stale phonemizer backend after language switching is unrelated to inconsistent `from_pretrained` behavior inside Python classes." }, { "left": "issue:45588", "right": "issue:46851", "accept": false, - "reason": "Flash-attention sink_aux crash is a different path than Qwen3.5 position_ids handling." + "reason": "A flash-attention `s_aux=None` crash and a Qwen3.5 `position_ids` bug do not share the same code path." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both are Qwen3-MoE related, but one is LoRA compatibility across versions and the other is Qwen3VL model loading." + "reason": "Both mention Qwen3 MoE loading regressions, but one is LoRA compatibility across hf4.x/hf5.x and the other is Qwen3VL MoE loading on dev0; not the same concrete issue." }, { "left": "issue:45003", "right": "issue:46821", "accept": false, - "reason": "Generic modeling_utils sys.modules access does not match TimesFM window_size AttributeError." + "reason": "Unsafe `sys.modules` access and a TimesFM `window_size` AttributeError are unrelated." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "ANSI code emission in loading_report is unrelated to GLM-5 RoPE implementation." + "reason": "ANSI code output in loading report is unrelated to a GLM-5 RoPE implementation discussion/bug." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both concern auxiliary losses, but they are different models and different triggering options." + "reason": "Both concern conditional losses, but they affect different models and flags and do not look like one underlying bug." }, { "left": "issue:43550", "right": "issue:46506", "accept": false, - "reason": "torch.compile/SDPA failure is unrelated to CPU vs CUDA output divergence." + "reason": "A Bamba SDPA/`torch.compile` failure and CPU-vs-CUDA result divergence in PPDocLayoutV3 are different issues." }, { "left": "issue:42886", "right": "issue:46132", "accept": false, - "reason": "Offline tokenizer caching and AttentionInterface.register behavior change are different bugs." + "reason": "Offline tokenizer cache loading and `AttentionInterface.register` behavior changes are unrelated." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "DeepSeek tokenizer regression is unrelated to resize_token_embeddings not updating output embeddings." + "reason": "DeepSeek tokenizer regression and `resize_token_embeddings` not updating output embeddings are different code paths." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Different missing symbols from different packages: Qwen2.5-VLImageProcessor vs huggingface_hub.is_offline_mode." + "reason": "Both are import errors, but they concern different missing symbols from different packages and likely different fixes." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Docs mismatch for TextClassificationPipeline is unrelated to sys.modules access safety." + "reason": "Docs mentioning `return_all_scores` and unsafe `sys.modules` access are unrelated." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they involve different missing exports and likely different version breakages." + "reason": "Missing `Qwen2_5_VLImageProcessor` and missing `HybridCache` are separate import/export problems." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Qwen3-MoE LoRA incompatibility is unrelated to the DeepSpeed ZeRO3 deque error." + "reason": "Qwen3 MoE weight conversion with accelerate+deepspeed and a DeepSpeed ZeRO3 deque error are different failures." }, { "left": "issue:44496", "right": "issue:45003", "accept": false, - "reason": "Missing model_type config is a model-resolution issue, not a modeling_utils sys.modules bug." - }, - { - "left": "issue:44295", - "right": "issue:45003", - "accept": false, - "reason": "Reading a bufferized position_ids tensor is unrelated to unsafe sys.modules access." + "reason": "An unrecognized model/config.json problem is unrelated to unsafe `sys.modules` access." } ] }, @@ -13594,190 +13612,184 @@ "issue:43352|issue:45250", "issue:44479|issue:46170" ], - "prepared_review_unit_hash": "c7039ea9a42eb39a26f4fce9223e32fe6160134142ceac3b843d9feac3d25966", + "prepared_review_unit_hash": "83f75e21697a7401378a2be7dca1f59472bcb7ea59da77819874689828fd0335", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35004, - "estimated_input_tokens": 8751, - "estimated_eval_tokens": 17758 + "serialized_chars": 34999, + "estimated_input_tokens": 8750, + "estimated_eval_tokens": 17756 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items do not form a duplicate cluster: the soft pairs mostly share only broad keywords like video, Qwen, config, or embeddings, but they describe different code paths and fixes.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue fits this set; the candidate pairs are heterogeneous and do not point to one underlying bug or change.", + "summary": "These items are not a true duplicate cluster; they mostly share broad subsystem keywords (video/Qwen/config/modeling) but describe distinct bugs. No soft edge pair should be merged.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44479 is the closest umbrella candidate because it is the broadest, most recent video-regression report and the nearest thematic match to several other items; however, it still does not actually cover the other issues.", "canonical_pr_reason": null, - "best_issue_reason": "No issue is a good global representative for deduping here, because the items span unrelated regressions and subsystems.", + "best_issue_reason": "issue:44479 is the best representative issue for this set, but only as a loose thematic anchor, not as a real duplicate canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Both are runtime errors, but one is about video-classification processor lookup and the other about audio feature extraction; different subsystems and fixes." + "reason": "Different bugs: video-classification image-processor lookup vs audio feature extraction." }, { "left": "issue:43540", "right": "issue:46655", "accept": false, - "reason": "Both mention video inputs, but the models and failure mechanisms differ: Qwen3OmniMoe validation vs SmolVLM padding calculation." + "reason": "Both are video-related, but they hit different model/code paths and have different root causes." }, { "left": "issue:43262", "right": "issue:44246", "accept": false, - "reason": "Audio template sampling-rate handling and slow import behavior are unrelated problems." + "reason": "Audio chat-template sampling-rate default bug is unrelated to import slowness." }, { "left": "issue:44451", "right": "issue:46752", "accept": false, - "reason": "Model loading failure and empty-list chat template crash are different code paths with no shared bug." + "reason": "Model load failure vs empty-list chat template IndexError; not the same issue." }, { "left": "issue:44479", "right": "issue:46655", "accept": false, - "reason": "Both are video-related regressions, but one is a Qwen-family pipeline issue and the other is a SmolVLM padding bug; not the same fix." + "reason": "Same broad video area, but different models and distinct failures; not one concrete bug." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both involve Qwen multimodal configs, but one is a missing config field dropped by strict parsing and the other is from_config text_config unwrapping." + "reason": "Both touch Qwen multimodal config handling, but one is a missing field and the other is from_config text_config unwrapping." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Unsafe sys.modules access and a pytest-xdist file race are unrelated bugs." + "reason": "Unrelated: sys.modules access bug vs xdist race in testing utils." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "DeepSpeed ZeRO3 deque underflow and broken auto_mappings references are different failures in different areas." + "reason": "DeepSpeed ZeRO3 deque error and removed Sam3LiteText config references are different problems." }, { "left": "issue:45292", "right": "issue:47000", "accept": false, - "reason": "Token embedding resizing and RTDetr loss infeasibility are unrelated." + "reason": "Token embedding resize sync bug is unrelated to RTDetrLoss infeasible cost matrix." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "A position_ids buffer access error and a slowdown after upgrade are not the same underlying issue." + "reason": "Position_ids buffer error vs data-filtering slowdown after upgrade; different code paths." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage and Gemma 4 token-type defaults are different model-specific bugs." + "reason": "Different models and failures: ESM2 breakage vs Gemma 4 mm_token_type_ids defaulting." }, { "left": "issue:45859", "right": "issue:46399", "accept": false, - "reason": "Both are Qwen3.5-related, but one is CPU offload compatibility for rotary embeddings and the other is a kernel registration crash; different causes." + "reason": "Both mention Qwen3.5, but one is CPU offload compatibility and the other is kernel registration/decorator failure." }, { "left": "issue:43295", "right": "issue:45292", "accept": false, - "reason": "Custom processor/tokenizer regression and token embedding resizing are unrelated." + "reason": "Processor/tokenizer regression is unrelated to resize_token_embeddings not updating outputs." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "Missing Qwen2.5-VL image processor import and a DeepSpeed/BertModel index error share nothing beyond being exceptions." + "reason": "Missing Qwen2_5_VLImageProcessor import is unrelated to DeepSpeed/BertModel zero3 indexing." }, { "left": "issue:43475", "right": "issue:46369", "accept": false, - "reason": "Both mention SAM3, but one is a missing attribute in SAM3 Video outputs and the other is a memory overflow in sam3-litetext." - }, - { - "left": "issue:44466", - "right": "issue:45335", - "accept": false, - "reason": "Serialization of tied weights and decoder embedding updates after resize_token_embeddings are different bugs." + "reason": "SAM3 video encoder missing attribute vs SAM3 LiteText memory overflow are different issues." }, { "left": "issue:44623", "right": "issue:46726", "accept": false, - "reason": "Missing files from save_pretrained and an AutoProcessor backend-missing error are unrelated." + "reason": "save_pretrained missing files is not the same as unrecognized image processor due to missing vision backend." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "MoE router loss accounting and LwDetrImageLoss AMP crashes are not the same code path." + "reason": "Load-balancing loss omission and AMP image-loss crash are unrelated." }, { "left": "issue:43526", "right": "issue:46897", "accept": false, - "reason": "BeitImageProcessorFast label reduction and Florence2 training-loss double-shift are distinct issues." + "reason": "BEiT fast label reduction bug vs Florence2 training double-shift bug; different models and fixes." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Both involve config/strictness, but they affect different models and different missing fields." + "reason": "Strict config loading granite_speech is unrelated to Qwen config field handling." }, { "left": "issue:44038", "right": "issue:46506", "accept": false, - "reason": "Qwen3-VL-Moe bug reports and PPDocLayoutV3 CPU/CUDA divergence are unrelated." + "reason": "Qwen3-VL-Moe bug and PPDocLayoutV3 CPU/CUDA divergence are not the same change." }, { "left": "issue:29870", "right": "issue:30333", "accept": false, - "reason": "Bart generate() unbound local error and MLflow job-status reporting are completely unrelated." + "reason": "BART generation UnboundLocalError vs MLflow job status reporting; unrelated." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Loading a specific model and hiding tokenizer errors when protobuf is absent are different problems." + "reason": "Loading a model and masking tokenizer protobuf errors are different issues." }, { "left": "issue:43526", "right": "issue:46894", "accept": false, - "reason": "A Beit image-label bug and a CohereASR training-loss shift bug do not share the same root cause." + "reason": "BEiT label reduction bug is unrelated to CohereASR training-loss double-shift." }, { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "A concrete model not supporting Flash Attention 2.0 and a generic Flash Attention 2.0 issue are not the same underlying bug." + "reason": "Specific model not supporting Flash Attention 2.0 is not the same as the generic FA2 topic." }, { "left": "issue:44479", "right": "issue:46170", "accept": false, - "reason": "A video-input regression and a Windows/ROCm torch.distributed import problem are unrelated." + "reason": "Video-input regression vs Windows ROCm/torch.distributed import dependency issue; unrelated." } ] }, @@ -13854,8 +13866,8 @@ "issue:42199|issue:46851", "issue:44060|issue:44077", "issue:44368|issue:44977", - "issue:43653|issue:44162", "issue:45412|issue:46123", + "issue:43653|issue:44162", "issue:45397|issue:45656", "issue:43065|issue:43408", "issue:44877|issue:46170", @@ -13882,20 +13894,20 @@ "issue:44749|issue:45656", "issue:43701|issue:44749" ], - "prepared_review_unit_hash": "4813168dd0812fc36ea4b3eb0658a0daa8884583f970b5ae2b9a83fd038e3591", + "prepared_review_unit_hash": "38eae7d74dd354014f77d97152afe00124991125f57a06306404e0e689ac1f6d", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36876, + "serialized_chars": 36875, "estimated_input_tokens": 9219, "estimated_eval_tokens": 18694 }, @@ -13904,192 +13916,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are distinct Transformer bugs spanning tokenizers, cache/compile behavior, config/loading, model-specific runtime errors, and memory/performance issues. None of the soft edges looks like the same underlying bug or a mergeable PR duplicate.", - "confidence": 0.94, - "canonical_issue_reason": "No clear canonical issue: the items do not share one concrete root cause or code path, so there is no safe umbrella duplicate representative.", - "canonical_pr_reason": "No PRs are present in this cluster.", - "best_issue_reason": "No single issue is a strong global representative; the cluster is too mixed across unrelated failure modes.", - "best_pr_reason": "No PRs are present in this cluster.", + "summary": "The soft pairs are mostly broad topical similarities, not true duplicates; I rejected all of them because they describe different bugs, code paths, or model families.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42199", "right": "issue:46851", "accept": false, - "reason": "Both are model-specific, but one is a DETR cardinality error and the other is a Qwen3.5 variable-length position_ids bug; different code paths and symptoms." + "reason": "Different models and failures: DETR cardinality/background-class logic vs Qwen3.5 variable-length position_ids." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Both mention model/config warnings, but one is a Qwen3-Next tied-weights warning and the other is a PatchTSMixer post_init policy issue; not the same bug." + "reason": "Unrelated issues: tied-weights warning in Qwen3-Next vs PatchTSMixer post_init policy." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "Shared Qwen3.5 branding is not enough; one is a tied-embeddings warning during LoRA fine-tuning, the other is a flash-attention generation failure." + "reason": "Both mention Qwen3.5, but one is a LoRA warning and the other is a flash-attention generation bug." }, { - "left": "issue:43653", - "right": "issue:44162", + "left": "issue:45412", + "right": "issue:46123", "accept": false, - "reason": "Tokenizer special-token decoding issue vs a broad ESM2 regression; unrelated components and failure modes." + "reason": "Different subsystems: RT-DETR GC/memory release vs MaskGenerationPipeline batching/drop logic." }, { - "left": "issue:45412", - "right": "issue:46123", + "left": "issue:43653", + "right": "issue:44162", "accept": false, - "reason": "RT-DETR memory-release bug and MaskGenerationPipeline batch-finalization bug are different subsystems and causes." + "reason": "Tokenizer special-token registration bug vs ESM2 breakage; no shared code path." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma-4 zero3 loading issue vs deepspeed double optimizer step; different runtime behaviors and likely different fixes." + "reason": "Both involve deepspeed, but loading with ZeRO-3 and double optimizer stepping are distinct bugs." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "Both are SAM3-related, but one is a dummy Conv2d in a pixel decoder and the other is a model-type mismatch warning; not the same underlying bug." + "reason": "Both are SAM3-related, but one is a dummy layer implementation and the other is a model-type mismatch warning." }, { "left": "issue:44877", "right": "issue:46170", "accept": false, - "reason": "Strict config loading for granite_speech vs Windows/ROCm torch.distributed import errors are unrelated." + "reason": "Strict config loading vs Windows/ROCm import dependency issue; unrelated failures." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "SAM3 video attribute missing vs Qwen3OmniMoe video-input ValueError; different models and code paths." + "reason": "Different model families and symptoms: SAM3 vision encoder attribute error vs Qwen3OmniMoe video processing ValueError." }, { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Offline tokenizer cache loading vs Gemma4 multi-GPU CUDA OOM; one is I/O/cache logic, the other is memory pressure." + "reason": "Tokenizer cache/offline loading bug vs Gemma4 multi-GPU CUDA OOM; unrelated." }, { "left": "issue:46170", "right": "issue:46832", "accept": false, - "reason": "Import/dependency failure on Windows/ROCm vs RT-DETR feature-level mismatch in forward; unrelated." + "reason": "Import/dependency failure vs RT-DETR forward shape mismatch; no common underlying defect." }, { "left": "issue:43723", "right": "issue:46710", "accept": false, - "reason": "Tokenizer loading regression in v5 vs accuracy regression after upgrade; different symptoms and no shared concrete code path." + "reason": "Tokenizer-loading regression vs accuracy regression after upgrade; different problem classes." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 tied-embeddings warning are different model-loading concerns, not the same defect." + "reason": "Qwen3-VL weight-shape load error vs Qwen3.5 warning about tie_word_embeddings; unrelated." }, { "left": "issue:44188", "right": "issue:46202", "accept": false, - "reason": "Torch.compile attention-kernel divergence vs Gemma pipeline roundtrip failure on \u2581; these are distinct issues." + "reason": "Attention-kernel divergence under torch.compile vs Gemma pipeline roundtrip token issue; different code paths." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "Tied-embeddings warning during fine-tuning vs stale num_return_sequences warning in paged generate(); unrelated generate-time behavior." + "reason": "Qwen3.5 tie_word_embeddings warning vs stale num_return_sequences warning in paged generate." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "GLM MoE stale indexer cache on second forward pass vs Gemma4 mm_token_type_ids defaulting; different bugs." + "reason": "GlmMoeDsa stale indexer cache on second forward vs Gemma4 mm_token_type_ids training requirement." }, { "left": "issue:44483", "right": "issue:46202", "accept": false, - "reason": "Chat-completions API request handling vs Gemma pipeline token roundtrip failure; no shared underlying code path." + "reason": "Chat-completions API request rejection vs Gemma pipeline roundtrip failure; unrelated." }, { "left": "issue:43906", "right": "issue:46202", "accept": false, - "reason": "Reproduction of a specific external issue vs Gemma pipeline token roundtrip failure; the titles suggest different models and bugs." + "reason": "One is an isolated reproduction of another issue; the Gemma roundtrip bug is a different concrete failure." }, { "left": "issue:44188", "right": "issue:46421", "accept": false, - "reason": "Both involve compile/perf concerns, but one is attention-kernel divergence and the other is cache lazy initialization causing recompilation; not the same fix." + "reason": "Both touch cache/compile behavior, but one is divergent kernels and the other is spurious recompilation from lazy init." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "GLM MoE stale cache bug vs flash_attention s_aux=None crash; different layers of the stack and different root causes." + "reason": "Different bugs: stale cache in GlmMoeDsa vs flash_attention.py crashing on s_aux=None." }, { "left": "issue:45593", "right": "issue:47068", "accept": false, - "reason": "D-FINE auxiliary-loss behavior vs deepspeed host-synchronization in loss computation; different models and mechanisms." + "reason": "D-FINE auxiliary-loss behavior vs a deepspeed loss host-sync concern; not the same defect." }, { "left": "issue:45685", "right": "issue:46431", "accept": false, - "reason": "MPS histogram integer support issue vs missing chat template ValueError; unrelated functionality." + "reason": "MPS int histogram failure vs processor chat-template absence; no overlap." }, { "left": "issue:45987", "right": "issue:46082", "accept": false, - "reason": "StaticCache return-type contract bug vs LlamaConfig head_dim validation; one is cache API behavior, the other is config validation." + "reason": "Cache return-type contract bug vs LlamaConfig head_dim validation; different issues." }, { "left": "issue:45103", "right": "issue:46858", "accept": false, - "reason": "Docstring helper crash with future annotations vs cache.get_seq_length returning a device tensor on non-CUDA backends; different utilities and failures." + "reason": "Docstring helper crash with future annotations vs static-cache seq-length tensor bug; unrelated." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer task AttributeError vs a Transformer upgrade performance regression; unrelated symptoms and fix surface." + "reason": "Tokenizer task AttributeError vs performance regression after version upgrade; different symptoms and code paths." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "SAM3 pixel-decoder dummy layer issue vs MaskGenerationPipeline partial-batch result loss; unrelated code paths." + "reason": "SAM3 pixel-decoder implementation issue vs pipeline batching logic; unrelated." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Tokenizer class metadata mismatch on save vs processor.save_pretrained missing files; both are save-path issues but not the same defect." + "reason": "Both concern save_pretrained, but tokenizer_config class mismatch and processor missing files are distinct serialization bugs." }, { "left": "issue:43994", "right": "issue:46726", "accept": false, - "reason": "SigLIP2 nonsensical outputs with AutoModel/pipeline vs AutoProcessor missing-vision-backend error; different failure modes." + "reason": "Siglip2 wrong outputs vs missing vision-backend error in AutoProcessor; different failure modes." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Upgrade-related filtering slowdown vs deepspeed double optimizer step; no shared underlying bug." + "reason": "Data filtering slowdown vs double optimizer step under deepspeed; unrelated." }, { "left": "issue:43701", "right": "issue:44749", "accept": false, - "reason": "Checkpoint key mismatch vs post-upgrade performance slowdown; unrelated issues." + "reason": "Checkpoint key mismatch vs data filtering slowdown; no shared underlying bug." } ] }, @@ -14190,20 +14202,20 @@ "issue:43873|issue:44945", "issue:43784|issue:46399" ], - "prepared_review_unit_hash": "4439a50c20f1cab3011ae02fec0c8732af7dc4c9393d86cff53fbd48527f74fd", + "prepared_review_unit_hash": "15dc14f8634b7e849cd2c8666a7153e78dc8852c5d28491af817e6a4c63f6df8", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 34857, + "serialized_chars": 34860, "estimated_input_tokens": 8715, "estimated_eval_tokens": 17686 }, @@ -14212,180 +14224,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a loose cluster of distinct Transformers regression reports across v5 loading, generation, quantization, tokenizers, and processor/vision paths. One soft duplicate-looking pair stands out around AutoProcessor and missing vision backends; the rest are not the same underlying bug.", - "confidence": 0.57, - "canonical_issue_reason": "issue:43646 is the broadest and most representative report here: a general Transformers 5.0 breakage around model initialization, matching the cluster\u2019s repeated theme of v5 compatibility regressions.", + "summary": "The set is mostly a grab-bag of unrelated Transformers bugs, with one clear duplicate chain around Qwen3.5-MoE `from_pretrained` failures. The most representative issue is the one that names the concrete missing config field behind that regression.", + "confidence": 0.69, + "canonical_issue_reason": "Issue 45375 is the best canonical issue because it identifies the concrete root cause (`deepstack_visual_indexes` being dropped by `@strict`) behind the Qwen3.5-MoE loading regression.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the best single issue to anchor the cluster because it is general, centrally about v5 breakage, and closer to the repeated pattern than the narrowly scoped model-specific reports.", + "best_issue_reason": "Issue 45375 is the strongest global representative: it is specific enough to explain the failure mode, while the other Qwen3.5-MoE reports are mostly generic symptoms of the same load problem.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Different models and failure modes: generic custom model initialization vs ESM2-specific breakage." + "reason": "Both are v5-related regressions, but one is generic custom-model initialization and the other is an ESM2-specific failure path; they are not the same bug." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Different subsystems: missing files from processor.save_pretrained vs DeepSpeed optimizer stepping twice." + "reason": "`processor.save_pretrained(...)` missing files and a double optimizer step under DeepSpeed are different code paths and different failure modes." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Unrelated problems: eos_token_id config mismatch vs AttentionInterface.register behavior change." + "reason": "Gemma EOS config mismatch and `AttentionInterface.register` behavior change are unrelated issues in different subsystems." }, { "left": "issue:46752", "right": "issue:47020", "accept": false, - "reason": "Different code paths: empty apply_chat_template input vs XLMRobertaTokenizer/Unigram constructor type mismatch." + "reason": "An empty chat list crashing `apply_chat_template` is unrelated to `XLMRobertaTokenizer` passing the wrong type to `Unigram`." }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Different stack areas: multimodal model loading regression vs DeepSpeed ZeRO3 deque error." + "reason": "Loading a multimodal Phi-4 model and a DeepSpeed ZeRO3 deque pop are separate bugs with no clear shared code path." }, { "left": "issue:45375", "right": "issue:45478", - "accept": false, - "reason": "Both involve Qwen3.5-MoE, but one is a specific missing config field while the other is a generic from_pretrained failure; too little evidence they are the same bug." + "accept": true, + "reason": "Both reports point to the same Qwen3.5-MoE `from_pretrained` regression under Transformers v5.4.x; 45375 gives the concrete missing-field root cause." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Both relate to quantized models, but one is saving and the other is loading a specific GGUF model; not the same issue." + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different quantization/loading problems, not the same bug." }, { "left": "issue:39692", "right": "issue:46726", "accept": false, - "reason": "Both mention vision/processor usage, but one is a documentation example with model/processor mismatch and quantization errors; not the same underlying bug." + "reason": "A broken documentation example is not the same as AutoProcessor's missing-vision-backend error handling." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "Different attribute-missing failures in different objects and code paths; not the same root cause." + "reason": "SAM 3 video attribute loss and `trust_remote_code=True` returning incomplete config objects are different model/config issues." }, { "left": "issue:43475", "right": "issue:46821", "accept": false, - "reason": "Both are attribute errors, but they affect different models and different fields." + "reason": "These are distinct `AttributeError`s in different models (`Sam3VisionEncoderOutput` vs TimesFM `window_size`)." }, { "left": "issue:43986", "right": "issue:46726", - "accept": true, - "reason": "Both describe AutoProcessor failing when the required vision backend is missing, with the same unrecognized-image-processor symptom and same underlying fix area." + "accept": false, + "reason": "Both involve vision backend availability, but one is a video-model import crash without torchvision and the other is an unrecognized image processor error; not the same concrete bug." }, { "left": "issue:44188", "right": "issue:46851", "accept": false, - "reason": "Different bugs in different models: attention-kernel divergence under torch.compile vs Qwen3.5 variable-length position_ids handling." + "reason": "The first is a torch.compile attention-kernel divergence; the second is a Qwen3.5 variable-length `position_ids` bug." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "The first is a missing PyTorch dependency import error; the second is a loading hang. Similar area, different failure." + "reason": "Missing PyTorch import support and Sentence Transformers getting stuck loading are different symptoms and likely different fixes." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "No meaningful overlap; one is a placeholder bug report, the other is a Gemma-4 ZeRO3 loading issue." + "reason": "A placeholder bug report is not a duplicate of the Gemma-4 ZeRO3 `from_pretrained` failure." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "Different generation-time regressions: sync_gpus/update_model_kwargs ordering vs missing GenerationMixin cache in chunked_prefill." + "reason": "Both are generation-related, but one is about `_update_model_kwargs_for_generation` after `sync_gpus` and the other is a v5.0 cache regression during chunked prefill." }, { "left": "issue:46877", "right": "issue:47000", "accept": false, - "reason": "Different kinds of issues: doc-lint stdout noise at import time vs RTDetrLoss infeasible cost matrix." + "reason": "Import-time docstring lint output and an RTDetrLoss infeasible cost-matrix crash are unrelated." }, { "left": "issue:45310", "right": "issue:45375", - "accept": false, - "reason": "Same model family, but one is a generic from_pretrained error and the other is a specific config-field-strictness bug; not enough to merge." + "accept": true, + "reason": "Both describe the same Qwen3.5-MoE `from_pretrained` breakage in v5.4.x; 45375 is the more specific root-cause report." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Different models and failure modes; Qwen3-VL-Moe bug is not the same as processor.save_pretrained missing files." + "reason": "Qwen3-VL-Moe loading and processor save-pretrained file emission are different problems despite both touching processor/model code." }, { "left": "issue:46463", "right": "issue:46877", "accept": false, - "reason": "Incorrect decoder validation message vs docstring lint output; unrelated." + "reason": "A decoder-input validation message bug and stdout doc-lint noise at import time are not the same issue." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "Both are import-related, but they refer to different missing symbols and different modules." + "reason": "A missing `AutoImageProcessor` import and a non-existent `image_utils` import are separate API surface problems." }, { "left": "issue:44373", "right": "issue:46851", "accept": false, - "reason": "One is a docstring mistake; the other is a runtime position_ids bug in Qwen3.5." + "reason": "A docstring typo for `position_ids` is unrelated to the Qwen3.5 variable-length input failure." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized model saving vs missing pad_token_id in config are separate problems." + "reason": "Quantized model saving and a missing `pad_token_id` on `Llama4Config` are different failures." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Different tokenizer/text-processing failures in different model families." + "reason": "Whisper timestamp decoding and Kimi-K2.5 tokenizer codec handling are separate tokenizer/decoding regressions." }, { "left": "issue:43723", "right": "issue:46326", "accept": false, - "reason": "Tokenizer loading regression vs continue_final_message behavior for mistral_common tokenizers; not the same bug." + "reason": "Tokenizer loading in v5 and `continue_final_message` with `mistral_common` are different paths and different bugs." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 settings API guidance is unrelated to InternVL2 meta-initialization incompatibility." + "reason": "TF32 API guidance and InternVL2 meta-initialization incompatibility are unrelated issues." }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization offloading behavior and pipeline-parallel output correctness are different issues." + "reason": "Quantization/offloading behavior and incorrect pipeline-parallel LLM output are different runtime bugs." }, { "left": "issue:43784", "right": "issue:46399", "accept": false, - "reason": "Sentence-transformers import NameError vs Qwen3.5 kernel registration crash; unrelated." + "reason": "Sentence-transformers import failure and Qwen3.5 kernel-registration crash are unrelated code paths." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The canonical issue choice is reasonable, and the soft-edge verdicts avoid over-claiming duplicate status except for one clearly related AutoProcessor/vision-backend pair, which is supported by the titles." + "feedback": "The summary is well grounded in the packet: 45310 and 45478 are clearly the same Qwen3.5-MoE from_pretrained regression, and 45375 is the more specific root-cause report naming the missing `deepstack_visual_indexes` field. The soft-edge rejections are conservative and the accepted pairs look mergeable as a single bugfix." }, "evaluator_used": true, "retried": false, @@ -14481,228 +14493,240 @@ "issue:45561|issue:45600", "issue:43976|issue:44933", "issue:30990|issue:35141", - "issue:43479|issue:46829", "issue:42898|issue:44568", + "issue:43479|issue:46829", "issue:42175|issue:45712", "issue:44568|issue:47085", "issue:44623|issue:45310", "issue:45600|issue:47059", "issue:42673|issue:46693" ], - "prepared_review_unit_hash": "ac0532e041af74ce70a764d28a36e3188fffc6a649495a9530289653b67ca019", + "prepared_review_unit_hash": "7d7fd15c23d22216fe62d25c95d2be0def45162def84013ed6e50baeace052cf", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37178, - "estimated_input_tokens": 9295, - "estimated_eval_tokens": 18846 + "serialized_chars": 37174, + "estimated_input_tokens": 9294, + "estimated_eval_tokens": 18844 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous; I only see one clear duplicate pair: the Qwen3ForCausalLM VRAM leak reports. The rest are related only by broad subsystem or model family, not the same underlying bug/change.", - "confidence": 0.94, - "canonical_issue_reason": "issue:42673 is the clearest and most detailed report of the exact same Qwen3ForCausalLM VRAM leak as issue:43704, so it is the best canonical issue.", + "summary": "This is mostly a heterogeneous similarity cluster with one real duplicate pair and many false-positive matches across unrelated bugs/features. The only strong duplicate is the Qwen3ForCausalLM VRAM leak reports.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue for the whole cluster: most items are unrelated. For the one true duplicate pair, issue 42673 is the better anchor because it is earlier and more discussed.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42673 is the strongest issue to anchor this cluster because it matches its duplicate exactly and has the fuller reproduction/context.", + "best_issue_reason": "Issue 42673 is the best representative only for the genuine duplicate subcluster (paired with 43704), but it does not represent the rest of the mixed cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45561", "right": "issue:47059", "accept": false, - "reason": "Different failures: xdist capture-file race vs CLI completion attribute error." + "reason": "pytest-xdist file race vs a huggingface_hub CLI completion breakage; different code paths." }, { "left": "issue:45137", "right": "issue:47000", "accept": false, - "reason": "Different code paths: DeepSpeed ZeRO3 deque underflow vs RTDetr loss assignment infeasibility." + "reason": "DeepSpeed ZeRO3 deque error vs RTDetr loss infeasible cost-matrix crash; unrelated failures." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different missing symbols and unrelated APIs." + "reason": "Different missing imports: AutoImageProcessor vs HybridCache." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "Different missing imports: GenerationMixin vs AutoImageProcessor." + "reason": "GenerationMixin import failure is a different symbol/module than AutoImageProcessor." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Both affect generation, but one is pipeline-parallel output corruption and the other is a model/image-specific infinite loop." + "reason": "Pipeline-parallelism output bug vs Qwen3_5 image infinite loop; not the same bug." }, { "left": "issue:44857", "right": "issue:46877", "accept": false, - "reason": "Loss/AMP crash is unrelated to a docstring import-time stdout bug." + "reason": "CUDA AMP crash in LwDetrImageLoss vs auto_docstring stderr at import time." }, { "left": "issue:44661", "right": "issue:47059", "accept": false, - "reason": "Tokenizer-mapping registration failure is unrelated to the CLI completion regression." + "reason": "Tokenizer-mapping model registration issue vs CLI completion attribute error." }, { "left": "issue:45406", "right": "issue:46202", "accept": false, - "reason": "Gemma processor save/load attribute bug vs tokenizer roundtrip behavior; not the same bug." + "reason": "Gemma4Processor missing _tokenizer during serve vs Gemma pipeline roundtrip tokenization issue." }, { "left": "issue:42994", "right": "issue:43824", "accept": false, - "reason": "Quantized save failure and Qwen2.5-VL import failure are distinct issues." + "reason": "Quantized save_pretrained failure vs missing Qwen2_5_VLForConditionalGeneration import; different functionality." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Same broad model family, but different model variants and different loading failures." + "reason": "Both mention Qwen3 MoE, but the titles point to different model/version-specific loading problems without clear same code path." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Tokenizer patch crash and auto-docstring annotation crash are unrelated." + "reason": "Mistral tokenizer regex crash vs auto_docstring annotation crash; unrelated." }, { "left": "issue:45593", "right": "issue:46877", "accept": false, - "reason": "D-FINE auxiliary-loss behavior is unrelated to auto_docstring stderr/stdout output." + "reason": "D-FINE auxiliary-loss behavior vs auto_docstring import-time diagnostics; unrelated." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "FP8 save_pretrained failure is not the same as the Qwen3.5 MoE from_pretrained error." + "reason": "FP8 save_pretrained failure and Qwen3.5 MoE from_pretrained failure are different code paths." }, { "left": "issue:46202", "right": "issue:46701", "accept": false, - "reason": "Both involve Gemma, but one is tokenizer roundtrip and the other is pipeline output plumbing." + "reason": "Gemma pipeline roundtrip issue vs image-text-to-text pipeline output class issue; related model family but different bug." }, { "left": "issue:42673", "right": "issue:43704", "accept": true, - "reason": "Same exact bug and model path: Qwen3ForCausalLM VRAM leak when used from multiple dataloader threads." + "reason": "Same Qwen3ForCausalLM VRAM leak in multiple dataloader threads; these are duplicate reports of the same bug." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs update after sync_gpus is unrelated to the Janus image-generation test failure." + "reason": "Generation kwargs/update bug after sync_gpus vs janus image-generation test failure; different paths." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "FP8 accuracy regression and FP8 save_pretrained failure are different problems." + "reason": "FP8 accuracy regression vs FP8 save_pretrained failure; same broad feature area but different issues." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Backend dependency omission is unrelated to a missing image_utils import." + "reason": "Packaging backend omission vs missing image_utils import; unrelated." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Model config/import resolution bug vs xdist file race; not the same issue." + "reason": "Model config loading error vs testing-utils xdist race; unrelated." }, { "left": "issue:45405", "right": "issue:46877", "accept": false, - "reason": "Package version pinning issue is unrelated to docstring logging at import time." + "reason": "Dependency-version bump complaint vs docstring diagnostics printing; unrelated." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLflow reporting on failed training jobs is unrelated to a Deepspeed/Adafactor OOM." + "reason": "MLflow reporting on failed training vs deepspeed OOM; different failure modes." }, { "left": "issue:44743", "right": "issue:47068", "accept": false, - "reason": "Recurrent-state reset in Qwen3.5 module and host synchronization in deepspeed_sp_compute_loss are different bugs." + "reason": "Qwen3 recurrent-state reset bug vs host synchronization in deepspeed loss; unrelated." }, { "left": "issue:45561", "right": "issue:45600", "accept": false, - "reason": "Test race in patched utils vs stale auto_mappings references; not the same underlying failure." + "reason": "pytest-xdist race vs stale auto_mappings references; unrelated." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version compatibility issue is unrelated to a missing image_utils import." + "reason": "Python-version compatibility issue vs missing import from image_utils; unrelated." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "SentenceTransformers loading hang and embedding reinitialization are unrelated." + "reason": "Loading hang for Sentence Transformers vs embedding resize/post_init reinitialization; unrelated." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Token cleanup behavior change vs BOS/EOS insertion regression; both tokenizer-related but not the same bug." }, { "left": "issue:43479", "right": "issue:46829", "accept": false, - "reason": "Config default initialization bug vs output field mismatch in classifier output." + "reason": "Multimodal config default initialization vs VideoPrism output field mismatch; unrelated." }, { - "left": "issue:42898", - "right": "issue:44568", + "left": "issue:42175", + "right": "issue:45712", "accept": false, - "reason": "Tokenizer cleanup behavior change and BOS/EOS insertion regression are different tokenizer issues." + "reason": "TensorFlow extras packaging vs leftover dummy PT objects; unrelated." + }, + { + "left": "issue:44568", + "right": "issue:47085", + "accept": false, + "reason": "Special-tokens behavior regression vs tokenizer thread-safety regression; different defects." }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "Processor.save_pretrained file omission is unrelated to Qwen3.5 MoE loading failure." + "reason": "processor.save_pretrained missing files vs Qwen3.5 MoE loading error; different code paths." }, { "left": "issue:45600", "right": "issue:47059", "accept": false, - "reason": "Stale model mappings and CLI completion attribute error are unrelated." + "reason": "auto_mappings CI breakage vs CLI completion attribute error; unrelated." }, { "left": "issue:42673", "right": "issue:46693", "accept": false, - "reason": "Both are performance-related, but one is a VRAM leak and the other is a long-context slowdown regression." + "reason": "Qwen3ForCausalLM VRAM leak vs FA2 long-context performance regression; different issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted pair, issue:42673 and issue:43704, is an exact title match and clearly the same bug report. The false soft-edge verdicts are cautious and consistent with the titles; none overstate evidence beyond what is visible in the packet." + "feedback": "Grounded overall. The summary stays conservative: it identifies the identical Qwen3ForCausalLM VRAM-leak pair (42673/43704) as the only strong duplicate and treats the rest as heterogeneous false positives. The soft-edge verdicts are cautious and do not overclaim sameness where the titles point to different bugs or code paths." }, "evaluator_used": true, "retried": false, @@ -14775,8 +14799,8 @@ "soft_pairs": [ "issue:44186|issue:47085", "issue:46077|issue:46123", - "issue:43479|issue:44869", "issue:42898|issue:45812", + "issue:43479|issue:44869", "issue:43723|issue:44568", "issue:44016|issue:46231", "issue:39692|issue:46909", @@ -14787,8 +14811,8 @@ "issue:27957|issue:30333", "issue:42673|issue:47085", "issue:30333|issue:34689", - "issue:42222|issue:45200", "issue:42898|issue:43295", + "issue:42222|issue:45200", "issue:43874|issue:45092", "issue:44295|issue:45656", "issue:44734|issue:45103", @@ -14804,214 +14828,208 @@ "issue:42831|issue:44623", "issue:44060|issue:45125" ], - "prepared_review_unit_hash": "d9829e54923c6da19ba90b77e3f208eed2747e6ffce8d604b3becd604173665d", + "prepared_review_unit_hash": "e740abd294763e2fb60855dc235cf71f1b00605461e5bdf057726a67afa2b6be", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36913, - "estimated_input_tokens": 9229, - "estimated_eval_tokens": 18714 + "serialized_chars": 36908, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issues; reject all soft pairs. Several items share broad tokenizer or multimodal wording, but they point to different code paths, models, or documentation bugs rather than the same underlying defect.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue is safe here: the set is heterogeneous, and the apparent matches are only superficial similarities.", + "summary": "This cluster is not a true duplicate set: the items span many unrelated bugs and docs regressions across tokenizers, model loading, multimodal configs, training, and infrastructure. All soft pairs should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No clear canonical issue: the cluster is a grab bag of unrelated bugs, regressions, docs issues, and infrastructure problems.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue stands out because the cluster does not cohere around one bug or one subsystem.", + "best_issue_reason": "No single issue is a good global representative because the topics are heterogeneous and do not share one underlying code path.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44186", "right": "issue:47085", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash and fast tokenizer thread-safety are different bugs in different tokenizer stacks." + "reason": "Different problems: LayoutLMv2 NER/padding crash vs fast tokenizer thread-safety regression." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Gemma4 attention-mask construction and MaskGenerationPipeline batch handling are unrelated failure modes." + "reason": "Gemma4 attention-mask bug and MaskGenerationPipeline partial-batch dropping are unrelated code paths." }, { - "left": "issue:43479", - "right": "issue:44869", + "left": "issue:42898", + "right": "issue:45812", "accept": false, - "reason": "Phi4MultimodalConfig default-init bug and Whisper timestamp decode crash affect different components." + "reason": "Tokenizer cleanup behavior change is a different regression from Granite token-ID corruption." }, { - "left": "issue:42898", - "right": "issue:45812", + "left": "issue:43479", + "right": "issue:44869", "accept": false, - "reason": "Both are tokenizer regressions, but one is a general v5 cleanup behavior change and the other is a Granite model ID regression." + "reason": "Multimodal config default initialization and Whisper decode crash are unrelated." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "AutoTokenizer.from_pretrained loading issues and mdeberta add_special_tokens behavior are distinct bugs." + "reason": "Generic AutoTokenizer loading issue is not the same as mDeBERTa special-token insertion regression." }, { "left": "issue:44016", "right": "issue:46231", "accept": false, - "reason": "Notebook syntax error and a doc parameter-name mismatch are unrelated documentation problems." + "reason": "Notebook syntax error and a docs parameter-name mismatch are separate documentation issues." }, { "left": "issue:39692", "right": "issue:46909", "accept": false, - "reason": "Different documentation examples for different models; not the same broken example." + "reason": "Different broken examples for different models; no shared underlying bug." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Sentence-transformers import NameError and Qwen2.5-VL import error are separate import-path failures." + "reason": "Sentence-transformers import failure and missing Qwen2.5-VL export are distinct import issues." }, { "left": "issue:43479", "right": "issue:44079", "accept": false, - "reason": "A multimodal config initialization bug does not match a ModelOutput key-assignment bug." + "reason": "Config defaulting bug vs ModelOutput key assignment bug; unrelated internals." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both involve tokenizer regressions, but they target different models and different tokenizer behavior." - }, - { - "left": "issue:44843", - "right": "issue:47085", - "accept": false, - "reason": "Offline hub model_info call and tokenizer thread-safety are unrelated." + "reason": "Both involve tokenizers, but one is a global cleanup regression and the other is a Deepseek codec-handling regression." }, { "left": "issue:27957", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 on XLMRoberta and MLFlow reporting after failed training are unrelated." + "reason": "Flash Attention model-loading issue and MLflow reporting issue have nothing in common." }, { "left": "issue:42673", "right": "issue:47085", "accept": false, - "reason": "Both mention threading, but one is a VRAM leak in Qwen3ForCausalLM and the other is tokenizer concurrency safety." + "reason": "Qwen3 VRAM leak in dataloader threads is different from tokenizer thread-safety." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "MLFlow status handling and Llama 3.2 Vision model loading are unrelated." + "reason": "MLflow failure reporting and Llama 3.2 Vision loading regression are unrelated." }, { - "left": "issue:42222", - "right": "issue:45200", + "left": "issue:42898", + "right": "issue:43295", "accept": false, - "reason": "Vitpose model breakage and Gemma 4 mm_token_type_ids defaults are different model-specific issues." + "reason": "Tokenizer cleanup regression is not the same as processor.tokenizer/image-passing regression." }, { - "left": "issue:42898", - "right": "issue:43295", + "left": "issue:42222", + "right": "issue:45200", "accept": false, - "reason": "Both are v5 tokenizer regressions, but one concerns tokenizer cleanup/output semantics and the other processor/tokenizer image-handling behavior." + "reason": "Vitpose model breakage and Gemma4 multimodal token-type defaults are different model families and bugs." }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "Missing image-patch accessor and old InternVL2 meta-init incompatibility are different multimodal bugs." + "reason": "Missing image-processor method and remote-code meta-init incompatibility are different failures." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Position_ids buffer access and double optimizer stepping with deepspeed are unrelated." + "reason": "Position_ids buffer access error and double optimizer stepping under deepspeed are unrelated." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "Serve KV-cache tensor indexing and auto_docstring annotation handling are separate code paths." + "reason": "Serve KV-cache indexing crash and auto_docstring future-annotations crash are separate subsystems." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Tokenizer cleanup regression and Kimi-K2.5 codec/fix_mistral_regex regression are different tokenizer issues." + "reason": "Kimi-K2.5 tokenizer codec regression is a different issue from clean_up_tokenization_spaces behavior changes." }, { "left": "issue:30333", "right": "issue:35141", "accept": false, - "reason": "MLFlow reporting and embedding reinitialization after resize are unrelated." + "reason": "MLflow reporting and embedding reinitialization are unrelated bugs." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Missing transformers.onnx module and missing PyTorch dependency are different import errors." + "reason": "Missing transformers.onnx module and missing PyTorch dependency are both import errors but with different causes." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "FP8 save_pretrained moe and tokenizer.save_pretrained class metadata are different save-pretrained bugs." + "reason": "FP8 MoE save_pretrained bug and tokenizer save_pretrained config mismatch are different save paths." }, { "left": "issue:44075", "right": "issue:46231", "accept": false, - "reason": "Optimizer SGD args being ignored and a docs parameter-name typo are unrelated." + "reason": "Optimizer SGD args being ignored is unrelated to a docs parameter-name typo." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Example script multi-label output bug and local dataset loading failure are different issues in different scripts." + "reason": "Classification example returning empty results and image-classification dataset loading are separate example-script bugs." }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "A dependency-version bump and leftover dummy classes are unrelated maintenance issues." + "reason": "PEFT version bump and leftover dummy classes are unrelated maintenance issues." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Nonexistent image_utils import and dummy_pt_objects cleanup are not the same bug." + "reason": "Missing image_utils import and dummy_pt_objects cleanup are different problems." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save metadata mismatch and deepspeed double-step are unrelated." + "reason": "Tokenizer save_pretrained metadata mismatch has nothing to do with deepspeed optimizer stepping." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FineGrainedFP8 accuracy and processor.save_pretrained missing files are separate bugs." + "reason": "FineGrainedFP8 accuracy and processor.save_pretrained missing files are unrelated." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Wrong tied-weights warning and missing _tp_plan are different Qwen3-family issues with different fixes." + "reason": "Incorrect tied-weights warning and missing tensor-parallel plan are distinct Qwen3-Next issues." } ] }, @@ -15102,8 +15120,8 @@ "issue:43122|issue:46209", "issue:46614|issue:46932", "issue:44263|issue:45092", - "issue:43232|issue:46424", "issue:44295|issue:44623", + "issue:43232|issue:46424", "issue:46209|issue:46431", "issue:43232|issue:45834", "issue:30990|issue:38175", @@ -15117,220 +15135,220 @@ "issue:42831|issue:45310", "issue:22355|issue:27957" ], - "prepared_review_unit_hash": "b95d278044856e864bf653ab26a0875ccd3b45ee3a4e5cb3fbc0bb12c1a86f7c", + "prepared_review_unit_hash": "4a77ca958ad2d80bf4393b7a6d80e89ef0c23a2dba30408b7bcfe95a56e2d15e", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36538, - "estimated_input_tokens": 9135, - "estimated_eval_tokens": 18526 + "serialized_chars": 36527, + "estimated_input_tokens": 9132, + "estimated_eval_tokens": 18520 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is highly heterogeneous. None of the soft pairs look like the same underlying bug or change; most are only loosely similar by subsystem or wording. If one issue must serve as an anchor, issue:44263 is the most central by soft-link connectivity, but it is not a true duplicate representative of the whole set.", - "confidence": 0.35, - "canonical_issue_reason": "issue:44263 has the strongest soft-edge connectivity and the most concrete code-path bug surface (GlmMoeDsaIndexer), so it is the best anchor among otherwise unrelated items.", + "summary": "This cluster is not a duplicate set; it mixes many unrelated Transformers regressions and model-specific bugs (tokenization, config validation, quantization, attention, save/load, generation, and CLI/import issues).", + "confidence": 0.95, + "canonical_issue_reason": "Issue 45071 is the broadest representative of the cluster\u2019s dominant v5 regression theme, but there is no single underlying bug shared by most items.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44263 is the best representative because it is narrowly scoped, actionable, and sits at the center of the closest related cluster of DSA/indexer issues.", + "best_issue_reason": "Issue 45071 is the best global representative because it describes a general Transformers v5 compatibility regression rather than a narrowly model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "Different problems: Pegasus output regression vs attention-registration behavior/type semantics." + "reason": "Different bugs: Pegasus v4/v5 output change vs AttentionInterface.register altering callback behavior." }, { "left": "issue:42796", "right": "issue:43723", "accept": false, - "reason": "Both are tokenizer/import-related, but one is BPE marker handling and the other is AutoTokenizer loading in v5; not the same bug." + "reason": "Both are tokenizer-related, but one is a BPE-marker handling bug and the other is an AutoTokenizer import/load issue." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both mention save_pretrained, but one is FP8 MoE saving and the other is processor file serialization; different code paths." + "reason": "Different save_pretrained paths: FP8 MoE checkpoint saving vs processor artifact omissions." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Config default initialization bug vs non-persistent buffer corruption; unrelated mechanisms." + "reason": "Different code paths: multimodal config default initialization vs non-persistent buffer initialization." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category-limit runtime error and attention interface behavior are unrelated." + "reason": "Unrelated bugs: category-count runtime limit vs attention registration behavior." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols/packages and different root causes." + "reason": "Both are import errors, but for different symbols and different root causes." }, { "left": "issue:44360", "right": "issue:46762", "accept": false, - "reason": "DSA indexer ReLU discussion vs Minimax attention numerical differences; different implementations and failures." + "reason": "Different issues in attention/math: missing ReLU in DSA indexer vs numerical mismatch in Minimax MSA attention." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Wrong docstring for position_ids is documentation-only; GLM-5 RoPE is an implementation issue." + "reason": "Docstring mismatch is not the same as a RoPE implementation discussion." }, { "left": "issue:46132", "right": "issue:46710", "accept": false, - "reason": "AttentionInterface registration behavior vs model accuracy regression after upgrade; not the same change." + "reason": "AttentionInterface behavior change is not the same as a model accuracy regression after upgrade." }, { "left": "issue:44297", "right": "issue:45310", "accept": false, - "reason": "Tokenizer config serialization mismatch vs Qwen3.5 MoE from_pretrained error; different symptoms and likely fixes." + "reason": "Different tokenizer persistence/load failures with different symptoms and likely different fixes." }, { "left": "issue:43504", "right": "issue:47059", "accept": false, - "reason": "Beit legacy-field loading bug vs CLI completion attribute failure; unrelated subsystems." + "reason": "Unrelated failures: legacy field loading vs CLI breakage from huggingface_hub API changes." }, { "left": "issue:45071", "right": "issue:46082", "accept": false, - "reason": "Type checking breakage in PretrainedConfig vs LlamaConfig head_dim validation; different validations." + "reason": "Generic PretrainedConfig type-checking breakage is not the same as LlamaConfig head_dim validation." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Tokenizer loading issue vs flash-attention generation failure; not the same code path." + "reason": "Tokenizer load failure and flash-attention generation issue are different code paths." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "Both involve save_pretrained, but one is a serialization bug and the other is a shard-size question/value choice." + "reason": "Checkpoint-saving bug vs a question about the default max_shard_size." }, { "left": "issue:43122", "right": "issue:46209", "accept": false, - "reason": "Tokenizer-version output drift vs FP8 inference failure on B200; unrelated." + "reason": "Tokenizer behavior changed across versions, but the other is an FP8 inference failure on B200." }, { "left": "issue:46614", "right": "issue:46932", "accept": false, - "reason": "Stale phonemizer backend state vs Gemma4 dtype casting; completely different areas." + "reason": "Different multilingual phonemizer state bug vs Gemma4 quantized embedder dtype casting fix." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "DSA indexer split/return-value issue vs remote-code meta-init incompatibility; different bugs." + "reason": "torch.split return handling in a MoE indexer is unrelated to remote-code meta-initialization incompatibility." }, { - "left": "issue:43232", - "right": "issue:46424", + "left": "issue:44295", + "right": "issue:44623", "accept": false, - "reason": "Generation kwargs after sync_gpus vs cache-length honoring in quantized cache; different features." + "reason": "Buffer registration/read error is not the same as missing files in processor.save_pretrained." }, { - "left": "issue:44295", - "right": "issue:44623", + "left": "issue:43232", + "right": "issue:46424", "accept": false, - "reason": "Buffer read error for position_ids vs processor.save_pretrained missing files; unrelated." + "reason": "Both touch generation/cache logic, but they involve different behaviors and different cache settings." }, { "left": "issue:46209", "right": "issue:46431", "accept": false, - "reason": "FP8 B200 inference failure vs missing chat template on a processor; different failures." + "reason": "FP8 inference on B200 and missing chat template are unrelated problems." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation sync_gpus behavior vs OCR long-input index error; not the same issue." + "reason": "Generation kwarg update after sync_gpus is unrelated to long-OCR index overflow in Kosmos2.5." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence-transformers loading hang vs zero-probability output issue; different model behaviors." + "reason": "A loading hang and zero-probability outputs are different issues." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "DSA indexer bug vs docstring mismatch; not the same underlying change." + "reason": "MoE indexing bug vs docstring wording issue are not duplicates." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "Local dataset loading in a script vs LlamaConfig head_dim validation; unrelated." + "reason": "Dataset loading in a script is unrelated to LlamaConfig head_dim validation." }, { "left": "issue:46369", "right": "issue:46909", "accept": false, - "reason": "Memory overflow in sam3-litetext vs broken DIA documentation example; unrelated." + "reason": "Model memory overflow and a broken documentation example are different classes of problems." }, { "left": "issue:43475", "right": "issue:46305", "accept": false, - "reason": "Missing attribute in SAM3 video output vs Moonshine sliding-window off-by-one; different code paths." + "reason": "SAM3 missing attribute bug and Moonshine sliding-window off-by-one are different model paths." }, { "left": "issue:46305", "right": "issue:46655", "accept": false, - "reason": "Both are vision/video preprocessing bugs, but the off-by-one mask and padding calculation are distinct." + "reason": "Both are video-related, but the bugs are different: mask off-by-one vs padding calculation." }, { "left": "issue:42907", "right": "issue:46050", "accept": false, - "reason": "Dequantized save failure for Ministral/Devstral vs sinq quantized save/load breakage; different quantization flows." + "reason": "Both involve quantization/save-load, but they are different quantization flows and failure modes." }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "Same broad DSA/indexer area, but one is torch.split return handling and the other is a missing ReLU; not the same bug." + "reason": "Different DSA-related bugs: split return handling vs missing ReLU." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "FineGrainedFP8 accuracy issue vs Qwen3.5 MoE from_pretrained error; different root causes." + "reason": "FineGrainedFP8 accuracy issue is not the same as Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:22355", "right": "issue:27957", "accept": false, - "reason": "Missing transformers.onnx module vs XLMRoberta Flash Attention 2 support; unrelated." + "reason": "One is a missing transformers.onnx module import, the other is an XLMRoberta flash-attention question." } ] }, @@ -15434,20 +15452,20 @@ "issue:42907|issue:44945", "issue:30990|issue:33357" ], - "prepared_review_unit_hash": "0fea559a42d1633e82f088d70bc260e2c87cc90c877a9b6da1f67ae4f05d9c41", + "prepared_review_unit_hash": "81e67b80da83028743db2a9e6f062578d61974dc1fad01ef2d3909cbf98b6207", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 35924, + "serialized_chars": 35923, "estimated_input_tokens": 8981, "estimated_eval_tokens": 18218 }, @@ -15456,186 +15474,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated standalone issues rather than true duplicates. None of the soft pairs look like the same concrete bug/change, so I would not merge any of them as duplicates. The clearest standalone issue in the set is the tokenizer spacing bug (#28218), but it is not a canonical duplicate target for the rest.", - "confidence": 0.84, - "canonical_issue_reason": "Issue #28218 is the most self-contained, concrete bug report here: it has a specific tokenizer repro, active discussion, and inbound references. The rest are either broader regressions or unrelated model/version-specific failures.", + "summary": "The cluster is heterogeneous; the soft edges mostly connect superficially similar regression reports, but none are clearly the same underlying bug/change. I would not merge any of the proposed pairs.", + "confidence": 0.89, + "canonical_issue_reason": "If a representative issue is needed, 45278 is the broadest and most umbrella-like report, but the set is too mixed to treat it as a true canonical duplicate target.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #28218 is the best representative issue because it is narrowly scoped, reproducible, and actionable. The other issues are too heterogeneous to serve as a single canonical duplicate target.", + "best_issue_reason": "45278 is the best global representative because it captures the broad import-regression theme more than the others, even though it should not be used to collapse the whole cluster into one duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Both involve multimodal/model-loading regressions in v5, but they describe different concrete failures: remote-code meta-init incompatibility vs incorrect visual-encoder keys on save_pretrained." + "reason": "Both are multimodal/model-regression reports, but one is a meta-init incompatibility and the other is incorrect save_pretrained key handling." }, { "left": "issue:44279", "right": "issue:46328", "accept": false, - "reason": "Completely different problems: a generic dependency issue versus a SequenceFeatureExtractor.pad performance concern." + "reason": "A generic dependency issue is unrelated to a tokenizer padding performance regression." }, { "left": "issue:44483", "right": "issue:45406", "accept": false, - "reason": "Both mention serving/runtime failures, but one is a /v1/chat/completions request rejection and the other is a Gemma4Processor attribute crash. Different code paths." + "reason": "Serve request acceptance failure and a Gemma4Processor attribute crash are different code paths." }, { "left": "issue:44964", "right": "issue:47059", "accept": false, - "reason": "Unrelated failures: loading a Phi-4 multimodal model versus a CLI breakage in huggingface_hub." + "reason": "Model loading failure and CLI completion breakage are unrelated failures." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "Both are accuracy/regression-style reports, but they concern different models and different failure mechanisms." + "reason": "FP8 accuracy degradation and Qwen3-VL-Moe loading/behavior issues are not the same bug." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "Different bugs: flash_attention_3 causes degenerate training, while the other reports NaNs during checkpoint/init for classification heads." + "reason": "Flash-attention training degeneration and NaNs after checkout/init have different symptoms and likely different roots." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but for different missing symbols/classes; no evidence they share the same underlying broken export path." + "reason": "Both are import errors, but they involve different missing symbols and different model areas." }, { "left": "issue:45381", "right": "issue:46305", "accept": false, - "reason": "Both are off-by-one-style complaints, but one is about Qwen2.5-VL video position ids and the other is MoonshineStreaming sliding-window masking. Different subsystems." + "reason": "A wrong video position-id computation and an off-by-one sliding window mask are different tensor bugs." }, { "left": "issue:43408", "right": "issue:46402", "accept": false, - "reason": "Same model family (Sam3) but different issues: model-type mismatch warning versus missing text encoder weights on load." + "reason": "A model-type warning and missing text-encoder weights are not the same failure." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "Different failure modes: missing rms_norm_eps precision mismatch versus remote-code/meta-init incompatibility." + "reason": "Layernorm epsilon precision drift is unrelated to remote-code checkpoint meta-initialization incompatibility." }, { "left": "issue:43644", "right": "issue:46305", "accept": false, - "reason": "A broad Transformers v5 buffer-initialization regression versus a specific streaming mask off-by-one bug." + "reason": "Non-persistent buffer initialization junk and streaming mask off-by-one are unrelated." }, { "left": "issue:39379", "right": "issue:46172", "accept": false, - "reason": "Different model internals and different bugs: incorrect mllama mask handling versus a DeepSeek-V4 tensor operation issue." + "reason": "These affect different model internals: mllama mask handling vs a DeepSeek tensor transpose issue." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "Both are tokenizer-related, but one is an import failure for transformers.onnx and the other is an added-token spacing bug." + "reason": "Missing an ONNX module and adding an extra space after added tokens are unrelated tokenizer/export problems." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "Both are Qwen-related quality regressions, but one is a FineGrainedFP8 accuracy issue and the other is an infinite loop for a specific image." + "reason": "FP8 accuracy loss and a model looping on a specific image are different underlying bugs." }, { "left": "issue:44327", "right": "issue:47059", "accept": false, - "reason": "QA top-k bounds crash versus a CLI attribute error; no shared bug path." + "reason": "QA top-k bounds checking and a hub CLI completion attribute error do not share a concrete code path." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer behavior regressions, but they affect different tokenizers and different output properties (offset mappings vs BOS/EOS insertion)." + "reason": "Wrong offset mapping and missing BOS/EOS insertion are distinct tokenizer regressions." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "The broader import-error report may mention dependency breakage, but it is not the same concrete bug as a specific missing is_offline_mode import." + "reason": "Both mention import failures, but one is a specific missing symbol while the other is a broad post-update report; not enough evidence they are the same bug." }, { "left": "issue:43408", "right": "issue:46369", "accept": false, - "reason": "Same Sam3 family, but a warning about model type mismatch is not the same as memory overflow during sam3-litetext execution." + "reason": "A model-type warning and a memory overflow while calling the model are unrelated." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Torch.compile attention-kernel divergence is unrelated to a docstring issue." + "reason": "Attention-kernel branching under torch.compile and a docstring typo are not duplicates." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Both concern saving/loading, but one is offline cache loading for tokenizers and the other is quantized model save failure." + "reason": "Offline tokenizer cache loading and quantized model save failure are different subsystems." }, { "left": "issue:27957", "right": "issue:30990", "accept": false, - "reason": "Both mention loading-related issues, but one is XLMRoberta Flash Attention 2 and the other is Sentence Transformers hanging while loading." + "reason": "XLM-R Flash Attention 2 support and a sentence-transformers loading hang are not the same issue." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline tokenizer cache loading and wrong tokenizer decoder type in v5 are distinct bugs." + "reason": "Offline cache loading and wrong tokenizer decoder type are separate tokenizer regressions." }, { "left": "issue:42796", "right": "issue:43122", "accept": false, - "reason": "Both are tokenizer regressions, but one is a specific BPE-marker handling bug and the other is a broader tokenization difference across versions." + "reason": "Tokenizer BPE-marker handling and cross-version tokenization differences are related only broadly, not the same concrete bug." }, { "left": "issue:45381", "right": "issue:46909", "accept": false, - "reason": "Unrelated: Qwen2.5-VL video position-id bug versus a broken DIA documentation example." + "reason": "A model video-input bug and a broken documentation example are unrelated." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation kwargs synchronization and D-FINE auxiliary-loss behavior are unrelated code paths." + "reason": "Generation kwargs handling after sync_gpus and D-FINE auxiliary-loss behavior are different changes." }, { "left": "issue:43540", "right": "issue:46909", "accept": false, - "reason": "A memory overflow in sam3-litetext is unrelated to a documentation example bug." + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to a docs example bug." }, { "left": "issue:43646", "right": "issue:46002", "accept": false, - "reason": "Custom model initialization breakage in v5 is not the same as NaNs appearing in classification heads after checkout/init." + "reason": "Custom model initialization breakage and NaNs in classification heads do not look like the same code-path problem." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Quantized-model save failure versus incorrect output under pipeline parallelism; different symptoms and layers." + "reason": "Quantized-model save failures and pipeline-parallel wrong outputs are different failures." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Both are loading/runtime failures, but one is a hang and the other is a macOS bus error with a pretrained community CLIP model." + "reason": "A loading hang and a macOS bus error on a different model/version are not the same underlying bug." } ] }, @@ -15719,8 +15737,8 @@ "issue:42371|issue:44263", "issue:44117|issue:44704", "issue:42907|issue:43937", - "issue:39401|issue:44488", "issue:43784|issue:45600", + "issue:39401|issue:44488", "issue:39401|issue:44521", "issue:44625|issue:44829", "issue:44704|issue:45290", @@ -15741,214 +15759,220 @@ "issue:43122|issue:44779", "issue:42757|issue:44351" ], - "prepared_review_unit_hash": "58945f128fed8afc7385099bd8637adbe8b814633ed5be161c8d5b13142b309a", + "prepared_review_unit_hash": "03151fc76614f2c04ee5de94b1fca07c73b619a506aa5dadc732be8d890d7093", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36824, - "estimated_input_tokens": 9206, - "estimated_eval_tokens": 18668 + "serialized_chars": 36827, + "estimated_input_tokens": 9207, + "estimated_eval_tokens": 18670 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a heterogeneous mix of unrelated transformers issues; none of the soft edges look like true duplicates on titles alone, so I would not merge any pair. The broadest issue is the generic import-regression report (45278), but it is only an umbrella/meta ticket rather than the same bug as the more specific import failures.", - "confidence": 0.82, - "canonical_issue_reason": "Issue 45278 is the broadest, most umbrella-style report in the set ('many import errors after update from 4.57.0 to 5.5.0') and is the most plausible anchor if the import-related tickets were being organized, but it is not a duplicate of the more specific failures.", + "summary": "This cluster is noisy: most items are unrelated tokenizer, multimodal, or model-specific bugs that only share vague wording. The closest common thread is a set of import/export regressions in Transformers v5, so the broad import-regression report is the best representative, but most soft links should be rejected.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45278 is the broadest representative of the import-regression subset and can subsume several specific \"cannot import\" reports, whereas the others are narrower symptom reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45278 is the best overall issue candidate because it is the most general regression report and can subsume the broad import-error theme better than the narrowly scoped model/tokenizer bugs.", + "best_issue_reason": "It is the most general user-facing report in the cluster and best captures the shared failure mode of top-level import breakage after upgrade.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44295", "right": "issue:44297", "accept": false, - "reason": "Different subsystems: one is a position_ids buffer/readback bug, the other is tokenizer_class metadata written by save_pretrained." + "reason": "Different bugs: buffer access error vs tokenizer_config mismatch." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Different model/loading paths: dequantized Ministrals/Devstrals save failure vs loading Phi-4 multimodal instruct." + "reason": "Different model/save/load paths and different failure modes." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "One is the transformers env CLI, the other is an import-time NameError in sentence-transformers integration." + "reason": "Both are import/runtime regressions, but they concern different symbols and code paths." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Unrelated problems: Python version compatibility vs incorrect outputs under pipeline parallelism." + "reason": "Unrelated issues: Python compatibility vs pipeline-parallel output correctness." }, { "left": "issue:43329", "right": "issue:47059", "accept": false, - "reason": "Different failures entirely: multimodal token counting video-path bug vs a CLI/huggingface_hub incompatibility." + "reason": "Multimodal token counting bug vs CLI completion breakage from huggingface_hub." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "The latter is a broad import-regression umbrella, but this specific AutoImageProcessor import error is a distinct concrete failure." + "reason": "Related by broad import breakage, but 45278 is an umbrella report and not the same concrete missing symbol." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Different model-specific logic: RoPE layer dropping vs an incorrect tied-weights warning in Qwen3-Next." + "reason": "Different model-specific architectural bugs, not the same code-path." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 API guidance and a torch.split return-value issue in a DSA indexer are unrelated." + "reason": "TF32 configuration guidance vs a DSA indexer tensor-splitting issue." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES/ from_pretrained fallback logic is a different code path from AutoProcessor argument forwarding." + "reason": "Both touch from_pretrained-related code, but the defects are different and not mergeable into one fix." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Different bugs: save/dequantization failure vs invalid GenerationConfig for GLM-5." + "reason": "Different GLM-5 generation/configuration failures." }, { - "left": "issue:39401", - "right": "issue:44488", + "left": "issue:43784", + "right": "issue:45600", "accept": false, - "reason": "Qwen3 offset_mapping corruption is unrelated to loading cjvt/sleng-bert failing in the current version." + "reason": "Importing sentence-transformers failing on nn vs stale auto_mappings references." }, { - "left": "issue:43784", - "right": "issue:45600", + "left": "issue:39401", + "right": "issue:44488", "accept": false, - "reason": "Importing sentence-transformers with latest transformers and auto_mappings.py CI breakage are different import-time issues." + "reason": "Different tokenizer/model issues for different models." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Tokenizer offset_mapping bug and apply_chat_template assistant_mask bug are separate tokenizer features." + "reason": "Tokenizer offset mapping bug vs assistant mask generation for multimodal chat templates." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Qwen3.5 num_labels propagation and flash_attention_3 training degeneration are unrelated model/training issues." + "reason": "Config propagation bug vs training instability under flash_attention_3." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Passing kwargs to cached_file and crashing on assistant tool-call messages are different apply_chat_template/loader behaviors." + "reason": "cached_file kwargs forwarding vs chat template crash on tool-call messages." }, { "left": "issue:46172", "right": "issue:46762", "accept": false, - "reason": "Deepseek-v4 transpose/type issue and Minimax attention numerical mismatch are different model implementations." + "reason": "Different model implementations and different numerical discrepancy reports." }, { "left": "issue:43122", "right": "issue:46326", "accept": false, - "reason": "Tokenizer version drift vs continue_final_message on mistral_common are distinct tokenizer behaviors." + "reason": "Both tokenizer-related, but one is a version regression and the other is a mistral_common chat-template behavior bug." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "Wrong model type warning for sam3_tracker and missing fpn_position_embeddings are separate SAM3 video bugs." + "reason": "Same model family, but a config warning is not the same bug as a missing output attribute." }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "A missing ReLU in the DSA indexer and GLM-5 RoPE implementation are unrelated algorithmic issues." + "reason": "Completely different model/algorithm bugs." }, { "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "Different tokenizer features: general tokenization drift vs multimodal assistant_mask generation." + "reason": "Tokenization regression vs multimodal assistant mask handling." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe loading and tokenizer save_pretrained metadata mismatch are not the same failure." + "reason": "Qwen3-VL-Moe load bug vs tokenizer_class metadata mismatch." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Dequantized model save failure is separate from gguf_file breaking from_pretrained loading." + "reason": "Dequantized model save failure vs gguf_file loading breakage." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "Module import failure for transformers.onnx and slow/fast tokenizer inconsistency are unrelated issues." + "reason": "Different modules and different tokenizer behavior problems." }, { "left": "issue:44038", - "right": "issue:44222", + "right": "issue:44295", "accept": false, - "reason": "Qwen3-VL-Moe load failure and FP8 save_pretrained for moe are different code paths." + "reason": "Different subsystems: model loading vs buffer read/write." }, { "left": "issue:45216", "right": "issue:45478", "accept": false, - "reason": "Both mention Qwen3.5 regressions, but one is save_pretrained checkpoint corruption and the other is from_pretrained load failure; too distinct to merge." + "reason": "Related Qwen3.5 family, but one is save_pretrained regression and the other is from_pretrained loading error." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Different bugs: model load failure vs FP8 save_pretrained issue." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Different missing symbols: Qwen2_5_VLForConditionalGeneration vs HybridCache import errors." + "reason": "Both are import errors, but for different symbols and likely different fixes." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Example script empty multi-label outputs and Gemma 4 mm_token_type_ids defaults are unrelated training/inference issues." + "reason": "Example script multi-label output bug vs Gemma 4 mm_token_type_ids defaulting." }, { "left": "issue:39401", "right": "issue:46614", "accept": false, - "reason": "Qwen3 tokenizer offset_mapping bug and stale phonemizer language state in Wav2Vec2PhonemeCTCTokenizer are unrelated." + "reason": "Different tokenizer implementations and different statefulness bugs." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and symptoms; not the same underlying bug." + "reason": "Different tokenization regressions affecting different models/paths." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they are for different missing names from different packages and do not share a concrete code path." + "reason": "Both are import failures, but one is huggingface_hub and the other is transformers; not the same underlying bug." } ] }, @@ -16028,8 +16052,8 @@ "issue:42175|issue:44938", "issue:39379|issue:43232", "issue:43493|issue:45753", - "issue:44188|issue:46328", "issue:46231|issue:46877", + "issue:44188|issue:46328", "issue:38617|issue:45278", "issue:43937|issue:44485", "issue:42898|issue:46614", @@ -16051,20 +16075,20 @@ "issue:46328|issue:47116", "issue:43116|issue:43825" ], - "prepared_review_unit_hash": "245fcca9e783e02207406461d517c85cc67aaab174d6b37ca845264122b46b92", + "prepared_review_unit_hash": "bc57161992e5b4c4af5698861416fd88103d0004c50d46d5ee06f42c2a14a775", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35731, + "serialized_chars": 35732, "estimated_input_tokens": 8933, "estimated_eval_tokens": 18122 }, @@ -16073,180 +16097,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab-bag of mostly unrelated Transformers issues across different models, versions, and failure modes. None of the soft pairs look like true duplicates or mergeable same-bug reports.", - "confidence": 0.82, - "canonical_issue_reason": "Issue 45278 is the broadest umbrella-style report in the set ('many import errors after update from 4.57.0 to 5.5.0'), so it is the closest thing to a canonical issue among otherwise unrelated reports.", + "summary": "This cluster is highly heterogeneous: the paired items are mostly unrelated regressions, import errors, model-specific bugs, docs issues, and pipeline problems. I found no soft pair that looks like the same underlying bug/change, so no duplicate merge is warranted.", + "confidence": 0.94, + "canonical_issue_reason": "issue:42371 is the broadest and most discussed item in the set, but the cluster is not a true duplicate group; it is only a weak stand-in representative.", "canonical_pr_reason": null, - "best_issue_reason": "45278 is the most general and cluster-like issue here; if one issue must stand in for the set, it best represents the broad regression/import-error theme.", + "best_issue_reason": "issue:42371 has the highest discussion activity and the most general framing among the issues, making it the best representative if one must be chosen.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor parallel training bug vs multi-label classification example failure; different code paths and symptoms." + "reason": "Different bugs: tensor-parallel training behavior vs. multi-label example script returning empty results." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "Embedding dtype drift between train/eval is unrelated to incorrect output under pipeline parallelism." + "reason": "Both involve training/eval behavior, but one is a dtype regression and the other is incorrect output under pipeline parallelism." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "Missing transformers.onnx module and a LayoutLMv3 error-message clarity issue are unrelated." + "reason": "Unrelated: missing transformers.onnx module vs. LayoutLMv3 error-message clarity." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "Backend/installation packaging problem vs Python 3.14 load failure; different compatibility issues." + "reason": "Different failures: pip extra backend composition vs. Python 3.14 import/load breakage." }, { "left": "issue:39379", "right": "issue:43232", "accept": false, - "reason": "mllama mask handling bug and generation kwargs after sync_gpus are different generation-path bugs." + "reason": "mllama mask handling bug and generation kwargs update after sync_gpus are different code paths." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 HF/JAX discrepancy and Qwen3_5 infinite loop are different models and different behaviors." + "reason": "Different model-specific regressions: SigLIP2 implementation mismatch vs. Qwen3_5 image-specific infinite loop." }, { - "left": "issue:44188", - "right": "issue:46328", + "left": "issue:46231", + "right": "issue:46877", "accept": false, - "reason": "torch.compile attention-kernel branching vs SequenceFeatureExtractor.pad performance; unrelated subsystems." + "reason": "Docs-only mismatch versus import-time stdout logging from auto_docstring; not the same issue." }, { - "left": "issue:46231", - "right": "issue:46877", + "left": "issue:44188", + "right": "issue:46328", "accept": false, - "reason": "Documentation parameter mismatch is not the same as import-time doc-lint stdout logging." + "reason": "Attention-kernel divergence under torch.compile is unrelated to SequenceFeatureExtractor padding inefficiency." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "Both are import-related, but one is a specific missing symbol while the other is a broad upgrade regression; not the same concrete bug." + "reason": "Both are import-related, but one is a specific missing symbol and the other is a broad post-upgrade import breakage." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Invalid GenerationConfig for GLM-5 is distinct from the GLM-5 RoPE implementation discussion." + "reason": "GenerationConfig validation error and GLM-5 RoPE discussion do not describe the same bug." }, { "left": "issue:42898", "right": "issue:46614", "accept": false, - "reason": "Tokenizer behavior change in v5 and stale phonemizer backend state are different tokenizer issues." + "reason": "Tokenization cleanup behavior change and stale espeak backend state are separate tokenizer issues." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version incompatibility and an incorrect PyTorch version check are separate compatibility bugs." + "reason": "Version-support problem vs. incorrect PyTorch version gate for AuxRequest import; different fixes needed." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "base_model_prefix loading regression and NaNs in classification heads are unrelated failures." + "reason": "Model loading with base_model_prefix is unrelated to NaNs in classification heads after checkout/init." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Generic Qwen3-VL-Moe bug vs specific Qwen3.5 Moe from_pretrained error; not the same change or fix." + "reason": "Both mention Qwen3.5/Qwen3-VL-Moe, but they are different regressions in different versions and code paths." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Both concern tokenizers in v5, but they affect different models and different concrete defects." + "reason": "Tokenizer regressions in v5, but for different models and different concrete failures." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in multi-threaded dataloaders is unrelated to bfloat16 dtype mismatches in inference." + "reason": "VRAM leak in multithreaded Qwen3ForCausalLM use is unrelated to dtype mismatches in inference tests." }, { "left": "issue:44355", "right": "issue:46231", "accept": false, - "reason": "Compiled Python file errors and a documentation parameter typo do not share the same underlying bug." + "reason": "Compiled Python file errors are unrelated to a documentation parameter-name typo." }, { "left": "issue:44625", "right": "issue:46002", "accept": false, - "reason": "Qwen3.5 num_labels propagation and NaNs in classification heads are different model/config issues." + "reason": "Qwen3.5 num_labels propagation and classification-head NaNs are separate bugs." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "save_pretrained key regression and from_pretrained load failure are related model family but different concrete bugs." + "reason": "Both are Qwen3.5-related, but one is save_pretrained key regression and the other is from_pretrained loading failure." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Multi-label classification script output issue is unrelated to SigLIP2 implementation discrepancy." + "reason": "Different scopes: example script output handling versus model implementation discrepancy." }, { "left": "issue:44610", "right": "issue:46909", "accept": false, - "reason": "OmDet-Turbo processor size mismatch is unrelated to a broken documentation example." + "reason": "Processor input-size mismatch is unrelated to a broken documentation example." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "qwen3_omni_moe feature-length helper bug is unrelated to the AuxRequest PyTorch version check." + "reason": "Feature-length helper behavior and PyTorch version checking are not the same underlying change." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API guidance and a missing image-patch helper in GLM46V are unrelated." + "reason": "TF32 API guidance and missing image-patch method in GLM46V are unrelated." }, { "left": "issue:46726", "right": "issue:46981", "accept": false, - "reason": "Missing vision-backend messaging and HTTP 429 retry behavior are both AutoProcessor-related but not the same bug." + "reason": "Both involve AutoProcessor, but one is missing vision backend messaging and the other is retry behavior on HTTP 429." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "SGD args being ignored is a different problem from trainer.evaluate() failing after trainer.train()." + "reason": "Optimizer argument handling is unrelated to trainer.evaluate failing after train." }, { "left": "issue:43618", "right": "issue:46829", "accept": false, - "reason": "Missing attentions in CLIPOutput and wrong hidden_states in VideoPrism output are different model-output bugs." + "reason": "Both are output-field issues, but they affect different model classes and different attributes." }, { "left": "issue:46328", "right": "issue:47116", "accept": false, - "reason": "SequenceFeatureExtractor.pad performance and streaming pipeline regression are unrelated." + "reason": "Padding performance and streaming pipeline regression are separate concerns." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Classification example returning empty results is unrelated to pipeline translation-task error messaging." + "reason": "Example-script multi-label behavior is not the same as pipeline translation-task support messaging." } ] }, @@ -16350,20 +16374,20 @@ "issue:44704|issue:47110", "issue:45071|issue:47059" ], - "prepared_review_unit_hash": "9b8d6291aab26bbc5b38574fb938e42f3b6639c134e3d88e3568fd5a9d3d81ef", + "prepared_review_unit_hash": "435b082902a2d580ca484b02ad6f27f6d0794adf72ab05dce4702607f234ec93", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36166, + "serialized_chars": 36168, "estimated_input_tokens": 9042, "estimated_eval_tokens": 18340 }, @@ -16372,198 +16396,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous Transformer bugs rather than true duplicates. None of the soft pairs look like the same underlying issue; the most representative standalone report is the tokenizer/import regression in issue 43824.", - "confidence": 0.68, - "canonical_issue_reason": "issue:43824 is the clearest, most actionable regression report in the set and has the strongest discussion/inbound-references signal, so it is the best single issue to anchor this cluster even though the cluster itself is not a duplicate set.", + "summary": "No soft pair looks like a true duplicate; most are only loosely related by broad themes like tokenizer regressions, FlashAttention, or v5 behavior changes. I\u2019d use issue:44779 as the best representative issue because it has the clearest concrete regression report, but I would not merge any of these pairs.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44779 is the strongest standalone bug report here: it describes a concrete tokenizer regression in v5 with a specific model and clear behavioral change, making it the best representative issue for this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43824 has a concrete failure mode, clear user impact, and enough discussion to make it the strongest issue for triage and follow-up.", + "best_issue_reason": "issue:44779 has the most actionable, model-specific tokenizer regression description and is the best single issue to anchor related triage.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Different code paths: missing encoder weight loading vs AutoProcessor kwarg forwarding." + "reason": "Different bugs: one is missing encoder weights during UMT5 loading, the other is AutoProcessor forwarding kwargs incorrectly." }, { "left": "issue:43066", "right": "issue:46710", "accept": false, - "reason": "Tokenizer decoder-type regression is unrelated to a DeepSeek accuracy regression after upgrade." + "reason": "Both are v5-related, but one is a tokenizer decoder-type issue and the other is an accuracy regression in a model; not the same code-path bug." }, { "left": "issue:28348", "right": "issue:46640", "accept": false, - "reason": "One is a GPT2 FlashAttention-2 feature request; the other is a generic flash-attention bug report." + "reason": "FlashAttention is the only overlap; one is a feature request for GPT2 support and the other is an unspecified FA2 issue." }, { "left": "issue:42673", "right": "issue:42796", "accept": false, - "reason": "VRAM leak in multi-threaded causal LM use is unrelated to tokenizer BPE-marker handling." + "reason": "Unrelated model areas: VRAM leak in Qwen3ForCausalLM versus tokenizer BPE marker handling in Ministral." }, { "left": "issue:44995", "right": "issue:46697", "accept": false, - "reason": "GlmMoeDsa stale indexer cache and Swin2SR image over-padding are different subsystems and bugs." + "reason": "Different subsystems entirely: GLM MoE cache crash versus Swin2SR image processor padding." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Docs mentioning pipeline() after removals is unrelated to a PyTorch version check in flex_attention." + "reason": "Docs still mentioning pipeline() is unrelated to a PyTorch version check bug in flex_attention." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Both are tokenizer problems, but one is a loading regression while the other is a wrong offset_mapping bug in Qwen3." + "reason": "Both are tokenizer regressions, but one is a generic loading regression and the other is a Qwen3 offset_mapping bug; not the same issue." }, { "left": "issue:45092", "right": "issue:46762", "accept": false, - "reason": "Meta-initialization incompatibility for remote-code checkpoints is unrelated to Minimax attention numerical drift." + "reason": "Meta-init incompatibility for old InternVL2 checkpoints is not the same as numerical differences in Minimax MSA attention." }, { "left": "issue:44263", "right": "issue:46762", "accept": false, - "reason": "torch.split return behavior in GlmMoeDsaIndexer does not match an attention implementation accuracy issue." + "reason": "A torch.split return-value bug in GlmMoeDsaIndexer is unrelated to Minimax attention numerical differences." }, { "left": "issue:43824", "right": "issue:46981", "accept": false, - "reason": "Importing a missing class is a different failure mode from retry behavior on HTTP 429 during config downloads." + "reason": "ImportError for a missing class and repeated HTTP 429 retries are different failure modes in different layers." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "Optimizer-state corruption under FSDP2/PEFT is unrelated to generation cache handling in chunked prefill." + "reason": "Optimizer-state corruption in Trainer/FSDP2/PEFT is unrelated to missing GenerationMixin cache during chunked prefill." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "VitPose model breakage is a different model-family issue than Ministral FlashAttention failures." + "reason": "Vitpose breakage and Ministral FlashAttention failure are separate model-specific bugs." }, { "left": "issue:43673", "right": "issue:46328", "accept": false, - "reason": "Generation cache missing is unrelated to a padding performance inefficiency in SequenceFeatureExtractor." + "reason": "Generation cache behavior and SequenceFeatureExtractor padding performance are unrelated." }, { "left": "issue:39379", "right": "issue:43688", "accept": false, - "reason": "mllama mask handling and auxiliary-loss normalization are distinct training/math bugs in different models." + "reason": "mllama mask handling and OLMoE/GPT Oss auxiliary-loss normalization affect different code paths." }, { "left": "issue:42222", "right": "issue:47000", "accept": false, - "reason": "VitPose breakage and RTDetr loss infeasibility are unrelated model-specific failures." + "reason": "Vitpose model breakage is unrelated to RTDetrLoss infeasible cost-matrix errors." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "A bumped PEFT minimum version and an AuxRequest import version gate are separate dependency/versioning issues." + "reason": "A dependency-version bump issue is unrelated to an incorrect PyTorch version check in flex_attention." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "Invalid GenerationConfig validation and a stale warning for num_return_sequences are different generation-path bugs." + "reason": "Invalid GenerationConfig and a stale paged-generate warning are different generation issues." }, { "left": "issue:42222", "right": "issue:46752", "accept": false, - "reason": "VitPose breakage is unrelated to apply_chat_template crashing on an empty list." + "reason": "Vitpose breakage is unrelated to apply_chat_template crashing on an empty list." }, { "left": "issue:44380", "right": "issue:46077", "accept": false, - "reason": "GPT2 attention-scaling being ignored and Gemma4 bidirectional attention mask construction are different attention bugs." + "reason": "Both touch attention, but GPT2 attention scaling under SDPA/FlashAttention is not the same bug as Gemma4 building causal masks when bidirectional attention is requested." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "Unexpected _is_hf_initialized argument handling is a different incompatibility than PretrainedConfig type checking." + "reason": "One is an unexpected init_empty_weights argument, the other is a PretrainedConfig type-checking break; related to v5, but not the same defect." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer-mapping None handling is unrelated to a flash_attention AttributeError on s_aux=None." + "reason": "Tokenizer-mapping None handling and a flash_attention crash on s_aux=None are unrelated." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "lm_head.weight serialization differences and NaNs in classification heads are not the same bug." + "reason": "Serialization of tied weights versus NaNs after checkout/init are different problems." }, { "left": "issue:39379", "right": "issue:46897", "accept": false, - "reason": "The mllama mask issue is not the same as the Florence2 double-shift training-loss bug." + "reason": "Both mention training-loss style bugs, but the affected models and failure mechanisms differ." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and failure modes; not one concrete bug." + "reason": "Both are tokenizer regressions, but one is about offset_mapping and the other about tokenization correctness on DeepSeek; they do not look like the same concrete bug." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Missing qkv/gate projections in layer 0 is a model-architecture issue, not the same as a Qwen3.5 Moe load error." + "reason": "Missing qkv/gate projections in qwen3next and a Qwen3.5 MoE from_pretrained error are not the same code-path." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Batched video-processor shape handling and missing torchvision crash are distinct video-processing problems." - }, - { - "left": "issue:45071", - "right": "issue:47059", - "accept": false, - "reason": "PretrainedConfig type checking and a cli completion attribute error in huggingface_hub are unrelated." + "reason": "Video processor output-shape bugs and AutoProcessor import/crash behavior without torchvision are distinct." }, { "left": "issue:44704", "right": "issue:47110", "accept": false, - "reason": "AutoProcessor cached_file kwargs forwarding is unrelated to special_tokens being a list instead of a mapping." + "reason": "Keyword forwarding in AutoProcessor and a special_tokens list-vs-dict type error are different bugs." }, { - "left": "issue:43937", - "right": "issue:44977", + "left": "issue:45071", + "right": "issue:47059", "accept": false, - "reason": "GLM-5 GenerationConfig validation and DeepSeek tokenizer output drift are different failure classes." + "reason": "PretrainedConfig type checking and CLI completion breakage from huggingface_hub are unrelated." }, { - "left": "issue:44704", - "right": "issue:47110", + "left": "issue:43066", + "right": "issue:46762", "accept": false, - "reason": "AutoProcessor download-argument forwarding and special_tokens shape/type handling do not share a code path." + "reason": "A tokenizer/decoder-type issue is unrelated to Minimax attention numerical discrepancies." }, { - "left": "issue:45071", - "right": "issue:47059", + "left": "issue:44977", + "right": "issue:46697", "accept": false, - "reason": "PretrainedConfig typing and huggingface_hub CLI completion compatibility are different issues." + "reason": "DeepSeek tokenizer regression and Swin2SR padding behavior are unrelated." } ] }, @@ -16665,20 +16683,20 @@ "issue:43208|issue:46894", "issue:46752|issue:47116" ], - "prepared_review_unit_hash": "7bf453f6e79a294586b7a54c49bf15ab2d337a10fb76df7a0fdee2ff6e6d0e26", + "prepared_review_unit_hash": "7b6eb736882329f1441df19f0437f645a7f5997d89fcbc4943943bc74e6102ae", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35434, + "serialized_chars": 35435, "estimated_input_tokens": 8859, "estimated_eval_tokens": 17974 }, @@ -16687,174 +16705,174 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster does not look like a duplicate set: it mixes import errors, tokenizer regressions, model-specific config bugs, training issues, and test-utils problems. I found no soft-edge pair that safely collapses to one underlying bug/change.", - "confidence": 0.92, - "canonical_issue_reason": null, + "summary": "This set is highly heterogeneous: most items are unrelated bugs/regressions across tokenizer, model loading, config, optimizer, and testing utilities. I found no soft-edge pair that is clearly the same underlying issue; all candidate pairs are too broad or target different code paths.", + "confidence": 0.93, + "canonical_issue_reason": "No single issue cleanly represents the whole set because it mixes many unrelated bugs. If one must be chosen as the best canonical anchor, issue 43066 is the most concrete and self-contained regression report.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the broadest umbrella-like report in the set ('many import errors after update from 4.57.0 to 5.5.0'), but even it does not explain the rest of the heterogeneous issues, so it is only the nearest cluster representative, not a true canonical duplicate target.", + "best_issue_reason": "Issue 43066 is the strongest triage anchor: it describes a specific, user-visible tokenizer regression with a clear symptom and likely narrow code path.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45341", "right": "issue:46328", "accept": false, - "reason": "A small testing_utils bug and a SequenceFeatureExtractor.pad performance complaint are unrelated code paths." + "reason": "One is a testing_utils bug; the other is a performance issue in SequenceFeatureExtractor.pad. Different modules and failure modes." }, { "left": "issue:45071", "right": "issue:46512", "accept": false, - "reason": "Type-checking breakage in PreTrainedConfig is separate from from_config(dtype=...) leaving config.dtype stale." + "reason": "PretrainedConfig type-checking regression vs stale config.dtype after from_config; related to config handling but not the same bug." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "Flash Attention 3 degenerate training and Qwen3.5 save_pretrained checkpoint corruption are different bugs in different flows." + "reason": "Flash-attention training degeneration and Qwen3.5 save_pretrained regression are unrelated code paths." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Both involve testing helpers, but one is a docs/utility bug and the other is a xdist race on captured_info.txt." + "reason": "A testing_utils bug versus a pytest-xdist race condition; different symptoms and fixes." }, { "left": "issue:45632", "right": "issue:46981", "accept": false, - "reason": "Local trust_remote_code cache path collisions and AutoProcessor HTTP 429 retry behavior are distinct failure modes." + "reason": "Cache-path collision for trust_remote_code is unrelated to repeated HTTP 429 retries in AutoProcessor.from_pretrained." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "A sam3_video/sam3_tracker model-type warning is not the same as a MobileLLM config key error." + "reason": "SAM3 model-type warning and MobileLLM key error are different model-loading problems." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "3d_parallel.py startup failure and a testing_utils bug do not share the same underlying cause." + "reason": "3d_parallel.py execution failure and a small testing_utils bug do not share the same underlying defect." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES returning None and _patch_mistral_regex crashing on backend_tokenizer are different tokenizer-loading paths." + "reason": "Tokenizer mapping None-handling and Mistral regex patch backend_tokenizer access are distinct tokenizer issues." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they involve different missing symbols and different root causes." + "reason": "Both are import errors, but for different symbols and likely different compatibility breaks." }, { "left": "issue:43122", "right": "issue:46298", "accept": false, - "reason": "Tokenization output changes across v5 and a noisy logging message are unrelated." + "reason": "Tokenizer output regression and noisy logging are unrelated." }, { "left": "issue:42222", "right": "issue:46178", "accept": false, - "reason": "Broken vitpose models and a BACKENDS_MAPPING tensorflow_text error are separate model/backend issues." + "reason": "Broken vitpose models and unsupported tensorflow_text backend mapping are separate failures." }, { "left": "issue:39401", "right": "issue:46209", "accept": false, - "reason": "Qwen3 offset_mapping corruption and FP8 inference failure on B200 are unrelated." + "reason": "Qwen3 offset_mapping bug and FP8 inference failure on B200 are unrelated." }, { "left": "issue:44877", "right": "issue:46231", "accept": false, - "reason": "Strict config loading for granite_speech is not the same as a set_training() documentation parameter mismatch." + "reason": "Strict config loading for granite_speech and a docs/parameter-name mismatch are different issues." }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 API behavior guidance and a missing rms_norm_eps in MLA q_a_layernorm are different problems." + "reason": "TF32 API-settings guidance and MLA rms_norm_eps precision behavior are unrelated." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "save_pretrained shard-size concerns and a Qwen3.5 MoE from_pretrained error do not share the same code path." + "reason": "save_pretrained shard-size question and Qwen3.5 Moe from_pretrained error are different concerns." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "An invalid GenerationConfig and an unrecognized model_type in config.json are distinct config-loading failures." + "reason": "GenerationConfig validation and unrecognized model/config.json loading are separate code paths." }, { "left": "issue:44933", "right": "issue:45278", "accept": false, - "reason": "A specific nonexistent image_utils import is not the same bug as broad import failures after upgrading to 5.5.0." + "reason": "A missing image_utils import and broad import errors after upgrade are too different to treat as the same bug." }, { "left": "issue:44261", "right": "issue:46762", "accept": false, - "reason": "Both mention numerical differences, but they concern different models/implementations and likely different defects." + "reason": "Both mention numerical behavior, but they affect different attention implementations and likely different fixes." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "Importing layer_type_validation and _torch_extract_fbank_features are unrelated errors." + "reason": "ImportError for layer_type_validation is unrelated to _torch_extract_fbank_features()." }, { "left": "issue:42371", "right": "issue:46617", "accept": false, - "reason": "TF32 settings guidance and a trust_remote_code cache FileNotFoundError are separate issues." + "reason": "TF32 behavior settings and trust_remote_code cache FileNotFoundError are unrelated." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Missing Qwen3next projections and a Qwen3.5 MoE from_pretrained error are not the same underlying bug." + "reason": "Missing qwen3next projections and Qwen3.5 Moe from_pretrained error are different model-definition/loading bugs." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "SGD arguments being ignored and optimizer.step() being called twice under deepspeed are different optimizer bugs." + "reason": "SGD args ignored and optimizer.step() called twice with deepspeed are optimizer-related but not the same concrete defect." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Removal of tie_embeddings_and_encoder_decoder and unexpected position-id keys are different regression surfaces." + "reason": "Removed tie_embeddings_and_encoder_decoder and unexpected position-id keys are distinct regression reports." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Wrong tokenizer decoder type and tokenization output differences across versions may both involve tokenizers, but not the same concrete defect." + "reason": "Both are tokenizer v5 regressions, but one is decoder-type mismatch and the other is changed tokenization output; not clearly the same code-path bug." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "A torch.split return-value bug in GlmMoeDsaIndexer is separate from GLM-5 RoPE implementation concerns." + "reason": "torch.split return-value issue and GLM-5 RoPE implementation problem are different bugs." }, { "left": "issue:43208", "right": "issue:46894", "accept": false, - "reason": "xLSTM training blockers and a CohereASR double-shift loss bug are unrelated model-specific training issues." + "reason": "xLSTM training blockers and CohereASR double-shift loss bug are unrelated model-specific issues." }, { "left": "issue:46752", "right": "issue:47116", "accept": false, - "reason": "apply_chat_template crashing on empty input and a streaming pipeline regression are different code paths." + "reason": "Empty-list apply_chat_template crash and streaming pipeline regression are different failure modes." } ] }, @@ -16960,220 +16978,226 @@ "issue:43064|issue:44315", "issue:43208|issue:46897" ], - "prepared_review_unit_hash": "eccaada626d87bcd020d0f73f636de3d2b6dcc84e95105c806ce02909aabf504", + "prepared_review_unit_hash": "8d375073e9e479f0e90eb1fde7e1d31788a5950b3493d5bb054d6e3c567eb607", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36555, - "estimated_input_tokens": 9139, - "estimated_eval_tokens": 18534 + "serialized_chars": 36552, + "estimated_input_tokens": 9138, + "estimated_eval_tokens": 18532 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are mostly unrelated bug reports spanning different subsystems and models. I found no soft-edge pair that is clearly the same underlying bug/change, so all candidate merges should be rejected. For a canonical issue, the broad API-regression report about `tie_embeddings_and_encoder_decoder` removal is the strongest representative.", - "confidence": 0.9, - "canonical_issue_reason": "issue:43097 is the most broadly impactful and clearly framed regression in the set: it describes an API removal with user-visible breakage and a concrete compatibility gap, making it a better canonical representative than the many model-specific reports.", + "summary": "This cluster is not a duplicate set; it mixes unrelated regressions across imports, model loading, Flash Attention, config/schema issues, and pipeline/runtime behavior.", + "confidence": 0.98, + "canonical_issue_reason": "No single issue is a clear canonical duplicate target; the items describe different failure modes and code paths.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43097 is the best overall issue candidate because it is specific, actionable, and has wider applicability than the model-family-specific or one-off import/load failures.", + "best_issue_reason": "No global best issue fits the cluster, because there is no one underlying bug shared by the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44077", "right": "issue:46981", "accept": false, - "reason": "Different problems: model-specific `post_init` policy versus repeated HTTP 429 retry behavior in `AutoProcessor.from_pretrained`." + "reason": "Different bugs: a patchtsmixer API restriction vs AutoProcessor retry behavior on 429/HF_ENDPOINT." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Both mention chat/completions failures, but one is a general v5.3 API rejection and the other is a streaming inference failure for Qwen3.5-0.8B; not the same concrete bug." + "reason": "Both involve chat/completions, but one is request validation and the other is streaming inference on a specific Qwen model." }, { "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Different Qwen model variants and different failures: shape mismatch on load versus missing attention projections in layer 0." + "reason": "Different Qwen model failures: weight-shape mismatch vs missing attention projections in qwen3next." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Unrelated areas: generation cache during chunked prefill versus a testing utility bug." + "reason": "Unrelated: GenerationMixin chunked_prefill cache issue vs a testing_utils bug." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Different code paths and symptoms: missing `AutoImageProcessor` import versus `PretrainedConfig` type checking regression." + "reason": "Different regressions: AutoImageProcessor import failure vs PreTrainedConfig type-checking breakage." }, { "left": "issue:27957", "right": "issue:43352", "accept": false, - "reason": "Different models and different Flash Attention 2 issues; not the same underlying bug." + "reason": "Both mention Flash Attention 2, but they affect different models and different code paths." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "Unrelated: shard-size default behavior versus Qwen3.5 MoE loading failure." + "reason": "Unrelated: save_pretrained shard-size question vs Qwen3.5 MoE loading error." }, { "left": "issue:43874", "right": "issue:44261", "accept": false, - "reason": "Different multimodal components and different failures: missing image-patch helper versus RMS norm epsilon precision issue." + "reason": "Different model internals: missing image-patch helper vs missing rms_norm_eps in MLA q_a_layernorm." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "Both are dtype/precision related, but one is a compiler warning on bfloat16 and the other is a Blip2 submodule dtype propagation bug; not the same issue." + "reason": "Both involve dtype/precision, but one is a PyTorch warning during compile and the other is a Blip2 dtype mismatch on load." }, { "left": "issue:44263", "right": "issue:46172", "accept": false, - "reason": "Different GLM-family bugs: `torch.split()` handling in the DSA indexer versus a DeepSeek-v4 transpose/dtype path." + "reason": "Different code paths: GlmMoeDsaIndexer split handling vs deepseek-v4 tensor transpose/cast." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Both touch Flash Attention, but one is model support failure and the other is degenerate training with flash_attention_3; not mergeable as one fix." + "reason": "Both mention FlashAttention, but one is a model support failure and the other is training degeneracy with flash_attention_3." }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "Different models and code paths: DSA indexer ReLU bug versus DeepSeek-v4 tensor conversion/transpose behavior." + "reason": "Different bugs in different models: DSA indexer activation issue vs deepseek-v4 tensor handling." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "Same broad model family, but the concrete failures are different enough that this is not clearly one bug." + "reason": "Both are Qwen3-VL-related, but one is a specific shape mismatch and the other is a separate Qwen3-VL-Moe v5 issue." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "API regression from removed config helper versus `gguf_file` loading breakage; unrelated." + "reason": "Different problems: removed tie_embeddings API vs gguf_file/from_pretrained loading breakage." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "Video processor/torchvision import failure is unrelated to the testing utility bug." + "reason": "Unrelated: AutoProcessor/torchvision dependency crash vs testing_utils bug." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "`sys.modules` access bug in modeling utils is unrelated to the testing utility issue." + "reason": "Different areas: sys.modules access in modeling_utils vs testing_utils." }, { "left": "issue:42886", "right": "issue:46981", "accept": false, - "reason": "Both involve Hub/network behavior, but one is offline cache loading and the other is retry logic on 429 responses with `HF_ENDPOINT`; different bugs." + "reason": "Both are hub/network-related, but offline cache loading and HTTP 429 retry behavior are distinct bugs." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Different model internals: MLA layer norm epsilon omission versus GLM-5 RoPE implementation." + "reason": "Different model math issues: MLA layernorm epsilon bug vs GLM-5 RoPE implementation question." }, { "left": "issue:45753", "right": "issue:46369", "accept": false, - "reason": "Different models and symptoms: infinite loop on a specific image versus memory overflow on a SAM3 text model." + "reason": "Different symptoms and models: Qwen3_5 infinite-loop on one image vs SAM3 memory overflow." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Sliding-window bug in Qwen3-MoE is unrelated to the testing utility issue." + "reason": "Unrelated: Qwen3-MoE sliding_window behavior vs testing_utils bug." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "Different problems: config API removal versus flash attention crash when `s_aux=None`." + "reason": "Different issues: API removal in config handling vs flash_attention.py crashing on sink-less models." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Both are loading-related, but one is a tokenizer mapping assumption and the other is a specific model load failure; too little evidence they share the same root cause." + "reason": "Both are load-time failures, but TOKENIZER_MAPPING_NAMES=None and a specific model load regression are not the same bug." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Both are GLM-related, but one is an image-processor helper omission and the other is a `torch.split()` indexing bug." + "reason": "Different GLM subsystems: image-patch count helper vs DSA torch.split return handling." }, { "left": "issue:43673", "right": "issue:46123", "accept": false, - "reason": "Different pipeline stages: generation cache handling versus mask-generation batching logic." + "reason": "Different generation bugs: missing cache during chunked prefill vs MaskGenerationPipeline dropping the final partial batch." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Different model families and failures: missing attention projections versus a config field dropped by strict validation." + "reason": "Both are Qwen-family issues, but one is missing projections and the other is a config field silently dropped by @strict." }, { "left": "issue:43097", "right": "issue:46050", "accept": false, - "reason": "Unrelated: API removal regression versus quantized model save/load corruption." + "reason": "Different behaviors: removed tie_embeddings API vs quantized model save/load breakage." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Different DeepSeek issues: divergence from remote implementation vs a tensor cast/transpose bug." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are ImportErrors, but for different missing symbols from different packages; not the same bug." + "reason": "Both are ImportError reports, but they involve different missing symbols from different dependencies." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Different Qwen model variants and different implementation gaps." + "reason": "Related model family, but the reported failures are different and not the same concrete bug." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Different failures: missing torchvision path for video models versus `ReasoningEffort` import breakage in Gemma 4 processor loading." + "reason": "Different import/dependency problems affecting different model processors." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "Different training bugs in different features: FSDP2/PEFT optimizer state corruption versus Liger Kernel not being applied with `model_init`." + "reason": "Different training issues: FSDP2/PEFT optimizer-state corruption vs Liger Kernel not applied with model_init." }, { "left": "issue:43208", "right": "issue:46897", "accept": false, - "reason": "Different model families and distinct training bugs; only the general idea of a training issue overlaps." + "reason": "Different model-specific training bugs; only loosely similar as loss/training regressions." } ] }, @@ -17274,246 +17298,246 @@ "issue:43992|issue:44484", "issue:43352|issue:43976", "issue:43208|issue:43550", - "issue:43352|issue:44380", "issue:43825|issue:44655", "issue:43066|issue:43906", + "issue:43352|issue:44380", "issue:44734|issue:47116", "issue:43097|issue:46298", "issue:45584|issue:46365", "issue:44380|issue:46132" ], - "prepared_review_unit_hash": "e86af9bb77c111db862af62cf2bbe84e80e7c5ac5e84581d7764b844e76c05a7", + "prepared_review_unit_hash": "74df4f8dec0eb806afcb3869b60f521cf0ef52065130a6751051ef4de50108ad", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37733, - "estimated_input_tokens": 9434, - "estimated_eval_tokens": 19124 + "serialized_chars": 37730, + "estimated_input_tokens": 9433, + "estimated_eval_tokens": 19122 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the soft pairs mostly share only broad Transformer/v5 surface area, not the same underlying bug or change. I rejected all soft edges; if an anchor is needed, issue 47116 is the broadest representative of the loose v5 regression theme, but it is not a true duplicate target.", - "confidence": 0.98, - "canonical_issue_reason": "No pair in this set looks like a true duplicate cluster. If forced to choose one issue as an anchor, 47116 is the broadest, most general open v5 regression report, but the set is too diverse for a real canonical duplicate.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports. A few pairs share a subsystem or version theme, but none look like the same underlying bug/change closely enough to merge; no soft edge is accepted.", + "confidence": 0.71, + "canonical_issue_reason": "issue:47116 is the broadest open v5/pipeline regression item and the closest thing to an umbrella representative here, though the overall set is still mostly unrelated.", "canonical_pr_reason": null, - "best_issue_reason": "issue:47116 is the best loose representative because it is broad, current, and regression-oriented, but it should be treated as a thematic anchor only, not a duplicate canonical.", + "best_issue_reason": "issue:47116 is the most general current issue in the set, but it should not be treated as a duplicate of the other reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46655", "right": "issue:46664", "accept": false, - "reason": "Same model family, but one is a padding bug and the other is a num_frames semantic mismatch; different underlying processor defects." + "reason": "Same video processor area, but one is about padding math and the other about the meaning of num_frames; different bugs." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "Unrelated failures: AutoProcessor/torchvision loading vs save_pretrained shard-size behavior." + "reason": "Different problems: a crash loading a video model via AutoProcessor vs a question about save_pretrained shard sizing." }, { "left": "issue:46298", "right": "issue:46752", "accept": false, - "reason": "Noisy logging vs empty-list IndexError in apply_chat_template; not the same bug." + "reason": "No overlap in bug shape: noisy logging vs an IndexError on empty chat-template input." }, { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "v5 config-removal complaint vs quantization offloading failure; different code paths and symptoms." + "reason": "v5 embedding-flag removal is unrelated to quantization offloading behavior." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different symbols/modules and unrelated breakages." + "reason": "Both are import errors, but for different missing symbols and different code paths." }, { "left": "issue:42890", "right": "issue:46298", "accept": false, - "reason": "Missing set_seed in tests is unrelated to a logging-message issue." + "reason": "Seed-related test flakiness is unrelated to a noisy logging message." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "Tokenizer AddedToken kwarg conflict vs TOKENIZER_MAPPING_NAMES None handling; separate tokenizer bugs." + "reason": "Tokenizer construction argument collision vs tokenizer mapping returning None are distinct failure modes." }, { "left": "issue:43012", "right": "issue:46821", "accept": false, - "reason": "PyTorch bfloat16 warning and TimesFM window_size AttributeError are unrelated." + "reason": "PyTorch warning during bfloat16 compilation is unrelated to a TimesFM AttributeError on window_size." }, { "left": "issue:44117", "right": "issue:46701", "accept": false, - "reason": "Tokenizer mapping assumption bug vs DiffusionGemma pipeline failure; different subsystems." + "reason": "Tokenizer mapping fallback bug vs a pipeline failure for DiffusionGemma are not the same change." }, { "left": "issue:46172", "right": "issue:46710", "accept": false, - "reason": "DeepSeek tensor transpose bug vs model accuracy regression after upgrade; not the same issue." + "reason": "A specific tensor transpose bug is not the same as a broader accuracy regression after upgrading versions." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "local_files_only network leakage vs missing embed_tokens.weight on load; unrelated." + "reason": "Local-files-only network leakage and a missing embed_tokens.weight are unrelated loader issues." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "AutoProcessor kwargs forwarding vs auto_docstring annotation crash; different layers entirely." + "reason": "Passing kwargs to cached_file and auto_docstring crashes from future annotations are unrelated." }, { "left": "issue:44936", "right": "issue:46365", "accept": false, - "reason": "trainer.evaluate regression vs Trainer import blocked by datasets; not the same underlying problem." + "reason": "Trainer.evaluate failing after train is unrelated to an import conflict when importing Trainer." }, { "left": "issue:44451", "right": "issue:47116", "accept": false, - "reason": "Model loading failure for ScandiBERT vs streaming pipeline regression; no shared bug." + "reason": "Loading one specific model and a streaming pipeline regression are different concrete failures." }, { "left": "issue:44380", "right": "issue:46421", "accept": false, - "reason": "GPT2 attention scaling under SDPA/FlashAttention vs cache lazy-init recompilation; different behavior." + "reason": "GPT2 attention scaling being ignored is not the same as cache lazy-init causing recompilation." }, { "left": "issue:44938", "right": "issue:46178", "accept": false, - "reason": "Python 3.14 import/load failure vs unsupported tensorflow_text backend mapping; unrelated." + "reason": "Python 3.14 import/load failure and an invalid backend mapping error are different root causes." }, { "left": "issue:43986", "right": "issue:44016", "accept": false, - "reason": "Video-model loading without torchvision vs notebook syntax error; no overlap." + "reason": "Video-model import/runtime crash vs a notebook syntax error are unrelated." }, { "left": "issue:43064", "right": "issue:47116", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption vs streaming pipeline regression; different training/serving paths." + "reason": "FSDP2 optimizer-state corruption and streaming pipeline regression do not describe the same bug." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Dequantized model save failure vs Qwen image infinite loop; unrelated." + "reason": "Saving dequantized models and an infinite loop on a specific image are unrelated." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Both involve saving, but one is dequantized model serialization and the other pipeline save_pretrained support; not the same bug." + "reason": "Both mention saving, but one is dequantized models and the other is Pipeline.save_pretrained; different artifacts and symptoms." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "Warning about bfloat16 precision vs Bamba torch.compile SDPA failure; distinct issues." + "reason": "A compiler warning in bfloat16 code is not the same as a torch.compile failure on Bamba with SDPA." }, { "left": "issue:45712", "right": "issue:46298", "accept": false, - "reason": "Dummy PT classes leaking into dir/check_repo vs logging noise; unrelated." + "reason": "Left is about leftover dummy classes and repo checks; right is a noisy logging message." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Different warnings in different contexts: bfloat16 compilation warning vs tie_word_embeddings config warning." + "reason": "Both are warnings, but they concern different config topics and different code paths." }, { "left": "issue:43827", "right": "issue:47116", "accept": false, - "reason": "Docs still referencing pipeline() vs runtime streaming regression; documentation and execution bugs differ." + "reason": "Docs still referencing pipeline() after removals is a documentation issue, not the streaming regression itself." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed_tokens.weight vs max_shard_size question; unrelated model-loading vs saving behavior." + "reason": "Missing encoder weights vs a question about max_shard_size are unrelated." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "FlashAttention support gap vs Python-version compatibility issue; different root causes." + "reason": "FlashAttention support in one model is unrelated to the Python-version compatibility issue." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training bugs vs Bamba SDPA compile failure; different models and failure modes." - }, - { - "left": "issue:43352", - "right": "issue:44380", - "accept": false, - "reason": "FA2 unsupported warning vs GPT2 attention scaling ignored; not the same code path." + "reason": "xLSTM training bugs and a Bamba torch.compile SDPA failure are different model-specific defects." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "pipeline error message wording vs inability to save Pipeline objects; different issues." + "reason": "Incorrect v5 translation messaging and saving Pipeline objects are separate issues." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "Wrong tokenizer decoder type vs isolated reproduction of another issue; no evidence of same bug." + "reason": "Wrong tokenizer decoder type and a reproduction of another issue are not the same bug." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "FlashAttention support gating is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." }, { "left": "issue:44734", "right": "issue:47116", "accept": false, - "reason": "Serve KV-cache continuation indexing bug vs streaming pipeline regression; different components." + "reason": "KV-cache continuation crash in serve is a different code path from a streaming pipeline regression." }, { "left": "issue:43097", "right": "issue:46298", "accept": false, - "reason": "Removed config flag concern vs noisy logging; unrelated." + "reason": "API removal around tie_embeddings_and_encoder_decoder is unrelated to a noisy logging message." }, { "left": "issue:45584", "right": "issue:46365", "accept": false, - "reason": "Whisper empty-transcription generation bug vs Trainer import conflict; unrelated." + "reason": "Whisper empty-transcription failure and Trainer import problems are unrelated." }, { "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "GPT2 backend-specific attention scaling bug vs AttentionInterface.register behavior change; different functionality." + "reason": "GPT2 backend-specific attention scaling bug and AttentionInterface.register behavior change are distinct issues." } ] }, @@ -17592,8 +17616,8 @@ "issue:42890|issue:47059", "issue:44222|issue:45478", "issue:36683|issue:43582", - "issue:44380|issue:45479", "issue:42757|issue:43824", + "issue:44380|issue:45479", "issue:43064|issue:44380", "issue:42175|issue:46365", "issue:42490|issue:43012", @@ -17619,227 +17643,230 @@ "issue:43976|issue:44938", "issue:44938|issue:47059" ], - "prepared_review_unit_hash": "788ea7bbe95727319d57b97ed03805d021cc6e8ff3e200fe9450b69403a0ef3f", + "prepared_review_unit_hash": "79a46ed3d9a14caa822688aa18bd1dbf024f6cc74eb3cc4cd557242d228a799b", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36750, - "estimated_input_tokens": 9188, - "estimated_eval_tokens": 18632 + "serialized_chars": 36748, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is not a true duplicate cluster: the soft pairs mostly share broad keywords like import errors, loading, or model families, but each reports a different underlying failure mode. I would not merge any of the suggested soft edges.", - "confidence": 0.94, - "canonical_issue_reason": "No single issue is a valid canonical for this cluster; the items span unrelated regressions across imports, model configs, training, saving, and CLI behavior.", + "summary": "The cluster is mostly a grab bag of unrelated regressions and import/runtime errors. I only see one plausible duplicate pair: the two CLIP vision output regression reports. Everything else is too far apart in model, API, or failure mode to merge safely.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 43761 is the best representative because it describes the concrete CLIPVisionModel.forward v5 regression, and it is the closest match to the only plausible duplicate in the set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the strongest representative of a concrete Transformers v5 loading regression, but it still does not subsume the rest of the set.", + "best_issue_reason": "Issue 43761 is the clearest and most specific issue in the only likely duplicate pair, so it is the best global issue representative for this cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "Different bugs: ESM2 training/runtime breakage vs a deepseek-v4 dtype/transpose issue." + "reason": "Different models and failures: ESM2 training breakage vs a DeepSeek-V4 tensor operation bug." }, { "left": "issue:44077", "right": "issue:46365", "accept": false, - "reason": "Unrelated failures: patchtsmixer post_init policy vs importing Trainer with datasets installed." + "reason": "Unrelated issues: patchtsmixer post_init validation vs Trainer import failure tied to datasets." }, { "left": "issue:42890", "right": "issue:47059", "accept": false, - "reason": "One is a flaky test seed issue; the other is a CLI compatibility break with huggingface_hub." + "reason": "Both are test/runtime problems, but one is a seed-related SAM HQ test flake and the other is a CLI incompatibility with huggingface_hub 1.22.0." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "Both mention MoE/save/load, but one is FP8 save_pretrained and the other is a from_pretrained regression; not the same bug." + "reason": "Both involve MoE/quantization, but one is save_pretrained FP8 failure and the other is a from_pretrained loading error for Qwen3.5 Moe." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Gemma3 config attribute error and AppleSilicon allocator TypeError are unrelated." + "reason": "Completely different subsystems: Gemma3 config attribute error vs Apple Silicon allocator warmup TypeError." }, { - "left": "issue:44380", - "right": "issue:45479", + "left": "issue:42757", + "right": "issue:43824", "accept": false, - "reason": "Attention-scaling backend regression vs a degenerate zero-loss classification bug." + "reason": "Both are import errors, but they concern different missing symbols from different packages and code paths." }, { - "left": "issue:42757", - "right": "issue:43824", + "left": "issue:44380", + "right": "issue:45479", "accept": false, - "reason": "Both are import errors, but for different missing symbols and different code paths." + "reason": "Different bugs: GPT2 attention scaling is ignored in backend attention implementations, while the other is a zero-loss classification issue." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Distributed optimizer-state corruption and GPT-2 attention scaling are unrelated." + "reason": "Different areas and symptoms: distributed optimizer state corruption vs attention scaling configuration handling." }, { "left": "issue:42175", "right": "issue:46365", "accept": false, - "reason": "Packaging/backend dependency issue vs a Trainer import conflict with datasets." + "reason": "Package dependency/import problems, but one is backend extras installation and the other is Trainer import shadowing by datasets." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Local-path import warning and bfloat16 compile warning are different behaviors." + "reason": "Both mention warnings, but one is a local model path warning and the other is a PyTorch dtype precision warning." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "API request validation failure vs chat-template crash on batched VL input." + "reason": "Different chat-path failures: an HTTP API request rejection vs a processor crash on batched template application." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "ImportError for a missing internal helper and unexpected network access under local_files_only are different bugs." + "reason": "Different code paths: missing configuration_utils export vs local_files_only still allowing network requests." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "CLIP attention outputs missing and GPT-2 attention scaling are separate regressions." + "reason": "CLIP output assignment regression vs GPT2 attention scaling backend behavior; not the same bug." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Both mention RoPE, but one is missing layers and the other is a GLM-5 RoPE implementation discussion/bug." + "reason": "Same broad RoPE/model family theme, but one is a Smollm3 layer drop bug and the other is GLM-5 RoPE implementation discussion." }, { "left": "issue:45276", "right": "issue:46077", "accept": false, - "reason": "Gemma4 token embedding resize bug and Gemma4 bidirectional attention masking bug are different code paths." + "reason": "Different model behaviors: token embedding resize propagation vs causal-vs-bidirectional attention masking." }, { "left": "issue:43618", "right": "issue:43761", - "accept": false, - "reason": "Related CLIP regressions, but one drops attentions while the other drops hidden_states; not enough evidence for one bug." + "accept": true, + "reason": "Both look like the same v5 CLIP vision regression where forward output fields are no longer populated as expected." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are unrelated." + "reason": "Unrelated: xLSTM training blockers vs a BeitImageProcessorFast label reduction bug." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Both are import failures, but for different modules and missing names." + "reason": "Both are import errors, but they involve different removed/moved symbols from different modules." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Qwen3VL MoE loading breakage and sorted-state_dict load errors are different loading failures." + "reason": "Different loading failures: Qwen3VL-Moe model loading regression vs a state_dict sorting-related load error." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Both involve save_pretrained, but one is quantized-model saving and the other is Pipeline serialization." + "reason": "Both involve save_pretrained, but one is quantized model saving and the other is Pipeline object serialization; too different to unify." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Both are Qwen3-VL/MoE/fp8 related, but one is a bug report and the other is fp8 inference support; not the same underlying change." + "reason": "Same model family, but one is a general v5 Qwen3-VL-Moe bug and the other is fp8 inference support; not the same concrete defect." }, { "left": "issue:44483", "right": "issue:45290", "accept": false, - "reason": "Chat-completions API rejection and apply_chat_template tool-call crash are different layers of the stack." + "reason": "Both touch chat/template flows, but one is API request validation and the other is assistant tool-call handling during tokenization." }, { "left": "issue:44625", "right": "issue:46365", "accept": false, - "reason": "Qwen3.5 num_labels propagation and Trainer import conflict are unrelated." + "reason": "Different issues: Qwen3.5 num_labels propagation vs a Trainer import conflict with datasets." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "FlashAttention2 support gating and GenerationConfig validation are different model/feature bugs." + "reason": "Nemotron Flash Attention support and GLM-5 GenerationConfig validation are separate model-specific problems." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "A config propagation bug and a Gemma4 attention-mask bug do not share the same root cause." + "reason": "Different model/config bugs: label propagation vs Gemma4 attention mask construction." }, { "left": "issue:44355", "right": "issue:46877", "accept": false, - "reason": "Compiled-Python loading errors and import-time docstring diagnostics are unrelated." + "reason": "Different import-time behaviors: compiled Python file errors vs docstring lint output at import time." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "Base-model-prefix loading regression vs Pipeline save_pretrained failure are different save/load paths." + "reason": "Different regressions: base_model_prefix loading breakage vs Pipeline save_pretrained failure." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Different attribute errors on different models and objects; no shared underlying bug." + "reason": "Different attribute errors on different models: Gemma3Config vocab_size vs Sam3VisionEncoderOutput fpn_position_embeddings." }, { "left": "issue:44655", "right": "issue:46050", "accept": false, - "reason": "Pipeline serialization and sinq quantized model round-trip failures are separate save/load regressions." + "reason": "Both involve save/load, but one is Pipeline serialization and the other is sinq quantized model round-tripping." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both are version-specific load failures, but Python 3.9+/3.10+ and Python 3.14 compatibility issues are not the same bug." + "reason": "Both are Python-version compatibility issues, but they are about different interpreter versions and different failure modes." }, { "left": "issue:44938", "right": "issue:47059", "accept": false, - "reason": "Python 3.14 import/load failure and a CLI completion attribute error are unrelated." + "reason": "Both are environment compatibility problems, but one is Python 3.14 import failure and the other is a CLI break with huggingface_hub 1.22.0." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and the conservative rejections are well supported by the titles. The lone accepted soft pair, issue:43618 and issue:43761, is plausibly the same CLIP v5 regression and could be fixed in one PR. No obvious overstatement beyond that." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -17938,20 +17965,20 @@ "issue:43502|issue:44062", "issue:42898|issue:46664" ], - "prepared_review_unit_hash": "b270ecbd9096de312f4f3c219fb7ceefdbb1bdeea0c4ac0ef326b546f2d9ec9a", + "prepared_review_unit_hash": "c1d2b7c52143cf672d8c258cf8cfe10a5a84c3217b515ae9fe31c74ee25f2f0c", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36405, + "serialized_chars": 36406, "estimated_input_tokens": 9102, "estimated_eval_tokens": 18460 }, @@ -17960,214 +17987,211 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are unrelated despite superficial wording overlap. The only credible duplicate is the import-time slowdown pair (41669/44246).", - "confidence": 0.94, - "canonical_issue_reason": "issue:41669 is the clearest root-cause report for the import slowdown, with a specific explanation and impact; issue:44246 is the broader symptom report and fits as a duplicate.", + "summary": "This cluster is mostly a loose collection of unrelated Transformers issues (import/load regressions, model-config attribute errors, training bugs, tokenizer behavior, and performance complaints). The closest common thread is upgrade-related import/load breakage, but the individual reports are not the same underlying bug.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45278 is the broadest and most representative report in the set: a version-upgrade-induced spike in import failures. It best captures the cluster\u2019s loose import/regression theme without overfitting to one model-specific failure.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41669 is the best representative because it identifies the likely cause of the slowdown, not just the user-visible symptom.", + "best_issue_reason": "issue:45278 is the best single issue to anchor this cluster because it is broad, user-facing, and closest to the shared import/regression complaints seen elsewhere in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43867", "right": "issue:45216", "accept": false, - "reason": "Both are model-loading/checkpoint problems, but one is about sorted state_dict handling and the other about incorrect Qwen3.5 save_pretrained output; different failure modes." + "reason": "Both are load/save-related, but one is a generic state_dict loading error and the other is a Qwen3.5 save_pretrained regression; not the same concrete bug." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Gemma3Config missing vocab_size is unrelated to patchtsmixer post_init validation." + "reason": "Different models and different failures: missing vocab_size vs patchtsmixer post_init validation. No shared bug path." }, { "left": "issue:46077", "right": "issue:46512", "accept": false, - "reason": "Bidirectional attention mask construction and stale config dtype are different code paths and symptoms." + "reason": "One is about Gemma4 attention-mask semantics, the other about stale config dtype after from_config; unrelated behavior." }, { "left": "issue:36683", "right": "issue:47110", "accept": false, - "reason": "Missing Gemma3Config vocab_size and special_tokens list/dict handling are unrelated." + "reason": "Gemma3Config missing vocab_size and special_tokens list/type handling are unrelated issues." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIPOutput attentions assignment and Qwen3.5 save_pretrained checkpoint corruption are different bugs." + "reason": "CLIPOutput attentions assignment and Qwen3.5 save_pretrained checkpoint correctness are different code paths." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Gemma3Config missing vocab_size is not the same as Llama4Config missing pad_token_id." + "reason": "Both mention missing config attributes, but they are different models with different missing fields and likely different fixes." }, { "left": "issue:29870", "right": "issue:36296", "accept": false, - "reason": "A BART generate() decoder_attention_mask bug is unrelated to a tensor-parallel training bug." + "reason": "BART generate decoder mask bug and tensor-parallel training bug are unrelated." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption is unrelated to a typo in cache strategies." + "reason": "FSDP/PEFT optimizer state corruption and a typo in cache strategy text are not the same problem." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "Cache update no_grad behavior and broad import errors after version upgrade are not the same underlying issue." + "reason": "One is a no_grad decorator request on cache updates; the other is a broad import-error regression report." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder.from_pretrained is unrelated to Qwen3.5 Moe from_pretrained errors." + "reason": "UMT5Encoder missing embed_tokens.weight and Qwen3.5 Moe from_pretrained failure are different model-loading defects." }, { "left": "issue:44075", "right": "issue:46365", "accept": false, - "reason": "Unused SGD args and Trainer import failing with datasets installed are different problems." + "reason": "Optimizer argument handling and Trainer import depending on datasets are unrelated." }, { "left": "issue:44908", "right": "issue:46851", "accept": false, - "reason": "Scheduler kwargs being ignored is unrelated to Qwen3.5 variable-length position_ids handling." + "reason": "Scheduler kwargs propagation and Qwen3.5 variable-length position_ids are different subsystems." }, { "left": "issue:29870", "right": "issue:33453", "accept": false, - "reason": "BART generate masking failure and tokenizer loading regression do not describe the same bug." + "reason": "BART generate unbound variable and tokenizer-loading regression are not the same failure." }, { "left": "issue:44393", "right": "issue:47000", "accept": false, - "reason": "Qwen3-VL 2D bbox output hallucination/error is not the same as RTDetrLoss cost-matrix infeasibility." + "reason": "Qwen3-VL bounding-box output errors and RTDetrLoss infeasible cost matrix are different model/task failures." }, { "left": "issue:43867", "right": "issue:46202", "accept": false, - "reason": "Sorted state_dict loading failure and Gemma pipeline roundtrip failure on \u2581 are different issues." + "reason": "State-dict loading issues and Gemma pipeline roundtrip failure on \u2581 are not the same bug." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Compiled Python file import errors and a Python 3.13 IndentationError from torch.jit parsing are related to imports, but not the same concrete bug." + "reason": "Both involve import/runtime errors, but one is generic compiled-file failures and the other is a Python 3.13 torch.jit parsing issue." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "State_dict sorting/load failure is unrelated to missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig." + "reason": "Generic load_model/state_dict sorting issue and Qwen3.5 vision-config field loss are different problems." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "A missing set_seed in SamHQ tests is a different test flakiness cause than xdist races on captured_info.txt." + "reason": "Missing set_seed in tests and xdist race on captured_info.txt are both test-flakiness reports but not the same defect." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "Decorating cache update with no_grad and a broad 'breaks too many models' complaint are not the same bug." + "reason": "A specific no_grad/cache-update concern is not the same as a vague complaint that many models broke since v5.0." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Tensor-parallel training bug and loading a specific Qwen2-57B model causing OOM are not sufficiently the same underlying issue." + "reason": "Tensor parallel training bug and Qwen2-57B-A14B-Instruct OOM are different runtime failures." }, { "left": "issue:41669", "right": "issue:44246", - "accept": true, - "reason": "Both report the same import-time slowdown in transformers; 41669 gives the specific root cause ('import *' in models) while 44246 is the generic symptom." + "accept": false, + "reason": "Both concern import performance, but one is a specific import-star cleanup/perf issue and the other is a vague intermittent slow-import report; not enough to treat as the same underlying bug." }, { "left": "issue:43618", "right": "issue:44077", "accept": false, - "reason": "CLIPOutput attentions assignment and patchtsmixer post_init restrictions are unrelated." + "reason": "CLIP attentions assignment and patchtsmixer post_init allowance are unrelated." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "Cache update no_grad and Python version compatibility for transformers 5.1.0 are unrelated." + "reason": "Cache update no_grad and Python-version compatibility are unrelated." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Sorted state_dict loading failure and incorrect visual encoder keys on save_pretrained are different Qwen3.5/Qwen issues." + "reason": "State_dict sorting/loading problems are not the same as Qwen3.5 visual-encoder key serialization." }, { "left": "issue:43992", "right": "issue:45478", "accept": false, - "reason": "UMT5Encoder missing embed_tokens.weight is unrelated to the Qwen3.5 Moe from_pretrained regression." + "reason": "UMT5Encoder missing weights and a repeated Qwen3.5 Moe from_pretrained error are different model families and likely different code paths." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "PyTorch ImportError and tokenizer loading regression are different classes of failures." + "reason": "Missing PyTorch environment import error and tokenizer-loading regression are unrelated." }, { "left": "issue:43618", "right": "issue:46285", "accept": false, - "reason": "CLIPOutput attentions assignment and CLIPTextModel structural inconsistency are related to CLIP but not the same code-path problem." + "reason": "CLIPOutput attentions assignment and CLIP text-model structure inconsistency are different CLIP-side issues." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer ReLU discussion and DeepseekV3MoE divergence are unrelated." + "reason": "DSA indexer ReLU discussion and DeepseekV3MoE divergence are unrelated model implementation concerns." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Seed-related test flakiness and Phi-4-multimodal loading failure are unrelated." + "reason": "Test seeding flakiness and Phi-4 multimodal load failure are different issues." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "SamHQ test seeding issues and GraniteSpeech stale device override failures are different CI bugs." + "reason": "SAM HQ test flakiness and GraniteSpeech device-override CI failure are both test issues but not the same underlying bug." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "local_files_only still making API requests is unrelated to AddedToken receiving duplicate 'special' kwargs." + "reason": "local_files_only network leakage and AddedToken keyword collision are different API bugs." }, { "left": "issue:42898", "right": "issue:46664", "accept": false, - "reason": "clean_up_tokenization_spaces behavior change and SmolVLM num_frames semantics are unrelated tokenizer/video processor issues." + "reason": "Tokenizer cleanup-space behavior and SmolVLM num_frames semantics are unrelated." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet. The only accepted pair, issue:41669/issue:44246, is a conservative duplicate call based on one being the specific root-cause slowdown report and the other the generic symptom report. The other soft-edge rejections are appropriately cautious and do not overclaim mergeability or sameness." - }, - "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 }, @@ -18240,8 +18264,8 @@ "issue:43502|issue:44485", "issue:43824|issue:46617", "issue:43010|issue:43065", - "issue:43874|issue:46617", "issue:45850|issue:47110", + "issue:43874|issue:46617", "issue:43576|issue:46132", "issue:44387|issue:44556", "issue:30333|issue:33453", @@ -18253,8 +18277,8 @@ "issue:43502|issue:44222", "issue:43502|issue:44961", "issue:43827|issue:44961", - "issue:43010|issue:43232", "issue:46664|issue:46726", + "issue:43010|issue:43232", "issue:30333|issue:36296", "issue:43299|issue:44556", "issue:29870|issue:31795", @@ -18274,277 +18298,277 @@ "issue:43475|issue:46617", "issue:46143|issue:46144" ], - "prepared_review_unit_hash": "39e4b236f80d6da537ce38f173faf8d00802e3535f05d38984b6eebe629f00cb", + "prepared_review_unit_hash": "0dba2dfd84d8543ce9814863610c17028a96b71feae0fe0ee311a37702a9f600", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 39, - "serialized_chars": 38959, - "estimated_input_tokens": 9740, - "estimated_eval_tokens": 19736 + "serialized_chars": 38964, + "estimated_input_tokens": 9741, + "estimated_eval_tokens": 19738 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous based on titles/body metadata alone. There are several soft-similarity loading/import regressions, but most look like distinct concrete bugs. The closest duplicate candidate is the Qwen3.5 MoE loading pair (45125/45478), which is plausibly the same underlying failure, though the evidence here is limited.", - "confidence": 0.64, - "canonical_issue_reason": "Issue 45125 is the best anchor for the plausible Qwen3.5 MoE duplicate pair because it is the more concrete report: it names Qwen3_5MoeForConditionalGeneration and a missing `_tp_plan`, while 45478 is the broader from_pretrained failure on the same model family/version.", + "summary": "This cluster is mostly heterogeneous Transformers issues (loading/import regressions, docs, multimodal/model-specific bugs). Only the two RoFormer reports appear to describe the same underlying bug in the model\u2019s attention path.", + "confidence": 0.78, + "canonical_issue_reason": "issue:46144 is the best anchor because it states the likely root cause more directly: RoFormer attention is not using the attention interface, which matches the kwargs-passthrough symptom in the paired issue.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45125 is the strongest cluster representative among the loading-related reports because it is specific enough to anchor the likely Qwen3.5 MoE issue without overgeneralizing beyond the available metadata.", + "best_issue_reason": "issue:46144 is the most suitable representative of the only true duplicate pair and reads like the root-cause report rather than the downstream symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44734", "right": "issue:46617", "accept": false, - "reason": "Different bugs: KV-cache continuation tensor indexing vs loading a trust_remote_code model from cache." + "reason": "Different failures: KV-cache continuation tensor indexing vs loading a trust_remote_code model from cache." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Both are loading-related, but one is tokenizer loading in v5 and the other is checkpoint reload after upgrade; not the same concrete failure." + "reason": "Both are loading-related, but one is tokenizer loading in v5 and the other is a checkpoint reload/version-compatibility problem." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 save_pretrained/moe storage behavior vs missing `_tp_plan` for tensor parallelism; different code paths." + "reason": "Different model features: FP8 save_pretrained for MoE vs tensor-parallel plan missing for Qwen3.5 MoE." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Local-files-only network leak vs GLM-5 RoPE implementation; no shared underlying bug." + "reason": "Unrelated bugs: local_files_only still triggering network requests vs GLM-5 RoPE implementation." }, { "left": "issue:43824", "right": "issue:46617", "accept": false, - "reason": "Import error for Qwen2.5 VL class vs cache FileNotFoundError on trust_remote_code loading." + "reason": "Different root causes: missing Qwen2_5_VL export vs cache FileNotFoundError when loading trust_remote_code models." }, { "left": "issue:43010", "right": "issue:43065", "accept": false, - "reason": "Cache/layer `update()` no_grad issue is unrelated to the dummy Conv2d in Sam3PixelDecoder." + "reason": "Cache/layer no_grad decoration is unrelated to the Sam3PixelDecoder dummy Conv2d issue." }, { - "left": "issue:43874", - "right": "issue:46617", + "left": "issue:45850", + "right": "issue:47110", "accept": false, - "reason": "Missing image-patch helper in GLM46V is a model-specific AttributeError, not a cache FileNotFoundError." + "reason": "One is a broad v5 regression complaint; the other is a specific special_tokens list/key error." }, { - "left": "issue:45850", - "right": "issue:47110", + "left": "issue:43874", + "right": "issue:46617", "accept": false, - "reason": "Broad complaint about v5 breakage vs a specific `special_tokens` attribute error." + "reason": "Missing GLM46V image-processor method vs cache loading FileNotFoundError are different bugs." }, { "left": "issue:43576", "right": "issue:46132", "accept": false, - "reason": "CLI/env command breakage and AttentionInterface.register behavior are different issues." + "reason": "Different areas and symptoms: transformers env CLI breakage vs AttentionInterface.register behavior change." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "Int4 quantization memory regression is unrelated to checkpoint reload incompatibility after upgrade." + "reason": "Memory/OOM under int4 quantization is unrelated to checkpoint reload failures after upgrade." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLflow reporting failure and tokenizer loading regression are unrelated." + "reason": "MLflow job-status reporting vs tokenizer loading regression are unrelated." }, { "left": "issue:45125", "right": "issue:45478", - "accept": true, - "reason": "Same model family and same loading path: 45478 reports a from_pretrained failure on Qwen3.5 MoE, and 45125 adds a concrete tensor-parallel loading gap; these plausibly describe the same bug." + "accept": false, + "reason": "Same model family, but one is missing _tp_plan and the other is a from_pretrained error; not enough to treat as the same bug." }, { "left": "issue:42548", "right": "issue:44351", "accept": false, - "reason": "Cannot import `PreTrainedModel` and cannot import `HybridCache` are separate import regressions." + "reason": "Different missing imports: PreTrainedModel vs HybridCache." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Sentence Transformers loading hang is unrelated to docs confusion about `input_ids` and `past_key_values`." + "reason": "Sentence Transformers loading hang and forward-doc confusion are different kinds of issues." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled Python file execution errors and data-filtering slowdown after upgrade are not the same bug." + "reason": "Compiled Python file errors are unrelated to a performance regression after version upgrade." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Both involve Qwen3.5 MoE, but one is a missing tensor-parallel plan and the other is a dropped config field; different concrete fixes." + "reason": "Tensor-parallel plan missing is unrelated to a config field being dropped by @strict." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "Local-files-only networking bug and FP8 `save_pretrained` behavior are unrelated." + "reason": "Network access despite local_files_only is unrelated to FP8 save_pretrained MoE behavior." }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "Local-files-only networking bug and a placeholder issue are unrelated." + "reason": "No meaningful relation between local_files_only network requests and the racoon issue." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs still referencing `pipeline()` after removals is unrelated to the placeholder issue." + "reason": "Docs still referencing pipeline() is unrelated to the racoon issue." }, { - "left": "issue:43010", - "right": "issue:43232", + "left": "issue:46664", + "right": "issue:46726", "accept": false, - "reason": "Different generation/cache problems: no_grad decoration vs `_update_model_kwargs_for_generation` after sync_gpus." + "reason": "Both are vision-related, but num_frames semantics and missing image-processor backend are different bugs." }, { - "left": "issue:46664", - "right": "issue:46726", + "left": "issue:43010", + "right": "issue:43232", "accept": false, - "reason": "Different video/processor issues: inconsistent `num_frames` semantics vs missing vision backend handling." + "reason": "Both mention generation/cache internals, but one is about no_grad decoration and the other about sync_gpus handling." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLflow reporting failure is unrelated to tensor parallel training bugs." + "reason": "MLflow reporting and tensor-parallel training are unrelated." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Qwen3VL MoE loading breakage and checkpoint reload regressions are different model/version failures." + "reason": "Both involve loading after version changes, but the affected models and failure modes differ." }, { "left": "issue:29870", "right": "issue:31795", "accept": false, - "reason": "BART generate-time `decoder_attention_mask` error is unrelated to documentation wording." + "reason": "BART generation UnboundLocalError and forward-doc confusion are unrelated." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Tokenizer-loading regression and Llama 3.2 Vision model-loading breakage are different concrete failures." + "reason": "Tokenizer-loading regression and Llama 3.2 vision model-loading breakage are distinct issues." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled-file execution errors and `PretrainedConfig` type checking are unrelated." + "reason": "Compiled-file errors are unrelated to PretrainedConfig type checking in v5.4.0." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Missing `is_offline_mode` import from huggingface_hub is unrelated to the v5 env CLI issue." + "reason": "Missing huggingface_hub export and broken transformers env CLI are separate breakages." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "CLI/env command breakage and `PretrainedConfig` type-checking regression are different code paths." + "reason": "CLI breakage and PretrainedConfig type-checking regression are not the same bug." }, { "left": "issue:43576", "right": "issue:47059", "accept": false, - "reason": "Both touch CLI behavior, but one is the `transformers env` command and the other is a hub version-specific Typer completion attribute error." + "reason": "Different CLI failures with different external-version triggers and different missing attributes." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Docs confusion in `model.forward` is unrelated to the `layer_type_validation` import error." + "reason": "Docs confusion around model.forward is unrelated to a missing configuration_utils import." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "Env command breakage and Qwen2.5 VL import failure are unrelated." + "reason": "CLI env command breakage is unrelated to Qwen2.5-VL import failure." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger Kernel application with `model_init` and a DSA indexer ReLU discussion are unrelated." + "reason": "Liger Kernel application with model_init is unrelated to a DSA indexer/ReLU discussion." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "Placeholder issue and Gemma-4 audio positional encoding bug are unrelated." + "reason": "No relation between the racoon issue and Gemma-4 audio positional encoding." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." + "reason": "GLM-5 RoPE implementation and the racoon issue are unrelated." }, { "left": "issue:44351", "right": "issue:46617", "accept": false, - "reason": "`HybridCache` import failure and cache FileNotFoundError are different import/load problems." + "reason": "Missing HybridCache import and trust_remote_code cache FileNotFoundError are different failures." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor parallel training bug is unrelated to docs mentioning `return_all_scores`." + "reason": "Tensor parallel training bug and TextClassificationPipeline docs mismatch are unrelated." }, { "left": "issue:43352", "right": "issue:44961", "accept": false, - "reason": "Flash Attention support error for Nemotron is unrelated to the placeholder issue." + "reason": "Flash Attention support error for Nemotron and the racoon issue are unrelated." }, { "left": "issue:43475", "right": "issue:46617", "accept": false, - "reason": "SAM3 vision encoder attribute error and cache FileNotFoundError are unrelated." + "reason": "Missing SAM3 vision encoder attribute and cache FileNotFoundError are different bugs." }, { "left": "issue:46143", "right": "issue:46144", - "accept": false, - "reason": "They are related to RoFormer behavior, but one is missing `**kwargs` passthrough and the other is attention-interface integration; not clearly the same concrete bug." + "accept": true, + "reason": "Both report the same RoFormer attention-path bug: kwargs are not being propagated because the attention implementation is not using the attention interface." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The clustering is mostly conservative, but the canonicalization is a bit overconfident. The packet only gives title/body-length metadata, so calling issue 45125 a \"root-cause report\" for 45478 goes beyond the evidence available here. At most, you can say they are plausibly related Qwen3.5 MoE loading reports. Also, \"only credible duplicate pair\" is stronger than the packet supports given there are multiple soft-similarity pairs in the cluster. Please keep the summary more tentative and evidence-bounded." + "accept": true, + "feedback": "The summary is grounded in the packet, and the only positive duplicate call (issue:46143/issue:46144) is stated conservatively and fits the titles. The other soft-edge rejections are also cautious and supported by the issue titles, so I don\u2019t see evidence of overstating similarity." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -18653,20 +18677,20 @@ "issue:41669|issue:45341", "issue:43976|issue:46218" ], - "prepared_review_unit_hash": "5dfb9fccfbab7d1312291c27b62c0ee4445e03e667e5014ca98f3a8659cb8274", + "prepared_review_unit_hash": "3c1365398e1a7c671af5955b614546d417891f71bf421274e3e4894d363ea469", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 40, - "serialized_chars": 38961, + "serialized_chars": 38963, "estimated_input_tokens": 9741, "estimated_eval_tokens": 19738 }, @@ -18675,256 +18699,259 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a mixed set of issue reports, but the strongest duplicate theme is a lazy-loading / top-level import-export regression in `transformers`. Most other soft pairs are only loosely related by subsystem or are clearly unrelated.", - "confidence": 0.66, - "canonical_issue_reason": "`issue:44273` is the broadest and most central report for the import/lazy-loading regression theme, and it can reasonably absorb several of the nearby import-related symptoms.", + "summary": "This cluster is noisy and mostly groups broad import/lazy-loading/runtime-regression reports rather than one concrete defect. The best representative is the open lazy-loading issue, but none of the soft pairs are safe duplicate merges.", + "confidence": 0.63, + "canonical_issue_reason": "issue:44273 is the broadest and most central representative: it directly targets lazy loading behavior, which is the closest shared theme across many of the import-related reports in this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "`issue:44273` is the best single issue to represent this cluster because it is the most umbrella-like description of the shared failure mode.", + "best_issue_reason": "issue:44273 is the best issue to anchor the cluster because it is open, clearly describes the core lazy-loading problem, and can serve as the umbrella for adjacent import-related symptoms.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41628", "right": "issue:42548", - "accept": true, - "reason": "Both are top-level `from transformers import ...` failures; likely the same broken export/lazy-loading regression." + "accept": false, + "reason": "Both are import-name errors, but they concern different symbols (AutoImageProcessor vs PreTrainedModel) and likely different broken import paths." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "Same broad FP8/MoE area, but one is about saving and the other about inference support; not the same concrete bug." + "reason": "Both mention FP8/moe, but one is about saving pretrained checkpoints and the other about inference support; different concrete failures." }, { "left": "issue:45375", "right": "issue:47110", "accept": false, - "reason": "Different failures: a missing config field vs a `special_tokens` attribute error." + "reason": "One is a missing config field in Qwen3_5MoeVisionConfig; the other is a special_tokens type error. Not the same bug." }, { "left": "issue:43867", "right": "issue:47110", "accept": false, - "reason": "Unrelated model-loading error versus `special_tokens` shape/typing error." + "reason": "State-dict loading/sorting error is unrelated to the special_tokens list/keys failure." }, { "left": "issue:44315", "right": "issue:46209", "accept": false, - "reason": "Different code paths and symptoms: `model_init`/Liger Kernel versus FP8 inference on B200." + "reason": "Liger Kernel/model_init application is unrelated to FP8 inference on B200." }, { "left": "issue:43525", "right": "issue:47110", "accept": false, - "reason": "Both are attribute errors, but on unrelated objects and features." + "reason": "Llama4Config missing pad_token_id is a config attribute issue, not the special_tokens list/dict mismatch." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "One is a specific chat crash for Qwen3.5-35B; the other is a generic broad complaint about v5 breaking models." + "reason": "Qwen3.5 chat crash and the generic v5 regression report are too broad and do not point to the same concrete code path." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "Chunked-prefill cache regression and a cache-strategy typo are not the same bug." + "reason": "Chunked prefill cache regression and a cache-strategy typo are different cache-related problems." }, { "left": "issue:44485", "right": "issue:46664", "accept": false, - "reason": "RoPE implementation and video `num_frames` semantics are unrelated." + "reason": "GLM-5 RoPE implementation and SmolVLM num_frames semantics involve different model families and different functionality." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Lazy loading regression and TensorBoard/TF backend detection are different problems." + "reason": "Lazy loading problems are unrelated to the TensorBoard-without-TensorFlow backend error." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "A specific `processor.save_pretrained` file bug versus a placeholder generic bug report." + "reason": "Missing processor save_pretrained files is not the same as a generic placeholder bug report." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "Both involve Qwen3/MoE, but one is a VL bug and the other is missing tensor-parallel metadata." + "reason": "Qwen3-VL/Moe bug and missing _tp_plan are distinct model-specific issues." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Both touch import internals, but the reported failures are different and not clearly the same underlying defect." + "reason": "Both touch import/loading internals, but one is a user-visible lazy-loading failure and the other is unsafe sys.modules access; not enough to call them the same bug." }, { "left": "issue:44360", "right": "issue:46664", "accept": false, - "reason": "Completely unrelated discussion topics." + "reason": "DSA indexer/ReLU discussion and video num_frames semantics are unrelated." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Import-star cleanup/performance and cache-missing generation behavior are not the same change or bug." + "reason": "Import-star cleanup/performance and chunked-prefill cache missing are different changes and symptoms." }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "Vision bbox hallucination/error and `PretrainedConfig` type-checking regression are unrelated." + "reason": "Qwen3-VL bounding-box hallucination is unrelated to PretrainedConfig type-checking regressions." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "Both are Qwen3-VL issues, but one is bbox output and the other is an infinite loop on a specific image; too different." + "reason": "Both are Qwen3.5/VL issues, but one is bounding-box output and the other is an infinite-loop image case; different failure modes." }, { "left": "issue:43576", "right": "issue:44556", "accept": false, - "reason": "Both are v5 regressions, but one is the `transformers env` command and the other is checkpoint reload compatibility." + "reason": "Both are version-upgrade regressions, but one is the env command and the other is checkpoint reload compatibility; not the same bug." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "Specific missing config field versus a generic complaint about v5 breaking many models." + "reason": "Missing pad_token_id in Llama4Config is specific and unrelated to a broad 'v5 breaks too many models' complaint." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both involve v5 loading regressions, but one is generic checkpoint reload and the other is a specific remote-code/meta-init case." + "reason": "Both concern v5 migration, but one is checkpoint reload compatibility and the other is remote-code/meta-init incompatibility; different code paths." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "Specific config attribute regression versus a broad release-quality complaint." + "reason": "StableLmConfig pad_token_idx missing is a specific config regression, not the generic v5 breakage report." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "FP8 inference support and weight conversion tooling are different issues." + "reason": "FP8 inference support for Qwen models is unrelated to a deepseekv4 weight-conversion question." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "A cache-strategy typo and a broad v5 complaint do not match." + "reason": "Cache-strategy typo is a narrow naming/doc bug, not the broad v5 regression report." }, { "left": "issue:45850", "right": "issue:47059", "accept": false, - "reason": "Generic v5 breakage versus a CLI compatibility failure with `huggingface_hub`." + "reason": "Generic v5 breakage and a cli completion failure with huggingface_hub 1.22.0 are different issues." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 settings and cache-strategy typo are unrelated." + "reason": "TF32 API guidance and cache-strategy typo are unrelated." }, { "left": "issue:41669", "right": "issue:46218", "accept": false, - "reason": "Import-star cleanup/performance is not the same as a missing module import error." + "reason": "Import-star cleanup/performance does not match the Qwen3_5ForCausalLM import-module failure." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "Model-type warning and Liger Kernel application bug are unrelated." + "reason": "sam3_video/sam3_tracker warning and Liger Kernel/model_init are different model-initialization problems." }, { "left": "issue:44246", "right": "issue:44273", - "accept": true, - "reason": "Same import/lazy-loading regression: one reports slow import, the other says lazy loading is broken." + "accept": false, + "reason": "Import taking long sometimes and lazy loading not working properly overlap on imports, but they describe different symptoms and likely different fixes." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "Test seeding issue and scheduler kwargs handling are unrelated." + "reason": "Missing set_seed in a test is unrelated to inverse_sqrt scheduler kwargs handling." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "CLI/environment command breakage and scheduler kwargs bug are unrelated." + "reason": "env command breakage and scheduler kwargs propagation are unrelated subsystems." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash-Attention support limitation versus Qwen3-VL bbox error; different subsystems." + "reason": "Flash Attention 2.0 support gating for Nemotron and Qwen3-VL bounding-box errors are unrelated model-specific bugs." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python version compatibility bug versus vision-model bbox issue." + "reason": "Python version compatibility and Qwen3-VL bounding-box errors are different regressions." }, { "left": "issue:44295", "right": "issue:45230", "accept": false, - "reason": "Specific `position_ids` buffer access bug versus a generic bug report." + "reason": "Reading position_ids after buffer registration is a concrete tensor/buffer bug; the other item is just a generic bug report placeholder." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler keyword handling and PretrainedConfig type checking are unrelated." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer save metadata mismatch versus a generic bug report." + "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to the generic bug report placeholder." }, { "left": "issue:44492", "right": "issue:46328", "accept": false, - "reason": "Cache-strategy typo and `SequenceFeatureExtractor.pad` inefficiency are unrelated." + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad performance issue are different areas and different fixes." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-star cleanup/performance and missing `HybridCache` export are not the same underlying bug." + "reason": "Import-star cleanup/performance and missing HybridCache import are different import-related bugs." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "Both mention kwargs, but the affected APIs and failures are different." + "reason": "AutoProcessor kwargs forwarding and inverse_sqrt scheduler kwargs are unrelated code paths." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Import-star cleanup issue and testing-utils bug are unrelated." + "reason": "Import-star cleanup/performance is unrelated to the testing_utils bug." }, { "left": "issue:43976", "right": "issue:46218", "accept": false, - "reason": "Python compatibility bug and missing `Qwen3_5ForCausalLM` import are unrelated." + "reason": "Python 3.9/3.10 compatibility and Qwen3_5ForCausalLM import failure are not the same issue." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The cluster summary stays grounded in the packet: the strongest shared theme is indeed the lazy-loading/top-level import-export regression around `transformers`, and the two accepted soft edges are conservative and supported by the issue titles. The rejections avoid overclaiming beyond the evidence shown." - }, - "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 }, @@ -19033,256 +19060,280 @@ "issue:44230|issue:44623", "issue:28348|issue:43352" ], - "prepared_review_unit_hash": "3c1b738b1eb42789b2968ca29c78630cbb4ea7a629afcf197ff519abfdbc2797", + "prepared_review_unit_hash": "72212d6bd36ddd2bf79af05d4eec842cd98efb590e6cf749bc64b8f328a144c1", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 41, - "serialized_chars": 39102, - "estimated_input_tokens": 9776, - "estimated_eval_tokens": 19808 + "serialized_chars": 39107, + "estimated_input_tokens": 9777, + "estimated_eval_tokens": 19810 }, "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. The soft pairs share broad keywords like loading, tokenizers, or v5 regressions, but they describe different concrete failures and code paths, so none should be merged as duplicates.", - "confidence": 0.37, - "canonical_issue_reason": "issue:33453 is the strongest standalone issue in the set: it has the most discussion and a clear, specific regression report around tokenizer loading, making it the best anchor among otherwise unrelated issues.", + "summary": "This cluster is mostly a set of unrelated issue reports that only share broad Transformers themes like loading, tokenizers, model internals, or Flash Attention. I don\u2019t see any true duplicate pair among the soft edges, so all of them should stay split.", + "confidence": 0.59, + "canonical_issue_reason": "issue:45850 is the closest thing to an umbrella report because it broadly describes v5 breaking many models, but it is still too vague to be a strong duplicate anchor.", "canonical_pr_reason": null, - "best_issue_reason": "issue:33453 is the most actionable and well-described problem here; it looks like a real regression rather than a vague or generic report.", + "best_issue_reason": "issue:45850 is the broadest representative of the mostly heterogeneous set, though it is not a precise bug report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Different failures: optional post_init validation vs an AttributeError in _get_tied_weight_keys." + "reason": "Different bugs: optional post_init validation vs _get_tied_weight_keys crashing on a list." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "One is documentation confusion; the other is a tokenizer loading regression." + "reason": "Docs confusion about forward args vs a tokenizer loading regression." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken env CLI command is unrelated to lazy loading not working." + "reason": "CLI env command breakage vs lazy loading behavior; different code paths." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "model_init Liger application bug is a different code path from flash_attention_3 training degeneration." + "reason": "Liger kernel application through model_init vs flash_attention_3 training instability." }, { "left": "issue:43519", "right": "issue:46655", "accept": false, - "reason": "Different processor bugs: Qwen3VL timestamp calculation vs SmolVLM padding calculation." + "reason": "Different processor bugs in different models: timestamp math vs video padding math." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "DSA indexer ReLU math issue is unrelated to Gemma-4 audio positional encoding." + "reason": "Unrelated model internals: DSA indexer ReLU vs Gemma-4 audio positional encoding." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading problem is not the same as inverse_sqrt scheduler kwargs being ignored." + "reason": "Lazy loading vs inverse_sqrt scheduler kwargs; no shared underlying bug." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "Generic bug report is not the same concrete issue as a Qwen3.5 MoE from_pretrained error." + "reason": "Generic placeholder report vs a concrete Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Timestamp calculation in Qwen3VL and GLM-5 RoPE implementation are different bugs." + "reason": "Processor timestamp calculation and GLM-5 RoPE are different subsystems." }, { "left": "issue:45468", "right": "issue:46894", "accept": false, - "reason": "Audio positional encoding issue is unrelated to the CohereASR double-shift training-loss bug." + "reason": "Different audio/model bugs: Gemma-4 positional encoding vs CohereASR double-shift training loss." }, { "left": "issue:43576", "right": "issue:46218", "accept": false, - "reason": "Broken transformers env command is unrelated to a missing Qwen3_5ForCausalLM module import." + "reason": "CLI env command issue vs Qwen3_5 import failure; unrelated symptoms." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Pipeline save_pretrained serialization failure is not the same as the tied-weights helper crash." + "reason": "Pipeline save_pretrained serialization vs tied-weight-key lookup crash." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Processor timestamp math vs scheduler parameter handling." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import-star cleanup/perf issue is unrelated to the Qwen3ForCausalLM VRAM leak." + "reason": "Codebase import-star cleanup/perf discussion vs a runtime VRAM leak." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Documentation wording in forward is unrelated to Llama 3.2 Vision model loading breaking." + "reason": "Documentation confusion vs a concrete model-loading regression." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "GLM-5 RoPE implementation and Gemma-4 audio positional encoding are different model-specific bugs." + "reason": "Different model implementations: GLM-5 RoPE vs Gemma-4 audio positional encoding." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leakage in dataloader threads is unrelated to GLM-5 RoPE behavior." + "reason": "Memory leak in Qwen3ForCausalLM vs positional-encoding math in GLM-5." }, { "left": "issue:46018", "right": "issue:46172", "accept": false, - "reason": "Shared expert gating logic and a comb.to(...).transpose(...) bug are different DeepSeek-v4 issues." + "reason": "DeepSeek-V4 expert gating vs a tensor dtype/transpose bug; not the same fix." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "local_files_only network access bug is unrelated to Qwen3VL timestamp calculation." + "reason": "local_files_only download enforcement vs processor timestamp calculation." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API settings guidance is unrelated to a Qwen3ForCausalLM VRAM leak." + "reason": "TF32 configuration API change vs a VRAM leak; different concerns." }, { "left": "issue:45850", "right": "issue:46218", "accept": false, - "reason": "Broad complaints about v5 breakage do not match the specific Qwen3_5ForCausalLM import failure." + "reason": "Broad v5 breakage report vs a specific import error for Qwen3_5ForCausalLM." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Timestamp calculation bug vs missing files in processor.save_pretrained." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return-value issue is unrelated to Gemma-4 audio positional encoding." + "reason": "torch.split/indexer logic vs Gemma-4 audio positional encoding." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "rms_norm_eps precision mismatch is a different model bug from the audio positional encoding problem." + "reason": "Missing rms_norm_eps in MLA q_a_layernorm vs an audio positional-encoding issue." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching audio support is unrelated to DeepSeek-v4 weight conversion." + "reason": "Continuous batching audio support vs a DeepSeek-V4 weight conversion question." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Qwen3VL timestamp math and Qwen3ForCausalLM VRAM leaks are different failures." + "reason": "Processor timestamp math vs a model VRAM leak." }, { "left": "issue:44351", "right": "issue:46218", "accept": false, - "reason": "HybridCache import failure is not the same as importing Qwen3_5ForCausalLM." + "reason": "Both are import errors, but for different missing symbols and different root causes." }, { "left": "issue:43352", "right": "issue:46218", "accept": false, - "reason": "Unsupported Flash Attention 2.0 on NemotronH is unrelated to a module import error." + "reason": "Nemotron flash-attention support gap vs Qwen3_5 import failure; different problems." }, { "left": "issue:44861", "right": "issue:47110", "accept": false, - "reason": "tied-weight helper crash and special_tokens.keys AttributeError are different issues." + "reason": "Different attribute/type errors in unrelated utilities." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both mention tokenizer v5, but one is a broad behavior regression and the other is a specific decoder-type bug; not the same concrete issue." + "reason": "Both concern tokenizer v5 behavior, but one is a broad regression and the other is a specific decoder-type bug." }, { "left": "issue:44861", "right": "issue:47059", "accept": false, - "reason": "tied-weights AttributeError is unrelated to the CLI failure from huggingface_hub 1.22.0." + "reason": "Tied-weight-key crash vs a CLI completion incompatibility with huggingface_hub." }, { "left": "issue:28348", "right": "issue:45250", "accept": false, - "reason": "Specific GPT2LMHeadModel flash-attn support request is not the same as the generic Flash Attention 2.0 issue." + "reason": "Specific GPT2LMHeadModel support request vs a generic Flash Attention 2.0 issue; too broad to merge." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs bug is unrelated to the placeholder 'Unemployment' issue." + "reason": "Scheduler kwarg bug vs an unrelated placeholder issue." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "A typo in cache strategies is unrelated to the placeholder issue." + "reason": "Cache-strategy typo vs an unrelated placeholder issue." }, { "left": "issue:44998", "right": "issue:45850", "accept": false, - "reason": "Placeholder issue is not the same as broad v5 breakage complaints." + "reason": "Placeholder/non-technical report vs a broad v5 regression report." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to the placeholder issue." + "reason": "CLIPOutput attentions regression vs an unrelated placeholder issue." }, { "left": "issue:44938", "right": "issue:44998", "accept": false, - "reason": "Python 3.14 load failure is unrelated to the placeholder issue." + "reason": "Python 3.14 import/load failure vs an unrelated placeholder issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weight-key bookkeeping crash vs chat template tool-call parsing crash." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow from_pretrained checkpoint loading vs an ImportError for layer_type_validation." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "fp8 DeepSeek/Qwen support request is unrelated to processor.save_pretrained missing files." + "reason": "FP8 support request vs missing files on processor.save_pretrained." }, { "left": "issue:28348", "right": "issue:43352", "accept": false, - "reason": "GPT2 flash-attn support request is unrelated to NemotronH not supporting Flash Attention 2.0 yet." + "reason": "Same broad Flash Attention topic, but different models and different concrete fixes; not mergeable into one PR." } ] }, @@ -19344,142 +19395,142 @@ "issue:37934|issue:43668", "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "62a189a8603229592cd07a954a55703741250372a64c6f39a32b69b89741f7e5", + "prepared_review_unit_hash": "a68e2429ace1029cc8a6aea5b6bc06c4a487c8aac0e298e1d95483345956034d", "component_budget": { "node_count": 513, "item_count": 513, "soft_pair_count": 1886, - "serialized_chars": 978490, - "estimated_input_tokens": 244623, - "estimated_eval_tokens": 489502 + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 }, "budget": { "node_count": 17, "item_count": 17, "soft_pair_count": 18, - "serialized_chars": 14950, - "estimated_input_tokens": 3738, - "estimated_eval_tokens": 7732 + "serialized_chars": 14946, + "estimated_input_tokens": 3737, + "estimated_eval_tokens": 7730 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes unrelated bugs, docs issues, config/type-hint fixes, performance questions, and feature requests. No soft pair looks like the same underlying defect or change.", + "summary": "This cluster is heterogeneous: it mixes runtime bugs, performance complaints, documentation issues, config/type-hint problems, and model-specific feature requests. None of the soft edges look like the same underlying bug/change, so they should all be rejected.", "confidence": 0.95, - "canonical_issue_reason": "No clear canonical issue: the items span different code paths and problem types, so collapsing them into one duplicate cluster would be unsafe.", + "canonical_issue_reason": "No safe canonical issue stands out; the items span unrelated topics rather than one duplicate bug or change.", "canonical_pr_reason": null, - "best_issue_reason": "None stands out as a global best issue; the cluster is too diverse for a single representative to cover the others.", + "best_issue_reason": "No single issue is globally best suited as the representative; the cluster is too mixed and the soft links are only superficial similarity.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:29870", "right": "issue:31515", "accept": false, - "reason": "Bart generation UnboundLocalError and slow checkpoint loading are unrelated failures." + "reason": "BART generate() UnboundLocalError vs slow from_pretrained checkpoint loading: different code paths and failure modes." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLFlow run-state reporting on failure is unrelated to from_pretrained checkpoint load latency." + "reason": "MLflow run-state reporting after failed training is unrelated to checkpoint load performance." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Slow checkpoint loading and tensor-parallel training bugs are different performance/runtime problems." + "reason": "Model loading slowdown and tensor-parallel training bug are distinct problems in different execution paths." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Model loading performance and forward-argument documentation confusion are not the same issue." + "reason": "Checkpoint load slowness is not the same issue as confusing forward-API documentation." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "FP8 inference support request and a Qwen3.5 MoE from_pretrained regression are different changes." + "reason": "Both involve Qwen3.5/MoE, but one is fp8 support and the other is a from_pretrained regression; not the same concrete fix." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Audio model continuous batching support and FP8 Qwen3-VL/Qwen3.5 support are separate feature gaps." + "reason": "Continuous batching audio-model support and fp8 Qwen inference support are unrelated feature requests." }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Mamba init_weights behavior and RoFormer attention-interface usage involve different model implementations." + "reason": "Mamba/Mamba-2 init_weights usage and RoFormer attention-interface implementation are separate model-specific issues." }, { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Mask2Former ignore_value handling and DeepSeek-V4 shared-expert gating are unrelated model bugs." + "reason": "Mask2Former ignore_value handling and DeepSeek-V4 shared-expert gating are unrelated code paths." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "A ModernBERT config type hint error is unrelated to tokenizer regex warning behavior." + "reason": "Incorrect type hint on ModernBERTConfig.norm_eps is unrelated to tokenizer regex warning behavior." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Mask2Former ignore_value usage and loss-function kwargs propagation are different training-loss issues." + "reason": "Unused ignore_value in Mask2Former and missing kwargs in loss calculation are different bugs in different functions." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Loss kwargs plumbing and RoFormer attention routing do not share the same code path." + "reason": "Loss-kwargs propagation and RoFormer attention-interface usage do not describe the same change." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "A model-specific ignore_value bug and a global tokenizer warning regression are not duplicates." + "reason": "Mask2Former ignore_value handling is unrelated to tokenizer regex pattern warnings." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "Mask2Former ignore_value handling and RoFormer attention-interface implementation are distinct model bugs." + "reason": "Different model implementations and different failure modes; no shared underlying bug." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Mask2Former ignore_value behavior and a quantization-option typo are unrelated." + "reason": "Ignore_value logic in Mask2Former is unrelated to a quantization-operation typo." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "ModernBERT norm_eps typing and an orig_conversion.quantization_operation typo are separate small fixes." + "reason": "A config type-hint issue and a typo in quantization code are not the same underlying change." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Passing kwargs into loss calculation and correcting a config type hint are not the same problem." + "reason": "Missing loss kwargs and an incorrect config type hint are unrelated." }, { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Questions about Llama4TextL2Norm semantics and ModernBERT norm_eps typing are unrelated." + "reason": "Question about Llama4TextL2Norm/RMSNorm is unrelated to ModernBERT norm_eps typing." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Mask2Former ignore_value logic and ModernBERT norm_eps typing are distinct issues." + "reason": "Mask2Former ignore_value handling and ModernBERT config typing are unrelated." } ] }, @@ -19512,52 +19563,52 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "a01fd6a2546de4bfa7a6d6dfbde5aa937180b78cbe4e0e9b38491a7b0d1719f7", + "prepared_review_unit_hash": "ecf5b3d41c47bf73ea7475bdc2ab72962db5cdb13682014c258abd10c67d6993", "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": "No duplicate cluster here: the three issues are unrelated feature requests/questions about tokenization truncation, learning-rate scheduling, and Wav2Vec2 word timing offsets.", + "summary": "These three issues are not duplicates and do not share the same underlying bug or feature request: tokenizer truncation behavior, an optimizer/scheduler feature request, and Wav2Vec2 word-boundary timestamps. The cluster is mixed-topic, so no soft edges should be accepted.", "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 10519 is the most concrete and broadly scoped of the three, but it still does not represent the others; the cluster has no true canonical issue because the items are unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Different subsystems and problem statements: tokenizer truncation behavior vs Wav2Vec2 word boundary timestamps." + "reason": "Different domains and problems: tokenizer truncation direction vs Wav2Vec2 word time offsets. No shared code-path or underlying bug." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "One is about truncation direction in tokenization; the other asks for ReduceLROnPlateau-like scheduler functionality." + "reason": "Unrelated feature requests: truncation behavior vs ReduceLROnPlateau-like scheduler functionality. Same library, but not the same issue." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "Completely different areas: training scheduler feature request vs audio model timing offsets." + "reason": "Completely different concerns: learning-rate scheduling vs speech model word boundary timestamps. No evidence of a common bug or change." } ] }, @@ -19589,7 +19640,7 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "e62ec9775631c47d3999c3fcaf01a8ef21672c1d92eb04e75b62db43a1fdeb01", + "prepared_review_unit_hash": "5a0bf21a6b1804fee52072e1fca44e5b84cbd1cdc076d5b649f1f66b7f1a9833", "component_budget": { "node_count": 3, "item_count": 3, @@ -19611,24 +19662,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three issues are only loosely related by model/tokenizer loading, but they describe different problems: tokenization mismatch, auth-related download errors, and a generic model load failure. I do not see a true duplicate pair among them.", - "confidence": 0.92, - "canonical_issue_reason": "issue #16618 is the broadest and most central report in the set, but it still does not match the others closely enough to treat as the same bug.", + "summary": "These issues do not look like duplicates. One concerns Tapas tokenization behavior, one concerns a bad authentication error for private model downloads, and one concerns failing to load bert-base-uncased. They touch different failure modes and likely different code paths.", + "confidence": 0.93, + "canonical_issue_reason": "No strong duplicate set here; if one issue must be chosen as the representative, #16618 is the broadest and most discussed model-loading report, but it is still distinct from the other two.", "canonical_pr_reason": null, - "best_issue_reason": "issue #16618 is the best cluster anchor because it is the most general user-facing model-loading complaint; the other two are separate failure modes and should not be merged into it as duplicates.", + "best_issue_reason": "#16618 is the best standalone issue because it is the clearest user-facing model loading failure with the most general scope among the three.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tapas tokenization behavior is a different code path and symptom than failing to load bert-base-uncased; this is not the same underlying bug." + "reason": "Tapas tokenization mismatch is a tokenizer/code-path problem, while #16618 is about loading bert-base-uncased; these are different underlying bugs." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "One report is about a bad login/download error message for private models, the other is a generic load-model failure; related area, but not the same issue." + "reason": "#15980 is specifically about an incorrect error message for private model access without login, not the same failure as being unable to load bert-base-uncased." } ] }, @@ -19658,7 +19709,7 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "b6be70a06339e221f667c2d05606b1dc5dedcd5fe45abe4302108386c1a18a13", + "prepared_review_unit_hash": "22884d158073538dc85ac613e3555c0587c8df0a6567698eedff35c8aae7c259", "component_budget": { "node_count": 2, "item_count": 2, @@ -19680,18 +19731,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two different community contribution requests in unrelated areas (Flash Attention 2 vs GGUF). They do not look like duplicate reports or the same underlying change.", + "summary": "These are two separate feature requests about different architecture-support workstreams: Flash Attention 2 support vs GGUF support. They share a generic 'community contribution' framing but not the same underlying bug or change.", "confidence": 0.97, - "canonical_issue_reason": null, + "canonical_issue_reason": "Issue 26350 is a concrete, long-running request for Flash Attention 2 support across more architectures, which is self-contained and distinct from the GGUF request.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 26350 is the stronger canonical pick only because it is the earlier, more established thread with substantial discussion; it should not be treated as a duplicate of 33260.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different feature requests for different model/runtime support areas; no shared bug or concrete change path." + "reason": "Different features and code paths: Flash Attention 2 support versus GGUF support. Same broad theme, but not the same underlying change or bug." } ] }, @@ -19756,214 +19807,214 @@ "issue:46222|issue:46931", "issue:46226|issue:46931" ], - "prepared_review_unit_hash": "0b47b18a0db04d3f60c9f260ebf8722a4b31fa72872c9d47011a1cded183beda", + "prepared_review_unit_hash": "7448c522d9ee16c722b40ea33b1d6ac87ea69b85b74a4083bfaff67491e5c655", "component_budget": { "node_count": 9, "item_count": 9, "soft_pair_count": 29, - "serialized_chars": 15979, - "estimated_input_tokens": 3995, - "estimated_eval_tokens": 8246 + "serialized_chars": 15984, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 }, "budget": { "node_count": 9, "item_count": 9, "soft_pair_count": 29, - "serialized_chars": 15979, - "estimated_input_tokens": 3995, - "estimated_eval_tokens": 8246 + "serialized_chars": 15984, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a mix of translation-doc request issues for different languages, which should not be deduped together. The only strong duplicate candidates are the exact placeholder-titled issues and the exact Thai-title pair.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43834 is the oldest exact match in the generic placeholder-titled translation-doc subcluster, so it is the best representative for that duplicate set.", + "summary": "This cluster is mostly separate docs-translation requests by language, not one shared issue. The only clear duplicates are the near-identical placeholder-template issues and the exact duplicate Thai-titled pair.", + "confidence": 0.83, + "canonical_issue_reason": "issue:43834 is the earliest generic template-style issue and is the best representative of the exact placeholder subgroup.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43834 is the cleanest representative issue: it is generic, earliest, and matches the other placeholder-titled duplicates closely.", + "best_issue_reason": "issue:43834 is the most canonical representative because it uses the generic translation-request template and predates the other near-identical placeholder issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Different language request (Turkish vs generic placeholder); same topic is not enough to dedupe." + "reason": "Different language request (Turkish vs generic placeholder), so not the same underlying issue." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Different language request; these are separate translation tasks." + "reason": "Different language request; only broadly similar translation intent." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Turkish-specific request does not match the generic placeholder issue as the same underlying change." + "reason": "Turkish request does not match the generic placeholder issue closely enough to be a duplicate." }, { "left": "issue:27088", "right": "issue:38435", "accept": false, - "reason": "Different language request (Turkish vs Romanian)." + "reason": "Different target language (Turkish vs Romanian)." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Different language request (Turkish vs Spanish)." + "reason": "Different target language (Turkish vs Spanish)." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish-specific request is not the same underlying issue as the generic placeholder issue." + "reason": "Spanish translation request is distinct from the generic placeholder issue." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Different language request; not a duplicate." + "reason": "Different language request; not the same issue." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish vs Romanian translation requests are distinct." + "reason": "Different target language (Spanish vs Romanian)." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Different language request; only broadly similar." + "reason": "Spanish request and generic placeholder issue are not clearly the same underlying request." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Romanian-specific request does not match the generic placeholder issue as the same underlying task." + "reason": "Romanian translation request is distinct from the generic placeholder issue." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Different language request; not the same issue." + "reason": "Romanian request and generic placeholder issue are not duplicates." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "Romanian request and generic placeholder issue are not the same concrete translation task." + "reason": "Romanian request differs from the generic template issue." }, { "left": "issue:38435", "right": "issue:46222", "accept": false, - "reason": "Different language/title; these are not clearly the same underlying request." + "reason": "Different titles and no evidence they are the same translation request." }, { "left": "issue:38435", "right": "issue:46226", "accept": false, - "reason": "Different language/title; do not merge." + "reason": "Different titles and no concrete sign of duplication." }, { "left": "issue:43834", "right": "issue:44370", "accept": true, - "reason": "Same generic placeholder translation-doc issue; bodies/titles are effectively the same duplicate request." + "reason": "Near-identical generic placeholder issues with extremely high body similarity." }, { "left": "issue:43834", "right": "issue:44365", "accept": true, - "reason": "Same generic placeholder translation-doc issue; appears to be an exact duplicate." + "reason": "Near-identical generic placeholder issues; looks like the same template issue duplicated." }, { "left": "issue:43834", "right": "issue:46222", "accept": false, - "reason": "Generic placeholder issue vs a distinct Thai-titled issue; not clearly the same request." + "reason": "Titles differ materially; body similarity alone is not enough to prove the same issue." }, { "left": "issue:43834", "right": "issue:46226", "accept": false, - "reason": "Generic placeholder issue vs a distinct Thai-titled issue; not clearly the same request." + "reason": "Different title/content signal; not safely the same issue." }, { "left": "issue:43834", "right": "issue:46931", "accept": false, - "reason": "Placeholder-titled translation request and Thai-titled issue are not the same underlying task." + "reason": "Different title and no clear shared request despite similar body template." }, { "left": "issue:44365", "right": "issue:44370", "accept": true, - "reason": "Same generic placeholder translation-doc issue; looks like a duplicate." + "reason": "Near-exact duplicate placeholder issues with matching translation-template wording." }, { "left": "issue:44365", "right": "issue:46222", "accept": false, - "reason": "Different titles/content language; not enough to treat as the same issue." + "reason": "Different title and not enough evidence of the same underlying request." }, { "left": "issue:44365", "right": "issue:46226", "accept": false, - "reason": "Different titles/content language; not a duplicate." + "reason": "Different title and no explicit link to the same issue." }, { "left": "issue:44365", "right": "issue:46931", "accept": false, - "reason": "Different language/title; distinct issue." + "reason": "Different title; similarity appears to be template-level only." }, { "left": "issue:44370", "right": "issue:46222", "accept": false, - "reason": "Generic placeholder issue vs Thai-titled issue; not the same concrete request." + "reason": "Different title and no clear proof of duplication." }, { "left": "issue:44370", "right": "issue:46226", "accept": false, - "reason": "Generic placeholder issue vs Thai-titled issue; not the same concrete request." + "reason": "Different title and no concrete same-request evidence." }, { "left": "issue:44370", "right": "issue:46931", "accept": false, - "reason": "Different titles and apparent language targets; do not merge." + "reason": "Different title and only loose body similarity." }, { "left": "issue:46222", "right": "issue:46226", "accept": true, - "reason": "Exact same Thai-titled issue text and body shape; strong duplicate." + "reason": "Exact same title and effectively identical issue text, so this is a duplicate." }, { "left": "issue:46222", "right": "issue:46931", "accept": false, - "reason": "Different title/content; not the same underlying issue." + "reason": "Both are Thai-titled, but the titles differ and duplication is not certain." }, { "left": "issue:46226", "right": "issue:46931", "accept": false, - "reason": "Different title/content; not a duplicate." + "reason": "Different title; similar body template is not enough to merge them." } ] }, "evaluator_result": { "accept": true, - "feedback": "The clustering is grounded in the packet: the Turkish, Spanish, and Romanian issues are plausibly separate language-specific requests, while the placeholder-titled issues form a tight duplicate group and the two identical Thai-titled issues form another. The canonical representative choice of issue:43834 is reasonable as the earliest placeholder-titled item, and the soft-edge verdicts stay conservative rather than overclaiming duplicates across different language targets." + "feedback": "The summary is grounded in the packet: the cluster is dominated by language-specific docs-translation requests, while the placeholder-template issues and the Thai exact-match pair are the only clearly duplicate-looking groups. The accepted soft-edge verdicts are conservative overall, and the rejected edges avoid overclaiming across distinct language requests or materially different titles." }, "evaluator_used": true, "retried": false, @@ -20056,12 +20107,12 @@ "issue:42503|issue:46653", "issue:45027|issue:46653" ], - "prepared_review_unit_hash": "2f0da15c07ed4f53e681a74f9db4b6ba6b76186c73657c0408809d4635c1b2ba", + "prepared_review_unit_hash": "bdc2d7cb70aecfa353116e38c305d9591a20538cf2598297bb5c6c6d2172560e", "component_budget": { "node_count": 21, "item_count": 21, "soft_pair_count": 48, - "serialized_chars": 28613, + "serialized_chars": 28616, "estimated_input_tokens": 7154, "estimated_eval_tokens": 14564 }, @@ -20069,7 +20120,7 @@ "node_count": 21, "item_count": 21, "soft_pair_count": 48, - "serialized_chars": 28613, + "serialized_chars": 28616, "estimated_input_tokens": 7154, "estimated_eval_tokens": 14564 }, @@ -20078,276 +20129,258 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated model-request issues. The titles all look similar because they follow the same \"Add/support X\" template, but they target different models, modalities, and code paths, so I do not see true duplicates here.", - "confidence": 0.9, - "canonical_issue_reason": "Issue 36879 is the most central generic model-add request in the set, but it is only a representative, not a true duplicate target; the surrounding issues are mostly different models or modalities.", + "summary": "This cluster is noisy: most items are separate model-request issues across different model families and modalities, not true duplicates. The soft similarities are mostly template-level ('Add/Support X') rather than the same underlying change. No PRs are present.", + "confidence": 0.93, + "canonical_issue_reason": "No single canonical duplicate stands out; the issues span unrelated requests (BERT-like configs, audio ASR/TTS, vision-language encoders, keypoint models, sparse transformers).", "canonical_pr_reason": null, - "best_issue_reason": "Issue 36879 is the best representative because it is the most connected soft-similarity hub and has the broadest generic \"Add model\" framing, even though the cluster is not actually duplicate-heavy.", + "best_issue_reason": "If a representative issue is needed, 46133 is the most detailed and discussion-heavy request, but it is still not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36718", "right": "issue:47004", "accept": false, - "reason": "Related keypoint tools, but RoMa vs LoMa are different model requests, not the same change." + "reason": "Both are keypoint requests, but RoMa matcher and LoMa detector/matcher are different models and changes." }, { "left": "issue:36718", "right": "issue:41115", "accept": false, - "reason": "Keypoint matcher vs MiniCPM3 are unrelated model requests." + "reason": "RoMa keypoint matcher vs MiniCPM3 architecture; unrelated changes." }, { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "EuroBert config addition and DEIMv2 are different model targets." + "reason": "EuroBert config addition and DEIMv2 are different model families." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "BERT config work vs Qwen3-TTS support are different domains." + "reason": "EuroBert config vs Qwen3-TTS support; different modality and code path." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "Both are model additions, but EuroBert config and RF-DETR are distinct models and code paths." + "reason": "EuroBert config and RF-DETR are unrelated model additions." }, { "left": "issue:36817", "right": "issue:42954", "accept": false, - "reason": "EuroBert config and XiaoMi MiMo are different model requests." + "reason": "EuroBert config vs XiaoMi MiMo; same request pattern, different model." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are separate vision models, not one underlying bug/change." + "reason": "RF-DETR and DEIMv2 are distinct detection/vision model requests." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR and LFM2.5 Audio are different model families and implementations." + "reason": "RF-DETR vs LFM2.5 Audio; unrelated modality." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR and circuit-sparsity/CircuitGPT are unrelated." + "reason": "RF-DETR vs weight-sparse transformer; different code paths." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR and Qwen3-ASR are different modality/model support requests." + "reason": "RF-DETR and Qwen3-ASR are unrelated model-support requests." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR and Qwen3-TTS are separate requests." + "reason": "RF-DETR vs Qwen3-TTS; different model family and task." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR are unrelated model additions." + "reason": "NeoBERT and Qwen3-ASR are unrelated additions." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT and circuit-sparsity/CircuitGPT do not share the same change." + "reason": "NeoBERT vs sparse transformer; same broad request style, not same change." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT and LFM2.5 Audio are different model requests." + "reason": "NeoBERT vs LFM2.5 Audio; different subsystem." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are distinct models." + "reason": "RF-DETR and NeoBERT are different model requests." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are separate model support requests." + "reason": "NeoBERT vs ModernVBERT models; both BERT-ish, but distinct requests." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 and Qwen3-ASR are different model/mode changes." + "reason": "MiniCPM3 architecture vs Qwen3-ASR support; unrelated." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 and circuit-sparsity/CircuitGPT are unrelated." + "reason": "MiniCPM3 architecture and sparse transformer are different changes." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are distinct models." + "reason": "MiniCPM3 and ModernVBERT target different model families." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 and Voxtral TTS support are different code paths." + "reason": "MiniCPM3 vs Voxtral TTS support; different tasks." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are different requests." + "reason": "DEIMv2 and Qwen3-TTS are unrelated model-support requests." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are not the same model or bug." + "reason": "ModernVBERT models vs Qwen3-ASR; different model and modality." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are unrelated model support requests." + "reason": "ModernVBERT and Qwen3-TTS are not the same change." }, { "left": "issue:42503", "right": "issue:42954", "accept": false, - "reason": "ModernVBERT and XiaoMi MiMo are different models." + "reason": "ModernVBERT models vs XiaoMi MiMo; unrelated." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "RoPE-capable BERT-like models and Voxtral TTS are unrelated." + "reason": "BERT-like models with RoPE vs Voxtral TTS support; different code paths." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT and BERT-like models with RoPE are broader category overlap, but not the same concrete request." + "reason": "Both are BERT-related, but ModernVBERT and RoPE support are separate requests." }, { "left": "issue:42738", "right": "issue:42954", "accept": false, - "reason": "BERT-like RoPE support and MiMo are different model additions." + "reason": "BERT-like RoPE support vs MiMo request; unrelated." }, { "left": "issue:42738", "right": "issue:46653", "accept": false, - "reason": "BERT-like RoPE support and a person-named issue are not the same underlying change." + "reason": "BERT/RoPE feature request vs a person-named issue; no shared change." }, { "left": "issue:42954", "right": "issue:44205", "accept": false, - "reason": "XiaoMi MiMo and SAM3-LiteText are different model requests." + "reason": "MiMo and SAM3-LiteText are different model additions." }, { "left": "issue:42954", "right": "issue:43671", "accept": false, - "reason": "MiMo and Qwen3-TTS are different model domains." + "reason": "MiMo vs Qwen3-TTS; different model families." }, { "left": "issue:42954", "right": "issue:45177", "accept": false, - "reason": "MiMo and DeepSeek-OCR2 are unrelated model additions." + "reason": "MiMo and DeepSeek-OCR2 are unrelated requests." }, { "left": "issue:42954", "right": "issue:46653", "accept": false, - "reason": "MiMo and a person-named issue are clearly not the same change." + "reason": "MiMo vs person-named issue; clearly unrelated." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are related by family, but they are distinct support requests and code paths." + "reason": "Both are speech-related, but TTS and ASR support are different changes." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS and Voxtral TTS are different TTS model support requests." + "reason": "Both are TTS-ish model requests, but Qwen3-TTS and Voxtral TTS are different models." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated." + "reason": "Qwen3-TTS vs DeepSeek-OCR2; unrelated." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS and circuit-sparsity/CircuitGPT are unrelated." + "reason": "Qwen3-TTS and sparse transformer are different feature areas." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio are both audio, but they are different models and implementations." + "reason": "Both touch audio, but Qwen3-ASR and LFM2.5 Audio are different model-specific requests." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR and EUPE are different model requests." + "reason": "Qwen3-ASR vs EUPE; unrelated." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR and circuit-sparsity/CircuitGPT are unrelated." - }, - { - "left": "issue:43909", - "right": "issue:44205", - "accept": false, - "reason": "LFM2.5 Audio and SAM3-LiteText are different model requests." - }, - { - "left": "issue:43909", - "right": "issue:46645", - "accept": false, - "reason": "LFM2.5 Audio and Granite SWA+sink are not the same code-path problem." - }, - { - "left": "issue:44121", - "right": "issue:45322", - "accept": false, - "reason": "Circuit-sparsity/CircuitGPT and EUPE are unrelated model requests." + "reason": "ASR support vs sparse transformer; different code paths." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral TTS are different audio models and code paths." + "reason": "Qwen3-ASR and Voxtral TTS are different speech-model requests." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral TTS are unrelated." + "reason": "NeoBERT vs Voxtral TTS; unrelated model families." }, { "left": "issue:43837", @@ -20359,19 +20392,19 @@ "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE and TIPSv2 are both vision-language requests, but they are different model additions, not one mergeable fix." + "reason": "Both are vision-language/perception model requests, but EUPE and TIPSv2 are distinct models and changes." }, { "left": "issue:42503", "right": "issue:46653", "accept": false, - "reason": "ModernVBERT and a person-named issue are clearly not duplicates." + "reason": "ModernVBERT models vs person-named issue; no shared change." }, { "left": "issue:45027", "right": "issue:46653", "accept": false, - "reason": "Voxtral TTS support and a person-named issue are unrelated." + "reason": "Voxtral TTS support vs person-named issue; unrelated." } ] }, @@ -20401,7 +20434,7 @@ "soft_pairs": [ "issue:38740|issue:46443" ], - "prepared_review_unit_hash": "a14c81fa9f0a79e7adb0ccfdeb230fcee3f3ed7dc89281d7e311aa37bf6b50a2", + "prepared_review_unit_hash": "6d1f0b8215a8bbdb298a07b5b4ed40bd8aa1b4929666e549d65c9168b044d331", "component_budget": { "node_count": 2, "item_count": 2, @@ -20423,18 +20456,18 @@ "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, the other is a Vision RoPE refactor. They do not describe the same underlying problem.", + "summary": "These two issues are unrelated: one is a closed docs request about adding `pruna` as an optimization framework, the other is an open technical issue about a Vision RoPE refactor. They should not be treated as duplicates or merged into one cluster.", "confidence": 0.98, - "canonical_issue_reason": "If forced to pick an anchor, issue 46443 is the substantive technical topic; issue 38740 is a separate documentation request and not the same problem.", + "canonical_issue_reason": "No canonical issue: the titles and contexts point to different problems in different areas (documentation request vs. vision model refactor).", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46443 is the better standalone issue because it describes an actual code refactor, but it is not a duplicate of 38740.", + "best_issue_reason": "Issue 46443 is the more substantive active technical issue, while 38740 is a closed documentation request and not a close match.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:38740", "right": "issue:46443", "accept": false, - "reason": "Different domains and intent: documentation/framework listing versus Vision RoPE code refactor. No shared bug or change." + "reason": "Reject: these do not describe the same underlying bug or change. One is a docs enhancement request; the other is a Vision RoPE refactor." } ] }, @@ -20464,7 +20497,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "b717b459afcfb4e929b52a39cd888f066e02c0c3af2ada8a0dabb95955e86c83", + "prepared_review_unit_hash": "7f0c79329c28c999d75fce71ef63435d87765e2c29415f54def2c0fc45846424", "component_budget": { "node_count": 2, "item_count": 2, @@ -20486,18 +20519,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are not duplicates: one is about block decoding, the other about DeepSeek v3.2 model support. They target different feature areas with no clear shared underlying bug or change.", + "summary": "These two issues are not duplicates: one is about block decoding support, while the other is about DeepSeek v3.2 model support. They appear to be unrelated feature requests.", "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good canonical choice for the other; they describe unrelated requests.", - "best_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": null, + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different concerns: block decoding vs DeepSeek v3.2 support. No evidence they describe the same underlying bug or change." + "reason": "Different underlying requests with no shared target or code path: block decoding vs. DeepSeek v3.2 support." } ] }, @@ -20527,40 +20560,40 @@ "soft_pairs": [ "issue:41306|issue:41826" ], - "prepared_review_unit_hash": "62a3662a99dab25dd2541cd07ff75dd6d45ea950da269bc94a1d8cce1c6dcab6", + "prepared_review_unit_hash": "6fa336f35c2683d18964d54fd7f8e974e2eb51b98697d1d714634662133f280d", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1585, - "estimated_input_tokens": 397, - "estimated_eval_tokens": 1050 + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1585, - "estimated_input_tokens": 397, - "estimated_eval_tokens": 1050 + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a very broad \u2018tiled model components\u2019 level, but they describe different feature requests in different code paths: FLCE/`from_pretrained()` flag handling versus TiledMLP memory footprint integration. They should not be treated as duplicates.", + "summary": "The two issues are about different model components and different changes: one requests a `from_pretrained()` flag for TiledFusedLogitsLoss/FLCE, the other asks for TiledMLP integration to reduce memory usage. They are not duplicates.", "confidence": 0.97, - "canonical_issue_reason": "#41306 is the more concrete, implementation-facing request centered on a specific API path (`from_pretrained()` flag for TiledFusedLogitsLoss/FLCE).", + "canonical_issue_reason": "Issue 41306 is the more specific, self-contained request, with a clear API target (`from_pretrained()`) and a concrete feature flag change.", "canonical_pr_reason": null, - "best_issue_reason": "#41306 is the best representative issue because it is narrowly scoped and actionable; #41826 is a separate feature request about TiledMLP memory reduction, not the same underlying problem.", + "best_issue_reason": "Issue 41306 is the best representative issue because it names an exact integration point and a narrowly scoped user-facing change, making it easier to triage and implement.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41306", "right": "issue:41826", "accept": false, - "reason": "Different underlying requests and code paths: one concerns FLCE/TiledFusedLogitsLoss flag support in `from_pretrained()`, the other asks to integrate TiledMLP for lower memory usage." + "reason": "Different underlying requests: FLCE/TiledFusedLogitsLoss flag handling in `from_pretrained()` versus adding TiledMLP for lower memory footprint. No evidence they describe the same bug or change." } ] }, @@ -20590,12 +20623,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "d26a1addd7685b67065af97c68161847f88275a68525d9488bfa85811421ab62", + "prepared_review_unit_hash": "7f5c737f958b8035fe8472e168a1a5c0d70488ac5704c25bf49b3b1b9383e701", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, + "serialized_chars": 1582, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -20603,7 +20636,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1584, + "serialized_chars": 1582, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -20612,18 +20645,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are only loosely related through Trainer customization, but they target different problems: one asks to revisit `remove_unused_column`, the other asks to handle custom trainers in FLOPs estimation. They should not be treated as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 43598 is the better cluster anchor because it has a concrete Trainer behavior target (`remove_unused_column`) and a clearer change request than the more general FLOPs-estimation concern in 43600.", + "summary": "The two closed issues are related only at a broad Trainer/custom-trainer level, but they describe different problems: one is about `remove_unused_column` customizability, the other about FLOPS estimation for custom trainers. Not duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43598 is the broader, more general Trainer behavior discussion; issue 43600 is a separate FLOPS-estimation bug and should not be merged into it.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43598 is the most actionable standalone item: it identifies a specific customization regression/behavior to revisit, whereas 43600 is a separate metrics-accounting request.", + "best_issue_reason": "43598 is the best single representative of the cluster because it centers on Trainer customizability, which is the broader theme, while 43600 is narrower and distinct.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Different underlying problems: `remove_unused_column` customizability vs estimating FLOPs for custom trainers. Related subsystem, but not the same bug or change." + "reason": "Reject: they share a custom-trainer theme, but the concrete problems differ (`remove_unused_column` behavior vs FLOPS counting). They do not look like the same underlying bug." } ] }, @@ -20653,7 +20686,7 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "e5409eaee1f036331809bd20d3d7223cb8667e6bb043acf9de0c963120d85dda", + "prepared_review_unit_hash": "efa2909250967205293f9a0046c42ab4985c12aa85e88035cf2de7845c1fdfa2", "component_budget": { "node_count": 2, "item_count": 2, @@ -20675,18 +20708,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are distinct feature requests: one asks for Qwen3-Omni registration in AutoModel/AutoModelForConditionalGeneration, while the other asks for AutoModelForSequenceClassification support for Qwen3.5. They should not be deduplicated.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "These are related only at a very broad \u201cQwen model support\u201d level, but they target different model families and different API surfaces: one requests registration for Qwen3-Omni in base/generative auto classes, the other requests sequence-classification support for Qwen3.5. They should not be treated as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43684 is the broader, more central request: add AutoModel / AutoModelForConditionalGeneration registration for Qwen3-Omni. It represents the core model-registration gap more directly than the narrower task-head request.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 43684 is the best representative of the cluster because it is the broader model-registration item and more likely to generalize as the main support request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different model families and different auto-class targets; this is not the same underlying bug/change." + "reason": "Different underlying changes: Qwen3-Omni base/generative auto-registration versus Qwen3.5 sequence-classification support. They involve different configs, different auto classes, and different code paths." } ] }, @@ -20723,12 +20756,12 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "cf768d9ed1f3e939db22993ca1bf8c26e8f16d0ef40df07c104bbdad2195cd97", + "prepared_review_unit_hash": "12b0a4205b4ff9e4c8d7328fd85e0fe3f8dd12b63edbd4c7c7e709c1cdb1f616", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4266, + "serialized_chars": 4267, "estimated_input_tokens": 1067, "estimated_eval_tokens": 2390 }, @@ -20736,7 +20769,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4266, + "serialized_chars": 4267, "estimated_input_tokens": 1067, "estimated_eval_tokens": 2390 }, @@ -20745,54 +20778,54 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "One issue (#43835, \"Fett\") appears unrelated. The three \"huggingface\" issues (#43845, #43846, #43859) look like duplicates of the same report, with #43845 the best canonical candidate because it is the earliest of the group.", - "confidence": 0.78, - "canonical_issue_reason": "#43845 is the earliest of the near-identical \"huggingface\" issues and is the best representative for that duplicate set.", + "summary": "Three issues share the exact same title, \"huggingface \", and are likely duplicates of one another; the \"Fett\" issue appears unrelated.", + "confidence": 0.82, + "canonical_issue_reason": "Issue 43845 is the earliest of the matching \"huggingface \" reports, so it is the best canonical representative for that duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "#43845 is the strongest issue to keep as the cluster representative: same title as the other duplicate candidates, earliest timestamp, and no evidence suggesting a different underlying problem.", + "best_issue_reason": "Issue 43845 is the strongest overall issue candidate because it is the earliest clear duplicate cluster member and has the same exact title as the other matching reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Different titles and no body/content evidence tying \"Fett\" to the \"huggingface\" reports." + "reason": "Different titles and no other evidence of the same underlying problem." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Different titles and no body/content evidence tying \"Fett\" to the \"huggingface\" reports." + "reason": "Different titles and no shared issue-specific details tying them to the same bug." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Different titles and no body/content evidence tying \"Fett\" to the \"huggingface\" reports." + "reason": "Different titles; this looks unrelated to the repeated \"huggingface \" reports." }, { "left": "issue:43845", "right": "issue:43846", "accept": true, - "reason": "Same title, very close creation time, and no evidence of a distinct underlying issue." + "reason": "Same exact title and very close creation times, with no evidence suggesting different problems." }, { "left": "issue:43845", "right": "issue:43859", "accept": true, - "reason": "Same title and same apparent report theme; likely duplicate submissions of the same issue." + "reason": "Same exact title and likely the same duplicate report family." }, { "left": "issue:43846", "right": "issue:43859", "accept": true, - "reason": "Same title and same apparent report theme; likely duplicate submissions of the same issue." + "reason": "Same exact title and same apparent underlying issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The assessment is grounded in the packet and stays reasonably conservative: #43835 is clearly the outlier by title and low similarity, while the three \"huggingface\" issues share the same title and stronger soft-similarity signals, making #43845 a plausible canonical representative by earliest timestamp. The soft-edge verdicts do not overstate the evidence." + "feedback": "The summary is grounded in the packet: 43845, 43846, and 43859 all share the exact title \"huggingface \", while 43835 is clearly different. The canonical choice of 43845 is conservative and supported by it being the earliest of the matching set. The soft-edge verdicts are also appropriately cautious given only soft-similarity evidence." }, "evaluator_used": true, "retried": false, @@ -20834,12 +20867,12 @@ "issue:44573|issue:47036", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "a76ffcb8f1f4faf98e8ae455d100e4eb0db02c3ab7ca48b9d50367c5ea78ba7c", + "prepared_review_unit_hash": "66516f9754ac9d7750d8e7a7ffe49da188fbf8b435fb32a32005169dd0ef6ebd", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 12, - "serialized_chars": 7352, + "serialized_chars": 7349, "estimated_input_tokens": 1838, "estimated_eval_tokens": 3932 }, @@ -20847,7 +20880,7 @@ "node_count": 6, "item_count": 6, "soft_pair_count": 12, - "serialized_chars": 7352, + "serialized_chars": 7349, "estimated_input_tokens": 1838, "estimated_eval_tokens": 3932 }, @@ -20856,84 +20889,84 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster does not show a real duplicate pattern; it is mostly spam/placeholder issues with only superficial textual similarity.", - "confidence": 0.94, - "canonical_issue_reason": "No clear canonical issue: the items are short, mostly spam-like, and do not describe the same underlying bug or request.", + "summary": "These issues do not form a duplicate cluster; they look like unrelated spam/low-information posts mixed with a few generic suggestions.", + "confidence": 0.96, + "canonical_issue_reason": "No canonical issue: the set mixes spam placeholders, a docs suggestion, a benchmark tip, and a Google LLC mention with no shared concrete bug/request.", "canonical_pr_reason": null, - "best_issue_reason": "issue:47036 is the most plausible legitimate report in the set, but it is still too thin to serve as a strong duplicate anchor.", + "best_issue_reason": "44418 is the only minimally descriptive non-spam issue, but it still does not match the others closely.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are spammy placeholders with no substantive shared problem beyond generic similarity." + "reason": "Both are just \"\" placeholders; no concrete shared bug or change is described." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Same: short spam-like issues, not the same bug or request." + "reason": "Both are just \"\" placeholders; no concrete shared bug or change is described." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam, the other is a documentation suggestion; no shared underlying issue." + "reason": "Spam placeholder vs. documentation enhancement suggestion; unrelated." }, { "left": "issue:44371", "right": "issue:47036", "accept": false, - "reason": "Spam placeholder vs. a community tip; unrelated topics." + "reason": "Spam placeholder vs. community benchmark tip; unrelated." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder vs. a named-topic issue; not the same report." + "reason": "Spam placeholder vs. \"Check out \\\"Google LLC\\\"\"; unrelated." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "Documentation suggestion and spam placeholder are not duplicates." + "reason": "Documentation enhancement vs. spam placeholder; no shared underlying request." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "Different intent and no concrete shared bug or change." + "reason": "Documentation enhancement vs. spam placeholder; no shared underlying request." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are spam-like placeholders; similarity is purely superficial, not a shared underlying issue." + "reason": "Both are spam placeholders, but that is not enough to establish the same underlying issue." }, { "left": "issue:44572", "right": "issue:47036", "accept": false, - "reason": "Spam placeholder vs. a specific community tip; unrelated." + "reason": "Spam placeholder vs. community benchmark tip; unrelated." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "No evidence they describe the same problem; titles/body are unrelated." + "reason": "Spam placeholder vs. \"Check out \\\"Google LLC\\\"\"; unrelated." }, { "left": "issue:44573", "right": "issue:47036", "accept": false, - "reason": "Spam placeholder and community tip are not duplicates." + "reason": "Spam placeholder vs. community benchmark tip; unrelated." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "No concrete overlap in bug, request, or code path." + "reason": "Spam placeholder vs. \"Check out \\\"Google LLC\\\"\"; unrelated." } ] }, @@ -20963,7 +20996,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "b86acf0532dce37a46e58f0fd2cbbd33f9261e5dd3462299cd61c6eebd4fd751", + "prepared_review_unit_hash": "3d492ccd4205fb5795844eb35d862144dc698003e7036abcfa29204ffdf1e7bd", "component_budget": { "node_count": 2, "item_count": 2, @@ -20985,28 +21018,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues share the exact title and very similar body size, but there is no stronger linkage (no explicit cross-references, shared targets, or files), so the evidence supports only loose similarity rather than a confident duplicate call.", - "confidence": 0.7, - "canonical_issue_reason": "If a single issue must be chosen, #45151 is the slightly better canonical thread because it has the only discussion activity (1 vs 0) while otherwise matching #45145 closely.", + "summary": "The two issues are highly similar and likely report the same underlying problem, but the evidence is limited to identical titles, similar body lengths, and soft-similarity signals. #45145 is the better canonical thread because it already has an inbound reference.", + "confidence": 0.81, + "canonical_issue_reason": "#45145 is a likely duplicate pairmate to #45151 and has the slight edge as the more established thread, with one inbound reference already pointing to it.", "canonical_pr_reason": null, - "best_issue_reason": "#45151 is the best representative for the same reason: it is marginally more active, but the pair is not strong enough to treat as a definitive duplicate set.", + "best_issue_reason": "#45145 is the best issue candidate because it appears to be the more central thread for this likely duplicate report and already has external attention.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": false, - "reason": "Same title and similar body length are suggestive, but there is no stronger evidence tying them to the same underlying issue; with no explicit links, shared targets, or other distinct signals, this is too weak for a duplicate accept." + "accept": true, + "reason": "The titles are identical and the bodies are similarly sized, so they likely describe the same issue report; however, the match is based on limited evidence rather than a confirmed link." } ] }, "evaluator_result": { "accept": false, - "feedback": "The pair is plausibly related, but the analyst overstates the evidence in a few places. In particular, issue 45145 does not have more discussion signal than 45151 (it has 0 vs 1), and the packet only supports soft similarity plus matching titles/body lengths\u2014not \u201calmost exactly\u201d or a definitive duplicate conclusion. The soft-edge accept is not conservative enough for this packet." + "feedback": "The packet supports a strong similarity claim, but the analyst overstates certainty. We only have identical titles, similar body lengths, and a soft-similarity signal; there is no body text comparison, explicit link, or other hard evidence that these are definitively the same issue. The duplicate and canonical-thread conclusions should be phrased more conservatively as likely/possible rather than clear." }, "evaluator_used": true, "retried": true, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -21083,12 +21116,12 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "8351f08121fca2e5538b5d6ebc88e8c9ac7da761bdef6cdb69ca96d0ae27c862", + "prepared_review_unit_hash": "72f99de081f23f7720df49f59fbc5373cad39c85200061cc54974bd8411db1e4", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1476, + "serialized_chars": 1475, "estimated_input_tokens": 369, "estimated_eval_tokens": 994 }, @@ -21096,7 +21129,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1476, + "serialized_chars": 1475, "estimated_input_tokens": 369, "estimated_eval_tokens": 994 }, @@ -21105,18 +21138,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues are not duplicates: they are separate personal-name/date entries with no shared bug report content, targets, or discussion context.", + "summary": "The two closed issues are only superficially similar by title format; they refer to different names and different dates, with no shared context or linkage. They should not be treated as duplicates.", "confidence": 0.97, - "canonical_issue_reason": "No canonical issue: the items do not describe the same underlying problem.", + "canonical_issue_reason": "No canonical issue: the issues appear to be separate entries about different individuals, with no body text or evidence connecting them.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good duplicate representative because they appear unrelated to each other.", + "best_issue_reason": "Neither issue is a good canonical duplicate target; there is no substantive overlap beyond a similar naming pattern.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Different names and birthdates, with no shared issue content or explicit linkage; not the same underlying bug/report." + "reason": "Different names and dates with no contextual overlap; this is similarity in format only, not the same underlying issue." } ] }, @@ -21146,40 +21179,40 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "abcade60ace1943784eed56c272e5193914f080e3d6d70b69f70e1ac22d7eacd", + "prepared_review_unit_hash": "c6dd9cf117a2ba9100ebba25c336dee99b4065d67f34253e3e7aa44cfcb2f2b7", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, - "estimated_input_tokens": 370, - "estimated_eval_tokens": 996 + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, - "estimated_input_tokens": 370, - "estimated_eval_tokens": 996 + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two closed issues with only personal-name/date titles and no body or linked context; they do not appear to describe the same underlying bug or request.", - "confidence": 0.98, - "canonical_issue_reason": "No clear canonical issue: neither issue contains problem details, steps, or references that would establish a shared duplicate target.", + "summary": "Two closed issues with different personal-name/date titles and no body or linkage evidence; they do not appear to describe the same underlying problem.", + "confidence": 0.93, + "canonical_issue_reason": "No strong canonical choice: both issues are content-free and only contain distinct name/date strings, so neither is a better duplicate representative.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a good cluster anchor because both are content-free and unrelated on their face.", + "best_issue_reason": "Neither issue is a good global best representative because there is no bug report content, reproduction, or shared subject matter to establish a canonical topic.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "Only weak title similarity from both being name/date strings; no evidence they describe the same issue or any shared code/problem context." + "reason": "Reject: the titles refer to different people and different dates, with no shared description, targets, or evidence of the same underlying issue." } ] }, @@ -21209,7 +21242,7 @@ "soft_pairs": [ "issue:45810|issue:46295" ], - "prepared_review_unit_hash": "59b63663bc7b0c4e097db21498c2925eb961e0ba1d66281a52748e3b37d1039c", + "prepared_review_unit_hash": "cadb012a1858009c1b65f0524eaa44ac3352a0af434b80e89a5f70fdb296595d", "component_budget": { "node_count": 2, "item_count": 2, @@ -21231,18 +21264,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are unrelated feature requests: one asks for a new Qwen3_5 token-classification model class, the other for FP8 training support. They should not be treated as duplicates or merged.", - "confidence": 0.98, - "canonical_issue_reason": "Issue #46295 is the broader, higher-level request and has more inbound discussion, so it is the better representative if a single issue must be chosen.", + "summary": "These two issues are not duplicates: one asks for a new Qwen3_5 token-classification/value-model class, while the other requests FP8 training support. They target different features and code paths.", + "confidence": 0.96, + "canonical_issue_reason": "Issue #45810 is the more concrete, narrowly scoped feature request and is the better representative of the cluster if one issue must be picked.", "canonical_pr_reason": null, - "best_issue_reason": "#46295 is the more general product capability request, while #45810 is a narrow model-class addition.", + "best_issue_reason": "#45810 is the most actionable and specific item; #46295 is a separate training-capability request with different implementation scope.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45810", "right": "issue:46295", "accept": false, - "reason": "These are different feature areas: adding Qwen3_5ForTokenClassification versus training support for FP8. They do not describe the same underlying bug or change." + "reason": "Reject: these do not describe the same underlying bug or change. One is about adding a Qwen3_5 token classification/value-model class, the other is about FP8 training support." } ] }, @@ -21282,7 +21315,7 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "07b52ddd0e651acccbb6628624b52a896a777b06694383363cf4322c630ef6d3", + "prepared_review_unit_hash": "fe1aa7d9dea593e37968a33cebf8e1cdb8d460aee0146511a63f33c419466d2b", "component_budget": { "node_count": 5, "item_count": 5, @@ -21304,69 +21337,69 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This set is not one duplicate cluster: it splits into two exact-duplicate issue pairs (45870/45871 and 45872/45873), while 45869 is a separate dataset request and not the same underlying issue.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "This cluster appears to contain two highly similar issue pairs\u201445870/45871 and 45872/45873\u2014plus 45869 as a related but separate item. The evidence supports treating the two pairs as likely duplicates, but not calling them exact duplicates from the packet alone. 45869 is not the same underlying request.", + "confidence": 0.85, + "canonical_issue_reason": "45870 is a strong canonical anchor for the \"Example: Financial time series prediction with DataForge\" subcluster because it has an identical-title counterpart and sits in the most clearly duplicate-looking pair, while 45869 is unrelated and 45872/45873 form a separate similar pair.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "45870 is the best overall issue representative because it anchors one of the two strongest duplicate-like issue pairs and is more representative of the cluster than the unrelated 45869.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Different topics: a financial data bundle request versus an example for time-series prediction." + "reason": "Different artifacts: a dataset bundle promo versus a financial time-series example; related finance theme only, not the same issue." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Different topics: a financial data bundle request versus an example for time-series prediction." + "reason": "Different artifacts: dataset bundle promo versus an example issue; the shared financial-data context is too broad to treat as a duplicate." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Same title and apparent same content; this looks like a straight duplicate issue." + "reason": "Identical title and matching body length make this a likely duplicate pair representing the same underlying issue." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Same financial domain, but different requested artifact: example tutorial versus scraping-service/dataforge feature." + "reason": "They share financial-time-series/DataForge wording, but one is an example and the other is a scraping-service topic; not the same underlying request." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Same financial domain, but different requested artifact: example tutorial versus scraping-service/dataforge feature." + "reason": "Same broad domain, but the concrete topic differs from 45870's example/prediction request." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Different requested change; only broad theme overlaps." + "reason": "Related wording only; the concrete issue described is different." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Different requested change; only broad theme overlaps." + "reason": "Similar theme, but not the same specific issue or change request." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Same title and apparent same content; this looks like a straight duplicate issue." + "reason": "Identical title and matching body length make this a likely duplicate pair representing the same underlying issue." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 are the only clearly duplicate pairs, while 45869 is a different dataset request and does not match the others. The soft-edge verdicts are conservative and consistent with the titles/body sizes provided." + "accept": false, + "feedback": "The clustering summary is directionally consistent, but it overstates the evidence by calling 45870/45871 and 45872/45873 \"exact duplicate\" pairs. The packet only provides soft-similarity signals, identical titles, and matching body lengths\u2014not the bodies themselves\u2014so a more conservative phrasing would be \"highly similar\" or \"likely duplicates.\" The 45869 exclusion looks reasonable, but the duplicate certainty is stronger than the packet supports." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -21390,12 +21423,12 @@ "soft_pairs": [ "issue:46439|issue:46441" ], - "prepared_review_unit_hash": "601661b6a53f6e1c5917215136fecaeabcc050d546403eb28a8adca7454d7556", + "prepared_review_unit_hash": "792c6b893d7675c0c12aa65cdee999e5ed68ed466906c18143822d64195cc456", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1639, + "serialized_chars": 1640, "estimated_input_tokens": 410, "estimated_eval_tokens": 1076 }, @@ -21403,7 +21436,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1639, + "serialized_chars": 1640, "estimated_input_tokens": 410, "estimated_eval_tokens": 1076 }, @@ -21412,28 +21445,25 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both issues appear to describe the same static-cache/linear-attention initialization bug: one reports a missing `linear_attention` entry, the other reports corrupted shapes and a runtime error during `early_initialization`.", - "confidence": 0.78, - "canonical_issue_reason": "Issue 46439 is the better canonical issue because it describes the broader root cause (`early_initialization` corrupting linear-attention layers) rather than one model-specific symptom.", + "summary": "Both issues involve static-cache failures in linear-attention paths, but they describe different concrete symptoms: one is shape corruption during `early_initialization`, the other is a `KeyError` for `linear_attention` in Qwen3.5.", + "confidence": 0.87, + "canonical_issue_reason": "Issue 46439 is the better canonical representative because it describes the core static-cache/linear-attention failure in a more general, subsystem-level way and includes the concrete shape-corruption runtime error.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46439 is the strongest representative of the underlying bug since it frames the failure as a general static-cache initialization problem affecting linear-attention layers.", + "best_issue_reason": "Issue 46439 is the strongest standalone issue because it is more broadly applicable and more diagnostic of the underlying bug than the product-specific Qwen3.5 `KeyError` report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46439", "right": "issue:46441", - "accept": true, - "reason": "Both reports point to the same static cache / linear-attention initialization path, with different downstream failures (`KeyError` vs wrong shapes/runtime error)." + "accept": false, + "reason": "Reject: these look related to the same static-cache/linear-attention area, but the reported failures are different concrete bugs/symptoms (`RuntimeError` from wrong shapes vs `KeyError: 'linear_attention'`) and there is not enough evidence that one fix would plausibly cover both." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is supported by the packet: both issues are about static cache and linear-attention failures, and the soft-pair verdict stays reasonably conservative given the title-level evidence. There is no overstatement beyond the available signal." - }, - "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 }, @@ -21456,12 +21486,12 @@ "soft_pairs": [ "issue:46551|issue:46552" ], - "prepared_review_unit_hash": "14fc1683fbb90ad56e198acdba702d64be59667395357f42b7756871d5143d19", + "prepared_review_unit_hash": "d4ca78fd840ab4230e9e286d08e99f3d32d4202e942f346d1d1e658c65056f4d", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1706, + "serialized_chars": 1705, "estimated_input_tokens": 427, "estimated_eval_tokens": 1110 }, @@ -21469,7 +21499,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1706, + "serialized_chars": 1705, "estimated_input_tokens": 427, "estimated_eval_tokens": 1110 }, @@ -21478,24 +21508,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issue reports are effectively the same bug report with identical titles and body content; the open issue is the better canonical tracker.", + "summary": "The two issues are near-exact duplicates with the same title and body describing the same Wav2Vec2CTCTokenizer v5 special-token bug. The open issue with inbound references is the better canonical tracker.", "confidence": 0.99, - "canonical_issue_reason": "Issue 46552 is the same report as 46551 but remains open and has inbound references, so it is the better canonical issue to keep as the active tracker.", + "canonical_issue_reason": "Issue 46552 is the better canonical tracker because it is still open and already has inbound references, while 46551 is closed.", "canonical_pr_reason": null, - "best_issue_reason": "46552 is the best overall issue because it is the active open report for the exact same Wav2Vec2CTCTokenizer v5 special-token leak bug.", + "best_issue_reason": "Issue 46552 is the strongest representative of the cluster: it is open, matches the same problem statement exactly, and appears to be the active report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46551", "right": "issue:46552", "accept": true, - "reason": "Same title, same body, same underlying bug report; 46552 is just the open duplicate of 46551." + "reason": "Same title and same body describing the same tokenizer bug; these are duplicates of the same underlying issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The conclusion is grounded in the packet: the two issues have identical titles and body lengths, and the open issue 46552 has inbound references while 46551 is closed. The soft-edge judgment is reasonable for a same-issue match, though the wording would be slightly safer if it said \u201clikely duplicate\u201d rather than asserting duplication as explicit fact." + "feedback": "The verdict is grounded in the packet: the two issues have identical titles, identical body length, and a soft_similarity score of 1.0, so treating them as duplicates is conservative. Choosing the open issue with inbound references as the canonical tracker is also supported." }, "evaluator_used": true, "retried": false, @@ -21522,7 +21552,7 @@ "soft_pairs": [ "issue:46657|issue:46692" ], - "prepared_review_unit_hash": "74a38f2dc6adb9ae9ffa9efcbb8fc93e7f9caf0431358ec535cd84c4d16cf106", + "prepared_review_unit_hash": "70edf92b0407f7f7eb31553f758d6a9e4be7dc72664abed57964ca07f8bc59e9", "component_budget": { "node_count": 2, "item_count": 2, @@ -21544,18 +21574,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are only broadly related to object-detection training, but they describe different failure modes: a tensor shape mismatch versus out-of-memory after epochs in RT-DETR v2. They do not look like the same underlying bug.", - "confidence": 0.9, - "canonical_issue_reason": "Issue 46657 is the more specific and actionable report: it names the concrete error class (tensor size mismatch) and likely gives a clearer reproduction target.", + "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 training. The symptoms, likely failure modes, and model context are different.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 46657 is the better canonical issue because it describes a concrete training-time tensor shape mismatch, which is a specific bug report with clearer failure semantics than the broader OOM symptom in 46692.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 46657 is the best representative for the cluster only because it is the clearer bug report; however, the two issues are not duplicates.", + "best_issue_reason": "Issue 46657 is the best issue anchor for this cluster because it is more specific and likely easier to map to a single underlying code-path problem.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46657", "right": "issue:46692", "accept": false, - "reason": "Different symptoms and likely different root causes: one is a tensor-shape mismatch during training, the other is a memory leak/OOM after some epochs in RT-DETR v2. Shared training context is too broad to treat as the same bug." + "reason": "Reject: tensor size mismatch during object-detection training and post-epoch OOM in RT-DETR v2 are different failure modes; the titles do not indicate the same underlying bug." } ] }, @@ -21585,12 +21615,12 @@ "soft_pairs": [ "issue:46822|issue:46823" ], - "prepared_review_unit_hash": "ba0e8976f0c6ae06d007bc5d00957fc5094c1e7c164df2b2c91dd5097d44c8a7", + "prepared_review_unit_hash": "3459b776c9c289e70658a1ad5b6f4725eca1cd1c9429dc72c47e66d69e7393c9", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1672, + "serialized_chars": 1671, "estimated_input_tokens": 418, "estimated_eval_tokens": 1092 }, @@ -21598,7 +21628,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1672, + "serialized_chars": 1671, "estimated_input_tokens": 418, "estimated_eval_tokens": 1092 }, @@ -21607,18 +21637,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two distinct model-loading/device-placement bugs: one is a ZeRO-3 `zero.Init` partitioning failure for a composite multimodal submodule causing OOM, and the other is a 4-bit `device_map='auto'` / `max_memory` dispatch issue that incorrectly places modules on CPU and raises `ValueError`. They share only a broad theme of loading behavior.", - "confidence": 0.96, + "summary": "These are two different bug reports in different model-loading paths: ZeRO-3 zero.Init partitioning vs 4-bit device_map/max_memory dispatch. I would not merge them as duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 46823 is the cleaner representative of a concrete, user-facing loading regression with a precise repro path and clear expected-vs-actual behavior.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46822", "right": "issue:46823", "accept": false, - "reason": "Different underlying bugs and code paths: ZeRO-3 partitioning of a composite submodule versus 4-bit auto device placement with `max_memory`. They are not the same fix and would not plausibly merge into one issue." + "reason": "Different underlying problems and subsystems: 46822 is about ZeRO-3 failing to partition a composite submodule during multi-GPU load, while 46823 is about 4-bit loading with device_map='auto' and max_memory incorrectly placing modules on CPU. They do not look like the same bug." } ] }, @@ -21669,260 +21699,273 @@ "issue:46948|issue:46998", "issue:46969|issue:46985", "issue:46969|issue:47037", - "issue:46969|issue:47102", + "issue:46969|issue:47091", "issue:46985|issue:47037", - "issue:46985|issue:47102", + "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:47102", "issue:47037|issue:47049", "issue:47037|issue:47091", + "issue:47037|issue:47053", "issue:47049|issue:47053", - "issue:47049|issue:47102", - "issue:46985|issue:47049", + "issue:47049|issue:47091", "issue:47053|issue:47091", "issue:47053|issue:47102", - "issue:47037|issue:47053", "issue:47091|issue:47102", - "issue:46985|issue:47091" + "issue:46985|issue:47091", + "issue:47037|issue:47102", + "issue:46985|issue:47102", + "issue:47049|issue:47102" ], - "prepared_review_unit_hash": "5199e8c8bed19f1247148a51a56095dbcb91271d6b41ffb7dbefd0b084d9f071", + "prepared_review_unit_hash": "3f93be8d8c99d759bd199fb97f05fe24ac5a0507681364d860000ec1f7227540", "component_budget": { "node_count": 11, "item_count": 11, - "soft_pair_count": 33, - "serialized_chars": 18050, - "estimated_input_tokens": 4513, - "estimated_eval_tokens": 9282 + "soft_pair_count": 34, + "serialized_chars": 18420, + "estimated_input_tokens": 4605, + "estimated_eval_tokens": 9466 }, "budget": { "node_count": 11, "item_count": 11, - "soft_pair_count": 33, - "serialized_chars": 18050, - "estimated_input_tokens": 4513, - "estimated_eval_tokens": 9282 + "soft_pair_count": 34, + "serialized_chars": 18420, + "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 repeated daily \"[serge] integration failure triage\" issues, but the available evidence suggests separate triage reports by date rather than the same underlying bug. I would not merge any of the soft edges.", - "confidence": 0.8, - "canonical_issue_reason": "issue:46906 is the earliest issue in the series and one of the most-referenced, so it is the best anchor if a single representative issue is needed.", + "summary": "These are a dated series of separate [serge] integration failure triage issues, not duplicate reports of one underlying bug. The soft similarities reflect near-identical triage templates and recurring wording, but there\u2019s no evidence they should be merged.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 46906 is the earliest item in the dated triage series and is the cleanest anchor, though the set looks like separate daily reports rather than duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:46906 is the most suitable representative issue: it is earliest and heavily referenced, making it the clearest cluster anchor.", + "best_issue_reason": "Issue 47037 has the most discussion activity, so it is the strongest representative if one issue must stand in for the series.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46906", "right": "issue:46940", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Same triage-series title, but these are different daily reports; no evidence of one shared bug." }, { "left": "issue:46906", "right": "issue:46948", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Near-identical template, but different dates and no concrete sign they describe the same failure instance." }, { "left": "issue:46906", "right": "issue:46969", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "They are separate dated triage entries, not the same underlying issue." }, { "left": "issue:46906", "right": "issue:46985", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Same broad wording, but each issue is a distinct daily triage report." }, { "left": "issue:46906", "right": "issue:47037", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Recurring triage series with no evidence these are duplicate bug reports." }, { "left": "issue:46940", "right": "issue:46948", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Adjacent daily triage reports; similarity comes from the shared template, not a shared defect." }, { "left": "issue:46940", "right": "issue:46969", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Different dates in the same triage series; not enough to merge as duplicates." }, { "left": "issue:46940", "right": "issue:46985", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "These look like separate daily check-ins on integration failures, not one issue." }, { "left": "issue:46940", "right": "issue:47049", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Same recurring title, but no evidence of the same concrete bug or change." }, { "left": "issue:46948", "right": "issue:46969", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Both are triage reports, but they are distinct dated entries." }, { "left": "issue:46948", "right": "issue:46985", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Shared subsystem wording is not enough to treat them as duplicates." }, { "left": "issue:46948", "right": "issue:46998", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Separate daily triage issues; no same-failure evidence." }, { "left": "issue:46969", "right": "issue:46985", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Different daily triage reports with similar wording only." }, { "left": "issue:46969", "right": "issue:47037", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "These are distinct dated triage entries, not one bug report." }, { "left": "issue:46969", - "right": "issue:47102", + "right": "issue:47091", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Same report series, but no sign they reference the same underlying incident." }, { "left": "issue:46985", "right": "issue:46998", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Near-duplicate template text, but separate daily triage items." }, { "left": "issue:46985", "right": "issue:47037", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "No concrete shared bug; just recurring triage wording." }, { "left": "issue:46985", "right": "issue:47049", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." - }, - { - "left": "issue:46985", - "right": "issue:47091", - "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "These appear to be independent daily integration-failure triage reports." }, { - "left": "issue:46985", - "right": "issue:47102", + "left": "issue:46998", + "right": "issue:47053", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Similar triage framing, but different dates and no shared failure details." }, { "left": "issue:46998", "right": "issue:47037", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Distinct daily reports; similarity is only superficial." }, { "left": "issue:46998", "right": "issue:47049", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Same template, but not the same underlying bug." }, { "left": "issue:46998", - "right": "issue:47053", + "right": "issue:47091", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "These are separate triage records, not mergeable duplicates." }, { - "left": "issue:46998", + "left": "issue:46985", "right": "issue:47091", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Repeated triage-series wording, but each issue is a different day\u2019s report." }, { "left": "issue:47037", "right": "issue:47049", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." - }, - { - "left": "issue:47037", - "right": "issue:47053", - "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Adjacent triage issues; no evidence of the same concrete change or bug." }, { "left": "issue:47037", "right": "issue:47091", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Same report family, but not the same incident." }, { "left": "issue:47037", - "right": "issue:47102", + "right": "issue:47053", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Different dated triage entries with shared boilerplate only." }, { "left": "issue:47049", "right": "issue:47053", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Similar wording is due to the daily triage template, not a duplicate issue." }, { "left": "issue:47049", - "right": "issue:47102", + "right": "issue:47091", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "These are separate daily integration-failure reports." }, { "left": "issue:47053", "right": "issue:47091", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "No concrete shared bug; just the same triage series." }, { "left": "issue:47053", "right": "issue:47102", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." + "reason": "Different daily triage issues, not one underlying failure." }, { "left": "issue:47091", "right": "issue:47102", "accept": false, - "reason": "Same generic triage template, but different dates indicate separate daily incidents; no concrete evidence of the same bug." - } - ] - }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, - "error_kind": null, + "reason": "Same series, but each issue is a distinct day\u2019s triage note." + }, + { + "left": "issue:46985", + "right": "issue:47091", + "accept": false, + "reason": "Repeated triage-series wording, but still separate daily reports." + }, + { + "left": "issue:47037", + "right": "issue:47102", + "accept": false, + "reason": "No evidence these two track the same bug; they are dated triage entries." + }, + { + "left": "issue:46985", + "right": "issue:47102", + "accept": false, + "reason": "Different daily reports with similar boilerplate only." + }, + { + "left": "issue:47049", + "right": "issue:47102", + "accept": false, + "reason": "Same recurring triage format, but not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, "error_message": null }, { @@ -21944,7 +21987,7 @@ "soft_pairs": [ "issue:46983|issue:46984" ], - "prepared_review_unit_hash": "c528de25f27a4c7dc2c97816ddaeec47bac50fb65a6eeb7c669a563086f978f0", + "prepared_review_unit_hash": "d7283f66b2f94c1a6a2229be03193ec9be2b7133b2a5ca52444459f718e9d5a5", "component_budget": { "node_count": 2, "item_count": 2, @@ -21966,28 +22009,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues have the same title and very high textual similarity, indicating they are duplicate reports for the same RWKV7 support request.", - "confidence": 0.98, - "canonical_issue_reason": "Issue #46984 is the better canonical tracker because it is the open, later-created report and is more likely to be the active place for discussion and follow-up.", + "summary": "Two issues share the exact title \"Add RWKV7 model support\" and have a strong soft-similarity signal, so they likely represent the same feature request; however, the packet does not expose the issue bodies, so the evidence is not enough to claim identical text. The open issue is the better cluster representative.", + "confidence": 0.86, + "canonical_issue_reason": "#46984 is the better canonical issue because it is the open, active tracker for the same titled request.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #46984 is the best issue representative for this cluster for the same reason: it is open and current.", + "best_issue_reason": "#46984 best represents the cluster as the current open issue for this request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:46983", "right": "issue:46984", - "accept": true, - "reason": "Same title and highly similar issue content; both are clearly about adding RWKV7 model support, so they are duplicates of the same underlying request." + "accept": false, + "reason": "The pair looks likely related based on the identical title and high soft-similarity signal, but the packet does not show the underlying issue text, so there is not enough evidence here to confidently mark them as the same underlying request." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The assessment is well grounded in the packet: both issues share the same title, have very high text similarity, and the open later-created issue is a reasonable canonical tracker. The soft-edge duplicate verdict is conservative and supported by the similarity evidence." + "accept": false, + "feedback": "Reject: the cluster is plausibly the same feature request, but the analyst overstates the evidence. The packet only shows identical titles and a soft_similarity signal (with Jaccard), not the actual issue text, so \u201chighly similar issue text\u201d is unsupported. Also, \u201cno distinguishing metadata\u201d is a bit too strong because the issues differ in body length and activity. A more conservative verdict would be: likely duplicate/same request based on identical title and soft similarity." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -22010,12 +22053,12 @@ "soft_pairs": [ "issue:47007|issue:47015" ], - "prepared_review_unit_hash": "e897d8a744243e8adf8ad66e4d14834ac88f2db96b9b2696314d6a40e2d5070d", + "prepared_review_unit_hash": "e0c9657be4f95e7418395212c87c98296a36c43218a3bb7a16c20877d43788dc", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1626, + "serialized_chars": 1625, "estimated_input_tokens": 407, "estimated_eval_tokens": 1070 }, @@ -22023,7 +22066,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1626, + "serialized_chars": 1625, "estimated_input_tokens": 407, "estimated_eval_tokens": 1070 }, @@ -22032,18 +22075,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two closed issues are thematically similar AI/cybersecurity writeups, but they do not describe the same concrete bug, feature, or code path, so they should not be deduplicated.", - "confidence": 0.89, - "canonical_issue_reason": null, + "summary": "The two issues are topically similar (both about AI/LLM-based vulnerability discovery), but they read like different writeups rather than the same underlying bug or change. I would not merge them as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "No clear duplicate canonical emerges from the titles alone; if forced, 47015 is the more general/central phrasing, but the pair does not look like a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "No single canonical issue: both are broad, article-like AI vulnerability-discovery topics rather than the same underlying repository problem.", + "best_issue_reason": "Issue 47015 is the broader, more representative title for the shared AI vulnerability-discovery theme, so it is the better cluster anchor if one must be chosen.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:47007", "right": "issue:47015", "accept": false, - "reason": "The titles are related at a high level, but they describe different conceptual writeups with no shared code path, explicit target, or concrete duplicate symptom." + "reason": "They share a broad subject area, but the titles describe different angles: one is about lessons from using advanced LLMs, the other about a novel transformer-based cybersecurity discovery system. That is not enough to treat them as the same underlying issue." } ] }, @@ -22057,52 +22100,36 @@ { "label": "pull_request", "component_index": 1, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, - "review_unit_count": 3, + "review_unit_count": 2, "status": "reviewed", "reason": null, "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-57", + "cluster_id": "cluster-40546-41", "nodes": [ "pull_request:43451", "pull_request:44517", + "pull_request:45694", "pull_request:46196", "pull_request:46266", "pull_request:46658", "pull_request:46707", + "pull_request:46836", "pull_request:46903", - "pull_request:46907", - "pull_request:46908", "pull_request:46912", - "pull_request:46926", - "pull_request:46927", - "pull_request:46928", - "pull_request:46941", - "pull_request:46949", - "pull_request:46950", - "pull_request:46951", "pull_request:46954", "pull_request:46957", "pull_request:46958", - "pull_request:46961", "pull_request:46968", - "pull_request:46970", - "pull_request:46971", "pull_request:46974", - "pull_request:46986", "pull_request:46995", - "pull_request:47006", + "pull_request:47003", "pull_request:47025", "pull_request:47034", "pull_request:47035", - "pull_request:47054", - "pull_request:47055", - "pull_request:47056", - "pull_request:47057", - "pull_request:47058", "pull_request:47069", "pull_request:47072", "pull_request:47077", @@ -22110,55 +22137,32 @@ "pull_request:47082", "pull_request:47089", "pull_request:47090", + "pull_request:47092", "pull_request:47093", + "pull_request:47096", + "pull_request:47100", "pull_request:47107", "pull_request:47108", "pull_request:47113", - "pull_request:47114" + "pull_request:47114", + "pull_request:47117", + "pull_request:47118", + "pull_request:47119" ], "soft_pairs": [ "pull_request:46903|pull_request:47090", "pull_request:46903|pull_request:46958", - "pull_request:46950|pull_request:46951", - "pull_request:46950|pull_request:47057", - "pull_request:46927|pull_request:47054", - "pull_request:46927|pull_request:46951", - "pull_request:46951|pull_request:46971", - "pull_request:46949|pull_request:46971", - "pull_request:46951|pull_request:47057", - "pull_request:46941|pull_request:46970", - "pull_request:47056|pull_request:47058", - "pull_request:47006|pull_request:47058", - "pull_request:46908|pull_request:47006", - "pull_request:46927|pull_request:47055", - "pull_request:46908|pull_request:46928", - "pull_request:46907|pull_request:46926", - "pull_request:46926|pull_request:46928", - "pull_request:46928|pull_request:47057", - "pull_request:46927|pull_request:46986", - "pull_request:46986|pull_request:47058", - "pull_request:46926|pull_request:46941", - "pull_request:46908|pull_request:47058", - "pull_request:46928|pull_request:47058", - "pull_request:46928|pull_request:47054", - "pull_request:46941|pull_request:46986", - "pull_request:46941|pull_request:47006", - "pull_request:46928|pull_request:46941", - "pull_request:46954|pull_request:46961", - "pull_request:46958|pull_request:46961", "pull_request:46954|pull_request:46958", "pull_request:46954|pull_request:46968", - "pull_request:46961|pull_request:46974", "pull_request:46958|pull_request:46974", - "pull_request:46954|pull_request:46971", - "pull_request:46971|pull_request:46974", - "pull_request:46968|pull_request:46974", - "pull_request:46961|pull_request:46968", + "pull_request:47089|pull_request:47119", "pull_request:46958|pull_request:46968", - "pull_request:46961|pull_request:46971", - "pull_request:46949|pull_request:46968", + "pull_request:46968|pull_request:46974", + "pull_request:47072|pull_request:47119", "pull_request:47072|pull_request:47089", "pull_request:47077|pull_request:47078", + "pull_request:46968|pull_request:47114", + "pull_request:46707|pull_request:46968", "pull_request:46266|pull_request:46957", "pull_request:44517|pull_request:47082", "pull_request:47034|pull_request:47035", @@ -22167,1044 +22171,653 @@ "pull_request:47093|pull_request:47114", "pull_request:46707|pull_request:47114", "pull_request:46707|pull_request:47108", - "pull_request:46266|pull_request:46707", "pull_request:47089|pull_request:47113", - "pull_request:46707|pull_request:46961", - "pull_request:47082|pull_request:47108", + "pull_request:46266|pull_request:46707", "pull_request:47108|pull_request:47114", + "pull_request:47082|pull_request:47108", + "pull_request:46912|pull_request:47118", + "pull_request:47107|pull_request:47118", + "pull_request:47117|pull_request:47118", "pull_request:47077|pull_request:47114", + "pull_request:47100|pull_request:47118", "pull_request:46912|pull_request:47025", - "pull_request:46968|pull_request:47114", + "pull_request:46196|pull_request:47118", + "pull_request:47100|pull_request:47117", + "pull_request:47114|pull_request:47118", "pull_request:46658|pull_request:47114", "pull_request:46707|pull_request:47082", "pull_request:46266|pull_request:47114", - "pull_request:46961|pull_request:47082", "pull_request:47025|pull_request:47107", "pull_request:46912|pull_request:47082", + "pull_request:43451|pull_request:47082", "pull_request:46912|pull_request:47078", - "pull_request:46961|pull_request:47108", + "pull_request:47092|pull_request:47117", "pull_request:47035|pull_request:47089", "pull_request:43451|pull_request:46995", - "pull_request:46196|pull_request:47025" + "pull_request:46196|pull_request:47025", + "pull_request:46836|pull_request:46995", + "pull_request:46707|pull_request:47025", + "pull_request:45694|pull_request:47069", + "pull_request:46266|pull_request:47034", + "pull_request:47034|pull_request:47108", + "pull_request:45694|pull_request:47107", + "pull_request:46995|pull_request:47096", + "pull_request:47100|pull_request:47113", + "pull_request:47035|pull_request:47113", + "pull_request:47035|pull_request:47069", + "pull_request:46957|pull_request:47108", + "pull_request:46995|pull_request:47003", + "pull_request:47069|pull_request:47113", + "pull_request:46912|pull_request:47077", + "pull_request:47113|pull_request:47117", + "pull_request:46707|pull_request:47077", + "pull_request:47113|pull_request:47118", + "pull_request:43451|pull_request:46266", + "pull_request:47069|pull_request:47082", + "pull_request:45694|pull_request:47035", + "pull_request:43451|pull_request:47107", + "pull_request:43451|pull_request:47078", + "pull_request:47078|pull_request:47096", + "pull_request:43451|pull_request:47117", + "pull_request:46707|pull_request:46957", + "pull_request:47003|pull_request:47078", + "pull_request:47092|pull_request:47107" ], - "prepared_review_unit_hash": "610b5146dbe34edc7b8a869d0bc76261f3ecbeda9e8bdf8ba944a0717e31e96e", + "prepared_review_unit_hash": "f50e831e038b96a317cf229f62ba926afa45c62c46524500aad585ab0a85dbf8", "component_budget": { - "node_count": 57, - "item_count": 57, - "soft_pair_count": 182, - "serialized_chars": 204273, - "estimated_input_tokens": 51069, - "estimated_eval_tokens": 102394 + "node_count": 41, + "item_count": 41, + "soft_pair_count": 119, + "serialized_chars": 146957, + "estimated_input_tokens": 36740, + "estimated_eval_tokens": 73736 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 69, - "serialized_chars": 130620, - "estimated_input_tokens": 32655, - "estimated_eval_tokens": 65566 + "node_count": 38, + "item_count": 38, + "soft_pair_count": 72, + "serialized_chars": 113726, + "estimated_input_tokens": 28432, + "estimated_eval_tokens": 57120 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous PRs; the only clear duplicate subcluster is the CohereASR training-loss double-shift fix, with a smaller chance of overlap around the Whisper cache-init change. Most other pairs are broad subsystem overlaps or unrelated fixes and should stay separate.", - "confidence": 0.61, + "summary": "Only the CohereASR training-loss PRs look like true duplicates of the same underlying bug. The rest mostly share broad subsystems or filenames but fix different issues, so they should stay separate.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "Use pull_request:47090 as the canonical PR: it is the cleanest single-bug fix for CohereASR\u2019s training-loss double shift, with a direct issue target and narrower scope than the broader 46903 variant that also bundles PPFormulaNet changes.", + "canonical_pr_reason": "PR 46903 is the best canonical PR: it fixes the CohereASR double-shift loss bug and also carries the related PPFormulaNet change, with an explicit issue link and the most complete scope among the duplicates.", "best_issue_reason": null, - "best_pr_reason": "pull_request:47090 is the best overall representative because it is the most focused, concrete code-path fix in the cluster; it cleanly addresses one loss-alignment bug instead of mixing in unrelated model/doc/test edits.", + "best_pr_reason": "PR 46903 is the strongest representative of the duplicate set because it is the broadest, has the issue reference, and addresses the same loss-shift bug on the shared CohereASR code path.", "soft_edge_verdicts": [ { "left": "pull_request:46903", "right": "pull_request:47090", "accept": true, - "reason": "Same CohereASR loss double-shift bug in the same forward path; 46903 just bundles an extra PPFormulaNet fix." + "reason": "Same CohereASR training-loss double-shift bug in the same code path; 47090 is just an alternate implementation of the same fix." }, { "left": "pull_request:46903", "right": "pull_request:46958", "accept": true, - "reason": "Same CohereASR training-loss double-shift fix; the surrounding PR scopes differ, but the underlying bug is the same." - }, - { - "left": "pull_request:46950", - "right": "pull_request:46951", - "accept": false, - "reason": "Mamba2 OOM/runtime fix vs PEFT loading mismatch; same ecosystem, different bugs." - }, - { - "left": "pull_request:46950", - "right": "pull_request:47057", - "accept": false, - "reason": "Mamba2 model-kernel change vs DeepSeekV32 load_error test patch; unrelated fixes." - }, - { - "left": "pull_request:46927", - "right": "pull_request:47054", - "accept": false, - "reason": "GLM46V type-cast bug vs MiniMax-M3-VL offload test tweak; different code paths." - }, - { - "left": "pull_request:46927", - "right": "pull_request:46951", - "accept": false, - "reason": "GLM46V input handling vs PEFT loading behavior; no shared concrete bug." - }, - { - "left": "pull_request:46951", - "right": "pull_request:46971", - "accept": false, - "reason": "PEFT loader mismatch fix vs Mamba2 test/OOM workaround; same broad area only." - }, - { - "left": "pull_request:46949", - "right": "pull_request:46971", - "accept": false, - "reason": "Whisper/audioflamingo cache-init fix vs Mamba2 test dtype/OOM workaround; unrelated failures." - }, - { - "left": "pull_request:46951", - "right": "pull_request:47057", - "accept": false, - "reason": "PEFT integration change vs DeepSeekV32 test load-offload changes; different issues." - }, - { - "left": "pull_request:46941", - "right": "pull_request:46970", - "accept": false, - "reason": "DAC tensor-shape fix vs MusicGen melody output mismatch; different model bugs." - }, - { - "left": "pull_request:47056", - "right": "pull_request:47058", - "accept": false, - "reason": "Whisper cache-init fix vs Mamba2 OOM workaround; same test-fix genre, not same bug." - }, - { - "left": "pull_request:47006", - "right": "pull_request:47058", - "accept": false, - "reason": "Gemma3 generation logic vs Mamba2 test/OOM changes; unrelated code paths." - }, - { - "left": "pull_request:46908", - "right": "pull_request:47006", - "accept": false, - "reason": "DeepSeekV32 load-error tests vs Gemma3 generation logic; no concrete overlap." - }, - { - "left": "pull_request:46927", - "right": "pull_request:47055", - "accept": false, - "reason": "GLM46V type-cast bug vs EdgeTam import_or_config cleanup; different failures." - }, - { - "left": "pull_request:46908", - "right": "pull_request:46928", - "accept": false, - "reason": "DeepSeekV32 load-error workaround vs BLIP-2 tokenizer whitelist; unrelated." - }, - { - "left": "pull_request:46907", - "right": "pull_request:46926", - "accept": false, - "reason": "EdgeTam import_or_config fix vs MiniMax-M3-VL offload test fix; different bugs." - }, - { - "left": "pull_request:46926", - "right": "pull_request:46928", - "accept": false, - "reason": "MiniMax-M3-VL OOM/load_error workaround vs BLIP-2 tokenizer update; no shared bug." - }, - { - "left": "pull_request:46928", - "right": "pull_request:47057", - "accept": false, - "reason": "BLIP-2 tokenizer mapping vs DeepSeekV32 load-error test changes; unrelated." - }, - { - "left": "pull_request:46927", - "right": "pull_request:46986", - "accept": false, - "reason": "GLM46V input-casting fix vs GLM-OCR output mismatch fix; different model paths." - }, - { - "left": "pull_request:46986", - "right": "pull_request:47058", - "accept": false, - "reason": "GLM-OCR output mismatch fix vs Mamba2 OOM/test workaround; not the same change." - }, - { - "left": "pull_request:46926", - "right": "pull_request:46941", - "accept": false, - "reason": "MiniMax-M3-VL load_error workaround vs DAC runtime fix; different failures." - }, - { - "left": "pull_request:46908", - "right": "pull_request:47058", - "accept": false, - "reason": "DeepSeekV32 test load/offload fix vs Mamba2 OOM/test changes; unrelated." - }, - { - "left": "pull_request:46928", - "right": "pull_request:47054", - "accept": false, - "reason": "BLIP-2 tokenizer update vs MiniMax-M3-VL offload workaround; no duplicate bug." - }, - { - "left": "pull_request:46941", - "right": "pull_request:46986", - "accept": false, - "reason": "DAC fix vs GLM-OCR list-output mismatch; different concrete bugs." - }, - { - "left": "pull_request:46941", - "right": "pull_request:47006", - "accept": false, - "reason": "DAC tensor fix vs Gemma3 generation cache logic; unrelated code paths." - }, - { - "left": "pull_request:46928", - "right": "pull_request:46941", - "accept": false, - "reason": "BLIP-2 tokenizer mapping vs DAC runtime fix; same broad repo area only." - }, - { - "left": "pull_request:46954", - "right": "pull_request:46961", - "accept": false, - "reason": "Pytest-v9 fp32-module cleanup vs Serge workflow normalization; unrelated." - }, - { - "left": "pull_request:46958", - "right": "pull_request:46961", - "accept": false, - "reason": "CohereASR loss fix vs Serge workflow normalization; no shared bug." + "reason": "Both fix the CohereASR double-shift loss bug in the same files; 46903 just bundles an extra PPFormulaNet fix too." }, { "left": "pull_request:46954", "right": "pull_request:46958", "accept": false, - "reason": "Pytest-v9 fp32-module cleanup vs CohereASR loss double-shift fix; different issues." + "reason": "Different bugs: Pytest-v9 fp32-module behavior in GlmAsr vs CohereASR training loss shifting." }, { "left": "pull_request:46954", "right": "pull_request:46968", "accept": false, - "reason": "Pytest-v9 cleanup vs docs autolink fixes; unrelated." - }, - { - "left": "pull_request:46961", - "right": "pull_request:46974", - "accept": false, - "reason": "Serge workflow plumbing vs Musicgen generate typo; different concerns." + "reason": "Unrelated changes: GLM-ASR pytest/fp32 handling vs docs autolink fixes." }, { "left": "pull_request:46958", "right": "pull_request:46974", "accept": false, - "reason": "CohereASR loss fix vs Musicgen cache-length typo; different code paths." + "reason": "Same broad area only by name; one is CohereASR loss logic, the other is a Musicgen generate typo." }, { - "left": "pull_request:46954", - "right": "pull_request:46971", + "left": "pull_request:47089", + "right": "pull_request:47119", "accept": false, - "reason": "Pytest-v9 cleanup vs Mamba2 OOM test workaround; unrelated." + "reason": "Both are small validation fixes, but for different code paths: AqlmConfig type checks vs image padding handling." }, { - "left": "pull_request:46971", - "right": "pull_request:46974", + "left": "pull_request:46958", + "right": "pull_request:46968", "accept": false, - "reason": "Mamba2 test OOM fix vs Musicgen generate typo; different bugs." + "reason": "CohereASR loss bug vs documentation autolink cleanup; no shared underlying change." }, { "left": "pull_request:46968", "right": "pull_request:46974", "accept": false, - "reason": "Docs autolink cleanup vs Musicgen generation bug; unrelated." + "reason": "Docs autolink cleanup and Musicgen generate typo are unrelated." }, { - "left": "pull_request:46961", - "right": "pull_request:46968", - "accept": false, - "reason": "Workflow automation vs docs autolink cleanup; no duplicate change." - }, - { - "left": "pull_request:46958", - "right": "pull_request:46968", + "left": "pull_request:47072", + "right": "pull_request:47119", "accept": false, - "reason": "CohereASR loss fix vs docs autolink cleanup; unrelated." + "reason": "Chat-schema enum parsing and image padding are different utility bugs." }, { - "left": "pull_request:46961", - "right": "pull_request:46971", + "left": "pull_request:47072", + "right": "pull_request:47089", "accept": false, - "reason": "Workflow automation vs Mamba2 test workaround; different targets." + "reason": "Chat-template schema parsing vs AqlmConfig error-message/type checks are unrelated." }, { - "left": "pull_request:46949", - "right": "pull_request:46968", + "left": "pull_request:47077", + "right": "pull_request:47078", "accept": false, - "reason": "Whisper/audioflamingo cache-init fix vs docs autolink cleanup; unrelated." + "reason": "Both touch Japanese docs, but they change different content and wording, not the same change." }, { - "left": "pull_request:47072", - "right": "pull_request:47089", + "left": "pull_request:46968", + "right": "pull_request:47114", "accept": false, - "reason": "Chat schema crash fix vs AqlmConfig error-message wording; different bugs." + "reason": "Docs autolink cleanup vs weight-conversion docs update; same documentation area only." }, { - "left": "pull_request:47077", - "right": "pull_request:47078", + "left": "pull_request:46707", + "right": "pull_request:46968", "accept": false, - "reason": "Both are Japanese docs edits, but one adds DistilBERT translation pages and the other is a wording cleanup; not the same change." + "reason": "Different changes: native FSDP2 migration vs docs autolink fixes." }, { "left": "pull_request:46266", "right": "pull_request:46957", "accept": false, - "reason": "DINO conversion-mapping work vs finegrained FP8 decode optimization; unrelated model systems." + "reason": "Both touch generation/modeling internals, but one is DINO conversion mapping and the other is decode optimization for FP8/MoE." }, { "left": "pull_request:44517", "right": "pull_request:47082", "accept": false, - "reason": "Qwen3-TTS model addition vs RoMa keypoint matcher addition; same codegen pattern, different features." + "reason": "Both add model registry entries, but for different models and different features." }, { "left": "pull_request:47034", "right": "pull_request:47035", "accept": false, - "reason": "Neuron SFT experiment draft vs torch-compile backend default tweak; related platform, but not the same bug/change." + "reason": "Neuron SFT scripts/training args vs torch.compile backend default; related platform area, but different fixes." }, { "left": "pull_request:47069", "right": "pull_request:47089", "accept": false, - "reason": "num_labels property typing vs AqlmConfig error-message text; unrelated fixes." + "reason": "Different bugs: num_labels returning None vs AqlmConfig type-error wording." }, { "left": "pull_request:46196", "right": "pull_request:46995", "accept": false, - "reason": "Wav2vec2 SDPA dispatch vs MLU flash-linear-attention support; different backend support changes." + "reason": "Different accelerator/model-support additions: SDPA dispatch docs/model changes vs MLU support for flash linear attention." }, { "left": "pull_request:47093", "right": "pull_request:47114", "accept": false, - "reason": "OpenPangu-V2 model addition vs docs about custom-code conversion lookup; unrelated." + "reason": "New model support for OpenPangu-V2 vs documentation for custom-code conversion mapping." }, { "left": "pull_request:46707", "right": "pull_request:47114", "accept": false, - "reason": "Native FSDP2 module work vs docs conversion-mapping clarification; different concerns." + "reason": "FSDP2 migration work and conversion-mapping docs are unrelated." }, { "left": "pull_request:47089", "right": "pull_request:47113", "accept": false, - "reason": "AqlmConfig type-check wording vs audio feature/proc requirements cleanup; unrelated." - }, - { - "left": "pull_request:46707", - "right": "pull_request:47108", - "accept": false, - "reason": "FSDP2 migration vs doc-builder pin comment cleanup; no duplicate change." - }, - { - "left": "pull_request:47082", - "right": "pull_request:47108", - "accept": false, - "reason": "RoMa model addition vs doc-builder workflow comment tweak; unrelated." + "reason": "AqlmConfig validation messages vs audio feature/processor requirements are unrelated." }, { - "left": "pull_request:47077", - "right": "pull_request:47114", - "accept": false, - "reason": "Japanese DistilBERT docs vs conversion-mapping docs; different documentation topics." - }, - { - "left": "pull_request:46912", - "right": "pull_request:47025", - "accept": false, - "reason": "RoPE test/code sweep vs diffusion_gemma test skip; broad testing only, not same bug." - }, - { - "left": "pull_request:46968", - "right": "pull_request:47114", - "accept": false, - "reason": "Docs autolink cleanup vs conversion-mapping docs; unrelated docs changes." - }, - { - "left": "pull_request:46658", - "right": "pull_request:47114", - "accept": false, - "reason": "Step 3.7 Flash model/bootstrap work vs conversion-mapping docs; different features." - }, - { - "left": "pull_request:46707", - "right": "pull_request:47082", + "left": "pull_request:46266", + "right": "pull_request:46707", "accept": false, - "reason": "FSDP2 migration vs RoMa model addition; unrelated." + "reason": "Different core changes: DINO conversion mapping vs native FSDP2 support." }, { - "left": "pull_request:46266", + "left": "pull_request:47108", "right": "pull_request:47114", "accept": false, - "reason": "DINO conversion-mapping modernization vs docs conversion-mapping explanation; not the same artifact." + "reason": "Doc-builder pin comment cleanup vs conversion-mapping docs content; not the same change." }, { - "left": "pull_request:46961", + "left": "pull_request:47108", "right": "pull_request:47082", "accept": false, - "reason": "Workflow automation vs RoMa model addition; unrelated." + "reason": "Doc-builder workflow comment cleanup vs RoMa model support are unrelated." }, { - "left": "pull_request:47025", - "right": "pull_request:47107", - "accept": false, - "reason": "Diffusion-gemma test skip vs MoE decode optimization; different failures." - }, - { - "left": "pull_request:46912", - "right": "pull_request:47082", + "left": "pull_request:47082", + "right": "pull_request:47096", "accept": false, - "reason": "RoPE tests vs RoMa model addition; unrelated." + "reason": "Different model additions: RoMa keypoint matcher vs exportable kimi support." }, { "left": "pull_request:46912", - "right": "pull_request:47078", + "right": "pull_request:47118", "accept": false, - "reason": "RoPE tests vs Japanese docs wording cleanup; different scopes." + "reason": "RoPE test expansion across models vs cache dispatch simplification; not the same bug." }, { - "left": "pull_request:46961", - "right": "pull_request:47108", + "left": "pull_request:47107", + "right": "pull_request:47118", "accept": false, - "reason": "Workflow normalization vs doc-builder pin comment cleanup; unrelated." + "reason": "MoE decode optimization for submodel experts vs cache-layer dispatch simplification are distinct changes." }, { - "left": "pull_request:47035", - "right": "pull_request:47089", + "left": "pull_request:47117", + "right": "pull_request:47118", "accept": false, - "reason": "Neuron compile backend default vs AqlmConfig error text; different code paths." + "reason": "Output-record capture docs/code changes vs cache dispatch refactor are unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:46995", + "left": "pull_request:47100", + "right": "pull_request:47118", "accept": false, - "reason": "Molmo2 model addition vs MLU flash-linear-attention availability; unrelated features." + "reason": "kernels API compatibility fix vs cache dispatch simplification; different code paths." }, { - "left": "pull_request:46196", - "right": "pull_request:47025", - "accept": false, - "reason": "Wav2vec2 SDPA dispatch vs diffusion_gemma test skip; different model fixes." - } - ] - }, - "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": 3, - "status": "reviewed", - "reason": null, - "source": "llm", - "cache_hit": false, - "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-40546-57", - "nodes": [ - "pull_request:43451", - "pull_request:44517", - "pull_request:45694", - "pull_request:46196", - "pull_request:46266", - "pull_request:46658", - "pull_request:46707", - "pull_request:46836", - "pull_request:46880", - "pull_request:46895", - "pull_request:46907", - "pull_request:46908", - "pull_request:46912", - "pull_request:46926", - "pull_request:46927", - "pull_request:46949", - "pull_request:46950", - "pull_request:46951", - "pull_request:46957", - "pull_request:46958", - "pull_request:46961", - "pull_request:46971", - "pull_request:46995", - "pull_request:47003", - "pull_request:47025", - "pull_request:47034", - "pull_request:47035", - "pull_request:47054", - "pull_request:47055", - "pull_request:47056", - "pull_request:47057", - "pull_request:47058", - "pull_request:47069", - "pull_request:47077", - "pull_request:47078", - "pull_request:47082", - "pull_request:47089", - "pull_request:47090", - "pull_request:47092", - "pull_request:47096", - "pull_request:47100", - "pull_request:47107", - "pull_request:47108", - "pull_request:47113" - ], - "soft_pairs": [ - "pull_request:46836|pull_request:46995", - "pull_request:46707|pull_request:47025", - "pull_request:45694|pull_request:47069", - "pull_request:46266|pull_request:47034", - "pull_request:45694|pull_request:47107", - "pull_request:47034|pull_request:47108", - "pull_request:47035|pull_request:47113", - "pull_request:47100|pull_request:47113", - "pull_request:47035|pull_request:47069", - "pull_request:46995|pull_request:47096", - "pull_request:46995|pull_request:47003", - "pull_request:46836|pull_request:46961", - "pull_request:47069|pull_request:47113", - "pull_request:46957|pull_request:46961", - "pull_request:46961|pull_request:47035", - "pull_request:46912|pull_request:47077", - "pull_request:46707|pull_request:47077", - "pull_request:47069|pull_request:47082", - "pull_request:45694|pull_request:47035", - "pull_request:46196|pull_request:46961", - "pull_request:43451|pull_request:46266", - "pull_request:43451|pull_request:47107", - "pull_request:43451|pull_request:47078", - "pull_request:47078|pull_request:47096", - "pull_request:46707|pull_request:46957", - "pull_request:47003|pull_request:47078", - "pull_request:47092|pull_request:47107", - "pull_request:46658|pull_request:46961", - "pull_request:46912|pull_request:47113", - "pull_request:46836|pull_request:47107", - "pull_request:43451|pull_request:44517", - "pull_request:45694|pull_request:46707", - "pull_request:46912|pull_request:47100", - "pull_request:46961|pull_request:47113", - "pull_request:47096|pull_request:47107", - "pull_request:44517|pull_request:46836", - "pull_request:44517|pull_request:47096", - "pull_request:44517|pull_request:47003", - "pull_request:43451|pull_request:46957", - "pull_request:46957|pull_request:47096", - "pull_request:46658|pull_request:46895", - "pull_request:43451|pull_request:46961", - "pull_request:46961|pull_request:47096", - "pull_request:46961|pull_request:47003", - "pull_request:46961|pull_request:47034", - "pull_request:46961|pull_request:47092", - "pull_request:46912|pull_request:46961", - "pull_request:43451|pull_request:47089", - "pull_request:46196|pull_request:47100", - "pull_request:43451|pull_request:47100", - "pull_request:47096|pull_request:47100", - "pull_request:43451|pull_request:45694", - "pull_request:43451|pull_request:46895", - "pull_request:46895|pull_request:47096", - "pull_request:46880|pull_request:47092", - "pull_request:46895|pull_request:47090", - "pull_request:46895|pull_request:46958", - "pull_request:46958|pull_request:47090", - "pull_request:46907|pull_request:47055", - "pull_request:46949|pull_request:47056", - "pull_request:46950|pull_request:46971", - "pull_request:46950|pull_request:47058", - "pull_request:46971|pull_request:47058", - "pull_request:46926|pull_request:47054", - "pull_request:46908|pull_request:47057", - "pull_request:46949|pull_request:46950", - "pull_request:47054|pull_request:47057", - "pull_request:46949|pull_request:46951", - "pull_request:46927|pull_request:46949", - "pull_request:46907|pull_request:46927", - "pull_request:46908|pull_request:47054", - "pull_request:46926|pull_request:47057" - ], - "prepared_review_unit_hash": "7218ba276e9f267b31d9becff08cdc41c77e1b44f101e796265f30baadf1449d", - "component_budget": { - "node_count": 57, - "item_count": 57, - "soft_pair_count": 182, - "serialized_chars": 204273, - "estimated_input_tokens": 51069, - "estimated_eval_tokens": 102394 - }, - "budget": { - "node_count": 44, - "item_count": 44, - "soft_pair_count": 72, - "serialized_chars": 121069, - "estimated_input_tokens": 30268, - "estimated_eval_tokens": 60792 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": true, - "analyst_result": { - "summary": "This cluster is mostly a set of weakly similar but unrelated PRs, with a handful of true duplicate/follow-up chains in cohere_asr, edgetam, whisper, mamba2, minimax_m3_vl, and deepseek_v32.", - "confidence": 0.81, - "canonical_issue_reason": null, - "canonical_pr_reason": "No single PR cleanly represents the whole cluster because it mixes unrelated changes. If a single anchor is needed, 47090 is the best representative of the only clearly coherent duplicate chain (CohereASR double-shift), with the most focused fix and an explicit issue target.", - "best_issue_reason": null, - "best_pr_reason": "47090 is the strongest standalone PR in the cluster: it directly fixes the underlying CohereASR training-loss double-shift bug, has an explicit issue target, and is the cleanest version of that change family.", - "soft_edge_verdicts": [ - { - "left": "pull_request:46836", - "right": "pull_request:46995", + "left": "pull_request:47114", + "right": "pull_request:47118", "accept": false, - "reason": "Different features/models; only broad PR-title similarity." + "reason": "Weight-conversion docs update vs cache dispatch refactor are unrelated." }, { - "left": "pull_request:46707", - "right": "pull_request:47025", + "left": "pull_request:47096", + "right": "pull_request:47118", "accept": false, - "reason": "Unrelated subsystems and bugs." + "reason": "Exportable kimi support and cache dispatch simplification are different features." }, { - "left": "pull_request:45694", - "right": "pull_request:47069", + "left": "pull_request:47003", + "right": "pull_request:47118", "accept": false, - "reason": "Different changes: training-batch sizing vs num_labels typing." + "reason": "OpenVINO exporter support vs cache dispatch simplification are unrelated." }, { - "left": "pull_request:46266", - "right": "pull_request:47034", + "left": "pull_request:46995", + "right": "pull_request:47003", "accept": false, - "reason": "Different model families and unrelated fixes." + "reason": "Different hardware/runtime support areas; flash-linear-attention availability is not the same as OpenVINO exporter work." }, { - "left": "pull_request:45694", - "right": "pull_request:47107", + "left": "pull_request:47069", + "right": "pull_request:47070", "accept": false, - "reason": "No shared code-path; unrelated training_args vs MoE decode logic." + "reason": "Config property typing change and chat-template enum parsing are unrelated utility fixes." }, { - "left": "pull_request:47034", - "right": "pull_request:47108", + "left": "pull_request:47092", + "right": "pull_request:47089", "accept": false, - "reason": "Unrelated automation/workflow change vs Neuron training experiment." + "reason": "Test expectations for MiniCPM3 on XPU vs AqlmConfig validation messages are unrelated." }, { "left": "pull_request:47035", - "right": "pull_request:47113", + "right": "pull_request:47089", "accept": false, - "reason": "Different concerns: torch-compile backend default vs audio feature/proc requirements." + "reason": "Torch-compile backend defaulting and quantization-config error text are different changes." }, { - "left": "pull_request:47100", - "right": "pull_request:47113", + "left": "pull_request:47034", + "right": "pull_request:47108", "accept": false, - "reason": "Different library areas; not the same bug." + "reason": "Neuron training experiment scripts vs doc-builder workflow comment cleanup are unrelated." }, { - "left": "pull_request:47035", + "left": "pull_request:47034", "right": "pull_request:47069", "accept": false, - "reason": "Unrelated config default vs property typing fix." - }, - { - "left": "pull_request:46995", - "right": "pull_request:47096", - "accept": false, - "reason": "Different targets: MLU availability check vs exportable kimi." + "reason": "Neuron training scripts vs num_labels property typing are unrelated." }, { - "left": "pull_request:46995", - "right": "pull_request:47003", - "accept": false, - "reason": "Unrelated utility check vs exporter/docs PR." - }, - { - "left": "pull_request:46836", - "right": "pull_request:46961", + "left": "pull_request:47003", + "right": "pull_request:47078", "accept": false, - "reason": "Different model addition vs workflow normalization." + "reason": "OpenVINO exporter support and Japanese docs wording changes are unrelated." }, { - "left": "pull_request:47069", - "right": "pull_request:47113", + "left": "pull_request:47077", + "right": "pull_request:47114", "accept": false, - "reason": "No shared underlying bug." + "reason": "Japanese DistilBERT translation vs English weight-converter docs update are different documentation tasks." }, { "left": "pull_request:46957", - "right": "pull_request:46961", - "accept": false, - "reason": "Different code areas; weak title similarity only." - }, - { - "left": "pull_request:46961", - "right": "pull_request:47035", - "accept": false, - "reason": "Different automation vs compile-backend behavior." - }, - { - "left": "pull_request:46912", - "right": "pull_request:47077", + "right": "pull_request:47108", "accept": false, - "reason": "RoPE tests vs Japanese docs translation are unrelated." + "reason": "FP8/MoE decode optimization and doc-builder comment cleanup are unrelated." }, { - "left": "pull_request:46707", - "right": "pull_request:47077", + "left": "pull_request:47092", + "right": "pull_request:47107", "accept": false, - "reason": "Unrelated distributed/FSDP work vs docs translation." + "reason": "Test expectation updates for XPU vs MoE decode optimization are unrelated." }, { - "left": "pull_request:47069", + "left": "pull_request:43451", "right": "pull_request:47082", "accept": false, - "reason": "Unrelated config property typing vs new model addition." + "reason": "Molmo2 model addition and RoMa model addition are different model integrations." }, { - "left": "pull_request:45694", - "right": "pull_request:47035", - "accept": false, - "reason": "Different training-arg fixes, not one bug." - }, - { - "left": "pull_request:46196", - "right": "pull_request:47100", + "left": "pull_request:43451", + "right": "pull_request:46995", "accept": false, - "reason": "Audio/wav2vec SDPA support vs kernels API cleanup." + "reason": "Molmo2 model support vs flash-linear-attention availability are unrelated." }, { "left": "pull_request:43451", "right": "pull_request:46266", "accept": false, - "reason": "Different model additions and conversion paths." + "reason": "Different model addition vs Dino conversion-mapping modernization." }, { "left": "pull_request:43451", "right": "pull_request:47107", "accept": false, - "reason": "Molmo2 add PR vs MoE decode optimization are unrelated." + "reason": "Molmo2 model support and MoE decode optimization are unrelated." }, { "left": "pull_request:43451", "right": "pull_request:47078", "accept": false, - "reason": "Unrelated PRs: model addition vs docs wording." + "reason": "Molmo2 model addition vs Japanese docs wording changes are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47096", + "right": "pull_request:47117", "accept": false, - "reason": "No shared bug or code path." + "reason": "Molmo2 model support and output-recording docs/code changes are unrelated." }, { - "left": "pull_request:44517", - "right": "pull_request:46836", + "left": "pull_request:47003", + "right": "pull_request:47117", "accept": false, - "reason": "Different model/doc additions; not a duplicate." + "reason": "OpenVINO exporter support vs output-capturing changes are unrelated." }, { - "left": "pull_request:45694", - "right": "pull_request:46707", + "left": "pull_request:47072", + "right": "pull_request:47114", "accept": false, - "reason": "Training-args bug vs FSDP2 module migration." + "reason": "Schema parsing fix and conversion-mapping docs update are unrelated." }, { "left": "pull_request:46912", - "right": "pull_request:46961", + "right": "pull_request:47025", "accept": false, - "reason": "Different topics: RoPE tests vs cleanup workflow." + "reason": "RoPE test expansion vs Diffusion Gemma test adjustments are different model test changes." }, { - "left": "pull_request:46961", - "right": "pull_request:47096", + "left": "pull_request:47100", + "right": "pull_request:47117", "accept": false, - "reason": "Automation workflow vs exportability change." + "reason": "kernels API compatibility fix vs output-capturing changes are unrelated." }, { - "left": "pull_request:46961", - "right": "pull_request:47003", + "left": "pull_request:47113", + "right": "pull_request:47117", "accept": false, - "reason": "Workflow normalization vs exporter docs/code." + "reason": "Audio-feature/processor requirements vs output-capturing changes are unrelated." }, { - "left": "pull_request:46961", - "right": "pull_request:47034", + "left": "pull_request:47113", + "right": "pull_request:47118", "accept": false, - "reason": "Unrelated workflow and training experiment." + "reason": "Audio requirements fixes and cache dispatch refactor are unrelated." }, { - "left": "pull_request:46961", - "right": "pull_request:47092", + "left": "pull_request:47089", + "right": "pull_request:47096", "accept": false, - "reason": "No same underlying bug." + "reason": "Quantization-config type checking and exportable kimi support are unrelated." }, { - "left": "pull_request:46912", - "right": "pull_request:47113", + "left": "pull_request:47069", + "right": "pull_request:47082", "accept": false, - "reason": "RoPE tests vs audio requirement annotations." + "reason": "num_labels typing fix vs RoMa model addition are unrelated." }, { - "left": "pull_request:46836", - "right": "pull_request:47107", + "left": "pull_request:45694", + "right": "pull_request:47035", "accept": false, - "reason": "Different model addition vs decode optimization." + "reason": "TrainingArguments field formatting/respect-split-batches work and torch.compile backend defaulting are different fixes." }, { - "left": "pull_request:43451", - "right": "pull_request:44517", + "left": "pull_request:45694", + "right": "pull_request:47107", "accept": false, - "reason": "Both add models, but different models and implementations." + "reason": "TrainingArguments behavior change vs MoE decode optimization are unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:46957", + "left": "pull_request:46912", + "right": "pull_request:47082", "accept": false, - "reason": "Molmo2 addition vs FP8 generation optimization are unrelated." + "reason": "RoPE tests and RoMa model addition do not address the same underlying issue." }, { - "left": "pull_request:43451", - "right": "pull_request:47089", + "left": "pull_request:46912", + "right": "pull_request:47078", "accept": false, - "reason": "Different docs/model work vs AqlmConfig error-message cleanup." + "reason": "RoPE tests vs Japanese docs wording changes are unrelated." }, { - "left": "pull_request:46196", - "right": "pull_request:46961", + "left": "pull_request:47092", + "right": "pull_request:47117", "accept": false, - "reason": "Audio-model SDPA work vs workflow normalization." + "reason": "XPU test expectations and output-recording changes are unrelated." }, { - "left": "pull_request:46658", - "right": "pull_request:46961", + "left": "pull_request:47035", + "right": "pull_request:47113", "accept": false, - "reason": "Step 3.7 Flash work vs workflow normalization are unrelated." + "reason": "Torch-compile backend defaulting vs audio feature/processor requirement guards are unrelated." }, { - "left": "pull_request:46912", - "right": "pull_request:47054", + "left": "pull_request:47035", + "right": "pull_request:47069", "accept": false, - "reason": "Different models and failure modes." - }, - { - "left": "pull_request:46895", - "right": "pull_request:47090", - "accept": true, - "reason": "Same CohereASR double-shift fix; 47090 is a refined version of the same bug." + "reason": "Different TrainingArguments/configuration fixes with no shared bug." }, { - "left": "pull_request:46895", - "right": "pull_request:46958", - "accept": true, - "reason": "Same CohereASR training-loss double-shift issue and same files." - }, - { - "left": "pull_request:46958", - "right": "pull_request:47090", - "accept": true, - "reason": "Same concrete CohereASR loss-shift bug; mergeable into one PR." + "left": "pull_request:46957", + "right": "pull_request:47114", + "accept": false, + "reason": "FP8/MoE generation optimization and conversion-mapping docs are unrelated." }, { - "left": "pull_request:46907", - "right": "pull_request:47055", - "accept": true, - "reason": "Same EdgeTam forward-kwargs cleanup for post-processing keys." + "left": "pull_request:46707", + "right": "pull_request:47108", + "accept": false, + "reason": "FSDP2 migration and doc-builder comment cleanup are unrelated." }, { - "left": "pull_request:46949", - "right": "pull_request:47056", - "accept": true, - "reason": "Both patch the Whisper attention state update path; same underlying code-path bug." + "left": "pull_request:47077", + "right": "pull_request:47108", + "accept": false, + "reason": "Japanese translation changes and workflow comment cleanup are unrelated." }, { - "left": "pull_request:46950", - "right": "pull_request:46971", - "accept": true, - "reason": "Same Mamba2 integration/OOM-related test fix family." + "left": "pull_request:47078", + "right": "pull_request:47114", + "accept": false, + "reason": "Japanese wording edits and weight-converter docs update are unrelated." }, { - "left": "pull_request:46950", - "right": "pull_request:47058", - "accept": true, - "reason": "Same Mamba2 OOM/test-loading issue; compatible fixes." + "left": "pull_request:47093", + "right": "pull_request:47114", + "accept": false, + "reason": "New OpenPangu-V2 model support vs conversion-mapping documentation are unrelated." }, { - "left": "pull_request:46971", - "right": "pull_request:47058", - "accept": true, - "reason": "Same Mamba2 test memory/load problem, just different harness tweaks." + "left": "pull_request:47096", + "right": "pull_request:47114", + "accept": false, + "reason": "Exportable kimi support vs conversion-mapping docs are unrelated." }, { - "left": "pull_request:46926", - "right": "pull_request:47054", - "accept": true, - "reason": "Same MiniMax-M3-VL load_error/offload-folder workaround in the same test file." + "left": "pull_request:47117", + "right": "pull_request:47118", + "accept": false, + "reason": "Output-capture mapping changes and cache dispatch simplification are unrelated." }, { - "left": "pull_request:46908", - "right": "pull_request:47057", - "accept": true, - "reason": "Same DeepSeek-V3.2 load_error/offload-folder test fix." + "left": "pull_request:46958", + "right": "pull_request:47114", + "accept": false, + "reason": "CohereASR loss bug fix vs conversion-mapping docs are unrelated." }, { - "left": "pull_request:46949", - "right": "pull_request:46950", + "left": "pull_request:46903", + "right": "pull_request:47114", "accept": false, - "reason": "Different models and unrelated failures." + "reason": "CohereASR/PPFormulaNet loss fix vs conversion-mapping docs are unrelated." }, { - "left": "pull_request:47054", - "right": "pull_request:47057", + "left": "pull_request:47003", + "right": "pull_request:47096", "accept": false, - "reason": "Different models and unrelated test failures." + "reason": "OpenVINO exporter support and kimi exportability are different exporter features." }, { - "left": "pull_request:46949", - "right": "pull_request:46951", + "left": "pull_request:46954", + "right": "pull_request:47118", "accept": false, - "reason": "Whisper/audio model fix vs PEFT integration change are unrelated." + "reason": "Pytest-v9 GLM-ASR change vs cache dispatch simplification are unrelated." }, { - "left": "pull_request:46927", - "right": "pull_request:46949", + "left": "pull_request:47025", + "right": "pull_request:47107", "accept": false, - "reason": "GLM46V input-id cast fix is unrelated to Whisper/audio caching." + "reason": "Diffusion Gemma test fixes vs MoE decode optimization are unrelated." }, { - "left": "pull_request:46907", - "right": "pull_request:46927", + "left": "pull_request:46196", + "right": "pull_request:47118", "accept": false, - "reason": "Different models and bugs." + "reason": "Wav2Vec2 SDPA dispatch support and cache dispatch simplification are unrelated." }, { - "left": "pull_request:46908", - "right": "pull_request:47054", + "left": "pull_request:47089", + "right": "pull_request:47117", "accept": false, - "reason": "DeepSeek-V3.2 load fix vs MiniMax-M3-VL load fix are different models." + "reason": "AqlmConfig validation messages vs output-capturing changes are unrelated." }, { - "left": "pull_request:46926", - "right": "pull_request:47057", + "left": "pull_request:46957", + "right": "pull_request:47108", "accept": false, - "reason": "MiniMax-M3-VL vs DeepSeek-V3.2 are unrelated." + "reason": "FP8/MoE optimization and doc-builder workflow comment cleanup are unrelated." }, { - "left": "pull_request:47092", - "right": "pull_request:47107", + "left": "pull_request:46995", + "right": "pull_request:47003", "accept": false, - "reason": "Test expectations vs MoE decode optimization; unrelated." + "reason": "Hardware availability check and exporter addition are different features." }, { - "left": "pull_request:47096", - "right": "pull_request:47107", + "left": "pull_request:46954", + "right": "pull_request:47025", "accept": false, - "reason": "Exporter/codegen work vs MoE decode optimization are unrelated." + "reason": "GLM-ASR pytest compatibility and Diffusion Gemma test fixes are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The conclusions are grounded in the packet and the duplicate judgments are conservative. The CohereASR PRs are the only clearly same-bug cluster, and the rest of the pairwise rejections stay within evidence shown by titles/diffs. No obvious overstatement or non-mergeable pairing issue stands out." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, { "label": "pull_request", "component_index": 1, - "component_count": 34, - "review_unit_index": 3, - "review_unit_count": 3, + "component_count": 33, + "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-40546-57", + "cluster_id": "cluster-40546-41", "nodes": [ "pull_request:40546", "pull_request:43451", + "pull_request:44517", + "pull_request:45694", "pull_request:46196", "pull_request:46658", "pull_request:46707", "pull_request:46836", - "pull_request:46907", - "pull_request:46908", + "pull_request:46880", + "pull_request:46895", "pull_request:46912", - "pull_request:46926", - "pull_request:46927", - "pull_request:46970", - "pull_request:46971", - "pull_request:46986", + "pull_request:46957", + "pull_request:46958", "pull_request:47003", - "pull_request:47006", "pull_request:47025", "pull_request:47034", "pull_request:47035", - "pull_request:47054", - "pull_request:47055", - "pull_request:47056", - "pull_request:47057", - "pull_request:47058", "pull_request:47077", + "pull_request:47090", "pull_request:47092", - "pull_request:47096" + "pull_request:47096", + "pull_request:47100", + "pull_request:47107", + "pull_request:47113" ], "soft_pairs": [ - "pull_request:46927|pull_request:46971", - "pull_request:46908|pull_request:46926", - "pull_request:46986|pull_request:47006", - "pull_request:47054|pull_request:47055", - "pull_request:46970|pull_request:47006", - "pull_request:46970|pull_request:46986", - "pull_request:47057|pull_request:47058", - "pull_request:46907|pull_request:47054", - "pull_request:47055|pull_request:47056", - "pull_request:46907|pull_request:47056", - "pull_request:47054|pull_request:47056", - "pull_request:47054|pull_request:47058", - "pull_request:47055|pull_request:47057", - "pull_request:46907|pull_request:47006", - "pull_request:46907|pull_request:46970", - "pull_request:46970|pull_request:47055", - "pull_request:46907|pull_request:46986", + "pull_request:43451|pull_request:44517", + "pull_request:46836|pull_request:47107", + "pull_request:46912|pull_request:47113", + "pull_request:46912|pull_request:47100", + "pull_request:45694|pull_request:46707", + "pull_request:44517|pull_request:46836", + "pull_request:44517|pull_request:47096", + "pull_request:44517|pull_request:47003", + "pull_request:43451|pull_request:46957", + "pull_request:46957|pull_request:47096", + "pull_request:46957|pull_request:47003", + "pull_request:46658|pull_request:46895", + "pull_request:43451|pull_request:47100", + "pull_request:43451|pull_request:45694", + "pull_request:43451|pull_request:46895", + "pull_request:46895|pull_request:47096", + "pull_request:46880|pull_request:47092", + "pull_request:46895|pull_request:47090", + "pull_request:46895|pull_request:46958", + "pull_request:46958|pull_request:47090", "pull_request:47025|pull_request:47092", "pull_request:46707|pull_request:47034", "pull_request:46658|pull_request:46836", "pull_request:43451|pull_request:46836", "pull_request:46836|pull_request:47092", - "pull_request:43451|pull_request:47077", "pull_request:46836|pull_request:47096", + "pull_request:43451|pull_request:47077", + "pull_request:46836|pull_request:47003", + "pull_request:40546|pull_request:43451", "pull_request:43451|pull_request:46196", "pull_request:46196|pull_request:47096", - "pull_request:40546|pull_request:43451", "pull_request:46196|pull_request:47003", "pull_request:43451|pull_request:46658", "pull_request:46658|pull_request:47096", + "pull_request:46658|pull_request:47003", "pull_request:43451|pull_request:47092", "pull_request:43451|pull_request:47096", "pull_request:43451|pull_request:47003", @@ -23213,248 +22826,291 @@ "pull_request:47034|pull_request:47096", "pull_request:47092|pull_request:47096", "pull_request:43451|pull_request:47035", + "pull_request:47035|pull_request:47096", "pull_request:43451|pull_request:46912", - "pull_request:46836|pull_request:47003", - "pull_request:47003|pull_request:47034" + "pull_request:47003|pull_request:47034", + "pull_request:47003|pull_request:47092" ], - "prepared_review_unit_hash": "efa8588aa645432c08151d0aa0ee2050daff16c15be32d784851d108be6a283a", + "prepared_review_unit_hash": "f714fabfac04e2a6b3650546ccc473ba85e40f85d050bba923f20e0dd01ec0cf", "component_budget": { - "node_count": 57, - "item_count": 57, - "soft_pair_count": 182, - "serialized_chars": 204273, - "estimated_input_tokens": 51069, - "estimated_eval_tokens": 102394 + "node_count": 41, + "item_count": 41, + "soft_pair_count": 119, + "serialized_chars": 146957, + "estimated_input_tokens": 36740, + "estimated_eval_tokens": 73736 }, "budget": { - "node_count": 27, - "item_count": 27, - "soft_pair_count": 41, - "serialized_chars": 76898, - "estimated_input_tokens": 19225, - "estimated_eval_tokens": 38706 + "node_count": 24, + "item_count": 24, + "soft_pair_count": 47, + "serialized_chars": 76358, + "estimated_input_tokens": 19090, + "estimated_eval_tokens": 38436 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is heterogeneous: new model add-ons, exporter/runtime changes, training tweaks, docs updates, and many unrelated test-only fixes. None of the proposed soft edges look like the same underlying bug or change.", - "confidence": 0.98, + "summary": "Most pairs are not duplicates: they\u2019re unrelated model additions, exporter features, test tweaks, or doc-only changes. The only real duplicate cluster here is the CohereASR training-loss double-shift fixes, where all three PRs address the same loss-path bug.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the items span distinct models and subsystems, with only superficial overlap such as test failures or registration/docs work.", + "canonical_pr_reason": "pull_request:47090 is the clearest expression of the shared CohereASR bug fix: it names the exact training-loss double-shift problem and matches the same code path as 46895 and 46958.", "best_issue_reason": null, - "best_pr_reason": "No global best PR: there is no single representative duplicate because the cluster is not one coherent change.", + "best_pr_reason": "pull_request:47090 is the best representative PR for the duplicate set because it is the most explicit about the underlying bug and is directly interchangeable with the other two CohereASR fixes.", "soft_edge_verdicts": [ { - "left": "pull_request:46927", - "right": "pull_request:46971", + "left": "pull_request:43451", + "right": "pull_request:44517", "accept": false, - "reason": "Different model families and different fixes; one is GLM46V test/model behavior, the other is a Mamba2 OOM test patch." + "reason": "Both add different models (Molmo2 vs Qwen3-TTS); same kind of integration work, but not the same change." }, { - "left": "pull_request:46908", - "right": "pull_request:46926", + "left": "pull_request:46836", + "right": "pull_request:47107", "accept": false, - "reason": "Both are test-fix PRs, but for unrelated models and unrelated failure modes (DeepSeekV32 load error vs MiniMax-M3-VL load error)." + "reason": "Unrelated changes: Unlimited OCR model integration vs MoE decode optimization in generation." }, { - "left": "pull_request:46986", - "right": "pull_request:47006", + "left": "pull_request:46912", + "right": "pull_request:47113", "accept": false, - "reason": "Same broad symptom class, but different models and code paths (glm_ocr vs gemma3); not a duplicate change." + "reason": "RoPE test/model coverage work is unrelated to audio feature/processor requirement fixes." }, { - "left": "pull_request:47054", - "right": "pull_request:47055", + "left": "pull_request:46912", + "right": "pull_request:47100", "accept": false, - "reason": "Different models and different failure causes; only the triage label is similar." + "reason": "Different problems: RoPE tests vs a kernels API/deprecation compatibility fix." }, { - "left": "pull_request:46970", - "right": "pull_request:47006", + "left": "pull_request:45694", + "right": "pull_request:46707", "accept": false, - "reason": "Unrelated model fixes (musicgen_melody vs gemma3); shared output-mismatch wording is too broad." + "reason": "TrainingArguments batch-size semantics and FSDP2 migration are unrelated changes." }, { - "left": "pull_request:46970", - "right": "pull_request:46986", + "left": "pull_request:44517", + "right": "pull_request:46836", "accept": false, - "reason": "Different models and test targets; no common code-path bug." + "reason": "Different new model integrations; same broad auto-mapping/docs pattern, but not the same model or bug." }, { - "left": "pull_request:47057", - "right": "pull_request:47058", + "left": "pull_request:44517", + "right": "pull_request:47096", "accept": false, - "reason": "Both touch integration tests, but one is DeepSeekV32 load_error and the other is Mamba2 OOM; unrelated." + "reason": "Qwen3-TTS integration and Kimi exportability are separate features." }, { - "left": "pull_request:46907", - "right": "pull_request:47054", + "left": "pull_request:44517", + "right": "pull_request:47003", "accept": false, - "reason": "Different models and different issues; one patches EdgeTam import/config handling, the other MiniMax-M3-VL loading." + "reason": "Qwen3-TTS model support is unrelated to adding an OpenVINO exporter." }, { - "left": "pull_request:47055", - "right": "pull_request:47056", + "left": "pull_request:43451", + "right": "pull_request:46957", "accept": false, - "reason": "Different models and fixes; EdgeTam post-processing kwargs cleanup is unrelated to Whisper attention behavior." + "reason": "Molmo2 model support and finegrained FP8 decode optimization are different changes." }, { - "left": "pull_request:46907", - "right": "pull_request:47056", + "left": "pull_request:46957", + "right": "pull_request:47096", "accept": false, - "reason": "No shared concrete bug: EdgeTam import/config handling vs Whisper attention state handling." + "reason": "Both touch generation/export code, but they solve different issues in different paths." }, { - "left": "pull_request:47054", - "right": "pull_request:47056", + "left": "pull_request:46957", + "right": "pull_request:47003", "accept": false, - "reason": "Different model-specific test/code fixes; not the same underlying problem." + "reason": "Finegrained FP8 decode optimization is not the same as adding an OpenVINO exporter backend." }, { - "left": "pull_request:47054", - "right": "pull_request:47058", + "left": "pull_request:46658", + "right": "pull_request:46895", "accept": false, - "reason": "MiniMax-M3-VL loading/OOM issues are unrelated to the other model-specific fix." + "reason": "Step-3.7-Flash conversion/model support is unrelated to CohereASR loss-shift fixing." }, { - "left": "pull_request:47055", - "right": "pull_request:47057", + "left": "pull_request:43451", + "right": "pull_request:47100", "accept": false, - "reason": "EdgeTam import/config cleanup and DeepSeekV32 load_error are different bugs." + "reason": "New model support and a kernels API compatibility fix are unrelated." }, { - "left": "pull_request:46907", - "right": "pull_request:47006", + "left": "pull_request:43451", + "right": "pull_request:45694", "accept": false, - "reason": "Different model families and distinct failure surfaces; not mergeable as one PR." + "reason": "Molmo2 integration and batch-size semantics in TrainingArguments are unrelated." }, { - "left": "pull_request:46907", - "right": "pull_request:46970", + "left": "pull_request:43451", + "right": "pull_request:46895", "accept": false, - "reason": "EdgeTam import/config and MusicGen-Melody output mismatch are unrelated." + "reason": "Molmo2 support and CohereASR loss correction are unrelated." }, { - "left": "pull_request:46970", - "right": "pull_request:47055", + "left": "pull_request:46895", + "right": "pull_request:47096", "accept": false, - "reason": "Different models and unrelated fixes." + "reason": "CohereASR loss-shift fix and exportability work for Kimi are unrelated." }, { - "left": "pull_request:46907", - "right": "pull_request:46986", + "left": "pull_request:46880", + "right": "pull_request:47092", "accept": false, - "reason": "EdgeTam import/config handling does not overlap with glm_ocr output-mismatch fixes." + "reason": "Adding XPU expectations to tests and adding XPU expectations to different tests are separate test-only changes." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47090", + "accept": true, + "reason": "Same underlying CohereASR training-loss double-shift bug and same code path." + }, + { + "left": "pull_request:46895", + "right": "pull_request:46958", + "accept": true, + "reason": "Both fix CohereASR by correcting the training loss computation against the same labels path." + }, + { + "left": "pull_request:46958", + "right": "pull_request:47090", + "accept": true, + "reason": "These are the same CohereASR double-shift fix expressed with different implementation text." }, { "left": "pull_request:47025", "right": "pull_request:47092", "accept": false, - "reason": "One is DiffusionGemma test adjustments; the other adds XPU expectations for MiniCPM3. Different models and goals." + "reason": "Diffusion Gemma test adjustments and MiniCPM3 XPU expectations are unrelated test maintenance." }, { "left": "pull_request:46707", "right": "pull_request:47034", "accept": false, - "reason": "Both mention infrastructure, but FSDP2 migration and Neuron SFT experiment are unrelated changes." + "reason": "FSDP2 migration and Neuron SFT experiment scaffolding are different workstreams." }, { "left": "pull_request:46658", "right": "pull_request:46836", "accept": false, - "reason": "Both are model-adding PRs, but for completely different models and code paths." + "reason": "Step-3.7-Flash support and Unlimited OCR support are different model additions." }, { "left": "pull_request:43451", - "right": "pull_request:47077", + "right": "pull_request:46836", "accept": false, - "reason": "Molmo2 model integration vs Japanese DistilBERT docs translation are unrelated." + "reason": "Different model integrations, not the same underlying bug or change." }, { "left": "pull_request:46836", "right": "pull_request:47092", "accept": false, - "reason": "Unlimited OCR model integration and MiniCPM3 XPU test expectations are unrelated." + "reason": "Unlimited OCR integration and XPU test expectations for MiniCPM3 are unrelated." }, { - "left": "pull_request:40546", - "right": "pull_request:43451", + "left": "pull_request:46836", + "right": "pull_request:47096", "accept": false, - "reason": "Both add models, but VibeVoice and Molmo2 are separate model integrations with no shared bug/change." + "reason": "Unlimited OCR model support and exportability fixes for Kimi are unrelated." }, { - "left": "pull_request:46196", + "left": "pull_request:43451", + "right": "pull_request:47077", + "accept": false, + "reason": "Molmo2 support and Japanese translation updates are unrelated." + }, + { + "left": "pull_request:46836", "right": "pull_request:47003", "accept": false, - "reason": "wav2vec2 SDPA dispatch support and OpenVINO exporter work are unrelated." + "reason": "Unlimited OCR integration and OpenVINO exporter work are unrelated." + }, + { + "left": "pull_request:40546", + "right": "pull_request:43451", + "accept": false, + "reason": "VibeVoice and Molmo2 are distinct model integrations." }, { "left": "pull_request:43451", - "right": "pull_request:46658", + "right": "pull_request:46196", "accept": false, - "reason": "Different model additions; only broad registration/docs patterns overlap." + "reason": "Model integration work and SDPA dispatch support for wav2vec2 are unrelated." }, { - "left": "pull_request:46196", - "right": "pull_request:47096", + "left": "pull_request:43451", + "right": "pull_request:46658", "accept": false, - "reason": "Model-attn dispatch support vs exportability changes for Kimi are unrelated." + "reason": "Molmo2 addition and Step-3.7-Flash conversion/model work are unrelated." }, { "left": "pull_request:43451", "right": "pull_request:47092", "accept": false, - "reason": "Molmo2 integration and MiniCPM3 test expectation updates are unrelated." + "reason": "Molmo2 model support and MiniCPM3 XPU test expectations are unrelated." }, { "left": "pull_request:43451", "right": "pull_request:47096", "accept": false, - "reason": "Molmo2 model integration and exportable Kimi changes are different workstreams." + "reason": "Molmo2 model support and Kimi exportability fixes are unrelated." }, { - "left": "pull_request:46658", - "right": "pull_request:47096", + "left": "pull_request:43451", + "right": "pull_request:47003", "accept": false, - "reason": "Step-3.7 Flash integration and Kimi exportability are unrelated model changes." + "reason": "Molmo2 model support and OpenVINO exporter work are unrelated." }, { "left": "pull_request:43451", - "right": "pull_request:47003", + "right": "pull_request:47034", "accept": false, - "reason": "Molmo2 integration and OpenVINO exporter implementation are unrelated." + "reason": "Molmo2 model integration and Neuron SFT experiment code are unrelated." }, { - "left": "pull_request:47003", + "left": "pull_request:47034", + "right": "pull_request:47096", + "accept": false, + "reason": "Neuron SFT experiment scaffolding and Kimi exportability fixes are unrelated." + }, + { + "left": "pull_request:47092", "right": "pull_request:47096", "accept": false, - "reason": "OpenVINO exporter and Kimi exportability touch exporting, but they fix different code paths and are not one PR." + "reason": "Different concerns: test expectations for MiniCPM3 vs exporter compatibility for Kimi." }, { "left": "pull_request:43451", - "right": "pull_request:47034", + "right": "pull_request:47035", + "accept": false, + "reason": "Molmo2 integration and torch.compile backend defaults for Neuron are unrelated." + }, + { + "left": "pull_request:47035", + "right": "pull_request:47096", "accept": false, - "reason": "Molmo2 model integration and Neuron SFT experiment are unrelated." + "reason": "Torch.compile backend defaulting and Kimi exportability are unrelated changes." }, { "left": "pull_request:43451", "right": "pull_request:46912", "accept": false, - "reason": "Molmo2 integration and RoPE test coverage are unrelated." + "reason": "Molmo2 support and RoPE testing changes are unrelated." }, { - "left": "pull_request:46836", - "right": "pull_request:47003", + "left": "pull_request:47003", + "right": "pull_request:47034", "accept": false, - "reason": "Unlimited OCR model integration and OpenVINO exporter are separate features." + "reason": "OpenVINO exporter work and Neuron experiment scaffolding are unrelated." }, { - "left": "pull_request:43451", - "right": "pull_request:47035", + "left": "pull_request:47003", + "right": "pull_request:47092", "accept": false, - "reason": "Molmo2 model integration and torch-compile backend defaulting to neuron are unrelated." + "reason": "OpenVINO exporter changes and MiniCPM3 test expectations are unrelated." } ] }, @@ -23468,7 +23124,7 @@ { "label": "pull_request", "component_index": 2, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -23484,7 +23140,7 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "3af50daa7270ce0289f325b6305ae05079e799b05412a6a19745e11251b59768", + "prepared_review_unit_hash": "ae0228b294007c1df897730f5e3e6da0a3439be44e8a43f240af761654a34edf", "component_budget": { "node_count": 2, "item_count": 2, @@ -23506,18 +23162,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests are unrelated: one adds TimesFM covariate forecasting support, while the other adds DINOv3 image classification support. They do not fix the same code-path problem and should not be merged as duplicates.", + "summary": "These PRs are unrelated: one adds TimesFM covariate forecasting support, the other adds DINOv3 image classification support. They do not share a concrete bug or change target, so they should not be clustered as duplicates.", "confidence": 0.99, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 41224 is the more focused, self-contained change, but it is not a duplicate of PR 40755.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different model families and different features: TimesFM covariate forecasting vs DINOv3 image classification. No shared bug or mergeable code-path change." + "reason": "Different models and code paths (TimesFM forecasting vs DINOv3 image classification); no shared underlying bug or mergeable change." } ] }, @@ -23531,7 +23187,7 @@ { "label": "pull_request", "component_index": 3, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -23594,7 +23250,7 @@ { "label": "pull_request", "component_index": 4, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -23638,12 +23294,12 @@ "pull_request:43823|pull_request:45213", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "87ca94dfbedd554f3ca28e32b0110b19532ab32a5f118cbe222b80bfffcf6ebc", + "prepared_review_unit_hash": "eed9b0f61172e747b7743f0c157d77938f29cc2bee49926b57ec2f1f72eb55b8", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 18, - "serialized_chars": 34787, + "serialized_chars": 34786, "estimated_input_tokens": 8697, "estimated_eval_tokens": 17650 }, @@ -23651,7 +23307,7 @@ "node_count": 13, "item_count": 13, "soft_pair_count": 18, - "serialized_chars": 34787, + "serialized_chars": 34786, "estimated_input_tokens": 8697, "estimated_eval_tokens": 17650 }, @@ -23660,126 +23316,126 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Mostly unrelated PRs spanning tokenizer/model additions, docs, tests, and refactors. The only clear duplicate pair is the two `fixed_cross_entropy` changes in `loss_utils.py` for the same issue; the rest are not the same underlying bug/change.", - "confidence": 0.91, + "summary": "This cluster is mostly unrelated PRs that happen to touch Transformers internals; the only true duplicate pair is the two loss-utils PRs that both forward extra kwargs into fixed_cross_entropy. Everything else is just broad thematic overlap or different subsystems.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "`pull_request:43251` is the best representative of the only real duplicate cluster: it explicitly targets issue 43240, makes the concrete `cross_entropy` kwargs fix in `loss_utils.py`, and is slightly more complete/annotated than 43254.", + "canonical_pr_reason": "PR 43251 is the best canonical representative: it is the more complete, issue-linked fix for fixed_cross_entropy forwarding kwargs to nn.functional.cross_entropy, and it cleanly subsumes the overlapping 43254 change.", "best_issue_reason": null, - "best_pr_reason": "`pull_request:43251` is the strongest single PR in this set because it is explicitly issue-linked, fixes a specific code path, and cleanly subsumes the narrower variant in 43254.", + "best_pr_reason": "PR 43251 is the strongest standalone duplicate target in the cluster because it addresses the concrete cross-entropy kwargs bug directly and has the explicit issue link plus review activity.", "soft_edge_verdicts": [ { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Mistral tokenizer methods vs Executorch dynamic-shape export tests; unrelated changes." + "reason": "Tokenizer missing-methods work and executorch dynamic-shape export tests are unrelated changes." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Tokenizer API additions vs new MobileLLM model files; different features." + "reason": "Mistral tokenizer methods vs adding MobileLLM are different models and different fixes." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor auto-loading robustness vs adding MobileLLM; only broad area overlap, not same bug." + "reason": "Processor robustness for AudioFlamingo3 is unrelated to introducing MobileLLM." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "`fixed_cross_entropy` kwargs fix vs MobileLLM model addition; unrelated." + "reason": "Cross-entropy kwargs forwarding and MobileLLM model addition are unrelated code paths." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "Cross-entropy helper fix vs Mistral4 test/model changes; different code paths." + "reason": "Loss-function kwargs fix and Mistral4 test/model changes are not the same underlying bug." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Executorch export test vs Mistral4 tests/model fixes; not the same underlying issue." + "reason": "Executorch export test coverage and Mistral4 test fixes are different changes." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Executorch dynamic-shape test vs MobileLLM addition; unrelated." + "reason": "Executorch export testing and MobileLLM model addition do not target the same issue." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "Adding MobileLLM vs fixing `cls.config` resolution order; no shared bug/change." + "reason": "New model implementation and config-resolution ordering fix are unrelated." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM model addition vs Falcon output-collection refactor; separate work." + "reason": "MobileLLM addition and Falcon output-collection refactor are different model-level changes." }, { "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "Falcon interface refactor vs model-creation skill/docs; unrelated." + "reason": "Falcon forward/refactor work is unrelated to the model-creation skill PR." }, { "left": "pull_request:43995", "right": "pull_request:44827", "accept": false, - "reason": "Falcon output interface refactor vs Mistral4 tests; different models and changes." + "reason": "Falcon output-collection refactor and Mistral4 tests do not fix the same bug." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test fixes vs CLIP-like `layrnorm` renaming; unrelated despite both touching model code." + "reason": "Mistral4 test fixes and a layrnorm-to-layernorm rename are different issues." }, { "left": "pull_request:44827", "right": "pull_request:46267", "accept": false, - "reason": "Mistral4 tests vs chat parsing/serve updates; different subsystems." + "reason": "Mistral4 test fixes and chat-response parsing updates are unrelated subsystems." }, { "left": "pull_request:44827", "right": "pull_request:46048", "accept": false, - "reason": "Mistral4 tests vs TDT loss kernel integration; not the same issue." + "reason": "Mistral4 test cleanup and adding a TDT loss kernel are not the same change." }, { "left": "pull_request:44827", "right": "pull_request:45655", "accept": false, - "reason": "Mistral4 tests vs config resolution ordering fix; unrelated." + "reason": "Mistral4 tests and cls.config resolution order are unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 tests vs model-creation skill; no shared underlying change." + "reason": "Mistral4 test fixes and the model-creation skill are entirely different artifacts." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "MobileLLM model addition vs model-creation skill scaffolding; different artifact types and goals." + "reason": "MobileLLM model addition and the model-creation skill are not the same underlying code change." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs modify `loss_utils.py` to forward the same `cross_entropy` options for issue 43240; they are alternative implementations of the same fix and could be merged into one PR." + "reason": "Both PRs patch the same concrete code path in fixed_cross_entropy by forwarding supported kwargs to nn.functional.cross_entropy, and they could plausibly be merged into one PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "The assessment is grounded in the packet and stays conservative. The only supported duplicate cluster is PRs 43251 and 43254: both target issue 43240, edit the same `loss_utils.py` code path, and appear mergeable as alternative implementations of the same fix. The other pairwise rejections are also reasonable given the distinct files, features, and subsystems shown in the diffs." + "feedback": "The cluster summary is grounded in the packet: the only clearly mergeable same-bugfix pair is PR 43251/43254 on fixed_cross_entropy, while the rest are distinct models/subsystems. The soft-edge rejections are conservative and the canonical/duplicate reasoning is supported by the issue link and overlapping diff." }, "evaluator_used": true, "retried": false, @@ -23790,7 +23446,7 @@ { "label": "pull_request", "component_index": 5, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -23806,7 +23462,7 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "e4391716904425372860adfc1807bfb4ceb853487a8732f58518e10787d0f590", + "prepared_review_unit_hash": "60d54e4042a086551d05822b7ca30a4dae8e7097966f10f0f0cf93dd48f6f36f", "component_budget": { "node_count": 2, "item_count": 2, @@ -23828,18 +23484,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate workflow-maintenance PRs: #42975 upgrades several GitHub Actions for Node 24 compatibility, while #42976 bumps docker/build-push-action versions. They touch overlapping workflow files, but the underlying changes and failure modes are different, so they should not be deduplicated.", - "confidence": 0.95, + "summary": "Two open PRs that both bump GitHub Actions dependencies in workflow files, but they target different action families and different upgrade goals. They overlap in affected workflows, yet they are not the same concrete change and should not be merged as duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "#42975 is the better canonical representative only because it has the broader and more clearly stated purpose (Node 24 compatibility across multiple GitHub Actions). However, it is not a duplicate of #42976; the PRs update different actions and address different concerns.", + "canonical_pr_reason": "No issue is present. If one PR must represent the cluster, PR #42975 is the better anchor because it is the more specific compatibility-focused workflow action upgrade and better explains the shared automation maintenance theme.", "best_issue_reason": null, - "best_pr_reason": "#42975 is the best single PR to represent the cluster because it has the broader workflow-action upgrade scope and explicit compatibility motivation. #42976 is narrower and specifically about docker/build-push-action updates.", + "best_pr_reason": "PR #42975 is the better representative because its title and diff clearly center on a single compatibility objective (Node 24 support) across workflow actions, whereas PR #42976 is a separate, narrower set of action version bumps.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Reject: both are GitHub Actions dependency bumps, but they change different actions and different code paths. #42975 updates checkout/cache/upload-artifact for Node 24 compatibility; #42976 updates docker/build-push-action. They are related maintenance work, not the same underlying change." + "reason": "Reject as duplicate: both are workflow-maintenance PRs, but #42975 updates checkout/cache/upload-artifact for Node 24 compatibility while #42976 updates docker/build-push-action and a few other actions to latest versions. They touch some of the same workflows, but they do not appear to fix the same underlying bug or change the same code path." } ] }, @@ -23853,7 +23509,7 @@ { "label": "pull_request", "component_index": 6, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -23874,12 +23530,12 @@ "pull_request:43448|pull_request:45490", "pull_request:45490|pull_request:46562" ], - "prepared_review_unit_hash": "c822b45629263a85c18209645a2071ce0afc5efbce51149e1c5e99833cfe5cdf", + "prepared_review_unit_hash": "dfda101fa3db3c4fcead9e2eef1e6e69535d8c685b2626494e6a57b021e05910", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 13046, + "serialized_chars": 13047, "estimated_input_tokens": 3262, "estimated_eval_tokens": 6780 }, @@ -23887,7 +23543,7 @@ "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 13046, + "serialized_chars": 13047, "estimated_input_tokens": 3262, "estimated_eval_tokens": 6780 }, @@ -23896,36 +23552,36 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All four PRs are unrelated model-addition changes. They share template files like docs/_toctree.yml and auto-registry plumbing, but each introduces a different architecture (ViT NEPA, Molmo, CTSM, Step3p7) with distinct model codepaths.", - "confidence": 0.96, + "summary": "This cluster groups four independent model-addition PRs (ViT NEPA, Molmo, CTSM, Step3p7). They share some boilerplate docs/auto-registration files, but each introduces a different model family and codepath, so they are not duplicates.", + "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: these are separate feature PRs for different models, not duplicates of the same change.", + "canonical_pr_reason": "PR 45490 is the best representative only in a bookkeeping sense: it has the broadest review activity and touches the standard model-registration stack, but it is not a true duplicate of the others.", "best_issue_reason": null, - "best_pr_reason": "No best duplicate representative among them; the overlap is only in shared model-registration/documentation scaffolding, not the underlying model implementations.", + "best_pr_reason": "None of the PRs is a good global duplicate target; they add different models and would not be mergeable into one fix/change.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model families and codepaths: ViT NEPA vs Molmo. Shared docs/auto files are incidental to adding new models, not evidence of the same change." + "reason": "Different model additions (ViT NEPA vs Molmo); shared docs/auto files are generic scaffolding, not the same underlying change." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different architectures and targets: ViT NEPA vs CTSM. Similar registry/docs edits do not make these the same bug or feature." + "reason": "Different architectures and codepaths (ViT NEPA vs CTSM); overlap is only in registry/docs plumbing." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo and CTSM are distinct model additions with different source trees and tests; no shared concrete fix or implementation." + "reason": "Molmo and CTSM are unrelated model implementations; shared filenames do not indicate a duplicate bug or change." }, { "left": "pull_request:45490", "right": "pull_request:46562", "accept": false, - "reason": "Both touch auto-mapping/model registration, but they add different models (CTSM vs Step3p7) and would not plausibly merge into one PR as the same underlying change." + "reason": "CTSM and Step3p7 are separate model additions; despite some shared auto-mapping/timeseries files, they are not the same concrete change." } ] }, @@ -23939,7 +23595,7 @@ { "label": "pull_request", "component_index": 7, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -23974,133 +23630,133 @@ "pull_request:46539|pull_request:46922", "pull_request:46886|pull_request:46922" ], - "prepared_review_unit_hash": "6525799ae2d08ef4e37bf3af4094092af8b7cc2278d53aafb1e6a9a510105b90", + "prepared_review_unit_hash": "9a9ea4746a9369d4995f5c108adc338510e822d68a4310dfdac73aa714198a15", "component_budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 15, - "serialized_chars": 19604, - "estimated_input_tokens": 4901, - "estimated_eval_tokens": 10058 + "serialized_chars": 19590, + "estimated_input_tokens": 4898, + "estimated_eval_tokens": 10052 }, "budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 15, - "serialized_chars": 19604, - "estimated_input_tokens": 4901, - "estimated_eval_tokens": 10058 + "serialized_chars": 19590, + "estimated_input_tokens": 4898, + "estimated_eval_tokens": 10052 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: only the two YOLOS PRs appear to fix the same concrete device_map=\"auto\" device-mismatch bug. The rest are unrelated test, workflow, generation, quantization, or model-specific changes and should not be deduplicated.", + "summary": "The cluster contains one clear duplicate pair in YOLOS (46886/46922). The remaining PRs are not duplicates: they cover separate fixes in tests, FP8/loading, generation validation, and CI/workflow infrastructure, with substantive code changes in several of them.", "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46922 is the best canonical PR because it contains the direct YOLOS device-mismatch fix in the model forward path and also includes the no_split_modules adjustment already present in 46886, making it the more complete representative of that bug.", + "canonical_pr_reason": "PR 46922 is the best canonical representative for the YOLOS pair because it includes the concrete forward-path device fix and the matching no_split_modules update; 46886 only adds the metadata-side split guard.", "best_issue_reason": null, - "best_pr_reason": "PR 46922 is the best overall representative of the only real duplicate pair in this set: it addresses the YOLOS device placement bug more completely than 46886 and is the clearest stand-in for the underlying change.", + "best_pr_reason": "PR 46922 is the strongest standalone PR in the cluster because it addresses the actual YOLOS runtime mismatch directly and subsumes the smaller metadata-only change in 46886.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Different models and different test changes; no shared code-path bug." + "reason": "Both are test-related, but they target different models and different failures; no same underlying bug or change." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper tokenizer/test fix vs FP8 dequantization work; unrelated changes." + "reason": "Whisper tokenization tests and FP8/dequantization work are unrelated fixes." }, { "left": "pull_request:43291", "right": "pull_request:46380", "accept": false, - "reason": "Both touch Whisper-related files, but they address different problems in different code paths." + "reason": "Different subsystems and different issues; the overlap is only superficial test activity." }, { "left": "pull_request:43291", "right": "pull_request:46922", "accept": false, - "reason": "Whisper tokenizer/tests vs YOLOS device placement; clearly unrelated." + "reason": "Whisper test cleanup is unrelated to the YOLOS device-mapping fix." }, { "left": "pull_request:43665", "right": "pull_request:46539", "accept": false, - "reason": "Test-file edits versus CI infra and Phimoe integration setup; not the same bug." + "reason": "Both touch tests or CI, but they address different models and different problems." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "CLIP/ViT test noise is unrelated to the FP8 loading/dequant fix." + "reason": "CLIP/ViT test edits are unrelated to the FP8 loading/dequantization changes." }, { "left": "pull_request:43665", "right": "pull_request:46380", "accept": false, - "reason": "Different subsystems and different fixes; only superficial similarity in being PRs with test changes." + "reason": "Test edits versus generation/watermarking and whisper-related code are not the same change." }, { "left": "pull_request:43665", "right": "pull_request:46922", "accept": false, - "reason": "Unrelated test-only PR versus YOLOS model fix." + "reason": "Different model families and different fixes; the shared test framing is not enough." }, { "left": "pull_request:44815", "right": "pull_request:46539", "accept": false, - "reason": "FP8/dequantization changes are unrelated to CI memory patching and Phimoe tests." + "reason": "FP8 quantization/loading changes are unrelated to the Phimoe CI/runtime fix." }, { "left": "pull_request:44815", "right": "pull_request:46380", "accept": false, - "reason": "Different loading/quantization fix versus generation/watermarking and Whisper changes." + "reason": "No shared concrete bug: one is FP8/dequantization, the other is generation validation plus broader test updates." }, { "left": "pull_request:44815", "right": "pull_request:46922", "accept": false, - "reason": "Different model families and unrelated bugs." + "reason": "Different model code paths and different runtime issues; not the same fix." }, { "left": "pull_request:46380", "right": "pull_request:46539", "accept": false, - "reason": "Generation/watermarking fix versus CI/test-environment fix; not the same underlying issue." + "reason": "Generation watermarking validation and CI memory/device-map infrastructure are separate problems." }, { "left": "pull_request:46380", "right": "pull_request:46922", "accept": false, - "reason": "Whisper generation/config change and YOLOS device mismatch are unrelated." + "reason": "These are unrelated fixes in different subsystems." }, { "left": "pull_request:46539", "right": "pull_request:46922", "accept": false, - "reason": "CI memory/device-map workaround and YOLOS model-device fix are different problems." + "reason": "CI device/memory patching for Phimoe is unrelated to the YOLOS dispatch bug." }, { "left": "pull_request:46886", "right": "pull_request:46922", "accept": true, - "reason": "Both target the same YOLOS device_map=\"auto\" device mismatch in modeling_yolos; 46922 adds the missing forward-path device move and the same no_split_modules adjustment, so they are mergeable into one fix." + "reason": "Both target the same YOLOS device_map=\"auto\" mismatch in the same file and issue; 46922 adds the concrete forward-path fix on top of the metadata change." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The analysis is grounded in the packet and stays conservative on the non-YOLOS pairs. The only accepted duplicate pair, 46886 vs 46922, is well-supported by the shared issue target and shared YOLOS device-mismatch context, and the canonical PR choice is reasonable because 46922 includes the forward-path fix plus the no_split_modules change. No obvious overstatement beyond the evidence." + "accept": false, + "feedback": "The duplicate-pair call on 46886/46922 is grounded: they share the YOLOS file and issue target, and 46922 adds the forward-path fix on top of the metadata change. However, the cluster summary overstates the rest of the set as \"test-only\"; 44815, 46380, and 46539 all include substantive code and workflow changes beyond tests. Keep the unrelated-vs-duplicate distinction, but make the non-duplicate summary more conservative." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -24108,7 +23764,7 @@ { "label": "pull_request", "component_index": 8, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -24127,12 +23783,12 @@ "pull_request:43488|pull_request:43532", "pull_request:43532|pull_request:46747" ], - "prepared_review_unit_hash": "ff7f1f6e40410fce075b3e0ac1bda2d75441f55533d4fdb0dda8d7d7d9f26e0d", + "prepared_review_unit_hash": "f3b5dfa29b1eeeec5d73304b01518884a99a4686aab5f00d1d82b9fd3b95f9b0", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 6346, + "serialized_chars": 6348, "estimated_input_tokens": 1587, "estimated_eval_tokens": 3430 }, @@ -24140,7 +23796,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 6346, + "serialized_chars": 6348, "estimated_input_tokens": 1587, "estimated_eval_tokens": 3430 }, @@ -24149,30 +23805,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate synthetic \"don't merge\" PRs touching different code paths: packaging/setup, comment-driven CI workflows/testing utilities, and the PR security gate workflow. They share only broad repository/CI similarity, not the same underlying change.", - "confidence": 0.96, + "summary": "All three PRs are separate \u201c[don\u2019t merge]\u201d experiments touching different areas: packaging/test deps, CI/comment-workflow plumbing, and a security-gate workflow change. The shared similarity is superficial, so there is no duplicate/canonical PR here.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No canonical PR: the three pull requests address different concrete changes and do not look like the same underlying fix or feature.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "No single PR is a good duplicate target; the overlap is only broad subsystem/test-file similarity, not the same code-path problem.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:46747", "accept": false, - "reason": "Different changes: 43488 edits setup/dependency/test files for repo-bot formatting checks, while 46747 only changes the PR security-gate workflow from pull_request_target to pull_request. Same broad CI/test area, but not the same concrete bug or mergeable fix." + "reason": "Different changes: 43488 edits setup/dependency versions and a test file, while 46747 only changes the PR security-gate workflow. They do not fix the same bug or change the same code path." }, { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "43488 is a packaging/version/dependency test PR; 43532 rewrites several GitHub workflows plus testing/notification code to show diffs. They do not fix the same code-path problem." + "reason": "Only a loose shared test filename appears. 43488 is about packaging/dependency/version tweaks, whereas 43532 rewires multiple CI/workflow files and test utilities. Not the same underlying change." }, { "left": "pull_request:43532", "right": "pull_request:46747", "accept": false, - "reason": "43532 targets self-comment CI, scheduled/slack workflows, testing utils, and a Vit test; 46747 only touches the security-gate workflow trigger. Overlapping CI context is too broad to treat as the same underlying change." + "reason": "Both touch CI/workflow infrastructure, but the concrete edits differ: 43532 changes comment/scheduled/slack workflows plus testing utils, while 46747 only switches the security gate trigger. Too broad to treat as duplicates." } ] }, @@ -24186,7 +23842,7 @@ { "label": "pull_request", "component_index": 9, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -24202,12 +23858,12 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "fdf70469ba1335a3d77ad4a18b0d689fe6098eb7e34dca20393c00b1a350ac63", + "prepared_review_unit_hash": "6638dac6ea438fa5d496c92b7b353a885e0e663c03dff08256cdb5f5a22d90d9", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6081, + "serialized_chars": 6083, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -24215,7 +23871,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6081, + "serialized_chars": 6083, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -24224,18 +23880,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related pipeline-feature PRs with shared docs/registration scaffolding, but they implement different segmentation tasks and code paths: Promptable Concept Segmentation vs Promptable Visual Segmentation. Not duplicates.", - "confidence": 0.97, + "summary": "These are related segmentation pipeline additions, but they target different tasks and code paths: Promptable Concept Segmentation vs Promptable Visual Segmentation. They are not duplicates.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the two pull requests add different pipelines for different underlying tasks/model families, so neither is a duplicate of the other.", + "canonical_pr_reason": "No single canonical PR here; the two PRs introduce distinct pipelines with different model/test/doc surfaces.", "best_issue_reason": null, - "best_pr_reason": "No single PR is the global best representative because the cluster is not a duplicate set; both PRs are valid but distinct feature additions.", + "best_pr_reason": "Neither PR is a duplicate of the other. If forced to pick a representative, 43613 is broader, but it is not a substitute for 43612 because it serves a different segmentation task.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Shared documentation and registration files are generic scaffolding, but the actual feature targets are different pipelines with different task semantics and supporting models. They do not look like the same underlying bug/change." + "reason": "Reject: the PRs add different pipelines for different tasks and model stacks. Sharing docs/registry updates and a broad segmentation theme is not enough to treat them as the same fix/change." } ] }, @@ -24249,7 +23905,7 @@ { "label": "pull_request", "component_index": 10, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -24265,40 +23921,40 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "2fe348846f29e07a816c0e96d7d3f368241f04ebc6a4eccc3d0ae0fba41dcde2", + "prepared_review_unit_hash": "4c51409bc5b0480e7f7c782520ab43f8c7a2fc9461299737c0a1509ef9b26fea", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4293, - "estimated_input_tokens": 1074, - "estimated_eval_tokens": 2404 + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4293, - "estimated_input_tokens": 1074, - "estimated_eval_tokens": 2404 + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are unrelated trainer.py changes: one fixes gradient-accumulation loss scaling in training_step, the other extracts signature-column filtering for custom/PEFT trainers. Shared file location is incidental, not a shared underlying bug or change.", + "summary": "These are two unrelated Trainer changes in the same file: one adjusts gradient-accumulation loss scaling, the other refactors signature-column filtering for custom trainers/PEFT. They are not the same underlying bug or change.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No canonical PR: the PRs target different code paths and different behaviors in `trainer.py`, so they should not be deduplicated.", "best_issue_reason": null, - "best_pr_reason": "PR 43651 is the more concrete standalone fix if a single representative must be chosen; it changes runtime training behavior directly, whereas PR 43663 is an unrelated helper/refactor for column filtering.", + "best_pr_reason": "Neither PR is a duplicate representative; if a single PR must be picked for triage, 43651 is the more concrete fix, but it is not the same change as 43663.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Different code paths and goals: 43651 changes loss scaling in training_step, while 43663 adds _get_signature_columns for input-column filtering. They do not look like the same bug or a mergeable single PR." + "reason": "Different features in the same module: 43651 changes gradient accumulation loss normalization, while 43663 adds an overridable signature-column helper for input filtering. They do not fix the same bug and would not be merged as one PR." } ] }, @@ -24312,7 +23968,7 @@ { "label": "pull_request", "component_index": 11, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -24328,12 +23984,12 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "d3ed9902bb97d8072a8659026e386db8f53ed2e5c267aa1a3ab26e75f8184d71", + "prepared_review_unit_hash": "d6749780c17d4e9f43264adb638c2428b80258736dd04c4afc13bebc1fde2d2a", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3664, + "serialized_chars": 3663, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -24341,7 +23997,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3664, + "serialized_chars": 3663, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -24350,32 +24006,35 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs touch the same CLI file and both mention TypeAdapter NameError, but they appear to fix different concrete causes rather than the same bug.", - "confidence": 0.86, + "summary": "Both PRs address the same TypeAdapter-related runtime NameError in `src/transformers/cli/serve.py`, but `43842` is the narrower and more direct fix; `43656` includes broader typing/annotation cleanup around the same file.", + "confidence": 0.81, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43842 is the cleaner canonical PR: it directly addresses the TypeAdapter import/annotation problem with a minimal, targeted change under TYPE_CHECKING.", + "canonical_pr_reason": "PR 43842 is the best canonical PR: it directly fixes the `TypeAdapter` annotation/runtime import issue with a minimal change and a precise title/body tied to the pydantic-not-installed failure mode.", "best_issue_reason": null, - "best_pr_reason": "PR 43842 is the best standalone candidate because its diff is tightly scoped to the runtime/typing NameError and is easier to reason about as a single fix.", + "best_pr_reason": "PR 43842 is the best overall representative because it targets the exact crash condition with the smallest, most focused patch.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", - "accept": false, - "reason": "Reject: although both PRs involve TypeAdapter in src/transformers/cli/serve.py, 43656 appears to fix broader annotation/runtime typing issues, while 43842 specifically handles the pydantic-not-installed case by importing TypeAdapter only for type checking. They do not look like the same underlying bug or something that should be merged into one PR." + "accept": true, + "reason": "Same underlying bug in the same code path: a `TypeAdapter` NameError in the CLI serve module. The implementations differ, but both are plausible fixes for the same runtime annotation/import failure." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The comparison is grounded: both PR titles point to a TypeAdapter NameError in `src/transformers/cli/serve.py`, and the narrower `43842` looks like a plausible canonical representative. The soft-edge acceptance is conservative enough given the shared file and closely related fix patterns." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, { "label": "pull_request", "component_index": 12, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -24441,7 +24100,7 @@ { "label": "pull_request", "component_index": 13, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 2, "status": "reviewed", @@ -24489,8 +24148,8 @@ "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:44018|pull_request:44068", "pull_request:44066|pull_request:44068", "pull_request:44018|pull_request:44066", "pull_request:44019|pull_request:44722", @@ -24504,8 +24163,8 @@ "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:44068|pull_request:44086", "pull_request:44085|pull_request:44086", "pull_request:44098|pull_request:44101", "pull_request:44059|pull_request:44086", @@ -24554,112 +24213,112 @@ "pull_request:44024|pull_request:44076", "pull_request:44073|pull_request:44074" ], - "prepared_review_unit_hash": "d164b8c51b62d7480951e316492c0460234418d761a7dcf62e9730e5781970dd", + "prepared_review_unit_hash": "3905ff5780a46127b5d318441f6a9c6cdcd33bfc1e28ae673f4a0e5f2aa23c50", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 76, - "serialized_chars": 84115, - "estimated_input_tokens": 21029, - "estimated_eval_tokens": 42314 + "serialized_chars": 84109, + "estimated_input_tokens": 21028, + "estimated_eval_tokens": 42312 }, "budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 72, - "serialized_chars": 76937, - "estimated_input_tokens": 19235, - "estimated_eval_tokens": 38726 + "serialized_chars": 76930, + "estimated_input_tokens": 19233, + "estimated_eval_tokens": 38722 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a broad set of per-model output-tracing refactors. The packet directly supports duplicate/mergeable overlap in the GPT-J/CodeGen slice and the GPT-Neo slice, but not a Flaubert duplicate pair; the rest are only theme-level similarities across different model families.", - "confidence": 0.91, + "summary": "This cluster is mostly a broad output-tracing refactor theme spanning many different model implementations. I only accept soft edges when they are the same concrete model/code path (not just the same standardized-decorator pattern or the same tracking issue). The true duplicates here are the GPT-Neo slice, the GPT-J slices, and the Flaubert slice; most other pairs are model-specific refactors and should stay separate.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44722 is the best representative of the clearest supported duplicate subcluster because it captures the shared GPT-J/CodeGen output-tracing change with the strongest same-file overlap.", + "canonical_pr_reason": "PR 44722 is the clearest representative of the concrete GPT-J/CodeGen output-tracing change: it edits the same GPT-J/CodeGen code paths, explicitly targets the tracking issue, and is broader/more complete than the smaller follow-up slices.", "best_issue_reason": null, - "best_pr_reason": "PR 44722 is the strongest representative of the supported duplicate slice: it targets the GPT-J/CodeGen output-tracing refactor in the same concrete code path, while the remaining cluster items are mostly broader thematic matches across unrelated models.", + "best_pr_reason": "PR 44722 is the best single representative for the cluster because it directly captures the underlying GPT-J/CodeGen refactor rather than only the generic standardized-output theme.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different model families and code paths; same output-tracing theme is too broad to count as a duplicate." + "reason": "Same high-level output-tracing theme, but different model families and code paths (CVT/FNet vs GPT-J)." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Different model families and code paths; shared tracing refactor theme is not enough." + "reason": "Both are refactors around output tracing, but they target unrelated models and implementations." }, { "left": "pull_request:44066", "right": "pull_request:44085", - "accept": false, - "reason": "Same general GPT-J/CodeGen initiative, but not the same concrete PR-level change shown in the packet." + "accept": true, + "reason": "Both touch GPT-J output tracing in the same code path; this looks like the same underlying refactor and could plausibly be merged." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Different models and code paths; the shared standardized-output theme is only thematic." + "reason": "RegNet and EfficientNet are separate model implementations; same issue/theme is not enough." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "Different models and code paths; same initiative is not enough for duplication." + "reason": "EfficientNet and GPT-J are unrelated code paths despite both being standardized output-tracing work." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "Different models and code paths; only the standardized-output theme matches." + "reason": "Different model families and different forward paths; only the refactor style overlaps." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "Different model families; these are separate refactors, not one duplicate change." + "reason": "MobileNetV2 and DeBERTa-v2 are unrelated implementations." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "Different model families and code paths; not the same concrete fix." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44068", - "accept": true, - "reason": "Same GPT-Neo output-tracing refactor on the same model/file path." + "reason": "GPT-J/CodeGen and MGP-STR are unrelated; same decorator theme is too broad." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen versus MPT: different model implementations and code paths." + "reason": "GPT-J/CodeGen versus MPT is not the same concrete bug/change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both are GPT-Neo output-tracing refactors on the same model code path, so they look like the same underlying change." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen versus GPT-Neo: same initiative, but different concrete model paths." + "reason": "GPT-J/CodeGen and GPT-Neo are different model implementations." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are separate model-specific changes." + "reason": "Different models: GPT-Neo versus GPT-J/CodeGen." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet and GPT-J/CodeGen are unrelated concrete code paths." + "reason": "ResNet and GPT-J/CodeGen are unrelated code paths." }, { "left": "pull_request:44086", @@ -24671,322 +24330,328 @@ "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet and GPT-J are different concrete refactors." + "reason": "ResNet and GPT-J are unrelated; same issue label would not make them duplicates." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are unrelated model changes." - }, - { - "left": "pull_request:44019", - "right": "pull_request:44071", - "accept": false, - "reason": "ResNet and MPT are different concrete code paths." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44116", - "accept": false, - "reason": "GPT-Neo and Flaubert are separate model-specific refactors." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44722", - "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are related only by theme, not the same concrete change." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44116", - "accept": false, - "reason": "GPT-Neo and Flaubert are unrelated model migrations." - }, - { - "left": "pull_request:44071", - "right": "pull_request:44086", - "accept": false, - "reason": "MPT and MGP-STR are different models and code paths." + "reason": "GPT-Neo and ResNet are completely different models/code paths." }, { - "left": "pull_request:44018", - "right": "pull_request:44085", + "left": "pull_request:44019", + "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and GPT-J are separate concrete changes." + "reason": "ResNet versus MPT; only the refactor motif overlaps." }, { "left": "pull_request:44018", - "right": "pull_request:44086", + "right": "pull_request:44129", "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated model-specific changes." + "reason": "GPT-Neo and SpeechT5 are unrelated implementations." }, { "left": "pull_request:44068", - "right": "pull_request:44086", + "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo and MGP-STR are different model implementations." + "reason": "GPT-Neo and GPT-J/CodeGen are different model families." }, { "left": "pull_request:44071", - "right": "pull_request:44085", + "right": "pull_request:44086", "accept": false, - "reason": "MPT and GPT-J are unrelated concrete code paths." + "reason": "MPT and MGP-STR are unrelated model code paths." }, { "left": "pull_request:44085", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J and MGP-STR are unrelated model-specific changes." + "reason": "GPT-J and MGP-STR are different models; same theme is not enough." }, { "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "ViLT and Flaubert/XLM are unrelated changes." + "reason": "ViLT and Flaubert are unrelated model implementations." }, { "left": "pull_request:44059", "right": "pull_request:44086", "accept": false, - "reason": "GPT-2 and MGP-STR are separate model refactors." + "reason": "GPT-2 and MGP-STR are unrelated code paths." }, { "left": "pull_request:44059", "right": "pull_request:44722", "accept": false, - "reason": "GPT-2 output tracing is a different code path from GPT-J/CodeGen." + "reason": "GPT-2 and GPT-J/CodeGen are different model implementations." }, { "left": "pull_request:44018", "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and MPT are different model migrations." + "reason": "GPT-Neo and MPT are unrelated despite similar output-tracing cleanup." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "Segformer and LongT5 are different code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "Segformer and ViLT are unrelated model implementations." }, { "left": "pull_request:43997", "right": "pull_request:44017", "accept": false, - "reason": "RegNet and SegFormer are separate model implementations." + "reason": "RegNet and Segformer are separate models; same refactor theme is too broad." }, { "left": "pull_request:43997", "right": "pull_request:44056", "accept": false, - "reason": "RegNet and MPNet are different code paths." + "reason": "RegNet and MPNet are unrelated code paths." }, { "left": "pull_request:44071", "right": "pull_request:44076", "accept": false, - "reason": "MPT and ImageGPT are unrelated concrete changes." + "reason": "MPT and ImageGPT are unrelated implementations." }, { - "left": "pull_request:43997", - "right": "pull_request:44073", + "left": "pull_request:44074", + "right": "pull_request:44101", "accept": false, - "reason": "RegNet and VisualBERT are different model refactors." + "reason": "TextNet and Flaubert are different model code paths." }, { - "left": "pull_request:43997", + "left": "pull_request:44056", "right": "pull_request:44161", "accept": false, - "reason": "RegNet and LongT5 are unrelated model-specific changes." + "reason": "MPNet and LongT5 are unrelated model implementations." }, { - "left": "pull_request:44074", - "right": "pull_request:44101", + "left": "pull_request:43997", + "right": "pull_request:44018", "accept": false, - "reason": "TextNet and Flaubert/XLM are different models." + "reason": "RegNet and GPT-Neo are unrelated." }, { - "left": "pull_request:44056", + "left": "pull_request:43997", "right": "pull_request:44161", "accept": false, - "reason": "MPNet and LongT5 are unrelated." + "reason": "RegNet and LongT5 are unrelated code paths." }, { "left": "pull_request:44074", "right": "pull_request:44161", "accept": false, - "reason": "TextNet and LongT5 are different code paths." + "reason": "TextNet and LongT5 are unrelated models." }, { - "left": "pull_request:44010", - "right": "pull_request:44071", + "left": "pull_request:44071", + "right": "pull_request:44085", "accept": false, - "reason": "SqueezeBERT and MPT are unrelated model implementations." + "reason": "MPT and GPT-J are different model implementations." }, { - "left": "pull_request:44018", - "right": "pull_request:44129", + "left": "pull_request:43996", + "right": "pull_request:44161", + "accept": false, + "reason": "CVT/FNet and LongT5 are unrelated code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", "accept": false, - "reason": "GPT-Neo and SpeechT5 are separate concrete refactors." + "reason": "CVT/FNet and Segformer are different model implementations." }, { "left": "pull_request:44010", - "right": "pull_request:44073", + "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBERT and VisualBERT are different model-specific changes." + "reason": "SqueezeBERT and DeBERTa-v2 are unrelated models." }, { - "left": "pull_request:43997", + "left": "pull_request:44044", "right": "pull_request:44074", "accept": false, - "reason": "RegNet and TextNet are unrelated model migrations." + "reason": "DeBERTa-v2 and TextNet are unrelated code paths." }, { - "left": "pull_request:44010", + "left": "pull_request:44073", + "right": "pull_request:44161", + "accept": false, + "reason": "ViLT and LongT5 are unrelated model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are different models." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert are unrelated code paths." + }, + { + "left": "pull_request:44017", "right": "pull_request:44018", "accept": false, - "reason": "SqueezeBERT and GPT-Neo are different code paths." + "reason": "Segformer and GPT-Neo are unrelated models." }, { "left": "pull_request:44071", "right": "pull_request:44129", "accept": false, - "reason": "MPT and SpeechT5 are unrelated concrete changes." + "reason": "MPT and SpeechT5 are unrelated implementations." }, { - "left": "pull_request:43996", + "left": "pull_request:44018", "right": "pull_request:44129", "accept": false, - "reason": "CVT/FNet and SpeechT5 are unrelated model migrations." + "reason": "GPT-Neo and SpeechT5 are unrelated." }, { - "left": "pull_request:43996", - "right": "pull_request:44017", + "left": "pull_request:43997", + "right": "pull_request:44074", "accept": false, - "reason": "CVT/FNet and SegFormer are different model-specific changes." + "reason": "RegNet and TextNet are different model code paths." }, { - "left": "pull_request:44010", - "right": "pull_request:44044", + "left": "pull_request:43996", + "right": "pull_request:44073", "accept": false, - "reason": "SqueezeBERT and DeBERTa-v2 are unrelated implementations." + "reason": "CVT/FNet and ViLT are unrelated implementations." }, { - "left": "pull_request:44044", - "right": "pull_request:44074", + "left": "pull_request:44010", + "right": "pull_request:44018", "accept": false, - "reason": "DeBERTa-v2 and TextNet are different concrete refactors." + "reason": "SqueezeBERT and GPT-Neo are unrelated models." }, { - "left": "pull_request:44066", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J/CodeGen output-tracing refactor with direct file overlap." + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBERT and ViLT are different code paths." }, { - "left": "pull_request:44085", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J output-tracing refactor in the same file." + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model families." }, { - "left": "pull_request:44085", - "right": "pull_request:44129", + "left": "pull_request:44072", + "right": "pull_request:44154", "accept": false, - "reason": "GPT-J and SpeechT5 are unrelated model changes." + "reason": "EfficientNet and VITS are unrelated models." }, { - "left": "pull_request:43996", - "right": "pull_request:44007", + "left": "pull_request:44010", + "right": "pull_request:44101", "accept": false, - "reason": "CVT/FNet and ResNet are separate model refactors." + "reason": "SqueezeBERT and Flaubert are unrelated code paths." }, { - "left": "pull_request:43996", - "right": "pull_request:44129", + "left": "pull_request:44017", + "right": "pull_request:44073", "accept": false, - "reason": "CVT/FNet and SpeechT5 are unrelated." + "reason": "Segformer and ViLT are unrelated model implementations." }, { - "left": "pull_request:44044", - "right": "pull_request:44129", + "left": "pull_request:44017", + "right": "pull_request:44161", "accept": false, - "reason": "DeBERTa-v2 and SpeechT5 are different concrete code paths." + "reason": "Segformer and LongT5 are different code paths." }, { - "left": "pull_request:44044", - "right": "pull_request:44154", + "left": "pull_request:44018", + "right": "pull_request:44086", "accept": false, - "reason": "DeBERTa-v2 and ViLT are unrelated model refactors." + "reason": "GPT-Neo and MGP-STR are unrelated models." }, { - "left": "pull_request:44024", - "right": "pull_request:44076", + "left": "pull_request:44018", + "right": "pull_request:44059", "accept": false, - "reason": "FocalNet and ImageGPT are different model implementations." + "reason": "GPT-Neo and GPT-2 are different model implementations." }, { - "left": "pull_request:44073", - "right": "pull_request:44074", + "left": "pull_request:44018", + "right": "pull_request:44071", "accept": false, - "reason": "VisualBERT and TextNet are unrelated models." + "reason": "GPT-Neo and MPT are unrelated despite similar refactor style." }, { - "left": "pull_request:44066", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J/CodeGen output-tracing refactor with direct file overlap." + "left": "pull_request:44056", + "right": "pull_request:44071", + "accept": false, + "reason": "MPNet and MPT are unrelated model code paths." }, { - "left": "pull_request:44085", + "left": "pull_request:44068", "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J output-tracing refactor in the same file." + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model families." }, { "left": "pull_request:44085", - "right": "pull_request:44129", + "right": "pull_request:44086", "accept": false, - "reason": "GPT-J and SpeechT5 are unrelated model migrations." + "reason": "GPT-J and MGP-STR are unrelated models." }, { - "left": "pull_request:43996", - "right": "pull_request:44007", + "left": "pull_request:44066", + "right": "pull_request:44068", "accept": false, - "reason": "CVT/FNet and ResNet are separate model refactors." + "reason": "GPT-J/CodeGen and GPT-Neo are different model implementations." }, { - "left": "pull_request:43996", - "right": "pull_request:44129", - "accept": false, - "reason": "CVT/FNet and SpeechT5 are unrelated." + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor on the same concrete code paths; this is a plausible duplicate/follow-up." }, { - "left": "pull_request:44044", - "right": "pull_request:44129", - "accept": false, - "reason": "DeBERTa-v2 and SpeechT5 are different concrete changes." + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both are GPT-J output-tracing refactors on the same code path and look like the same underlying change." }, { - "left": "pull_request:44044", - "right": "pull_request:44154", - "accept": false, - "reason": "DeBERTa-v2 and ViLT are unrelated model-specific refactors." + "left": "pull_request:44101", + "right": "pull_request:44116", + "accept": true, + "reason": "Both modify the same Flaubert attention/output-tracing code path and appear to be the same refactor in different slices." }, { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "FocalNet and ImageGPT are different code paths." + "reason": "FocalNet and ImageGPT are unrelated model implementations." }, { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "VisualBERT and TextNet are unrelated models." + "reason": "ViLT and TextNet are separate models; same standardized-output pattern is not enough." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The per-pair verdicts are mostly conservative, but the summary overstates the evidence by implying true duplicate clusters \u201cespecially within GPT-J/CodeGen, GPT-Neo, and Flaubert.\u201d The packet only clearly supports same-file/mergeable duplication for GPT-J/CodeGen and GPT-Neo (and a few unrelated model-specific false pairs); it does not show a supported Flaubert duplicate pair. Please narrow the summary to what the packet directly demonstrates." + "accept": true, + "feedback": "The summary is grounded in the packet and stays mostly conservative. The accepted soft edges are all supported by shared model/code-path overlap, and the representative PR choice is reasonable. I don\u2019t see an overstatement beyond the evidence provided." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -24994,7 +24659,7 @@ { "label": "pull_request", "component_index": 13, - "component_count": 34, + "component_count": 33, "review_unit_index": 2, "review_unit_count": 2, "status": "reviewed", @@ -25017,58 +24682,58 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "d5d205a24138b7a1a61ad455373afb48b8d7d5656e0579bfa17e7ba2552fcc96", + "prepared_review_unit_hash": "3c914234c2212f60cbcc35f5023cfc1d5aa6dcc0ecd5b172e2349ff34bcf1e8c", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 76, - "serialized_chars": 84115, - "estimated_input_tokens": 21029, - "estimated_eval_tokens": 42314 + "serialized_chars": 84109, + "estimated_input_tokens": 21028, + "estimated_eval_tokens": 42312 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 4, - "serialized_chars": 12756, - "estimated_input_tokens": 3189, - "estimated_eval_tokens": 6634 + "serialized_chars": 12757, + "estimated_input_tokens": 3190, + "estimated_eval_tokens": 6636 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These PRs are related thematically as model-specific migrations to standardized output tracing/capture, but they are not duplicates of one another because each targets a different model and code path.", + "summary": "All items are model-specific refactors to the new standardized output-tracing utilities, but they target different architectures and code paths, so they are not duplicates of the same underlying change.", "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the best representative of the cluster because it has an explicit issue target and a concrete, model-specific refactor of the standardized output-capture flow.", + "canonical_pr_reason": "No issue artifact is present; among the PRs, pull_request:44056 is a strong representative because it cleanly shows the output-tracing refactor pattern with an explicit output-recording map in a single model file.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the strongest standalone PR in this set: it is well-scoped, explicitly linked to an issue, and makes a concrete output-tracing change rather than just matching the refactor pattern.", + "best_pr_reason": "pull_request:44056 is a good cluster representative because it captures the core migration pattern (adding capture_outputs/can_return_tuple and wiring _can_record_outputs) in a focused, model-local change.", "soft_edge_verdicts": [ { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "Both are output-tracing refactors, but they touch different models (MPNet vs VisualBert) and different concrete implementations; they are not the same bug/change and would not plausibly merge into one PR." + "reason": "Same refactor theme, but different models (MPNet vs VisualBert) and different forward paths/classes; too broad to treat as one concrete PR." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "Same refactor theme, but MPNet and TextNet are separate code paths with separate model files and behavior; this is not a duplicate change." + "reason": "Both migrate output tracing, but they apply to different architectures (MPNet vs TextNet) with distinct implementation details; not the same underlying change." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model implementations. The similar decorator migration is broad refactoring, not a shared underlying defect or single mergeable fix." + "reason": "RegNet and SqueezeBERT are unrelated model code paths; the similarity is only the shared utility migration, not a single fix/change." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 have different layer/output logic and only share the standardized tracing motif. The changes are model-specific and not one concrete fix." + "reason": "SqueezeBERT and SpeechT5 touch different model internals and output contracts; they are separate refactors, not one mergeable PR." } ] }, @@ -25082,7 +24747,7 @@ { "label": "pull_request", "component_index": 14, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25148,274 +24813,268 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "1807f9a411bd87d9f551fbd7f1f71588ae03bd220d3dd6e5418adce9845adf94", + "prepared_review_unit_hash": "675ccaa035afdd47710e3af7b9c3717d7fa937325243b86e24ae74743aea89c9", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40157, - "estimated_input_tokens": 10040, - "estimated_eval_tokens": 20336 + "serialized_chars": 40165, + "estimated_input_tokens": 10042, + "estimated_eval_tokens": 20340 }, "budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40157, - "estimated_input_tokens": 10040, - "estimated_eval_tokens": 20336 + "serialized_chars": 40165, + "estimated_input_tokens": 10042, + "estimated_eval_tokens": 20340 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs share a common output-tracing/refactor theme, but they target different model implementations and code paths. I do not see any true duplicate PRs here; they are better treated as separate model-specific refactors.", - "confidence": 0.93, + "summary": "All artifacts are separate, model-specific output-tracing refactors; they share a pattern, but not the same underlying change.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44003 is the most central and broadest of the set: it spans two related Mamba implementations and introduces shared output-capturing plumbing, making it the best representative of this cluster\u2019s refactor theme.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 44003 is the strongest single representative because it has the broadest scope and most infrastructure-level change among the candidates.", + "best_pr_reason": "pull_request:44025 is the broadest representative: it refactors both depth_anything and prompt_depth_anything with the new tracing decorator.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different model families and code paths; same refactor pattern, but not the same concrete change." + "reason": "Different model implementations and forward paths; same refactor theme only." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Backbone model vs vision-text dual encoder; unrelated implementations despite similar tracing refactor wording." + "reason": "Different architectures; not the same concrete change." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Different architectures and files; not a single mergeable bug fix." + "reason": "Different model code paths; only the tracing refactor pattern matches." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Both are output-tracing refactors, but they target different model stacks and code paths." + "reason": "Unrelated model families; not mergeable as one fix." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Unrelated model implementations; the similarity is only at the refactor-theme level." + "reason": "Different backbones and output paths; too broad to be duplicate." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNet V1 vs UnivNet are separate model paths; not the same underlying bug/change." + "reason": "Different models and output signatures; not the same bug/change." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "Different model families and behavior; no shared concrete fix." + "reason": "Different model targets; shared refactor style is insufficient." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "Different models and different output wrappers; too broad to treat as duplicates." + "reason": "MobileNetV1 vs Depth Anything are separate implementations." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "MobileNet V1 and VisionEncoderDecoder are unrelated code paths." + "reason": "Different wrapper/model code paths; not a duplicate change." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "Separate encoder-decoder wrapper and backbone model; same theme only." + "reason": "Separate encoder-decoder family and model-specific refactor." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder vs UnivNet; different concrete code paths." + "reason": "Vision-text dual encoder and UnivNet are unrelated code paths." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different model subsystems; not one underlying fix." + "reason": "Different model classes; same output-tracing theme only." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "Distinct model families; too generic to merge as one PR." + "reason": "Different model families and forward implementations." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different architectures and output types; not the same change." + "reason": "Distinct models; no shared underlying fix beyond refactor style." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are unrelated implementations." + "reason": "UnivNet and UperNet are different model paths." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Different model families and code paths; only the refactor pattern matches." + "reason": "Different architectures; not the same code-path problem." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "Different models; no evidence of a shared concrete bug." + "reason": "Different tasks and modules; not mergeable as one PR." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "UperNet and RWKV are unrelated code paths." + "reason": "UperNet and RWKV are unrelated implementations." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "Separate model stacks; not plausibly one combined fix." + "reason": "Different model families and output contracts." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Mamba/Falcon-Mamba output tracing vs SuperPoint output tracing; same theme, different code paths." + "reason": "Mamba/Falcon-Mamba and SuperPoint are unrelated code paths." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Different architectures and fixes; not the same concrete bug or change." + "reason": "Different models and concrete changes; only generic tracing refactor overlaps." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Timm backbone and CodeGen are unrelated model paths." + "reason": "Timm backbone and CodeGen fix different model code paths." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet audio model vs CodeGen decoder; not the same issue." + "reason": "UnivNet and CodeGen are unrelated changes." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet segmentation vs CodeGen attention/cache refactor are unrelated." - }, - { - "left": "pull_request:44000", - "right": "pull_request:44004", - "accept": false, - "reason": "Vision-text encoder-decoder vs CodeGen are different code paths." + "reason": "UperNet and CodeGen do not share the same underlying bug." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "Depth Anything and VisionEncoderDecoder are separate model families; same refactor theme only." + "reason": "Depth Anything and Vision Encoder-Decoder are separate model wrappers." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "Depth Anything vs SpeechEncoderDecoder are unrelated implementations." + "reason": "Different encoder-decoder implementations; not the same fix." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "Different models and output semantics; not one mergeable fix." + "reason": "Depth Anything and SuperPoint are unrelated model paths." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Depth Anything and RWKV do not share a concrete bug/change." + "reason": "Different models and forward logic; too broad to merge together." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "Depth Anything vs DPR are separate model code paths." + "reason": "Depth Anything and DPR are separate implementations." }, { "left": "pull_request:44026", "right": "pull_request:44029", "accept": false, - "reason": "VisionEncoderDecoder and RWKV are unrelated; similarity is only topical." + "reason": "Vision Encoder-Decoder and RWKV are unrelated code paths." }, { "left": "pull_request:44026", "right": "pull_request:44027", "accept": false, - "reason": "Different encoder-decoder wrappers; not the same concrete change." + "reason": "Vision and speech encoder-decoder wrappers are distinct changes." }, { "left": "pull_request:44026", "right": "pull_request:44028", "accept": false, - "reason": "VisionEncoderDecoder and SuperPoint are unrelated implementations." + "reason": "Different models; same tracing refactor theme only." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Different model families and files; no shared code-path bug." + "reason": "Vision Encoder-Decoder and DPR do not fix the same problem." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "SpeechEncoderDecoder and SuperPoint are unrelated model paths." + "reason": "Speech encoder-decoder and SuperPoint are unrelated." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Different architectures; same theme does not imply duplicate PRs." + "reason": "Different model families and code paths." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "SpeechEncoderDecoder and DPR are separate code paths." + "reason": "Speech Encoder-Decoder and DPR are separate model-specific refactors." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are unrelated; no common concrete fix." + "reason": "SuperPoint and RWKV have different forward implementations." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "SuperPoint and DPR target different model stacks." + "reason": "Different models and output shapes; not duplicate changes." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are unrelated implementations despite similar refactor language." + "reason": "RWKV and DPR are unrelated model paths." } ] }, @@ -25429,7 +25088,7 @@ { "label": "pull_request", "component_index": 15, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25445,7 +25104,7 @@ "soft_pairs": [ "pull_request:44662|pull_request:45401" ], - "prepared_review_unit_hash": "ce562142be6ea3e65d225d88975a86f10563150e37c23df3949c68a0f32353ce", + "prepared_review_unit_hash": "aed7a9dda648e230859867ef620c8e6b8a67853c8d2f37ce455cedbe54d14ff9", "component_budget": { "node_count": 2, "item_count": 2, @@ -25467,18 +25126,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both artifacts are separate model-addition PRs: PenguinVL and Voxtral TTS. They share boilerplate integration files, but they target different models, code paths, and release goals, so they are not duplicates.", - "confidence": 0.96, + "summary": "These are two separate model-addition PRs: PenguinVL implementation vs Voxtral TTS support. They share some boilerplate auto-mapping/docs paths, but they target different models and codepaths, so they should not be deduplicated.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR #45401 is the clearer cluster representative because it has an explicit issue target and a focused implementation for Voxtral TTS, but it is still not a duplicate of #44662.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "PR #44662 is the more substantial standalone PR in this cluster because it is a non-draft, broader implementation with more review activity. It best represents the PenguinVL side, though the two PRs are unrelated and should remain separate.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different underlying changes: PR 44662 adds PenguinVL, while PR 45401 adds Voxtral TTS. The shared files are generic Transformers registration/docs updates, not evidence of the same bug or implementation." + "reason": "Different models and changes: PenguinVL implementation versus Voxtral TTS support. Shared docs/auto-registry plumbing is generic and not enough to indicate the same underlying bug or mergeable change." } ] }, @@ -25492,7 +25151,7 @@ { "label": "pull_request", "component_index": 16, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25508,12 +25167,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "c2a8b14e51d41967e77174694fd8cddb2c7abc4f31c8f858bb81c8dad7ff0b48", + "prepared_review_unit_hash": "2adda2d8ad55b04474d3128861f0042600e18a90266982d6cc93f64f6fb583f8", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, + "serialized_chars": 4407, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -25521,7 +25180,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4408, + "serialized_chars": 4407, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -25530,18 +25189,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs are related in theme (module/kernel fusion) but are not the same change: #44979 adds a generic Module Fusion API, while #45363 implements n-to-1 kernel fusion via KernelConfig and integration hooks. They touch different code paths and represent different layers of functionality, so they should not be deduplicated.", + "summary": "The two PRs are related to fusion infrastructure, but they target different layers and goals: one introduces a generic Module Fusion API, while the other adds KernelConfig-based n-to-1 kernel fusion integration and an example. They are not the same underlying change.", "confidence": 0.9, "canonical_issue_reason": null, - "canonical_pr_reason": "#44979 is the more foundational, general-purpose fusion API change; #45363 is a separate feature built around KernelConfig and integration plumbing.", + "canonical_pr_reason": "PR #44979 is the more foundational change: it introduces the core Module Fusion API and tests, which the other PR appears to build on rather than duplicate.", "best_issue_reason": null, - "best_pr_reason": "#44979 is the better representative of the cluster because it defines the core fusion abstraction rather than one specific kernel-fusion use case.", + "best_pr_reason": "PR #44979 is the best representative of this cluster because it defines the underlying fusion abstraction; PR #45363 is a more specific application/integration of that idea.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Do not merge: #44979 introduces a generic module-fusion API, whereas #45363 adds a specific n-to-1 kernel-fusion path via KernelConfig and related integrations. Different concrete code paths and not plausibly one PR." + "reason": "They are related but not duplicates: #44979 adds a generic module fusion API, while #45363 wires KernelConfig-driven n-to-1 kernel fusion into model loading and includes an example. Different concrete code paths and purposes, so they would not plausibly be merged into one PR." } ] }, @@ -25555,7 +25214,7 @@ { "label": "pull_request", "component_index": 17, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25571,7 +25230,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "f668db5fdcc1e35aee4c5a89614ef2898ffc20180fed0930b66a85233bd18aa5", + "prepared_review_unit_hash": "5d5cec7046e6de36683a75dd189047000df17abf0e8622e8ebff5b1c9f82e38e", "component_budget": { "node_count": 2, "item_count": 2, @@ -25593,18 +25252,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests address the same Mistral regex patch bug: fixing an AttributeError by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer` in `_patch_mistral_regex`. The second PR is the more complete version because it includes tests and the same code fix.", + "summary": "Both pull requests patch the same AttributeError in `_patch_mistral_regex` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; the second PR adds test coverage but targets the same concrete fix.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45317 is the better canonical PR: it fixes the same concrete AttributeError in `_patch_mistral_regex` and also adds a test, making it the more complete change set.", + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same code fix as 45086 plus a test file, making it the more complete change.", "best_issue_reason": null, - "best_pr_reason": "PR 45317 is the strongest representative of the cluster because it covers the bug fix plus validation, and its scope is still the same single code-path change as PR 45086.", + "best_pr_reason": "PR 45317 best represents the cluster since it fixes the same bug and includes validation in tests, while 45086 is the narrower variant.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Accept: both PRs implement the same fix for the same AttributeError in `_patch_mistral_regex` on the same file, and the later PR appears to be the more complete version with tests." + "reason": "Same underlying bug and same code-path fix in the same file; the extra test in 45317 does not change the core change, so they are duplicates." } ] }, @@ -25618,7 +25277,7 @@ { "label": "pull_request", "component_index": 18, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25634,7 +25293,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "f5558433f0d29ef8658b63172c513e5830e47878de927a7a4f0b1cee496c7351", + "prepared_review_unit_hash": "3dda778432d4e2906eb503ea5233e2a3d4d3224f0299669e176d224d3395908e", "component_budget": { "node_count": 2, "item_count": 2, @@ -25656,18 +25315,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs touch NemotronH internals, but they target different changes: #45115 is a conversion/mapping and inheritance refactor toward GraniteMoeHybrid compatibility, while #45569 is a broader model split into NemotronH dense/sparse variants with docs and auto-mapping updates. They are related, but not the same underlying fix/change.", - "confidence": 0.88, + "summary": "Two related NemotronH PRs, but they address different layers of the work: one is a refactor/conversion-mapping inheritance change, the other splits NemotronH into dense/sparse models with docs/tests. They are not duplicates.", + "confidence": 0.86, "canonical_issue_reason": null, - "canonical_pr_reason": "#45569 is the better canonical representative because it is the broader, higher-level NemotronH architecture change and includes the public model/docs surface, whereas #45115 is a narrower conversion-mapping refactor.", + "canonical_pr_reason": "PR #45569 is the better canonical representative because it is the broader, more complete product change: it adds the new dense/sparse model split, updates auto-mapping/docs, and includes tests. PR #45115 is a narrower refactor focused on inheritance and conversion mapping.", "best_issue_reason": null, - "best_pr_reason": "#45569 is the strongest single PR in this cluster as it covers the main model restructuring end-to-end; #45115 looks more like a supporting compatibility refactor than the primary change.", + "best_pr_reason": "PR #45569 is the strongest standalone PR in the cluster because it covers the user-visible architecture/API change end to end and is backed by docs and tests. PR #45115 is more of a supporting refactor than the primary change.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "Related NemotronH work, but not the same concrete bug/change. #45115 is about checkpoint conversion and GraniteMoeHybrid-compatible renaming; #45569 is about splitting NemotronH into dense/sparse model variants and updating docs/auto-mapping. They would not be merged as one PR." + "reason": "Not the same concrete change. #45115 focuses on NemotronH refactoring and weight-conversion renames toward GraniteMoeHybrid compatibility, while #45569 introduces separate NemotronH dense/sparse model classes, docs, auto-mapping, and tests. They overlap in files but are not mergeable as one duplicate PR." } ] }, @@ -25681,7 +25340,7 @@ { "label": "pull_request", "component_index": 19, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25697,7 +25356,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "875fd6e7328e859973b99a3bf2630311de43ccba54225411bf24654089bc875d", + "prepared_review_unit_hash": "e1f538e3a8f67a93180df1f5d48a51400acb9a53ca7cd468ea8379c5e78ffdeb", "component_budget": { "node_count": 2, "item_count": 2, @@ -25719,18 +25378,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related but not duplicates: #45333 is a narrower per-layer configuration/doc update, while #45332 bundles that config work with broader heterogeneous modeling, masking, cache, and integration changes.", + "summary": "The two pull requests are closely related and share config/docs/test files, but they are not duplicates: one adds broader heterogeneous model support and the other focuses on per-layer configuration utilities. They look like complementary or layered work, not the same mergeable change.", "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "#45332 is the broader umbrella change and appears to subsume the configuration work while also adding the modeling/runtime pieces, so it is the better canonical PR in this cluster.", + "canonical_pr_reason": "PR #45332 is the broader, more complete change set: it includes heterogeneous modeling plus the config work, while #45333 is scoped to configuration utilities and docs/tests.", "best_issue_reason": null, - "best_pr_reason": "#45332 is the best representative because it covers the full heterogeneous-model feature, not just the config subset. #45333 is narrower and looks like a slice/follow-up rather than a separate duplicate.", + "best_pr_reason": "PR #45332 is the best representative of the cluster because it covers the main code-path change and subsumes the narrower configuration-focused PR.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "Related feature work, but not the same concrete change: #45332 adds heterogeneous modeling/runtime support plus config, whereas #45333 is config-only. They could not be cleanly merged as one duplicate PR without losing the extra functionality." + "reason": "Shared files indicate overlap in the heterogeneous config subsystem, but the PRs are not the same concrete fix/change: #45332 adds full heterogeneous model support and touches modeling/cache/integration code, while #45333 is limited to per-layer configuration utilities and docs/tests. They would not cleanly collapse into one PR without mixing distinct scopes." } ] }, @@ -25744,7 +25403,7 @@ { "label": "pull_request", "component_index": 20, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25760,12 +25419,12 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "8cfbb15d322abc08c3fa8f853bec9b208ecdea7272d778a254fdfe33fda823e8", + "prepared_review_unit_hash": "15c15eb216bdab8b16a5b8c993f7404196922ea36a85b778f8327309da9b02ea", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5513, + "serialized_chars": 5514, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -25773,7 +25432,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5513, + "serialized_chars": 5514, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -25782,18 +25441,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are both typing-related PRs, but they are not the same change: #45415 is a broad repository-wide type-checking sweep, while #45425 adds specific typing aliases/protocols for modeling_utils and related integrations. They overlap in a few files because both touch typing infrastructure, but they do not fix the same concrete code-path problem and would not naturally merge into one PR.", - "confidence": 0.94, + "summary": "Both PRs are typing-related, but they are not the same change: #45415 is a broad type-checking sweep across many Transformers modules plus pyproject type-ignore config, while #45425 is a targeted typing refactor adding shared aliases/protocols for specific modeling/PEFT paths. Their overlap is incidental rather than a shared concrete bug/fix.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "#45425 is the more focused, concrete typing change; #45415 is a much broader type-checking pass. They are related thematically but not duplicates.", + "canonical_pr_reason": "#45415 is the broader umbrella typing PR and best represents the cluster\u2019s general theme, but it is still distinct from the targeted typing refactor in #45425.", "best_issue_reason": null, - "best_pr_reason": "#45425 is the best representative PR here because it introduces a specific typing abstraction set tied to modeling_utils and adjacent modules, whereas #45415 is a large mechanical typing sweep.", + "best_pr_reason": "#45415 is the best representative PR here because it spans the largest set of affected files and includes the type-checking configuration change, making it the most canonical of the two typing cleanups.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Shared typing/filesystem infrastructure and a few overlapping filenames are not enough: #45415 is a broad type-checking cleanup, while #45425 adds targeted typing protocols/aliases for specific runtime interfaces. Different underlying changes." + "reason": "They share a typing/annotation theme and a few filenames, but #45415 is a broad type-checking sweep while #45425 adds specific protocols/aliases for modeling_utils/PEFT. They do not appear to fix the same concrete code-path problem and would not naturally merge as one PR." } ] }, @@ -25807,7 +25466,7 @@ { "label": "pull_request", "component_index": 21, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25823,7 +25482,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "afd9198b8b836c38a49187413aa00ca41db347eda09980b4f4843f713d654139", + "prepared_review_unit_hash": "5918793089f2df4fbb63e7ff2bf298f1862aaf6efaf3675f9b19321bb4bd5179", "component_budget": { "node_count": 2, "item_count": 2, @@ -25845,18 +25504,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two related but distinct EP-loading PRs: one fixes config/serialization for EP with DeepSpeed ZeRO-3 under accelerate launch, while the other fixes expert parameters being overwritten during FSDP2/rank-0 broadcast. They overlap in `modeling_utils.py` but address different failure modes and code paths.", + "summary": "These PRs are related at a high level (expert parallelism loading), but they fix different bugs on different code paths: one changes config serialization / EP detection for accelerate+ZeRO-3, the other prevents parameter overwrites during FSDP2 loading. They should not be treated as duplicates.", "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45662 is the broader and more concrete fix: it targets the parameter-loading path where expert tensors are incorrectly overwritten, with changes across loading and MoE/DTensor integration. PR #45548 is narrower and mainly preserves `distributed_config` plus an `has_ep` helper for launch/load compatibility.", + "canonical_pr_reason": "No single canonical PR: 45548 is about serializing distributed_config and exposing has_ep, while 45662 fixes shard-aware parameter assignment during loading; they do not resolve the same bug.", "best_issue_reason": null, - "best_pr_reason": "PR #45662 is the best representative PR for this cluster because it fixes the core EP parameter corruption bug in the loader, rather than only addressing configuration persistence and detection.", + "best_pr_reason": "PR 45662 is the stronger representative of a concrete runtime loading corruption fix, but it still does not subsume 45548's config/EP-detection changes.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "Both are EP-related, but they fix different bugs: #45548 is about keeping `distributed_config` during serialization and detecting EP for DeepSpeed ZeRO-3 loading, while #45662 prevents expert weights from being overwritten during FSDP2/rank-0 broadcast. Shared files alone are not enough to treat them as the same change." + "reason": "Different underlying failures: 45548 handles distributed_config serialization and EP detection for accelerate/ZeRO-3, while 45662 adds post-shard wrapping to stop experts being overwritten during FSDP2 loading." } ] }, @@ -25870,7 +25529,7 @@ { "label": "pull_request", "component_index": 22, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25886,40 +25545,40 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "e4c0d5568e5c8be6f8932f0c39c049fa9e0c02c3f5d5581f4019e9eb583bbe1d", + "prepared_review_unit_hash": "c4e5eaf3d6ca960dc1b218ca43a2f1788ab766531dfbf4ce95252b1d4cafd778", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4672, - "estimated_input_tokens": 1168, - "estimated_eval_tokens": 2592 + "serialized_chars": 4673, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4672, - "estimated_input_tokens": 1168, - "estimated_eval_tokens": 2592 + "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 address the same xdist-related captured_info artifact problem in transformers testing utilities; PR 45645 is the broader superset that also updates CI log collection, while 45639 is the core writer-side fix. They are close enough to treat as the same change cluster.", - "confidence": 0.86, + "summary": "Both PRs address the same xdist-related captured_info collision problem under issue 45561; 45645 is the more complete implementation because it covers both the writer and CI log collection paths.", + "confidence": 0.87, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the better canonical PR because it contains the core xdist-safe logging fix plus the CI workflow adjustment needed to read multiple captured_info files, making it the more complete implementation.", + "canonical_pr_reason": "PR 45645 is the broader, more complete fix: it makes the test debug output xdist-safe and also updates the workflow to collect multiple captured_info files without losing CI logs.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the best representative of the underlying change: it fixes the collision in testing_utils and also preserves/debug-prints the resulting logs in CI. PR 45639 is a narrower subset of that same fix.", + "best_pr_reason": "PR 45645 is the best representative because it resolves the artifact-collision problem end-to-end, including the CI-side consumer of those logs.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying xdist-safe captured_info logging problem, with overlapping testing_utils changes and shared target; 45645 is a superset that plausibly subsumes 45639." + "reason": "Same underlying xdist captured_info collision fix; 45645 appears to be a superset that adds the CI workflow handling needed for the new per-worker files." } ] }, @@ -25933,7 +25592,7 @@ { "label": "pull_request", "component_index": 23, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -25999,70 +25658,7 @@ { "label": "pull_request", "component_index": 24, - "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-45932-2", - "nodes": [ - "pull_request:45932", - "pull_request:46100" - ], - "soft_pairs": [ - "pull_request:45932|pull_request:46100" - ], - "prepared_review_unit_hash": "f61418c1a14beda88445646a5eec2b9338a12c7799f53272f49343f9c723da0e", - "component_budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 5161, - "estimated_input_tokens": 1291, - "estimated_eval_tokens": 2838 - }, - "budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 5161, - "estimated_input_tokens": 1291, - "estimated_eval_tokens": 2838 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": false, - "analyst_result": { - "summary": "These two PRs are not duplicates: one is a mixed model/infra change centered on Grafana/OpenTelemetry and AIMv2/model config updates, while the other is a test-migration PR adding kernel/DDP and vision model tests. They do not fix the same concrete bug or implement the same change.", - "confidence": 0.95, - "canonical_issue_reason": null, - "canonical_pr_reason": null, - "best_issue_reason": null, - "best_pr_reason": "PR #45932 is the better representative only because it is more substantive and self-contained than the mostly test-migration PR #46100; however, they should not be clustered as duplicates.", - "soft_edge_verdicts": [ - { - "left": "pull_request:45932", - "right": "pull_request:46100", - "accept": false, - "reason": "Different purposes and code paths: #45932 changes workflow secrets and several model/config files, while #46100 adds test infrastructure and model tests. They are not plausibly one mergeable fix/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": 25, - "component_count": 34, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26078,7 +25674,7 @@ "soft_pairs": [ "pull_request:46126|pull_request:46269" ], - "prepared_review_unit_hash": "7baf7cca38856aa9036992388fd022748ea4d13d22eb09dab94fad8f34204b3c", + "prepared_review_unit_hash": "64658a3aea57a890b911d9936c831fe223d3f3113ff2dfc3154d10a9f843b7ba", "component_budget": { "node_count": 2, "item_count": 2, @@ -26100,18 +25696,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs share some distributed code paths, but they are not the same concrete change: #46126 is a focused EP + Trainer integration, while #46269 is a much broader distributed-branch base touching many unrelated areas. I would not merge them as duplicates.", - "confidence": 0.89, + "summary": "These are related in the sense that both touch distributed/MoE code, but they are not the same underlying change. PR 46126 is a focused EP/router-parallel integration, while PR 46269 is a broad distributed branch base with many unrelated updates, including CI and multiple distributed subsystems.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical duplicate PR: the overlap is superficial (shared distributed files), but the scopes differ substantially and the larger branch PR is not the same underlying change as the EP/trainer integration PR.", + "canonical_pr_reason": "PR 46126 is the more specific, concrete code-path change: it targets router/EP behavior in tensor parallel, trainer integration, and Qwen3 MoE config. PR 46269 is a broad umbrella branch base touching many areas and is not a precise duplicate candidate.", "best_issue_reason": null, - "best_pr_reason": "#46269 is the better overall cluster representative because it is the broader umbrella distributed branch work; #46126 is a narrower feature PR layered on top of that area.", + "best_pr_reason": "PR 46126 is the best representative of the cluster because it is narrowly scoped and clearly describes a single functional change. PR 46269 is too broad and mixed to serve as the canonical change for duplicate triage.", "soft_edge_verdicts": [ { "left": "pull_request:46126", "right": "pull_request:46269", "accept": false, - "reason": "Shared filenames alone are not enough here. #46126 targets EP/router + trainer integration, while #46269 is a wide distributed refactor/branch-base PR with many extra changes; they do not look like one mergeable fix/change." + "reason": "Do not merge: they overlap in distributed filenames, but 46269 is a broad branch-base PR spanning many unrelated changes, not the same concrete bug fix or feature as the focused EP/router integration in 46126." } ] }, @@ -26124,8 +25720,8 @@ }, { "label": "pull_request", - "component_index": 26, - "component_count": 34, + "component_index": 25, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26179,220 +25775,220 @@ "pull_request:46504|pull_request:46716", "pull_request:46432|pull_request:46716" ], - "prepared_review_unit_hash": "8d3b9fbeab8b5c0f178c52619be4a573a7a2d34e06a4eac0681679c36a2056de", + "prepared_review_unit_hash": "01cb246bab2d5f472887bb3d28fc5e4f236aea8bcfbe85983cafd968c10e4ef2", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 31, - "serialized_chars": 26678, - "estimated_input_tokens": 6670, - "estimated_eval_tokens": 13596 + "serialized_chars": 26663, + "estimated_input_tokens": 6666, + "estimated_eval_tokens": 13588 }, "budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 31, - "serialized_chars": 26678, - "estimated_input_tokens": 6670, - "estimated_eval_tokens": 13596 + "serialized_chars": 26663, + "estimated_input_tokens": 6666, + "estimated_eval_tokens": 13588 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are mostly unrelated workflow-chore pull requests across different files. The pr-ci-security-gate edits are the closest cluster, but they still implement different security-hardening steps and should not be merged as duplicates.", - "confidence": 0.94, + "summary": "This cluster is heterogeneous: most PRs are unrelated workflow tweaks, and the pr-ci-security-gate items are similar in file but address different concerns. I found no true duplicate PR pair to merge.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 46554 is the most substantive and representative workflow change in the only tight subcluster: it adds concrete security-gate logic, while the nearby pr-ci-security-gate PRs make different, non-overlapping hardening edits.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "PR 46554 is the best overall representative because it contains the clearest underlying code-path change; the other PRs are separate workflow maintenance or security-hardening tweaks rather than duplicates.", "soft_edge_verdicts": [ { "left": "pull_request:46138", "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files and different fixes: stdin handling for self-comment CI vs adding checkout/diff checks to the PR security gate." + "reason": "Different workflows and different behavior; only a broad CI automation theme is shared." }, { "left": "pull_request:46138", "right": "pull_request:46558", "accept": false, - "reason": "Unrelated workflow edits; one changes self-comment CI argument passing, the other tightens PR security-gate permissions/approval flow." + "reason": "Different workflows; one updates self-comment CI stdin handling, the other changes PR CI security-gate permissions." }, { "left": "pull_request:46138", "right": "pull_request:46563", "accept": false, - "reason": "Different workflow and different purpose: self-comment CI input plumbing vs switching the PR security gate to a reusable workflow." + "reason": "Unrelated workflow files and different change goals." }, { "left": "pull_request:46138", "right": "pull_request:46716", "accept": false, - "reason": "Different workflow files and unrelated changes: self-comment CI vs changing security-gate secret handling." + "reason": "Different workflows; no shared underlying bug or code path." }, { "left": "pull_request:46138", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow files and unrelated code paths: self-comment CI stdin handling vs build-ci-docker-images message parsing." + "reason": "Different workflow files and different fixes; similarity is only superficial." }, { "left": "pull_request:46432", "right": "pull_request:46503", "accept": false, - "reason": "SSH runner hardening is unrelated to benchmark_v2 trigger logic." + "reason": "Different workflows with unrelated behavior changes." }, { "left": "pull_request:46432", "right": "pull_request:46501", "accept": false, - "reason": "Different workflow files and different concerns: SSH runner setup vs anti-slop permissions cleanup." + "reason": "Different workflow files; one is SSH runner hardening, the other anti-slop permissions." }, { "left": "pull_request:46432", "right": "pull_request:46502", "accept": false, - "reason": "Different workflow files and unrelated behavior: SSH runner setup vs assign-reviewers checkout ref." + "reason": "Different workflow files and unrelated changes." }, { "left": "pull_request:46432", "right": "pull_request:46554", "accept": false, - "reason": "SSH runner changes are unrelated to PR security-gate validation logic." + "reason": "Different workflow files and distinct security concerns." }, { "left": "pull_request:46432", "right": "pull_request:46558", "accept": false, - "reason": "Different workflows and different problems: SSH runner hardening vs security-gate permissions/approval gating." + "reason": "Different workflow files; no same underlying bug." }, { "left": "pull_request:46501", "right": "pull_request:46503", "accept": false, - "reason": "Same broad CI area, but different workflow logic: anti-slop permissions vs benchmark_v2 manual dispatch trigger." + "reason": "Both touch workflows, but they change different jobs and unrelated logic." }, { "left": "pull_request:46501", "right": "pull_request:46502", "accept": false, - "reason": "Different workflow files and unrelated fixes: anti-slop permissions vs reviewer assignment checkout ref." + "reason": "Different workflow updates in separate files; not the same fix." }, { "left": "pull_request:46501", "right": "pull_request:46504", "accept": false, - "reason": "Different workflows and different code paths: anti-slop permissions vs build-ci-docker-images commit-message parsing." + "reason": "Unrelated workflow changes; only generic CI maintenance overlaps." }, { "left": "pull_request:46501", "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files and unrelated changes: anti-slop permissions vs PR security-gate logic." + "reason": "Different files and unrelated security/CI behavior." }, { "left": "pull_request:46502", "right": "pull_request:46503", "accept": false, - "reason": "Different workflow behaviors: reviewer assignment checkout ref vs benchmark trigger conditions." + "reason": "Different workflow files and separate behavior changes." }, { "left": "pull_request:46502", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow files and unrelated fixes: reviewer assignment vs build-ci-docker-images message parsing." + "reason": "Different workflow files; no shared concrete bug." }, { "left": "pull_request:46502", "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files and unrelated concerns: assign-reviewers behavior vs security-gate logic." + "reason": "Different workflow files and distinct changes." }, { "left": "pull_request:46503", "right": "pull_request:46504", "accept": false, - "reason": "Different workflow files and unrelated changes: benchmark dispatch support vs build-ci-docker-images commit-message handling." + "reason": "Different workflow files and unrelated edits." }, { "left": "pull_request:46503", "right": "pull_request:46554", "accept": false, - "reason": "Different workflow files and unrelated fixes: benchmark trigger logic vs PR security-gate checks." + "reason": "Separate workflow concerns; not mergeable as one duplicate fix." }, { "left": "pull_request:46504", "right": "pull_request:46554", "accept": false, - "reason": "Different workflows and unrelated code paths: build-ci-docker-images parsing vs PR security-gate validation." + "reason": "Different workflow files and different operational behavior." }, { "left": "pull_request:46504", "right": "pull_request:46558", "accept": false, - "reason": "Different workflow files and different concerns: build-ci-docker-images parsing vs security-gate permissions/approval flow." + "reason": "Different workflow files; one is commit-message handling, the other PR CI permissions." }, { "left": "pull_request:46554", "right": "pull_request:46558", "accept": false, - "reason": "Same file, but different concrete changes: one adds checkout and file-change checks; the other changes permissions and approve-job gating." + "reason": "Same file, but different changes: one adds security checks, the other adjusts permissions/approval gating." }, { "left": "pull_request:46554", "right": "pull_request:46563", "accept": false, - "reason": "Same file, but different changes: reusable-workflow switch vs adding checkout and security checks inside the gate." + "reason": "Same workflow file, but the diffs address different concerns and are not the same code-path fix." }, { "left": "pull_request:46554", "right": "pull_request:46716", "accept": false, - "reason": "Same file, but different fixes: added checkout/diff checks vs replacing inherited secrets with empty secrets." + "reason": "Same workflow file, but one changes gate logic while the other changes reusable-workflow secrets handling." }, { "left": "pull_request:46558", "right": "pull_request:46563", "accept": false, - "reason": "Same file, but distinct security-gate changes: permissions/approval gating vs switching to a reusable workflow wrapper." + "reason": "Same file, but different modifications with different intent." }, { "left": "pull_request:46558", "right": "pull_request:46716", "accept": false, - "reason": "Same file, but not the same change: permissions and approve-job condition vs secrets handling." + "reason": "Same file, but permission/approval tweaks are distinct from secrets hardening." }, { "left": "pull_request:46563", "right": "pull_request:46716", "accept": false, - "reason": "Same workflow file, but different fixes: reusable-workflow invocation/comments vs changing secrets inheritance." + "reason": "Same file, but comment/documentation-only note versus secrets setting change." }, { "left": "pull_request:46504", "right": "pull_request:46563", "accept": false, - "reason": "Different workflows and unrelated changes: build-ci-docker-images parsing vs security-gate reusable workflow switch." + "reason": "Different workflow files and unrelated updates." }, { "left": "pull_request:46432", "right": "pull_request:46563", "accept": false, - "reason": "SSH runner hardening is unrelated to the PR security-gate reusable-workflow change." + "reason": "Different workflow files and no shared underlying issue." }, { "left": "pull_request:46504", "right": "pull_request:46716", "accept": false, - "reason": "Different workflows and unrelated code paths: build-ci-docker-images parsing vs security-gate secrets handling." + "reason": "Different workflow files and different behavior changes." }, { "left": "pull_request:46432", "right": "pull_request:46716", "accept": false, - "reason": "SSH runner changes are unrelated to PR security-gate secrets hardening." + "reason": "Different workflow files; no concrete duplicate change." } ] }, @@ -26405,8 +26001,8 @@ }, { "label": "pull_request", - "component_index": 27, - "component_count": 34, + "component_index": 26, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26468,8 +26064,8 @@ }, { "label": "pull_request", - "component_index": 28, - "component_count": 34, + "component_index": 27, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26485,12 +26081,12 @@ "soft_pairs": [ "pull_request:46281|pull_request:46890" ], - "prepared_review_unit_hash": "2a05af9eb7851567523da5f470bf3c2c4ee401819ebb86744f7ecdf7c17c9289", + "prepared_review_unit_hash": "6ade0bd89f46e66aafee8beb11e392e67120030763c0e422ebf72ad9c77b9053", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4280, + "serialized_chars": 4279, "estimated_input_tokens": 1070, "estimated_eval_tokens": 2396 }, @@ -26498,7 +26094,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4280, + "serialized_chars": 4279, "estimated_input_tokens": 1070, "estimated_eval_tokens": 2396 }, @@ -26507,18 +26103,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are security-oriented path-traversal fixes, but they target different code paths: sharded checkpoint shard resolution vs `transformers_weights` config handling.", - "confidence": 0.95, + "summary": "These are two separate security-hardened PRs that both block path traversal, but in different code paths: sharded checkpoint loading from an index (`hub.py`) versus resolving `transformers_weights` in model loading (`modeling_utils.py`). They are related by theme, not duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46281 is the cleaner representative of the cluster: it fixes path traversal in the sharded-checkpoint loader itself and includes a direct containment check; PR 46890 addresses a separate config-field traversal issue.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 46281 is the strongest standalone candidate: it has a narrowly scoped fix, explicit validation, and tests for the checkpoint shard loader.", + "best_pr_reason": "No single PR is the global duplicate-canonical for this cluster. PR 46281 fixes shard filename traversal in checkpoint index loading, while PR 46890 fixes traversal in `transformers_weights` resolution; they address different concrete bugs.", "soft_edge_verdicts": [ { "left": "pull_request:46281", "right": "pull_request:46890", "accept": false, - "reason": "Different underlying bugs and code paths. 46281 validates shard filenames from a checkpoint index in `hub.py`; 46890 validates `transformers_weights` in `modeling_utils.py`. They are related security themes, but not the same fix and not plausibly one merged PR." + "reason": "Both are path-traversal fixes, but they target different inputs and code paths (`weight_map` shard filenames vs `transformers_weights` config resolution), so they are not the same underlying change." } ] }, @@ -26531,8 +26127,8 @@ }, { "label": "pull_request", - "component_index": 29, - "component_count": 34, + "component_index": 28, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26548,7 +26144,7 @@ "soft_pairs": [ "pull_request:46574|pull_request:46580" ], - "prepared_review_unit_hash": "1e416f88d7d7177ccbc71c9ae2b6c06a337685bcf6d80844aa00155fe4d2def8", + "prepared_review_unit_hash": "cd5ed2136983f3635789ca040bd6f3390991d4f85d2823ab6d9e19c5d97cc7b5", "component_budget": { "node_count": 2, "item_count": 2, @@ -26570,18 +26166,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster contains two unrelated pull requests: one fixes CodeLlama decode-space stripping, the other fixes DeepSeek checkpoint tokenizer-class resolution. They do not look mergeable into a single PR.", - "confidence": 0.99, + "summary": "The two pull requests address different tokenizer problems: one fixes CodeLlama decode stripping real leading spaces, while the other fixes incorrect tokenizer-class selection for DeepSeek-Coder/DeepSeek-LLM checkpoints. They are not duplicates and should remain separate changes.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: the changes target different model families and different failure modes (decode behavior vs tokenizer-class selection).", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "Neither PR is a good cluster representative because they are unrelated fixes rather than duplicate implementations of the same bug.", + "best_pr_reason": "No single canonical PR stands out because these are separate bugs in different code paths: CodeLlama decoding behavior versus AutoTokenizer checkpoint-class routing.", "soft_edge_verdicts": [ { "left": "pull_request:46574", "right": "pull_request:46580", "accept": false, - "reason": "Different underlying bugs and code paths: CodeLlama decoder/leading-space handling vs DeepSeek auto-tokenizer class selection for specific checkpoint prefixes. They are not the same change and would not plausibly be merged into one PR." + "reason": "Different underlying bugs and fixes. PR 46574 changes CodeLlama decoder setup to preserve leading spaces on decode; PR 46580 changes AutoTokenizer checkpoint-prefix routing for DeepSeek models that load the wrong tokenizer class." } ] }, @@ -26594,8 +26190,8 @@ }, { "label": "pull_request", - "component_index": 30, - "component_count": 34, + "component_index": 29, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26611,7 +26207,7 @@ "soft_pairs": [ "pull_request:46773|pull_request:47087" ], - "prepared_review_unit_hash": "e8ea36081d455275c18d0280930d0306cc49a0ad0f70723717767649a28e4850", + "prepared_review_unit_hash": "b469ff071725075b88f6d73728065ef16d1fadc8564978681dcb85080aaa5169", "component_budget": { "node_count": 2, "item_count": 2, @@ -26633,18 +26229,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a high level (padding-mask behavior), but they fix different bugs in different code paths: batch-size-1 handling in linear-attention models vs. continued/cached recurrent forwards. I would not cluster them as duplicates.", - "confidence": 0.96, + "summary": "These are related padding-mask fixes, but they address different code paths: one fixes batch-size-1 masking in several linear-attention model implementations, while the other changes the shared recurrent masking utility for continued forwards/chunked prefill/cache continuation. They should not be deduplicated or merged as the same PR.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "47087 is the better standalone PR to anchor the cluster because it has an explicit issue link and a narrowly defined recurrent-forward masking fix.", + "canonical_pr_reason": "PR 47087 is the better canonical PR because it targets the shared masking utility, has an explicit issue link, and fixes a broader recurrent-forward masking bug with clearer scope.", "best_issue_reason": null, - "best_pr_reason": "47087 is the strongest PR candidate overall: it has explicit issue targeting, a clear single bug, and a tighter scope than 46773.", + "best_pr_reason": "PR 47087 is the strongest standalone PR: it fixes a concrete shared bug in recurrent masking behavior during continued forwards and is backed by tests and an issue target. PR 46773 is narrower and model-specific.", "soft_edge_verdicts": [ { "left": "pull_request:46773", "right": "pull_request:47087", "accept": false, - "reason": "Both concern padding masks, but the concrete defects differ: 46773 fixes masking being skipped for batch size 1 in linear-attention models, while 47087 fixes recurrent padding masks being skipped on continued/cache forwards. They are not the same underlying change." + "reason": "Both concern skipped padding masks, but the underlying bugs differ: 46773 fixes a batch-size-1 condition in linear-attention model code, while 47087 fixes recurrent mask handling for continued forwards in shared masking utilities. They are not the same concrete change and would not plausibly merge into one PR." } ] }, @@ -26657,8 +26253,8 @@ }, { "label": "pull_request", - "component_index": 31, - "component_count": 34, + "component_index": 30, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26676,7 +26272,7 @@ "pull_request:46781|pull_request:46993", "pull_request:46993|pull_request:46994" ], - "prepared_review_unit_hash": "bc1edff1d59883e2a03d1446c26541a2b4bfe0e1b9aa0b1d4482817f63edeec9", + "prepared_review_unit_hash": "bb3e7a2fbf81fc74f0c8295117fc2abd368826afbf0c44f1987643ec6460d3ce", "component_budget": { "node_count": 3, "item_count": 3, @@ -26698,24 +26294,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three pull requests are small but they address different bugs in different generation/cache code paths, so this cluster does not contain a duplicate pair.", - "confidence": 0.98, + "summary": "These are three separate pull requests touching different generation/cache validation paths. They share a broad Transformers-generation theme, but not the same underlying bug or mergeable change, so neither soft edge should be accepted.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: 46781 is about QuantizedCache layer-type validation, 46993 fixes prompt lookup decoding length budgeting, and 46994 tightens validation for empty bad_words_ids/sequence_bias inputs.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "No single best PR representative for a duplicate cluster, because these PRs are unrelated fixes rather than alternate patches for the same underlying bug.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:46781", "right": "pull_request:46993", "accept": false, - "reason": "Different problems and code paths: cache-layer support validation vs prompt lookup decoding max_length handling." + "reason": "Different code paths and failures: QuantizedCache layer-type validation vs. prompt lookup decoding max_length handling. They are unrelated bug fixes, not one mergeable PR." }, { "left": "pull_request:46993", "right": "pull_request:46994", "accept": false, - "reason": "Different generation bugs: prompt lookup candidate length capping vs validation of empty token lists in logits processing." + "reason": "One fixes candidate generation length capping, the other validates empty lists in logits processors. Same subsystem, but distinct issues and changes." } ] }, @@ -26728,8 +26324,8 @@ }, { "label": "pull_request", - "component_index": 32, - "component_count": 34, + "component_index": 31, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26745,7 +26341,7 @@ "soft_pairs": [ "pull_request:47026|pull_request:47046" ], - "prepared_review_unit_hash": "f9d51d0d51b1549424c13f12b43c1e9b2e251e403562a6e8e7d340faddcaf5c3", + "prepared_review_unit_hash": "7553ee12103cf181e5a8615c5dbfecdfeb53218819b003242476cf1f66cb514b", "component_budget": { "node_count": 2, "item_count": 2, @@ -26767,18 +26363,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same xLSTM train-mode crash path around `return_last_states=False` in `modeling_xlstm.py`. PR 47046 looks like the broader follow-up/final fix, while 47026 appears to be a narrower subset focused on the padding/train-path return-state handling.", - "confidence": 0.83, + "summary": "The two xLSTM PRs overlap on the same train-mode forward crash around `return_last_states=False`; PR 47046 appears to be the broader, more complete fix and likely subsumes PR 47026.", + "confidence": 0.84, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47046 is the better canonical PR because it addresses the same crash class but appears broader (`return_last_states=False` and `use_cache`) and includes tests, making it the more complete change set.", + "canonical_pr_reason": "PR 47046 is the more complete xLSTM fix: it targets the same train-mode crash path, adds `use_cache` handling as well, and includes tests/logging changes, so it is the better canonical PR for this bug family.", "best_issue_reason": null, - "best_pr_reason": "PR 47046 is the best representative because it covers the same underlying xLSTM train-mode failure and seems to subsume the narrower 47026 fix.", + "best_pr_reason": "PR 47046 is the best overall PR because it covers the shared crash scenario more comprehensively and appears to include validation coverage, whereas PR 47026 is a narrower subset fix.", "soft_edge_verdicts": [ { "left": "pull_request:47026", "right": "pull_request:47046", "accept": true, - "reason": "Both PRs fix the same xLSTM train-mode crash in the same file and under the same issue target; 47046 appears to be a broader superset of the narrower 47026 change." + "reason": "Both PRs address the same underlying xLSTM train-mode crash path involving `return_last_states=False` in `modeling_xlstm.py`; PR 47046 looks like a superset that also handles `use_cache`, so they are close enough to treat as the same change cluster." } ] }, @@ -26791,8 +26387,8 @@ }, { "label": "pull_request", - "component_index": 33, - "component_count": 34, + "component_index": 32, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26820,12 +26416,12 @@ "pull_request:47060|pull_request:47076", "pull_request:47074|pull_request:47076" ], - "prepared_review_unit_hash": "a0005eb6a949faeb7fab600dabbe1250596ada12005212a6c2be4892272915f3", + "prepared_review_unit_hash": "c5325b77d1c18a3b3875bc7f183df6205abb44f78404af2a6a262e7c852b8095", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 9, - "serialized_chars": 15865, + "serialized_chars": 15866, "estimated_input_tokens": 3967, "estimated_eval_tokens": 8190 }, @@ -26833,7 +26429,7 @@ "node_count": 6, "item_count": 6, "soft_pair_count": 9, - "serialized_chars": 15865, + "serialized_chars": 15866, "estimated_input_tokens": 3967, "estimated_eval_tokens": 8190 }, @@ -26842,66 +26438,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related only at a broad theme level: multimodal input/output handling. They fix different code paths in different components (pipelines, Voxtral processor, Qwen Omni post-processing, Gemma4 layout validation) and should not be deduplicated together.", + "summary": "These are six separate multimodal pipeline/processor fixes with overlapping themes but different concrete bugs and code paths. I do not see a true duplicate pair among them.", "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #47051 is the broadest and most central to the shared theme: it updates multimodal generation pipeline parameter routing in a way that spans both `any_to_any` and `image_text_to_text`. Among the set, it is the best representative of the cluster\u2019s pipeline-focused behavior changes.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR #47051 is the best fit as a representative patch for the cluster because it targets the common pipeline layer and addresses direct generation-kwarg forwarding across multimodal pipelines, unlike the others which are model-specific processor fixes or unrelated post-processing adjustments.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:47032", "right": "pull_request:47051", "accept": false, - "reason": "Both touch `image_text_to_text`, but they fix different problems: stop-sequence handling vs. forwarding direct generation kwargs. They are related changes, not the same underlying bug." + "reason": "Both touch image_text_to_text pipeline code, but 47032 changes stop_sequence -> stop_strings behavior while 47051 changes direct generation-kwarg forwarding. Different bug classes and not mergeable as one fix." }, { "left": "pull_request:47032", "right": "pull_request:47045", "accept": false, - "reason": "Different components and behaviors: image-text pipeline stop handling versus Voxtral audio format validation/encoding. Not the same bug." + "reason": "Different components and failures: image-text stop-sequence handling versus Voxtral audio-format validation/encoding." }, { "left": "pull_request:47045", "right": "pull_request:47076", "accept": false, - "reason": "Both are processor-side validation changes, but they address different inputs and models: Voxtral audio arrays vs. Gemma4 audio/image placeholder layout. Not duplicates." + "reason": "Both validate processor inputs, but one is Voxtral audio array formatting and the other is Gemma4 audio placeholder/layout validation. Distinct models and code paths." }, { "left": "pull_request:47045", "right": "pull_request:47074", "accept": false, - "reason": "Voxtral audio-format encoding and AnyToAny key normalization are unrelated fixes in different code paths." + "reason": "Unrelated fixes: Voxtral transcription audio handling versus AnyToAnyPipeline multimodal key normalization." }, { "left": "pull_request:47051", "right": "pull_request:47074", "accept": false, - "reason": "Both affect `any_to_any`, but one changes kwargs routing while the other fixes input key naming. Distinct issues and not safely mergeable as one duplicate fix." + "reason": "Both are in AnyToAnyPipeline, but 47051 forwards generation kwargs and 47074 fixes input-key naming for videos. Separate behaviors and not the same bug." }, { "left": "pull_request:47051", "right": "pull_request:47060", "accept": false, - "reason": "Multimodal pipeline parameter forwarding and Qwen Omni batched post-processing are different layers and different bugs." + "reason": "Different targets and problems: multimodal pipeline kwarg forwarding versus Qwen Omni batch post-processing decode shape handling." }, { "left": "pull_request:47060", "right": "pull_request:47074", "accept": false, - "reason": "The Qwen Omni batched decode fix and AnyToAny input-key normalization do not fix the same concrete code-path problem." + "reason": "Qwen Omni post-processing batch decoding is unrelated to AnyToAnyPipeline input key normalization." }, { "left": "pull_request:47060", "right": "pull_request:47076", "accept": false, - "reason": "Different model-specific issues: Qwen Omni output batching versus Gemma4 processor placeholder validation." + "reason": "Different models and code paths: Qwen Omni output decoding versus Gemma4 audio placeholder validation." }, { "left": "pull_request:47074", "right": "pull_request:47076", "accept": false, - "reason": "One normalizes pipeline input keys; the other validates Gemma4 audio placeholders. Similar theme, but not the same underlying change." + "reason": "Both are multimodal pipeline/processor changes, but one fixes AnyToAnyPipeline input keys and the other fixes Gemma4 placeholder validation. Not the same underlying bug or change." } ] }, @@ -26914,8 +26510,8 @@ }, { "label": "pull_request", - "component_index": 34, - "component_count": 34, + "component_index": 33, + "component_count": 33, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -26931,12 +26527,12 @@ "soft_pairs": [ "pull_request:47104|pull_request:47115" ], - "prepared_review_unit_hash": "6c2f9eb6bfe8d89dabf27d0ebb37760d16d1eb48d6358cd04be694e5223f6642", + "prepared_review_unit_hash": "2b8d757150617b14aa9fefd87d239d8bf6cc877b3462ad5447837e389a80584d", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6949, + "serialized_chars": 6950, "estimated_input_tokens": 1738, "estimated_eval_tokens": 3732 }, @@ -26944,7 +26540,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6949, + "serialized_chars": 6950, "estimated_input_tokens": 1738, "estimated_eval_tokens": 3732 }, @@ -26953,18 +26549,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests target the same underlying change: reducing processor-test CI memory by switching tests to tiny Hub repos. PR 47104 looks like the broader initial implementation, while 47115 is a narrower follow-up in the same area; they are similar enough to treat as the same change cluster.", - "confidence": 0.94, + "summary": "Both PRs target the same processor-test memory reduction work: switching processor tests to tiny Hub repos to lower CI memory usage. They overlap heavily in affected test files and appear to be alternative implementations of the same underlying change.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 47104 is the earlier, broader implementation of the same processor-test memory-reduction effort and includes the supporting workflow/tooling changes around the test updates.", + "canonical_pr_reason": "PR #47104 is the more comprehensive implementation of the same fix, bundling the processor-test tiny-repo migration with the supporting CI/memory-tracking changes and workflow updates.", "best_issue_reason": null, - "best_pr_reason": "PR 47115 is the clearest single-purpose fix for the shared problem, but 47104 is the better canonical representative because it captures the larger end-to-end change set.", + "best_pr_reason": "PR #47104 is the strongest representative of the cluster because it covers the core memory-reduction change plus the supporting infrastructure needed to validate and run it.", "soft_edge_verdicts": [ { "left": "pull_request:47104", "right": "pull_request:47115", "accept": true, - "reason": "Same title/theme and overlapping processor test files; both implement the same memory-reduction change by moving processor tests to tiny Hub repos, so they are effectively the same underlying PR change." + "reason": "Same underlying change: both convert processor tests to tiny Hub repos to reduce CI memory, with heavily overlapping test files and matching intent." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index c5570d7c47329ec1dabb8296720ae26c6b6f0878..a64adc3f364e0617b50de8e7989a7bd898aec160 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260707t000034z", + "analysis_id": "hybrid-model-20260707t060715z", "archived_artifacts": { - "hybrid": "snapshots/20260707T000034Z/analysis-runs/hybrid-model-20260707t000034z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260707T000034Z/analysis-runs/hybrid-model-20260707t000034z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-07-07T00:11:39Z", + "published_at": "2026-07-07T06:19:24Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260707T000034Z", + "snapshot_id": "20260707T060715Z", "variant": "hybrid" } diff --git a/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.json b/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..9d3fcbdcda1d647a0c21e972357f5f03e5c41d9a --- /dev/null +++ b/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.json @@ -0,0 +1,4448 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260707T060715Z", + "generated_at": "2026-07-07T06:19:22Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "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-44779-7", + "summary": "Cluster of 5 issues and 2 PRs centered on issue #45488.", + "status": "open", + "confidence": 0.6, + "canonical_issue_number": 45488, + "canonical_pr_number": 47017, + "issue_numbers": [ + 44779, + 45488, + 45812, + 45920, + 46489 + ], + "pr_numbers": [ + 46580, + 47017 + ], + "evidence_types": [ + "closing_reference", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46580, + "right_pr_number": 47017, + "code_similarity": 0.241, + "size_similarity": 0.352, + "file_overlap": 0.5, + "area_overlap": 0.0, + "patch_similarity": 0.136, + "shared_filenames": [ + "src/transformers/models/auto/tokenization_auto.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/auto/tokenization_auto.py", + "left_ranges": [ + [ + 361, + 367 + ], + [ + 399, + 421 + ], + [ + 724, + 739 + ] + ], + "right_ranges": [ + [ + 428, + 475 + ], + [ + 807, + 825 + ] + ] + } + ] + } + ] + }, + { + "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-46196-4", + "summary": "Cluster of 4 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 46196, + "issue_numbers": [], + "pr_numbers": [ + 46196, + 46658, + 47003, + 47096 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46196, + "right_pr_number": 46658, + "code_similarity": 0.163, + "size_similarity": 0.723, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.12, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46196, + "right_pr_number": 47003, + "code_similarity": 0.231, + "size_similarity": 0.995, + "file_overlap": 0.055, + "area_overlap": 0.006, + "patch_similarity": 0.092, + "shared_filenames": [ + "src/transformers/models/unispeech/modeling_unispeech.py", + "src/transformers/models/unispeech/modular_unispeech.py", + "src/transformers/models/wavlm/modeling_wavlm.py", + "src/transformers/models/wavlm/modular_wavlm.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/unispeech/modeling_unispeech.py", + "left_ranges": [ + [ + 29, + 35 + ], + [ + 45, + 52 + ], + [ + 337, + 343 + ], + [ + 376, + 382 + ], + [ + 419, + 440 + ], + [ + 451, + 458 + ], + [ + 472, + 486 + ], + [ + 532, + 550 + ], + [ + 563, + 570 + ], + [ + 583, + 600 + ], + [ + 677, + 686 + ], + [ + 934, + 954 + ], + [ + 964, + 977 + ], + [ + 1029, + 1041 + ], + [ + 1048, + 1062 + ], + [ + 1080, + 1085 + ], + [ + 1091, + 1096 + ], + [ + 1165, + 1178 + ], + [ + 1181, + 1196 + ], + [ + 1223, + 1228 + ], + [ + 1267, + 1281 + ], + [ + 1289, + 1310 + ], + [ + 1322, + 1327 + ] + ], + "right_ranges": [ + [ + 1138, + 1150 + ] + ] + }, + { + "filename": "src/transformers/models/unispeech/modular_unispeech.py", + "left_ranges": [ + [ + 22, + 31 + ], + [ + 147, + 156 + ], + [ + 241, + 261 + ], + [ + 271, + 284 + ], + [ + 336, + 348 + ], + [ + 355, + 369 + ], + [ + 387, + 392 + ] + ], + "right_ranges": [ + [ + 383, + 395 + ] + ] + }, + { + "filename": "src/transformers/models/wavlm/modeling_wavlm.py", + "left_ranges": [ + [ + 15, + 21 + ], + [ + 28, + 37 + ], + [ + 153, + 160 + ], + [ + 184, + 190 + ], + [ + 194, + 199 + ], + [ + 222, + 229 + ], + [ + 314, + 334 + ], + [ + 338, + 344 + ], + [ + 357, + 382 + ], + [ + 394, + 402 + ], + [ + 411, + 431 + ], + [ + 446, + 454 + ], + [ + 462, + 484 + ], + [ + 561, + 570 + ], + [ + 998, + 1018 + ], + [ + 1030, + 1049 + ], + [ + 1118, + 1131 + ], + [ + 1134, + 1149 + ], + [ + 1176, + 1181 + ], + [ + 1220, + 1234 + ], + [ + 1242, + 1263 + ], + [ + 1275, + 1280 + ], + [ + 1316, + 1330 + ], + [ + 1338, + 1359 + ], + [ + 1362, + 1367 + ], + [ + 1480, + 1494 + ], + [ + 1502, + 1523 + ], + [ + 1547, + 1552 + ] + ], + "right_ranges": [ + [ + 241, + 251 + ] + ] + }, + { + "filename": "src/transformers/models/wavlm/modular_wavlm.py", + "left_ranges": [ + [ + 10, + 17 + ], + [ + 81, + 88 + ], + [ + 112, + 118 + ], + [ + 122, + 127 + ], + [ + 150, + 157 + ], + [ + 222, + 242 + ], + [ + 246, + 252 + ], + [ + 265, + 290 + ], + [ + 302, + 310 + ], + [ + 319, + 339 + ], + [ + 354, + 362 + ], + [ + 370, + 392 + ] + ], + "right_ranges": [ + [ + 172, + 182 + ] + ] + } + ] + }, + { + "left_pr_number": 46196, + "right_pr_number": 47096, + "code_similarity": 0.044, + "size_similarity": 0.129, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.124, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47003, + "code_similarity": 0.174, + "size_similarity": 0.719, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.198, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 46658, + "right_pr_number": 47096, + "code_similarity": 0.035, + "size_similarity": 0.093, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.11, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 47003, + "right_pr_number": 47096, + "code_similarity": 0.071, + "size_similarity": 0.13, + "file_overlap": 0.078, + "area_overlap": 0.015, + "patch_similarity": 0.108, + "shared_filenames": [ + "src/transformers/exporters/exporter_dynamo.py", + "src/transformers/exporters/exporter_executorch.py", + "src/transformers/exporters/utils.py", + "tests/exporters/test_export.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/exporters/exporter_dynamo.py", + "left_ranges": [ + [ + 43, + 49 + ], + [ + 449, + 470 + ], + [ + 487, + 507 + ], + [ + 548, + 559 + ], + [ + 594, + 606 + ], + [ + 675, + 682 + ] + ], + "right_ranges": [ + [ + 350, + 357 + ], + [ + 381, + 388 + ] + ] + }, + { + "filename": "src/transformers/exporters/exporter_executorch.py", + "left_ranges": [ + [ + 336, + 342 + ], + [ + 347, + 390 + ], + [ + 397, + 412 + ], + [ + 446, + 457 + ], + [ + 471, + 503 + ], + [ + 633, + 761 + ], + [ + 809, + 1120 + ], + [ + 1212, + 1241 + ], + [ + 1377, + 1393 + ] + ], + "right_ranges": [ + [ + 418, + 442 + ] + ] + }, + { + "filename": "src/transformers/exporters/utils.py", + "left_ranges": [ + [ + 232, + 238 + ], + [ + 444, + 461 + ], + [ + 621, + 644 + ], + [ + 647, + 663 + ], + [ + 761, + 768 + ] + ], + "right_ranges": [ + [ + 432, + 441 + ], + [ + 444, + 455 + ], + [ + 464, + 490 + ], + [ + 708, + 720 + ] + ] + }, + { + "filename": "tests/exporters/test_export.py", + "left_ranges": [ + [ + 24, + 30 + ], + [ + 35, + 41 + ], + [ + 46, + 92 + ], + [ + 98, + 104 + ], + [ + 109, + 149 + ], + [ + 153, + 185 + ], + [ + 226, + 301 + ], + [ + 333, + 349 + ], + [ + 442, + 449 + ], + [ + 455, + 487 + ], + [ + 588, + 595 + ], + [ + 602, + 634 + ] + ], + "right_ranges": [ + [ + 181, + 188 + ] + ] + } + ] + } + ] + }, + { + "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": 0.763, + "size_similarity": 0.866, + "file_overlap": 1.0, + "area_overlap": 0.607, + "patch_similarity": 0.519, + "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": [ + [ + 175, + 213 + ] + ] + } + ] + }, + { + "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": 0.324, + "size_similarity": 0.35, + "file_overlap": 0.333, + "area_overlap": 0.364, + "patch_similarity": 0.179, + "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": 46958, + "right_pr_number": 47090, + "code_similarity": 0.838, + "size_similarity": 0.947, + "file_overlap": 1.0, + "area_overlap": 0.762, + "patch_similarity": 0.549, + "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": [ + [ + 22, + 28 + ], + [ + 638, + 645 + ] + ] + }, + { + "filename": "src/transformers/models/cohere_asr/modular_cohere_asr.py", + "left_ranges": [ + [ + 16, + 22 + ], + [ + 501, + 508 + ] + ], + "right_ranges": [ + [ + 16, + 22 + ], + [ + 501, + 508 + ] + ] + }, + { + "filename": "tests/models/cohere_asr/test_modeling_cohere_asr.py", + "left_ranges": [ + [ + 15, + 21 + ], + [ + 180, + 221 + ] + ], + "right_ranges": [ + [ + 175, + 213 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43240-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43240, + "canonical_pr_number": 43251, + "issue_numbers": [ + 43240 + ], + "pr_numbers": [ + 43251, + 43254 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43251, + "right_pr_number": 43254, + "code_similarity": 0.794, + "size_similarity": 0.64, + "file_overlap": 1.0, + "area_overlap": 0.667, + "patch_similarity": 0.882, + "shared_filenames": [ + "src/transformers/loss/loss_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/loss/loss_utils.py", + "left_ranges": [ + [ + 30, + 50 + ] + ], + "right_ranges": [ + [ + 30, + 43 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45647-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45647.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 45647, + "canonical_pr_number": 45738, + "issue_numbers": [ + 45647 + ], + "pr_numbers": [ + 45737, + 45738 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 45737, + "right_pr_number": 45738, + "code_similarity": 0.526, + "size_similarity": 0.11, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.024, + "shared_filenames": [ + "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/musicgen_melody/modeling_musicgen_melody.py", + "left_ranges": [ + [ + 491, + 497 + ] + ], + "right_ranges": [ + [ + 491, + 497 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-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-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-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-47013-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #47013.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 47013, + "canonical_pr_number": 47026, + "issue_numbers": [ + 47013 + ], + "pr_numbers": [ + 47026, + 47046 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 47026, + "right_pr_number": 47046, + "code_similarity": 0.308, + "size_similarity": 0.264, + "file_overlap": 0.5, + "area_overlap": 0.229, + "patch_similarity": 0.169, + "shared_filenames": [ + "src/transformers/models/xlstm/modeling_xlstm.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/xlstm/modeling_xlstm.py", + "left_ranges": [ + [ + 816, + 840 + ], + [ + 843, + 851 + ] + ], + "right_ranges": [ + [ + 25, + 39 + ], + [ + 789, + 795 + ], + [ + 808, + 817 + ], + [ + 823, + 846 + ], + [ + 849, + 857 + ], + [ + 1388, + 1394 + ], + [ + 1399, + 1405 + ], + [ + 1461, + 1473 + ], + [ + 1504, + 1522 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-45081-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #45081.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 45081, + "canonical_pr_number": 45317, + "issue_numbers": [ + 45081 + ], + "pr_numbers": [ + 45086, + 45317 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 45086, + "right_pr_number": 45317, + "code_similarity": 0.589, + "size_similarity": 0.34, + "file_overlap": 0.5, + "area_overlap": 1.0, + "patch_similarity": 0.136, + "shared_filenames": [ + "src/transformers/tokenization_utils_tokenizers.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/tokenization_utils_tokenizers.py", + "left_ranges": [ + [ + 1360, + 1370 + ], + [ + 1374, + 1380 + ] + ], + "right_ranges": [ + [ + 1360, + 1370 + ], + [ + 1374, + 1380 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43698-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43698.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43698, + "canonical_pr_number": 43779, + "issue_numbers": [ + 43698 + ], + "pr_numbers": [ + 43779, + 43816 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43779, + "right_pr_number": 43816, + "code_similarity": 0.418, + "size_similarity": 0.538, + "file_overlap": 1.0, + "area_overlap": 0.02, + "patch_similarity": 0.02, + "shared_filenames": [ + "src/transformers/integrations/integration_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/integrations/integration_utils.py", + "left_ranges": [ + [ + 2227, + 2241 + ], + [ + 2303, + 2309 + ] + ], + "right_ranges": [ + [ + 2278, + 2291 + ], + [ + 2309, + 2322 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43656-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 43656, + "issue_numbers": [], + "pr_numbers": [ + 43656, + 43842 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43656, + "right_pr_number": 43842, + "code_similarity": 0.405, + "size_similarity": 0.5, + "file_overlap": 1.0, + "area_overlap": 0.0, + "patch_similarity": 0.036, + "shared_filenames": [ + "src/transformers/cli/serve.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/cli/serve.py", + "left_ranges": [ + [ + 11, + 18 + ], + [ + 30, + 36 + ], + [ + 315, + 323 + ], + [ + 665, + 671 + ], + [ + 931, + 937 + ], + [ + 1843, + 1849 + ], + [ + 1868, + 1874 + ] + ], + "right_ranges": [ + [ + 54, + 61 + ], + [ + 587, + 593 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-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 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-47104-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 47115, + "issue_numbers": [], + "pr_numbers": [ + 47104, + 47115 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 47104, + "right_pr_number": 47115, + "code_similarity": 0.645, + "size_similarity": 0.47, + "file_overlap": 0.851, + "area_overlap": 0.711, + "patch_similarity": 0.309, + "shared_filenames": [ + "tests/models/altclip/test_processing_altclip.py", + "tests/models/aria/test_processing_aria.py", + "tests/models/aya_vision/test_processing_aya_vision.py", + "tests/models/clip/test_processing_clip.py", + "tests/models/colqwen2/test_processing_colqwen2.py", + "tests/models/csm/test_processing_csm.py", + "tests/models/dia/test_processing_dia.py", + "tests/models/donut/test_processing_donut.py", + "tests/models/ernie4_5_vl_moe/test_processing_ernie4_5_vl_moe.py", + "tests/models/gemma3n/test_processing_gemma3n.py", + "tests/models/glm46v/test_processor_glm46v.py", + "tests/models/glm4v/test_processor_glm4v.py", + "tests/models/glm_image/test_processor_glm_image.py", + "tests/models/grounding_dino/test_processing_grounding_dino.py", + "tests/models/idefics2/test_processing_idefics2.py", + "tests/models/idefics3/test_processing_idefics3.py", + "tests/models/internvl/test_processing_internvl.py", + "tests/models/janus/test_processing_janus.py", + "tests/models/kosmos2_5/test_processor_kosmos2_5.py", + "tests/models/lighton_ocr/test_processor_lighton_ocr.py", + "tests/models/llava/test_processing_llava.py", + "tests/models/llava_next_video/test_processing_llava_next_video.py", + "tests/models/minicpmv4_6/test_processing_minicpmv4_6.py", + "tests/models/mistral3/test_processing_mistral3.py", + "tests/models/mllama/test_processing_mllama.py", + "tests/models/owlv2/test_processing_owlv2.py", + "tests/models/parakeet/test_processing_parakeet.py", + "tests/models/pixtral/test_processing_pixtral.py", + "tests/models/qianfan_ocr/test_processing_qianfan_ocr.py", + "tests/models/qwen2_5_omni/test_processing_qwen2_5_omni.py", + "tests/models/qwen2_5_vl/test_processing_qwen2_5_vl.py", + "tests/models/qwen2_audio/test_processing_qwen2_audio.py", + "tests/models/qwen2_vl/test_processing_qwen2_vl.py", + "tests/models/qwen3_omni_moe/test_processing_qwen3_omni_moe.py", + "tests/models/qwen3_vl/test_processing_qwen3_vl.py", + "tests/models/seamless_m4t/test_processing_seamless_m4t.py", + "tests/models/smolvlm/test_processing_smolvlm.py", + "tests/models/trocr/test_processing_trocr.py", + "tests/models/video_llama_3/test_processing_video_llama_3.py", + "tests/test_processing_common.py" + ], + "shared_file_areas": [ + { + "filename": "tests/models/altclip/test_processing_altclip.py", + "left_ranges": [ + [ + 24, + 28 + ] + ], + "right_ranges": [ + [ + 24, + 28 + ] + ] + }, + { + "filename": "tests/models/aria/test_processing_aria.py", + "left_ranges": [ + [ + 33, + 39 + ] + ], + "right_ranges": [ + [ + 33, + 39 + ] + ] + }, + { + "filename": "tests/models/aya_vision/test_processing_aya_vision.py", + "left_ranges": [ + [ + 28, + 34 + ], + [ + 38, + 45 + ], + [ + 64, + 73 + ] + ], + "right_ranges": [ + [ + 28, + 39 + ] + ] + }, + { + "filename": "tests/models/clip/test_processing_clip.py", + "left_ranges": [ + [ + 27, + 31 + ] + ], + "right_ranges": [ + [ + 27, + 31 + ] + ] + }, + { + "filename": "tests/models/colqwen2/test_processing_colqwen2.py", + "left_ranges": [ + [ + 37, + 43 + ] + ], + "right_ranges": [ + [ + 37, + 43 + ] + ] + }, + { + "filename": "tests/models/csm/test_processing_csm.py", + "left_ranges": [ + [ + 33, + 39 + ], + [ + 194, + 201 + ] + ], + "right_ranges": [ + [ + 33, + 39 + ], + [ + 194, + 201 + ] + ] + }, + { + "filename": "tests/models/dia/test_processing_dia.py", + "left_ranges": [ + [ + 42, + 48 + ] + ], + "right_ranges": [ + [ + 42, + 48 + ] + ] + }, + { + "filename": "tests/models/donut/test_processing_donut.py", + "left_ranges": [ + [ + 21, + 27 + ] + ], + "right_ranges": [ + [ + 21, + 27 + ] + ] + }, + { + "filename": "tests/models/ernie4_5_vl_moe/test_processing_ernie4_5_vl_moe.py", + "left_ranges": [ + [ + 13, + 18 + ], + [ + 37, + 44 + ], + [ + 52, + 57 + ] + ], + "right_ranges": [ + [ + 13, + 18 + ], + [ + 37, + 44 + ], + [ + 52, + 57 + ] + ] + }, + { + "filename": "tests/models/gemma3n/test_processing_gemma3n.py", + "left_ranges": [ + [ + 33, + 39 + ] + ], + "right_ranges": [ + [ + 33, + 39 + ] + ] + }, + { + "filename": "tests/models/glm46v/test_processor_glm46v.py", + "left_ranges": [ + [ + 34, + 40 + ] + ], + "right_ranges": [ + [ + 34, + 40 + ] + ] + }, + { + "filename": "tests/models/glm4v/test_processor_glm4v.py", + "left_ranges": [ + [ + 33, + 45 + ] + ], + "right_ranges": [ + [ + 33, + 45 + ] + ] + }, + { + "filename": "tests/models/glm_image/test_processor_glm_image.py", + "left_ranges": [ + [ + 34, + 40 + ], + [ + 43, + 72 + ] + ], + "right_ranges": [ + [ + 24, + 30 + ], + [ + 34, + 45 + ] + ] + }, + { + "filename": "tests/models/grounding_dino/test_processing_grounding_dino.py", + "left_ranges": [ + [ + 32, + 38 + ] + ], + "right_ranges": [ + [ + 32, + 38 + ] + ] + }, + { + "filename": "tests/models/idefics2/test_processing_idefics2.py", + "left_ranges": [ + [ + 32, + 38 + ], + [ + 62, + 68 + ] + ], + "right_ranges": [ + [ + 32, + 38 + ], + [ + 63, + 74 + ], + [ + 86, + 92 + ], + [ + 109, + 116 + ] + ] + }, + { + "filename": "tests/models/idefics3/test_processing_idefics3.py", + "left_ranges": [ + [ + 27, + 35 + ] + ], + "right_ranges": [ + [ + 27, + 37 + ] + ] + }, + { + "filename": "tests/models/internvl/test_processing_internvl.py", + "left_ranges": [ + [ + 32, + 38 + ], + [ + 63, + 68 + ], + [ + 373, + 379 + ] + ], + "right_ranges": [ + [ + 32, + 38 + ], + [ + 63, + 68 + ], + [ + 373, + 379 + ] + ] + }, + { + "filename": "tests/models/janus/test_processing_janus.py", + "left_ranges": [ + [ + 24, + 30 + ] + ], + "right_ranges": [ + [ + 24, + 30 + ] + ] + }, + { + "filename": "tests/models/kosmos2_5/test_processor_kosmos2_5.py", + "left_ranges": [ + [ + 44, + 50 + ], + [ + 53, + 69 + ] + ], + "right_ranges": [ + [ + 44, + 68 + ] + ] + }, + { + "filename": "tests/models/lighton_ocr/test_processor_lighton_ocr.py", + "left_ranges": [ + [ + 37, + 43 + ] + ], + "right_ranges": [ + [ + 37, + 43 + ] + ] + }, + { + "filename": "tests/models/llava/test_processing_llava.py", + "left_ranges": [ + [ + 23, + 29 + ], + [ + 34, + 41 + ], + [ + 62, + 80 + ] + ], + "right_ranges": [ + [ + 23, + 29 + ], + [ + 34, + 41 + ], + [ + 62, + 80 + ] + ] + }, + { + "filename": "tests/models/llava_next_video/test_processing_llava_next_video.py", + "left_ranges": [ + [ + 32, + 38 + ] + ], + "right_ranges": [ + [ + 32, + 38 + ] + ] + }, + { + "filename": "tests/models/minicpmv4_6/test_processing_minicpmv4_6.py", + "left_ranges": [ + [ + 35, + 47 + ] + ], + "right_ranges": [ + [ + 35, + 49 + ] + ] + }, + { + "filename": "tests/models/mistral3/test_processing_mistral3.py", + "left_ranges": [ + [ + 32, + 38 + ], + [ + 55, + 61 + ], + [ + 83, + 89 + ], + [ + 147, + 153 + ], + [ + 200, + 206 + ], + [ + 261, + 267 + ], + [ + 280, + 286 + ] + ], + "right_ranges": [ + [ + 32, + 38 + ], + [ + 55, + 61 + ], + [ + 83, + 89 + ], + [ + 147, + 153 + ], + [ + 200, + 206 + ], + [ + 280, + 286 + ] + ] + }, + { + "filename": "tests/models/mllama/test_processing_mllama.py", + "left_ranges": [ + [ + 32, + 38 + ], + [ + 100, + 106 + ], + [ + 229, + 246 + ], + [ + 254, + 260 + ], + [ + 272, + 279 + ], + [ + 286, + 292 + ] + ], + "right_ranges": [ + [ + 32, + 38 + ], + [ + 100, + 106 + ], + [ + 229, + 248 + ], + [ + 256, + 262 + ], + [ + 274, + 281 + ], + [ + 288, + 294 + ] + ] + }, + { + "filename": "tests/models/owlv2/test_processing_owlv2.py", + "left_ranges": [ + [ + 9, + 13 + ] + ], + "right_ranges": [ + [ + 9, + 13 + ] + ] + }, + { + "filename": "tests/models/parakeet/test_processing_parakeet.py", + "left_ranges": [ + [ + 25, + 29 + ] + ], + "right_ranges": [ + [ + 25, + 29 + ] + ] + }, + { + "filename": "tests/models/pixtral/test_processing_pixtral.py", + "left_ranges": [ + [ + 30, + 36 + ], + [ + 79, + 85 + ], + [ + 143, + 149 + ], + [ + 196, + 202 + ] + ], + "right_ranges": [ + [ + 30, + 36 + ], + [ + 79, + 85 + ], + [ + 143, + 149 + ], + [ + 196, + 202 + ] + ] + }, + { + "filename": "tests/models/qianfan_ocr/test_processing_qianfan_ocr.py", + "left_ranges": [ + [ + 33, + 39 + ] + ], + "right_ranges": [ + [ + 33, + 39 + ] + ] + }, + { + "filename": "tests/models/qwen2_5_omni/test_processing_qwen2_5_omni.py", + "left_ranges": [ + [ + 44, + 50 + ] + ], + "right_ranges": [ + [ + 44, + 50 + ] + ] + }, + { + "filename": "tests/models/qwen2_5_vl/test_processing_qwen2_5_vl.py", + "left_ranges": [ + [ + 35, + 41 + ] + ], + "right_ranges": [ + [ + 35, + 41 + ] + ] + }, + { + "filename": "tests/models/qwen2_audio/test_processing_qwen2_audio.py", + "left_ranges": [ + [ + 23, + 43 + ], + [ + 73, + 79 + ] + ], + "right_ranges": [ + [ + 13, + 19 + ], + [ + 23, + 43 + ], + [ + 73, + 79 + ] + ] + }, + { + "filename": "tests/models/qwen2_vl/test_processing_qwen2_vl.py", + "left_ranges": [ + [ + 37, + 49 + ] + ], + "right_ranges": [ + [ + 37, + 49 + ] + ] + }, + { + "filename": "tests/models/qwen3_omni_moe/test_processing_qwen3_omni_moe.py", + "left_ranges": [ + [ + 45, + 51 + ] + ], + "right_ranges": [ + [ + 45, + 51 + ] + ] + }, + { + "filename": "tests/models/qwen3_vl/test_processing_qwen3_vl.py", + "left_ranges": [ + [ + 34, + 45 + ] + ], + "right_ranges": [ + [ + 34, + 45 + ] + ] + }, + { + "filename": "tests/models/seamless_m4t/test_processing_seamless_m4t.py", + "left_ranges": [ + [ + 29, + 35 + ] + ], + "right_ranges": [ + [ + 29, + 35 + ] + ] + }, + { + "filename": "tests/models/smolvlm/test_processing_smolvlm.py", + "left_ranges": [ + [ + 28, + 34 + ] + ], + "right_ranges": [ + [ + 28, + 34 + ], + [ + 77, + 86 + ] + ] + }, + { + "filename": "tests/models/trocr/test_processing_trocr.py", + "left_ranges": [ + [ + 3, + 8 + ], + [ + 15, + 26 + ] + ], + "right_ranges": [ + [ + 3, + 8 + ], + [ + 15, + 26 + ] + ] + }, + { + "filename": "tests/models/video_llama_3/test_processing_video_llama_3.py", + "left_ranges": [ + [ + 42, + 48 + ] + ], + "right_ranges": [ + [ + 42, + 48 + ] + ] + }, + { + "filename": "tests/test_processing_common.py", + "left_ranges": [ + [ + 97, + 109 + ], + [ + 142, + 180 + ], + [ + 186, + 204 + ], + [ + 379, + 404 + ], + [ + 414, + 427 + ] + ], + "right_ranges": [ + [ + 151, + 169 + ] + ] + } + ] + } + ] + } + ], + "duplicate_issues": [ + { + "cluster_id": "cluster-44779-7", + "canonical_issue_number": 45488, + "duplicate_issue_numbers": [ + 44779, + 45812, + 45920, + 46489 + ], + "reason": "Issues in cluster-44779-7 are treated as duplicates because they share closing_reference, soft_similarity evidence." + } + ], + "duplicate_prs": [ + { + "cluster_id": "cluster-41211-3", + "canonical_pr_number": 44339, + "duplicate_pr_numbers": [ + 41356 + ], + "target_issue_number": 41211, + "reason": "PRs in cluster-41211-3 are treated as duplicates because they converge on issue #41211 with closing_reference evidence." + }, + { + "cluster_id": "cluster-43240-3", + "canonical_pr_number": 43251, + "duplicate_pr_numbers": [ + 43254 + ], + "target_issue_number": 43240, + "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43656-2", + "canonical_pr_number": 43656, + "duplicate_pr_numbers": [ + 43842 + ], + "target_issue_number": null, + "reason": "PRs in cluster-43656-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-43698-3", + "canonical_pr_number": 43779, + "duplicate_pr_numbers": [ + 43816 + ], + "target_issue_number": 43698, + "reason": "PRs in cluster-43698-3 are treated as duplicates because they converge on issue #43698 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43979-11", + "canonical_pr_number": 44007, + "duplicate_pr_numbers": [ + 43996, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "target_issue_number": 43979, + "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-44018-2", + "canonical_pr_number": 44068, + "duplicate_pr_numbers": [ + 44018 + ], + "target_issue_number": null, + "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-44779-7", + "canonical_pr_number": 47017, + "duplicate_pr_numbers": [ + 46580 + ], + "target_issue_number": 45488, + "reason": "PRs in cluster-44779-7 are treated as duplicates because they converge on issue #45488 with closing_reference, 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-46196-4", + "canonical_pr_number": 46196, + "duplicate_pr_numbers": [ + 46658, + 47003, + 47096 + ], + "target_issue_number": null, + "reason": "PRs in cluster-46196-4 are treated as duplicates because they share 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-47013-3", + "canonical_pr_number": 47026, + "duplicate_pr_numbers": [ + 47046 + ], + "target_issue_number": 47013, + "reason": "PRs in cluster-47013-3 are treated as duplicates because they converge on issue #47013 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-47104-2", + "canonical_pr_number": 47115, + "duplicate_pr_numbers": [ + 47104 + ], + "target_issue_number": null, + "reason": "PRs in cluster-47104-2 are treated as duplicates because they share soft_similarity evidence." + } + ], + "best_issue": { + "cluster_id": "cluster-44779-7", + "issue_number": 45488, + "reason": "Issue #45488 is the strongest global issue candidate because it is open, belongs to a cluster with 7 artifacts, and carries 16 discussion comments plus 4 inbound references.", + "score": 206.0 + }, + "best_pr": { + "cluster_id": "cluster-43979-11", + "pr_number": 44007, + "reason": "PR #44007 is the strongest global PR candidate because it is open, belongs to a cluster with 11 artifacts, links to 1 issues, and carries 2 review/discussion events.", + "score": 238.886 + } +} diff --git a/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..f8eb7d18bfcc9b95febb2561c078b3e798c8dbc0 --- /dev/null +++ b/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,26575 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260707T060715Z", + "generated_at": "2026-07-07T06:19:22Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 1, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "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:43716|issue:46869", + "issue:24643|issue:30277", + "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:46468", + "issue:45507|issue:45685", + "issue:44292|issue:44918", + "issue:43716|issue:45685", + "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": "026a428721709cf69668b57ab28acbb5a76fc55fcd8398c28cee16ca1572e265", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 63, + "serialized_chars": 46565, + "estimated_input_tokens": 11642, + "estimated_eval_tokens": 23540 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster has three accepted soft-edge duplicate pairs, but they are separate families rather than one unified bug: the ZeRO-3/BERT zero-sized tensor IndexError (43596/43638), the Qwen3_5MoeTextRotaryEmbedding offload/materialization bug (45859/45902), and the torch 2.4.1 compatibility regression (45800/46515). Most other candidate links are correctly rejected because they only share a broad subsystem, model family, or version context, not the same concrete code path.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 43596 is the cleanest representative of the strongest duplicate family: the DeepSpeed ZeRO-3/BERT zero-sized tensor IndexError. It is broader and better-scoped than the narrower non-pretrained-BERT variant.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43596 is the best overall issue anchor because it corresponds to the clearest accepted duplicate pair and states the core failure mode succinctly.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different failures: missing config attribute vs streaming chat/completions API error. Same model family is not enough." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Different hardware/backend paths: AMD GPU failure vs MPS histogram Int limitation." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Both are model/device issues, but the concrete bugs differ: dtype mismatch in preprocessing vs AMD GPU runtime failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Different code paths: autocast dtype mismatch vs AMD GPU-specific failure." + }, + { + "left": "issue:45237", + "right": "issue:46468", + "accept": false, + "reason": "One is GPT-OSS on AMD GPUs, the other is a broken official example; no shared concrete bug." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different model classes and symptoms; too broad to be the same bug." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "Both involve runtime/model errors, but one is a collective mismatch and the other is ONNX export." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Different model/export failure vs FP8 configuration failure." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export and Qwen3 auto device map cudaErrorAssert are unrelated failures." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Both are documentation/example problems, but they affect different examples and different docs pages." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export bug is not the same as the GenerationMixin import regression." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are v5 docs cleanup items, but they concern different removed docs topics and not one identical change." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "Training/trainer embedding unpacking is unrelated to resuming local model training with trust remote code." + }, + { + "left": "issue:46165", + "right": "issue:46869", + "accept": false, + "reason": "Resume/trust-remote-code flow is unrelated to multi-GPU device-map failure." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask shape mismatch and Gemma4 PLE device mismatch are different forward-path bugs." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Both are shape/index errors, but one is a zero-sized tensor under ZeRO-3 and the other is a mask/index mismatch." + }, + { + "left": "issue:43638", + "right": "issue:46165", + "accept": false, + "reason": "DeepSpeed/BERT initialization bug is unrelated to resuming training with remote code." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "The zero-sized tensor IndexError is specifically tied to ZeRO-3/BERT init, not a generic mask indexing failure." + }, + { + "left": "issue:44805", + "right": "issue:46294", + "accept": false, + "reason": "Mask/index shape mismatch is unrelated to Qwen3.5 video token/template mismatch." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Different failures: tensor mask mismatch vs GPT-OSS MoE only-TP problem." + }, + { + "left": "issue:46165", + "right": "issue:46468", + "accept": false, + "reason": "Model resume/trust-remote-code issue is not the same as a broken official example." + }, + { + "left": "issue:44560", + "right": "issue:46165", + "accept": false, + "reason": "Video chat-template failure is unrelated to local resume/trust-remote-code training." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice preset handling and ModernBERT ONNX export are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Both are distributed-training errors, but the concrete mechanisms differ: collective mismatch vs zero-sized BERT init." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same underlying ZeRO-3/BERT zero-sized tensor IndexError; the second is just a narrower description of the same bug." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross-entropy routing and GPT-OSS MoE TP-only failure are different code paths." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Different problems: detokenization regression vs LoRA weight merge/save/reload mismatch." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Different model/runtime issue vs LoRA merge/reload output mismatch." + }, + { + "left": "issue:44841", + "right": "issue:46661", + "accept": false, + "reason": "Processor failure for Voxtral and remote-code model_class config_class error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:46869", + "accept": false, + "reason": "Different device-mismatch-style failures on different models and code paths." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Both are DeepSpeed-related, but one is a weight-dimension training error and the other is a collective mismatch on Jamba-v01." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation checkpointing/FSDP recompute mismatch is not the same as a mask shape indexing error." + }, + { + "left": "issue:45685", + "right": "issue:46869", + "accept": false, + "reason": "MPS histogram Int error is unrelated to YOLO multi-GPU device-map failure." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": true, + "reason": "Same underlying Qwen3_5MoeTextRotaryEmbedding/offload-materialization bug around inv_freq handling and CPU/offload semantics." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "FSDP checkpointing tensor-size mismatch is unrelated to CustomTrainer compute_loss signature breakage." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 BERT init IndexError and GPT-OSS TP-only bug are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "Same rationale: different model families and failure modes." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Different bug classes: dtype mismatch vs invalid method call." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime failure and GPT-OSS on AMD GPUs are different model/backend failures." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Autocast dtype mismatch and GPT-OSS MoE TP-only failure are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Mixtral torch dynamo tracing failure is unrelated to Mistral image preprocessor dtype mismatch." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Different model families and failure mechanisms; only both are runtime errors." + }, + { + "left": "issue:43716", + "right": "issue:46468", + "accept": false, + "reason": "Mistral image preprocessor dtype mismatch and broken official Nemotron example are not the same bug." + }, + { + "left": "issue:45507", + "right": "issue:45685", + "accept": false, + "reason": "Invalid method call in GraniteMoEHybrid and MPS histogram Int error are different code paths." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime failure and Qwen3.5 embedding unpacking with TRL are different issues." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Mistral dtype mismatch and MPS Int histogram error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Different model-specific runtime failures; no shared code path." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT unpacking failure is unrelated to GPT-OSS MoE TP-only failure." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy routing and TRL embedding unpacking are different training-stack issues." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading IndexError and BERT ZeRO-3 loading IndexError are similar only at a high level, but not clearly the same model/code path." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs issues, but one is about removed pipeline tasks and the other about the removed transformers run command." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Autocast dtype mismatch and MPS histogram Int error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Different code paths: Liger Kernel cross-entropy vs autocast dtype mismatch." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and CustomTrainer signature breakage are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation checkpointing/FSDP mismatch and GenerationMixin import regression are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "GPT-OSS on AMD GPUs and RAG docs are not the same kind of issue." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross-entropy dispatch and mask shape mismatch are unrelated." + }, + { + "left": "issue:45800", + "right": "issue:46515", + "accept": true, + "reason": "Same torch 2.4.1 compatibility regression; one title is the specific transformers 5.8.0 pairing and the other is the generic version incompatibility." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Device-map forward mismatch and expensive reverse embedding lookup are different issues." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "FSDP activation checkpointing mismatch is unrelated to Qwen3 auto device-map cudaErrorAssert." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Torch-version incompatibility and Qwen-3 NVFP4 runtime failure are different problems." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Autocast dtype mismatch and GPT-OSS MoE TP-only failure are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:43716", + "accept": false, + "reason": "Mixtral torch dynamo tracing failure is unrelated to Mistral image preprocessor dtype mismatch." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Different model families and failure mechanisms; only both are runtime errors." + }, + { + "left": "issue:43716", + "right": "issue:46468", + "accept": false, + "reason": "Mistral dtype mismatch and broken official Nemotron example are not the same bug." + }, + { + "left": "issue:45507", + "right": "issue:45685", + "accept": false, + "reason": "Invalid method call in GraniteMoEHybrid and MPS histogram Int error are different code paths." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime failure and Qwen3.5 embedding unpacking with TRL are different issues." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Mistral dtype mismatch and MPS Int histogram error are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44292", + "accept": false, + "reason": "Different model-specific runtime failures; no shared code path." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT unpacking failure is unrelated to GPT-OSS MoE TP-only failure." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy routing and TRL embedding unpacking are different training-stack issues." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading IndexError and BERT ZeRO-3 loading IndexError are similar only at a high level, but not clearly the same model/code path." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs issues, but one is about removed pipeline tasks and the other about the removed transformers run command." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Autocast dtype mismatch and MPS histogram Int error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Different code paths: Liger Kernel cross-entropy vs autocast dtype mismatch." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and CustomTrainer signature breakage are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation checkpointing/FSDP mismatch and GenerationMixin import regression are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "GPT-OSS on AMD GPUs and RAG docs are not the same kind of issue." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross-entropy dispatch and mask shape mismatch are unrelated." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The verdicts are mostly conservative, but the summary is not fully grounded in the packet: it says the ZeRO-3/BERT pair is the only clear duplicate and the rest are too broad, yet the analyst also accepted issue:45859 vs issue:45902 as a likely duplicate. That omission makes the aggregate summary imprecise. The accepted ZeRO-3/BERT pair and the torch 2.4.1 pair are well supported, but the summary should mention all accepted soft-edge pairs." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 2, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:36010", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:43039", + "issue:43296", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43572", + "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" + ], + "soft_pairs": [ + "issue:45800|issue:46304", + "issue:43296|issue:46304", + "issue:43866|issue:46402", + "issue:41762|issue:43596", + "issue:39290|issue:41720", + "issue:43901|issue:44512", + "issue:44589|issue:46304", + "issue:45237|issue:45901", + "issue:43039|issue:44560", + "issue:43975|issue:46015", + "issue:45823|issue:46294", + "issue:44841|issue:45823", + "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: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:43404|issue:43866", + "issue:45507|issue:46869", + "issue:41093|issue:41762", + "issue:45399|issue:46661", + "issue:43572|issue:45070", + "issue:45800|issue:45823" + ], + "prepared_review_unit_hash": "75041e5e4c1bac764fc899a19738544d9a1428d220cae03c156d446c5a0fa482", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 40040, + "estimated_input_tokens": 10010, + "estimated_eval_tokens": 20276 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous cluster. Keep only the AutoTokenizer v5 regression pair as a true duplicate; reject the ZeRO-3/BertModel IndexError pair as too generic and not sufficiently grounded. The rest are separate model-, version-, or docs-specific reports.", + "confidence": 0.68, + "canonical_issue_reason": "Issue 45812 is the clearest confirmed duplicate anchor: it states the broad AutoTokenizer v5 regression across multiple models, and 45920 is the same bug with an expanded model list.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45020 is the best global umbrella because it captures the broad recent-version / remote_code regression theme across otherwise heterogeneous reports, though it should not absorb unrelated runtime failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45800", + "right": "issue:46304", + "accept": false, + "reason": "Both are compatibility regressions, but they concern different breakages and different code paths." + }, + { + "left": "issue:43296", + "right": "issue:46304", + "accept": false, + "reason": "PaddleOCR-VL loading in vLLM is unrelated to USE_DISTRIBUTED=0 support breaking in pytorch." + }, + { + "left": "issue:43866", + "right": "issue:46402", + "accept": false, + "reason": "Different model-loading failures with different symptoms and no evidence of the same root cause." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Both mention ZeRO-3 and IndexError, but the models and failure contexts differ enough that they do not look like the same bug." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Gemma3 config attribute missing in vLLM and Qwen3 auto device-map CUDA assert are unrelated failures." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs-related, but they cover different documentation mismatches and should not be merged." + }, + { + "left": "issue:44589", + "right": "issue:46304", + "accept": false, + "reason": "Float8 storage lookup failure is a different runtime issue from the USE_DISTRIBUTED=0 compatibility break." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "AMD GPU model failure and table-question-answering crash are separate problems." + }, + { + "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:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization bug and outdated RAG docs are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46294", + "accept": false, + "reason": "Gemma4 device-mismatch with auto mapping and Qwen3.5 video chat-template mismatch are different model/input issues." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral and Gemma4 PLE device mismatch are not the same bug." + }, + { + "left": "issue:43866", + "right": "issue:46772", + "accept": false, + "reason": "Corrupted Ovis2 checkpoint and DiffusionGemma ModelOpt compatibility are unrelated." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading failure and RAG docs maintenance are unrelated." + }, + { + "left": "issue:45020", + "right": "issue:46304", + "accept": false, + "reason": "One is a broad remote_code regression; the other is a specific PyTorch distributed build incompatibility." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Both report the same AutoTokenizer v5 regression producing wrong token IDs across multiple models." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Ovis2 checkpoint corruption and NemotronH load failure are distinct model issues." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime errors and RAG docs are unrelated." + }, + { + "left": "issue:43872", + "right": "issue:46304", + "accept": false, + "reason": "bitsandbytes Int8Params signature break and USE_DISTRIBUTED=0 support are different incompatibilities." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "They both mention sliding-window-related terms, but affect different models and integration points." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "This is only a weak ZeRO-3/BertModel/IndexError similarity; the evidence does not support calling them the same underlying bug." + }, + { + "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": "GLM-4.7 unit-test loading issues and Ovis2 checkpoint corruption are different artifacts." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch 2.10 incompatibility and PretrainedConfig pydantic breakage are separate regressions." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization bug and table-question-answering crash are unrelated." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper load failure and pydantic PretrainedConfig breakage are different bugs." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Qwen3.5 chat crash and streaming chat/completions failure are separate issues." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "ZeRO-3 BertModel index error and torch/transformers version incompatibility do not share the same root cause." + }, + { + "left": "issue:43541", + "right": "issue:46869", + "accept": false, + "reason": "Mixtral grouped_mm tracing and YOLO multi-GPU device-map failure are unrelated." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Qwen NVFP4 runtime error and table-question-answering crash are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3 sliding_window_pattern missing and Gemma4 device mismatch are different model-specific failures." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3.5 streaming inference failure and MPS histogram_mps Int error are separate code paths." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "GenerationMixin import relocation and Gemma3 config attribute missing are different API breakages." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Mixtral grouped_mm tracing and MPS histogram_mps errors are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Grouped_mm tracing failure and autocast dtype mismatch are different bugs." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Voxtral processor failure and non-template-node compile error are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Mistral3 weight-tying bug and Ovis2 checkpoint corruption are different issues." + }, + { + "left": "issue:45507", + "right": "issue:46869", + "accept": false, + "reason": "GraniteMoEHybrid invalid method call and YOLO device-map auto failure are not the same bug." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are ZeRO-3 IndexErrors, but the error shapes and affected models differ enough that they do not look like the same underlying bug." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The writeup is mostly careful, but one soft-edge verdict is too strong for the packet: `issue:41093` vs `issue:43638` is only a weak ZeRO-3/BertModel/IndexError similarity, and the evidence shown does not support calling them the same underlying bug. Because the instruction is to stay conservative, that accepted duplicate looks overstated. The AutoTokenizer pair is well grounded, but the overall result should be rejected on precision grounds." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 3, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:39186", + "issue:40990", + "issue:41093", + "issue:42915", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43531", + "issue:43541", + "issue:43596", + "issue:43645", + "issue:43716", + "issue:43828", + "issue:43873", + "issue:43994", + "issue:44291", + "issue:44360", + "issue:44387", + "issue:44479", + "issue:44492", + "issue:44512", + "issue:44530", + "issue:44805", + "issue:44863", + "issue:44918", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45161", + "issue:45183", + "issue:45399", + "issue:45464", + "issue:45507", + "issue:45632", + "issue:45698", + "issue:45759", + "issue:45800", + "issue:45902", + "issue:46165", + "issue:46294", + "issue:46463", + "issue:46468", + "issue:46640", + "issue:46661", + "issue:46869" + ], + "soft_pairs": [ + "issue:43296|issue:43366", + "issue:44805|issue:44918", + "issue:43541|issue:45161", + "issue:43828|issue:45464", + "issue:43366|issue:45464", + "issue:44387|issue:45005", + "issue:44291|issue:45070", + "issue:43531|issue:45070", + "issue:41093|issue:43596", + "issue:45020|issue:45800", + "issue:43716|issue:46294", + "issue:43054|issue:43994", + "issue:44479|issue:46294", + "issue:46468|issue:46869", + "issue:45759|issue:45902", + "issue:43596|issue:46165", + "issue:46640|issue:46661", + "issue:43716|issue:44918", + "issue:39186|issue:44918", + "issue:43645|issue:45399", + "issue:40990|issue:44960", + "issue:45632|issue:45698", + "issue:42915|issue:45399", + "issue:43404|issue:45005", + "issue:43873|issue:45005", + "issue:44360|issue:44512", + "issue:44492|issue:44512", + "issue:44530|issue:45005", + "issue:44918|issue:45507", + "issue:44863|issue:45005", + "issue:45183|issue:46463" + ], + "prepared_review_unit_hash": "8dd474041cd1a0f0d54b97735e2b6fc962a1a09c8109095bb38df9f3f9077bbe", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36241, + "estimated_input_tokens": 9061, + "estimated_eval_tokens": 18378 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are mostly separate bugs across different model families and failure modes. The soft matches share broad themes (remote code, MoE, validation, dtype/masking), but none are close enough to treat as the same underlying issue. Best representative issue is the concrete remote-code loading bug in #45698.", + "confidence": 0.83, + "canonical_issue_reason": "#45698 is the most concrete, reproducible, and broadly relevant issue here: it describes a specific wrong-module loading bug in `from_pretrained` after `save_pretrained`, with a clear code path and likely useful anchor for nearby remote-code bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "#45698 is the strongest standalone issue in the set because it is precise, actionable, and describes a specific loader regression rather than a vague symptom or model-specific anomaly.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "Both involve model loading for GPT-OSS/Nemotron-like architectures, but one is a PaddleOCR-VL load failure in transformers/vLLM and the other is GGUF architecture support for gpt-oss. Different bug surfaces and fix paths." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Both mention an IndexError shape mismatch, but the first is a generic mask/tensor length mismatch while the second is Qwen3.5 embedding unpacking with TRL SFT trainer. Too different in trigger and code path." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "One is a torch dynamo tracing failure in Mixtral float32 grouped_mm, the other is TP-only behavior for GPT-OSS MoE. Same broad MoE area, but not the same concrete bug." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "These are different model families and error contexts: Phi-tiny-MoE dtype mismatch under autocast vs Qwen3.5 streaming chat/completions failure. Not the same underlying issue." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GPT-OSS GGUF architecture support and Qwen3.5 streaming inference are unrelated failures despite both involving recent model support." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Both concern v5 quantization/tied-weights regressions, but one is CUDA reserved memory OOM under int4 and the other is tied weights on translation models. Different symptom and remedy." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "Init-empty-weights TypeError from an unexpected `_is_hf_initialized` argument is a distinct initialization bug from the pydantic `PretrainedConfig` field regression." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and a pydantic model field regression are unrelated problems." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are IndexErrors, but one is a mask-vs-tensor shape mismatch during processing and the other is deepspeed zero3 init with BertModel and an empty-dimension access. Different causes." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Both are version/regression complaints, but one is about remote_code models breaking and the other is torch 2.4.1 vs transformers 5.8.0 incompatibility. Too broad to merge." + }, + { + "left": "issue:43716", + "right": "issue:46294", + "accept": false, + "reason": "Image preprocessor dtype mismatch in Mistral-3 is unrelated to Qwen3.5 video token/feature mismatch in chat template rendering." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Both are SIGLIP2-related quality complaints, but one is about text embeddings being worse and the other is nonsensical AutoModel/pipeline outputs. Not the same failure mode." + }, + { + "left": "issue:44479", + "right": "issue:46294", + "accept": false, + "reason": "Video input regression in Qwen vl models and a video feature/token mismatch in Qwen3.5-9B are related by domain but not clearly the same bug or fix." + }, + { + "left": "issue:46468", + "right": "issue:46869", + "accept": false, + "reason": "Official Nemotron example broken vs YOLO multi-GPU device_map auto are completely different issues and subsystems." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "One is `from_config` not unwrapping `text_config` for composite Qwen multimodal configs; the other is uninitialized memory in Qwen3_5MoeTextRotaryEmbedding after materialization. Related model family, different defect." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "Both touch model loading/training with remote/local models, but one is a zero3 init IndexError in BertModel and the other is resume-training with locally downloaded remote-code models. Not the same code path." + }, + { + "left": "issue:46640", + "right": "issue:46661", + "accept": false, + "reason": "Flash Attention 2 integration problem and an AutoModel remote-code `config_class` AttributeError are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Mistral-3 image preprocessing dtype mismatch is unrelated to Qwen3.5 input embedding unpacking with TRL." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP `'weight' must be 2-D` is a distributed training parameter-shape issue, not the same as Qwen3.5 embedding unpacking failure." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Notebook custom-model initialization in v5 and flash-attn2 fallback blocking are different regressions with different remediation." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "High perplexity on GPT-OSS-20B and the generic GLM5 issue are not the same underlying bug; one is a model-quality/regression report and the other is an umbrella model issue." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both involve remote-code/local custom module loading, but one is a cache-path collision for local models sharing a leaf name and the other is loading the wrong custom module after save/load. Similar area, different root cause." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "FineGrainedFP8Config failure for Qwen3Moe and flash-attn2 fallback gating are unrelated config/integration issues." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Untied lm_head in Mistral3 conditional generation is a model-weight tying bug, while the v5 translation-model issue is broader and different in scope." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization offloading behavior and translation-model tied weights are different features and failure modes; too broad to combine." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "A DSA indexer ReLU discussion and a docs typo about the removed run command are unrelated." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "Cache strategies typo and removed docs run command are both documentation/config text issues, but not the same issue." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash on Qwen3.5 models is a runtime cache/kernel incompatibility, not the same as tied-weight regressions." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Qwen3.5 input embedding unpacking with TRL SFT and GraniteMoEHybrid calling an invalid method are different model-specific bugs." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading failure and tied-weights regressions are separate model-integration problems." + }, + { + "left": "issue:45183", + "right": "issue:46463", + "accept": false, + "reason": "These both concern input validation errors, but one is a broad XOR logic bug across models and the other is T5 decoder-input error messaging. Similar pattern, not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 4, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "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:43782", + "issue:43824", + "issue:43828", + "issue:43856", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44292", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44910", + "issue:44928", + "issue:44960", + "issue:44991", + "issue:45020", + "issue:45399", + "issue:45520", + "issue:45663", + "issue:45701", + "issue:45859", + "issue:45902", + "issue:45920", + "issue:45987", + "issue:46489", + "issue:46491", + "issue:46858", + "issue:46899", + "issue:46932" + ], + "soft_pairs": [ + "issue:43975|issue:44292", + "issue:46489|issue:46491", + "issue:39692|issue:43054", + "issue:43950|issue:44534", + "issue:36331|issue:41762", + "issue:43644|issue:43950", + "issue:43872|issue:44589", + "issue:46899|issue:46932", + "issue:43828|issue:44928", + "issue:43054|issue:45701", + "issue:45920|issue:46489", + "issue:38175|issue:43054", + "issue:45987|issue:46858", + "issue:43950|issue:45902", + "issue:43054|issue:43493", + "issue:44910|issue:45859", + "issue:42947|issue:44387", + "issue:45663|issue:45902", + "issue:44403|issue:44451", + "issue:41553|issue:43883", + "issue:43975|issue:45520", + "issue:24643|issue:28530", + "issue:44403|issue:44488", + "issue:43643|issue:45020", + "issue:42947|issue:43856", + "issue:41762|issue:42915", + "issue:44991|issue:45399", + "issue:43824|issue:43872", + "issue:41553|issue:43881", + "issue:43782|issue:44960" + ], + "prepared_review_unit_hash": "f851bfae60cd62c7c3f50b44a534b2210b5e9091df7b1ccb7fc0fa5a5db1cae7", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36287, + "estimated_input_tokens": 9072, + "estimated_eval_tokens": 18400 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issue families overall, with one tight duplicate subcluster around the Transformers v5 non-persistent-buffer regression (#43950, plus later symptom reports like #44534 and #45902). The #43643 pairing is not the same bug and should be rejected.", + "confidence": 0.91, + "canonical_issue_reason": "#43950 is the best anchor: it explicitly describes the v5 non-persistent-buffer corruption regression and is broad enough to cover later symptom reports like #44534 and #45902.", + "canonical_pr_reason": null, + "best_issue_reason": "#43950 is the most representative issue overall because it names the concrete regression, identifies the affected code path, and matches the later buffer-corruption reports in the packet.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Different bugs: tokenizer detokenization regression vs Qwen NVFP4 load/runtime error." + }, + { + "left": "issue:46489", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and symptoms; not clearly the same underlying bug." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Docs/example mismatch and quantization failure vs a general text-embedding quality regression are not the same bug." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same non-persistent-buffer regression in Transformers v5; #44534 reads like a later symptom of the same bug." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Different areas and failure modes: trainer API change vs DeepSpeed ZeRO-3 model loading." + }, + { + "left": "issue:43643", + "right": "issue:43950", + "accept": false, + "reason": "#43643 is about trust_remote_code=True returning missing fields from AutoConfig.from_pretrained, while #43950 is a Transformers 5.x non-persistent-buffer corruption regression; only superficial similarity." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Unrelated incompatibilities: bitsandbytes parameter-constructor change vs missing Float8 storage." + }, + { + "left": "issue:46899", + "right": "issue:46932", + "accept": true, + "reason": "Same Gemma4 quantized vision/audio embedder dtype-casting bug; #46932 explicitly says it is the same issue." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Different model/code paths and symptoms: autocast dtype mismatch vs Qwen3.5 position-id/SDPA stability issues." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "Both mention tokenization/embeddings, but one is a quality regression and the other is a version-dependent tokenization change." + }, + { + "left": "issue:45920", + "right": "issue:46489", + "accept": true, + "reason": "Same tokenizer-output regression in v5; #46489 is a model-specific manifestation of the broader wrong-token-IDs issue." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Different underlying problems: zero probabilities on SigLIP2 vs a text-embedding quality complaint." + }, + { + "left": "issue:45987", + "right": "issue:46858", + "accept": true, + "reason": "Same cache sequence-length contract bug; #46858 is the backend-specific manifestation of the same return-type problem." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": true, + "reason": "Same non-persistent-buffer initialization/corruption regression; #45902 is a specific buffer instance exposed by the same bug." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Same model family, but different issues: quality regression vs HF/JAX implementation discrepancy." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Different failures: flash-attention packed-sequence misinterpretation vs CPU offload incompatibility in rotary embedding." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Both are memory-related, but one is LoRA checkpointing behavior and the other is int4 quantization reserve-memory growth." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Different code paths: Gemma4 FSDP2 per-layer keying issue vs non-persistent buffer materialization corruption." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic loading noise vs a specific model load failure; no shared underlying bug is evident." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Different model-loading bugs: bad AutoTokenizer error message for Voxtral vs missing all_tied_weights_keys on Molmo." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Different subsystems: tokenizer detokenization vs import_utils flash_attn mapping on Python 3.13." + }, + { + "left": "issue:24643", + "right": "issue:28530", + "accept": false, + "reason": "Unrelated training/runtime error vs early-stopping metric selection issue." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Both are loading-related, but the reported models and failures differ too much to be the same bug." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "#45020 is a broad remote_code regression report; #43643 is a specific AutoConfig field-loss symptom, not the same concrete bug." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Different memory problems: ineffective gradient checkpointing vs MoE training memory inefficiency." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Distinct model/config errors with no shared failing code path." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading for a specific model vs flash-attn2 fallback being blocked; unrelated issues." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Missing model import symbol vs bitsandbytes parameter-compatibility error." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Different load failures affecting different models and APIs." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Different problems: weight_only load failure for Qwen3VL vs a generic GLM5 issue without the same concrete failure mode." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Mostly grounded, but one soft-edge acceptance is too strong: issue:43643 vs issue:43950 are not supported as the same bug by the packet. 43643 is about `trust_remote_code=True` returning missing fields from `AutoConfig.from_pretrained`, while 43950 is a Transformers 5.x non-persistent-buffer corruption regression. The similarity is only superficial/soft, so that verdict should be conservative and rejected." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 5, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43404", + "issue:43792", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43872", + "issue:43881", + "issue:43883", + "issue:43940", + "issue:43950", + "issue:44220", + "issue:44291", + "issue:44368", + "issue:44479", + "issue:44509", + "issue:44560", + "issue:44661", + "issue:44805", + "issue:44863", + "issue:44928", + "issue:44991", + "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:45685|issue:45823", + "issue:45081|issue:45399", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:43828|issue:45823", + "issue:42491|issue:44368", + "issue:45020|issue:45356", + "issue:43881|issue:43883", + "issue:45715|issue:46315", + "issue:43872|issue:46869", + "issue:45759|issue:46661", + "issue:43872|issue:44291", + "issue:43827|issue:44509", + "issue:45488|issue:46489", + "issue:43404|issue:43950", + "issue:45020|issue:46661", + "issue:46894|issue:46897", + "issue:43792|issue:44220", + "issue:43257|issue:45305", + "issue:42947|issue:45923", + "issue:36331|issue:41093", + "issue:43940|issue:45313", + "issue:43854|issue:44863", + "issue:44928|issue:45588", + "issue:24643|issue:30064", + "issue:44479|issue:44560", + "issue:41093|issue:44805", + "issue:41553|issue:44991" + ], + "prepared_review_unit_hash": "c6eceecac739ace0137042fc0d0dc187d2517d12f5097022ac547e898abc5498", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36754, + "estimated_input_tokens": 9189, + "estimated_eval_tokens": 18634 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is still heterogeneous overall. A few pairs are tight duplicates worth keeping together\u2014especially the `_is_hf_initialized` / `init_empty_weights` regression, the v5 docs cleanup, and the DeepSeek tokenizer/pre-tokenizer bug\u2014but the Qwen/DeepSpeed and video-related pairs are only same-family or same-subsystem similarities, not proven single-root-cause duplicates.", + "confidence": 0.72, + "canonical_issue_reason": "No single issue cleanly represents the whole cluster. If an anchor is needed, issue 43872/44291 is the strongest exact duplicate pair because it names a precise shared compatibility failure with a clear common root cause.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43872/44291 is the best representative issue pair in the cluster: it is a concrete, well-supported duplicate with a narrowly defined failure mode, unlike the broader same-family regressions elsewhere in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "Different bugs: an MPS `histogram_mps` int error vs a Gemma4 device-map mismatch." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Different code paths: a Mistral regex patch crash vs flash-attn fallback gating." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Broad remote-code breakage is not the same concrete bug as the Gemma4 device mismatch." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "Tokenizer-mapping failure is a separate loader bug from the generic remote-code regression." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Different runtime failures: autocast dtype mismatch vs device-placement mismatch." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "Related Qwen/LoRA context, but one is a load-time incompatibility and the other is a warning about `tie_word_embeddings`." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "The Kimi tokenizer regression is more specific than the broad remote-code breakage." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Different model-loading failures for different architectures." + }, + { + "left": "issue:45715", + "right": "issue:46315", + "accept": false, + "reason": "Both are tokenizer slowdowns, but they affect different conversion paths; not the same bug." + }, + { + "left": "issue:43872", + "right": "issue:46869", + "accept": false, + "reason": "Bitsandbytes constructor incompatibility is unrelated to the YOLO multi-GPU device-map failure." + }, + { + "left": "issue:45759", + "right": "issue:46661", + "accept": false, + "reason": "Different failures in model construction/loading for different APIs and root causes." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Same `_is_hf_initialized` / `init_empty_weights` compatibility failure causing the constructor TypeError." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Same docs regression: both still reference removed v5 pipeline tasks." + }, + { + "left": "issue:45488", + "right": "issue:46489", + "accept": true, + "reason": "Same tokenizer/pre-tokenizer regression breaking DeepSeek-family tokenization on v5+." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Tied `lm_head` weights and non-persistent buffer corruption are different regression classes." + }, + { + "left": "issue:45020", + "right": "issue:46661", + "accept": false, + "reason": "Remote-code model breakage and missing `config_class` are different loader problems." + }, + { + "left": "issue:46894", + "right": "issue:46897", + "accept": false, + "reason": "Same high-level loss-shift pattern, but the packet does not establish one shared code-path bug." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": true, + "reason": "The Whisper failure is plausibly the concrete symptom of the `_torch_extract_fbank_features()` 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." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Gradient-checkpointing ineffective under PEFT LoRA is not the same as a missing `supports_gradient_checkpointing` flag." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Unexpected `num_items_in_batch` kwarg vs mask/tensor shape mismatch are unrelated trainer/runtime errors." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Same model family and loader subsystem, but the packet only shows separate ZeRO-3 weight-loading reports; a shared root cause is not established." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different model checkpoint-loading failures for different model families." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "Different runtime failures: RLHF NaNs from position IDs vs flash-attention crash on `s_aux=None`." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "Training DeepSpeed weight shape error is unrelated to void segmentation-map preprocessing." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Same general video area, but different model families and symptoms; the packet does not prove one shared bug." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "They share a shape-mismatch theme, but the actual failing paths and symptoms differ." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Bad Voxtral tokenizer messaging is unrelated to the EMBEDDIA tokenizer load failure." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster-level conclusion is mostly grounded, but a few soft-edge accepts are a bit too strong for the evidence in the packet. The exact duplicate pairs 43872/44291, 43827/44509, and 45488/46489 look well supported. However, 43940/45313 and 44479/44560 are only shown to be same-model-family, same-general-area regressions; the packet does not establish a shared root cause strongly enough to call them clear duplicate/mergeable bugfix pairs. The summary also slightly overstates certainty by calling those subgroups \"clear duplicate subgroups.\"" + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 6, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:41553", + "issue:42491", + "issue:42617", + "issue:42915", + "issue:43366", + "issue:43531", + "issue:43550", + "issue:43596", + "issue:43632", + "issue:43638", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43720", + "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:45030", + "issue:45325", + "issue:45584", + "issue:45663", + "issue:45672", + "issue:45685", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45902", + "issue:46402", + "issue:46489", + "issue:46515", + "issue:46858", + "issue:46899" + ], + "soft_pairs": [ + "issue:44530|issue:44863", + "issue:44863|issue:46402", + "issue:43366|issue:43828", + "issue:43366|issue:43531", + "issue:43596|issue:46515", + "issue:45325|issue:45902", + "issue:41553|issue:44488", + "issue:43366|issue:44589", + "issue:43638|issue:46515", + "issue:44155|issue:45584", + "issue:45663|issue:45823", + "issue:42915|issue:43957", + "issue:43824|issue:43957", + "issue:43716|issue:44560", + "issue:43950|issue:45834", + "issue:43645|issue:43646", + "issue:42617|issue:43366", + "issue:44910|issue:46858", + "issue:44560|issue:45685", + "issue:43550|issue:44910", + "issue:36010|issue:42915", + "issue:44683|issue:46515", + "issue:44560|issue:44918", + "issue:45812|issue:46489", + "issue:45823|issue:46515", + "issue:44661|issue:45823", + "issue:38175|issue:43994", + "issue:44560|issue:44805", + "issue:44877|issue:45030", + "issue:45672|issue:46899", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:42491|issue:43720", + "issue:43632|issue:45834", + "issue:44661|issue:44805" + ], + "prepared_review_unit_hash": "ba8b4303a7e9318b4364536a4153b463f0ce5438a232f33872d0d1bd83e94898", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38355, + "estimated_input_tokens": 9589, + "estimated_eval_tokens": 19434 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are clearly unrelated bug reports. I only see a few likely duplicate/near-duplicate edges: the custom-model initialization regression, the SigLIP2 wrong-output regression, and the Gemma4 4-bit multimodal dtype-cast regression.", + "confidence": 0.58, + "canonical_issue_reason": "issue:43646 is the broadest, most canonical phrasing of the custom-model-initialization regression; issue:43645 is a narrower Jupyter-notebook manifestation of the same problem.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best representative issue in this set because it describes the core regression without extra environment-specific constraints.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "Different models and failures: Qwen3.5 PagedAttentionCache group parsing vs NemotronH checkpoint loading." + }, + { + "left": "issue:44863", + "right": "issue:46402", + "accept": false, + "reason": "Both are loading bugs, but they affect different model families and different loading paths." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "gpt-oss GGUF support is unrelated to Phi-tiny-MoE autocast dtype mismatch." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "GGUF architecture support and Qwen3-MoE sliding_window behavior are different issues." + }, + { + "left": "issue:43596", + "right": "issue:46515", + "accept": false, + "reason": "A Bert/deepspeed zero3 index error is not the same as a broad torch-version incompatibility report." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Different Qwen variants and different bugs: rope index scaling vs uninitialized rotary embedding memory after materialization." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Both are model-loading complaints, but they concern different models and different root causes." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF support for gpt-oss is unrelated to Float8 storage lookup failures." + }, + { + "left": "issue:43638", + "right": "issue:46515", + "accept": false, + "reason": "Specific deepspeed/Bert zero3 failure is not the same as a general torch 2.4.1 incompatibility." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "AudioFlamingo3 track leakage and Whisper empty-transcription handling are different code paths." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but different failures: FSDP2 shared_kv_states rebuild vs device_map forward mismatch." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Qwen3Moe FP8 loading and meta-device model loading are different regression classes." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Missing Qwen2_5_VL symbol import is unrelated to meta-device initialization issues." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Different multimodal models and different failures: dtype mismatch vs StopIteration in video handling." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption during from_pretrained is unrelated to Kosmos2.5 long-OCR index errors." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Both describe the same transformers v5 custom-model initialization regression; 43645 is the notebook-specific variant." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "3d_parallel.py runtime failure and gpt-oss GGUF support are not the same underlying bug." + }, + { + "left": "issue:44910", + "right": "issue:46858", + "accept": false, + "reason": "Both involve caching/position arithmetic, but one is a Qwen3.5 Flash Attention crash and the other is a static-cache torch.compile backend issue." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3-vl video StopIteration and MPS histogram support are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Bamba SDPA torch.compile failure and Qwen3.5 Flash Attention position_ids crash are different concrete bugs." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "Importing GenerationMixin and Qwen3Moe FineGrainedFP8Config are unrelated failures." + }, + { + "left": "issue:44683", + "right": "issue:46515", + "accept": false, + "reason": "Torch>=2.9 flex_attention compilation breakage is not the same as a torch 2.4.1 incompatibility report." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Video embedding StopIteration and TRL SFT unpacking failure are different Qwen3.5-related bugs." + }, + { + "left": "issue:45812", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes." + }, + { + "left": "issue:45823", + "right": "issue:46515", + "accept": false, + "reason": "Gemma4 device placement mismatch is unrelated to a general torch-version incompatibility." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping-name validation and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Both report the same siglip2-base-patch16-224 wrong-output regression, just phrased as zero probabilities vs nonsensical results." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different failures: video StopIteration vs mask/tensor shape mismatch." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both mention config loading, but they concern different models and different validation failures." + }, + { + "left": "issue:45672", + "right": "issue:46899", + "accept": true, + "reason": "These look like the same Gemma4 4-bit multimodal dtype-cast bug; 46899 is the broader restatement of 45672's failure." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "Bert zero3 indexing and add-new-model-like tokenizer mapping validation are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Deepspeed zero3/Bert failure is not the same as tokenizer mapping-name validation logic." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "A qwen3_moe LoRA/hf4.x compatibility break is unrelated to BitNet packed-weight loading." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "The _is_hf_initialized regression and Kosmos2.5 OCR index error are different issues." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer mapping validation and mask shape mismatch are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the soft-edge decisions are conservative. The three accepted pairs are plausibly the same underlying regressions based on their titles, while the rejected pairs are generally supported as different bug classes or different model/path combinations. The canonical/best-issue reasoning for 43646 vs 43645 is also consistent with the issue titles." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 7, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:38175", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43644", + "issue:43645", + "issue:43708", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43994", + "issue:44075", + "issue:44164", + "issue:44291", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44756", + "issue:44863", + "issue:44910", + "issue:44960", + "issue:44977", + "issue:44991", + "issue:45276", + "issue:45292", + "issue:45313", + "issue:45335", + "issue:45357", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45507", + "issue:45588", + "issue:45812", + "issue:46050", + "issue:46459", + "issue:46468", + "issue:46512", + "issue:46846", + "issue:46897", + "issue:46901" + ], + "soft_pairs": [ + "issue:43317|issue:44756", + "issue:38175|issue:39692", + "issue:43644|issue:44534", + "issue:41720|issue:42915", + "issue:42915|issue:44756", + "issue:42491|issue:45276", + "issue:44960|issue:44977", + "issue:44451|issue:44488", + "issue:44863|issue:46846", + "issue:46459|issue:46512", + "issue:45313|issue:45357", + "issue:45507|issue:46468", + "issue:43644|issue:43645", + "issue:39692|issue:43994", + "issue:43257|issue:43854", + "issue:44488|issue:44991", + "issue:43856|issue:43873", + "issue:42491|issue:45292", + "issue:43854|issue:43931", + "issue:43856|issue:44387", + "issue:42491|issue:45335", + "issue:43708|issue:44075", + "issue:44960|issue:46050", + "issue:44164|issue:44291", + "issue:46897|issue:46901", + "issue:45276|issue:45335", + "issue:45488|issue:45812", + "issue:45362|issue:45406", + "issue:44910|issue:45588", + "issue:44402|issue:44488" + ], + "prepared_review_unit_hash": "765a726ebad0f46243760839ed41e0cc2c7139bb32a6693fe407d1f1cb326a94", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36053, + "estimated_input_tokens": 9014, + "estimated_eval_tokens": 18284 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous, but it contains a few true duplicate pairs: the v5 non-persistent-buffer corruption reports, the resize_token_embeddings propagation bug, and the training-loss double-shift bug. Most other pairs are only thematically similar and should stay separate.", + "confidence": 0.83, + "canonical_issue_reason": "issue:43644 is the clearest broad regression report in the set and has an obvious near-duplicate (issue:44534); it best represents the non-persistent-buffer corruption bug.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43644 is the most representative standalone issue here: concise, regression-focused, and tied to the clearest duplicate pair in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Different problems: device_map/offload loading failure vs mmap/OOM on Strix Halo." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "SigLIP2-related, but one is wrong probabilities and the other is a broken doc example; different root causes." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Same v5 regression about non-persistent buffers being filled with junk; titles are paraphrases of the same bug." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Both touch Qwen3, but one is auto device mapping CUDA assert and the other is FineGrainedFP8Config; not the same bug." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated issues: Qwen3 MoE FP8 failure vs hardware-specific mmap OOM." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "Different code paths and symptoms: LoRA model incompatibility across versions vs token embedding resize propagation." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different models and failures; both mention generation/flash-attention, but not the same underlying bug." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are loading failures for different BERT checkpoints, but the reports do not clearly point to the same root cause." + }, + { + "left": "issue:44863", + "right": "issue:46846", + "accept": false, + "reason": "Different model families and distinct failure modes: checkpoint loading vs TP plan missing entries." + }, + { + "left": "issue:46459", + "right": "issue:46512", + "accept": false, + "reason": "Both involve dtype handling, but one is checkpoint loading and the other is from_config initialization; not the same bug." + }, + { + "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 save_pretrained key handling." + }, + { + "left": "issue:45507", + "right": "issue:46468", + "accept": false, + "reason": "Different models and issues; no concrete shared bug path." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Both are v5 regressions, but buffer corruption and Jupyter custom-model initialization are separate problems." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Related SigLIP2 area, but doc-example mistakes vs bad model outputs are not the same bug." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different load/dispatch problems affecting different models and setups." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer/model loading regressions, but they affect different checkpoints and do not clearly share one root cause." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Different memory issues: MoE training inefficiency vs quantization offloading behavior." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Unrelated bugs: LoRA model compatibility vs resize_token_embeddings not updating output embeddings." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Different model-specific loading failures with different symptoms and likely causes." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both are memory-related, but one is training inefficiency and the other is int4 reserved-memory/OOM regression." + }, + { + "left": "issue:42491", + "right": "issue:45335", + "accept": false, + "reason": "Different models and separate resize/load issues; not the same code-path bug." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Training resumption max_steps math is unrelated to optimizer argument handling." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "Different quantization/generation issues with no clear shared root cause." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Both concern loading/saving internals, but extra_state handling and _is_hf_initialized are separate bugs." + }, + { + "left": "issue:46897", + "right": "issue:46901", + "accept": true, + "reason": "Same training-loss double-shift bug: both report labels being shifted incorrectly instead of training against the intended targets." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": true, + "reason": "Same resize_token_embeddings propagation bug; output embeddings/decoder embeddings are not updated across model variants." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different failure modes." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different issues: chat generation crash for Qwen3.5-35B vs Gemma4Processor missing _tokenizer in serve." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both are flash-attention crashes, but one is packed-sequence misinterpretation and the other is s_aux=None handling." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Token-vocab mismatch vs model loading failure; similar area, but not the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the accepted soft-edge pairs are supported by very close title-level overlap. The rejections are generally conservative and do not overstate duplicate evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 8, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:42947", + "issue:43335", + "issue:43381", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43645", + "issue:43854", + "issue:43873", + "issue:43957", + "issue:44079", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44589", + "issue:44756", + "issue:44863", + "issue:44877", + "issue:44912", + "issue:44928", + "issue:44960", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45859", + "issue:46326", + "issue:46468", + "issue:46829" + ], + "soft_pairs": [ + "issue:43425|issue:43957", + "issue:39692|issue:44960", + "issue:41720|issue:44155", + "issue:44291|issue:44756", + "issue:45663|issue:45759", + "issue:42947|issue:45704", + "issue:43404|issue:43454", + "issue:43873|issue:44387", + "issue:44079|issue:46829", + "issue:45663|issue:45859", + "issue:44292|issue:44912", + "issue:36010|issue:39404", + "issue:42947|issue:43381", + "issue:44155|issue:45823", + "issue:43957|issue:44756", + "issue:44387|issue:44756", + "issue:43335|issue:45663", + "issue:44863|issue:46468", + "issue:45276|issue:45292", + "issue:44479|issue:44877", + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:42796|issue:46326", + "issue:45588|issue:45663", + "issue:45684|issue:45698", + "issue:43854|issue:45216", + "issue:45292|issue:45335", + "issue:43645|issue:44164", + "issue:42175|issue:43957", + "issue:44589|issue:44756", + "issue:44402|issue:44451" + ], + "prepared_review_unit_hash": "1870a70a5a9ad5d2fc06a758b76aa172d80f194261f7d79df5d3eb6861e9ec52", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36443, + "estimated_input_tokens": 9111, + "estimated_eval_tokens": 18478 + }, + "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. The only clear duplicate-like pair is the resize_token_embeddings regression, where the Gemma4-specific report appears to be a narrower manifestation of the broader output_embeddings issue.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45292 is the broadest, model-agnostic report of resize_token_embeddings not updating output_embeddings, and it cleanly subsumes the Gemma4-specific variant in issue:45276.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the strongest global issue candidate because it describes a reusable API-level regression with broader impact than the model-specific duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Both are loading/runtime failures, but one is a Torch 2.10 incompatibility and the other is a meta-device model-loading regression; different bugs." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "SigLIP2 doc/example errors and a GLM5 issue are unrelated model-specific problems." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Qwen3 auto device-map CUDA asserts and AudioFlamingo3 batched inference token leaks are different failure modes." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized and Strix Halo mmap OOM are unrelated loading bugs." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Gemma4 FSDP shared_kv_states and Qwen multimodal text_config unwrapping are different code paths." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing with LoRA and a T5 apex RMSNorm memory leak are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both mention lm_head tying, but they affect different model implementations and don\u2019t look like the same concrete bug." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Both involve quantization/OOM, but one is offloading behavior and the other is increased CUDA reserved memory under int4; too broad to merge." + }, + { + "left": "issue:44079", + "right": "issue:46829", + "accept": false, + "reason": "Generic ModelOutput key handling is not the same bug as VideoPrism returning the wrong tensor in hidden_states." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different model families and different runtime paths: FSDP shared_kv_states vs CPU offload rotary embedding." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "init_empty_weights argument handling and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "GenerationMixin import failure and Whisper return_language pipeline regression are distinct issues." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is ineffective LoRA behavior and the other is eval-mode prohibition." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo3 batch leakage and Gemma4 device mismatch are different bugs." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading regressions and Strix Halo mmap OOM are not the same change." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Both can lead to OOM, but the causes are different: int4 reserved memory growth vs mmap on Strix Halo." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction and Gemma4 FSDP attention state rebuilding are unrelated." + }, + { + "left": "issue:44863", + "right": "issue:46468", + "accept": false, + "reason": "NemotronH checkpoint loading and a broken official Nemotron example are too different to treat as duplicates." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same underlying resize_token_embeddings bug: the Gemma4 report is a model-specific instance of the broader output_embeddings failure." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Video input regression for Qwen VL models and strict granite_speech config loading are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Eval-mode gradient checkpointing and Qwen3.5 NaN/SDPA fallback are different problems." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to the Strix Halo mmap OOM issue." + }, + { + "left": "issue:42796", + "right": "issue:46326", + "accept": false, + "reason": "Both concern tokenizers, but one is BPE-marker handling and the other is continue_final_message behavior; not the same bug." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Flash-attention sink-less crash and Gemma4 FSDP KeyError are unrelated." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Custom-model save permissions and wrong-module loading after save_pretrained are different issues in the same area, not the same bug." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "GLM-4.7-Flash loading in tests and Qwen3.5 save_pretrained regression are unrelated." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": false, + "reason": "Very similar theme, but the titles suggest different affected embeddings and no evidence they\u2019re the same exact regression beyond resize_token_embeddings." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Custom model initialization in Jupyter and extra_state save/load handling are different failure modes." + }, + { + "left": "issue:42175", + "right": "issue:43957", + "accept": false, + "reason": "Packaging/backend dependency omission and meta-device loading bugs are unrelated." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Float8 storage lookup failure and Strix Halo mmap OOM are not the same code-path problem." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer vocab mismatch and a model load failure are too general to call duplicates." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The only accepted duplicate pair, issue:45276 vs issue:45292, is supported by the shared resize_token_embeddings/output_embeddings regression, and the rest of the soft-edge rejections avoid overclaiming duplicate status." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 9, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:36032", + "issue:39404", + "issue:43299", + "issue:43425", + "issue:43452", + "issue:43596", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43856", + "issue:43901", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "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:45479", + "issue:45701", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46050", + "issue:46315", + "issue:46326", + "issue:46491", + "issue:46665", + "issue:46821", + "issue:46869", + "issue:46894", + "issue:46899", + "issue:46901" + ], + "soft_pairs": [ + "issue:43646|issue:43708", + "issue:46894|issue:46901", + "issue:43425|issue:43828", + "issue:46315|issue:46491", + "issue:45812|issue:46315", + "issue:39404|issue:45701", + "issue:44530|issue:45084", + "issue:43611|issue:43645", + "issue:43425|issue:46869", + "issue:44683|issue:45823", + "issue:39404|issue:46665", + "issue:43425|issue:43716", + "issue:43425|issue:45823", + "issue:44589|issue:45084", + "issue:44479|issue:44805", + "issue:43901|issue:44509", + "issue:43994|issue:44079", + "issue:43906|issue:44530", + "issue:44265|issue:44898", + "issue:44164|issue:45759", + "issue:36032|issue:43452", + "issue:46050|issue:46326", + "issue:43906|issue:46326", + "issue:39404|issue:44912", + "issue:44671|issue:44991", + "issue:44991|issue:45081", + "issue:45823|issue:46899", + "issue:45479|issue:46894", + "issue:43856|issue:44756", + "issue:44164|issue:46821", + "issue:43299|issue:43931", + "issue:43596|issue:44683" + ], + "prepared_review_unit_hash": "f8d75a346dffa3b207fa8b5928735c84c7615e13b4eb583eb5437b18311eed71", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37544, + "estimated_input_tokens": 9386, + "estimated_eval_tokens": 19028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated regression reports across tokenizer loading, multimodal inference, training, quantization, and docs. I don't see a single duplicate cluster or any soft edge that is strong enough to merge as the same underlying bug/change.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Different failure modes: custom model initialization in notebooks vs Trainer checkpoint-resume max_steps calculation." + }, + { + "left": "issue:46894", + "right": "issue:46901", + "accept": false, + "reason": "Same broad label of 'double-shift', but they affect different models and there is no evidence of one concrete shared code path." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version incompatibility and autocast dtype mismatch are distinct problems with different triggers." + }, + { + "left": "issue:46315", + "right": "issue:46491", + "accept": false, + "reason": "One is a performance regression in token ID lookup; the other is a CodeLlama encode/decode spacing bug." + }, + { + "left": "issue:45812", + "right": "issue:46315", + "accept": false, + "reason": "Wrong Granite token IDs is a correctness regression, not the same issue as slower convert_tokens_to_ids." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper return_language pipeline behavior and generic tokenization changes are not the same bug." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3.5 cache group-type crash and template-node compilation error are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both are v5 loading problems, but one is base_model_prefix handling and the other is custom model initialization in notebooks." + }, + { + "left": "issue:43425", + "right": "issue:46869", + "accept": false, + "reason": "Generic torch incompatibility is not the same as the multi-GPU device_map auto failure." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 PLE device mismatch are different code paths." + }, + { + "left": "issue:39404", + "right": "issue:46665", + "accept": false, + "reason": "Whisper pipeline return_language and from_pretrained inside Python classes do not describe the same defect." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch 2.10 incompatibility is unrelated to the Mistral-3 image preprocessor/model dtype mismatch." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage lookup and non-template-node compilation are separate errors." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "One is a Qwen VL video-input regression; the other is a mask/tensor shape IndexError." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs issues, but they cover different pipeline/task deprecations and should not be merged." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Wrong SigLIP2 outputs and ModelOutput key assignment are different bugs." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "A reproduction of issue #38071 is not the same as the Qwen3.5 linear_attention cache crash." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export with torch_compilable_check and Perceiver non-default-resolution failure are unrelated." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "save/from_pretrained extra_state handling and AutoModelForCausalLM.from_config text_config unwrapping are different loading bugs." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5Tokenizer add_special_tokens conflict and gguf_file loading failure are distinct tokenizer/load paths." + }, + { + "left": "issue:46050", + "right": "issue:46326", + "accept": false, + "reason": "Quantized save/load breakage and mistral_common continue_final_message behavior are unrelated." + }, + { + "left": "issue:43906", + "right": "issue:46326", + "accept": false, + "reason": "A reproduction of an earlier issue does not match the mistral_common conversation bug." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper return_language and MXFP4 quantization fallback are different subsystems." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "Both involve tokenizer/model regressions, but CamemBERT masked LM predictions and est-roberta tokenizer loading are different concrete failures." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer loading for est-roberta and Mistral regex patch crashes are not the same issue." + }, + { + "left": "issue:45823", + "right": "issue:46899", + "accept": false, + "reason": "Gemma4 device placement mismatch and Gemma4 4-bit multimodal dtype-cast breakage are different bugs." + }, + { + "left": "issue:45479", + "right": "issue:46894", + "accept": false, + "reason": "Sequence-classification zero-loss behavior and ASR training-loss double-shift are different label-handling issues." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "MoE training memory inefficiency and Strix Halo mmap OOM avoidance are unrelated." + }, + { + "left": "issue:44164", + "right": "issue:46821", + "accept": false, + "reason": "save/from_pretrained extra_state handling and TimesFM window_size AttributeError are different problems." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both are Qwen3-VL loading reports, but one is a dev0 breakage and the other is a weight-shape mismatch; not enough to treat as one bug." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Deepspeed ZeRO3/BertModel index error and compiled flex_attention torch>=2.9 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": 26, + "review_unit_index": 10, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42796", + "issue:42915", + "issue:43317", + "issue:43404", + "issue:43421", + "issue:43452", + "issue:43482", + "issue:43541", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:44206", + "issue:44220", + "issue:44451", + "issue:44462", + "issue:44488", + "issue:44530", + "issue:44661", + "issue:44671", + "issue:44683", + "issue:44737", + "issue:44811", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45479", + "issue:45507", + "issue:45584", + "issue:45859", + "issue:46178", + "issue:46285", + "issue:46326", + "issue:46402", + "issue:46489", + "issue:46897", + "issue:47050" + ], + "soft_pairs": [ + "issue:43632|issue:47050", + "issue:45479|issue:46897", + "issue:44488|issue:44987", + "issue:44488|issue:45081", + "issue:43611|issue:43644", + "issue:36032|issue:42915", + "issue:42796|issue:43404", + "issue:44462|issue:46489", + "issue:46285|issue:46402", + "issue:43716|issue:45084", + "issue:39404|issue:44671", + "issue:44661|issue:44683", + "issue:43643|issue:44530", + "issue:42915|issue:44530", + "issue:44451|issue:44987", + "issue:44737|issue:45859", + "issue:43541|issue:45084", + "issue:40990|issue:43421", + "issue:34567|issue:40444", + "issue:43644|issue:43646", + "issue:42175|issue:46178", + "issue:40444|issue:46326", + "issue:43582|issue:43632", + "issue:36032|issue:43645", + "issue:44206|issue:44220", + "issue:44811|issue:45584", + "issue:43452|issue:43482", + "issue:44987|issue:44991", + "issue:45084|issue:45507", + "issue:43317|issue:43482" + ], + "prepared_review_unit_hash": "0815f0a18d996f82438b34561c5c12dc7cfec27c3b2377b920d413f57889b2c8", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36746, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated Transformers regression reports across tokenizer loading, model initialization, backend setup, device/offload, and export paths. Only the audio feature-extraction pair looks like the same underlying bug; the rest are too different to merge as duplicates.", + "confidence": 0.64, + "canonical_issue_reason": "Issue 44206 is the best canonical anchor: it describes a concrete, localized regression in the audio feature-extraction path and is the only pair with a plausible same-codepath duplicate (44220).", + "canonical_pr_reason": null, + "best_issue_reason": "44206 is the clearest and most actionable report in the set, with a specific failure mode and code path; the other issues are broader or clearly about different subsystems.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43632", + "right": "issue:47050", + "accept": false, + "reason": "Both are regressions, but one is about hf-initialization state and the other is an fp16 ONNX attention dtype mismatch; different bugs and code paths." + }, + { + "left": "issue:45479", + "right": "issue:46897", + "accept": false, + "reason": "Both involve training loss behavior, but the symptoms and model families differ; not enough evidence they share the same fix." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are version-related loading failures, but they affect different models and likely different tokenizer/config paths." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "One is about ignoring tokenizer.json, the other is a Mistral regex patch crash; distinct failure modes." + }, + { + "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 state; not the same bug." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "Tokenizer method-name conflict vs FineGrainedFP8Config failure; unrelated code paths." + }, + { + "left": "issue:42796", + "right": "issue:43404", + "accept": false, + "reason": "Tokenizer BPE-marker handling and lm_head weight tying are separate issues in different model code." + }, + { + "left": "issue:44462", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions, but one is repo tokenizer resolution and the other is DeepSeek-Coder tokenization behavior; different root causes." + }, + { + "left": "issue:46285", + "right": "issue:46402", + "accept": false, + "reason": "CLIP structure inconsistency and Sam3 text-encoder weight loading are unrelated model-architecture bugs." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Image preprocessor dtype mismatch vs template compilation error; no shared code path." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline return_language regression and CamemBERT MLM prediction error are unrelated." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer-mapping bookkeeping bug vs torch flex_attention compilation failure; different layers entirely." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "Missing fields from trust_remote_code config vs PagedAttentionCache invalid group type; unrelated behavior." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "FineGrainedFP8Config failure and Qwen3.5 PagedAttentionCache crash are different model/runtime issues." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both concern loading failures, but one is a specific model load regression and the other is tokenizer loading on a different model." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both are device/offload related, but the models and failure mechanisms differ (CPU arange placement vs rotary embedding offload compatibility)." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo grouped_mm tracing failure is unrelated to the template-node compilation error." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "High perplexity on a benchmark and runtime special-token post-processor updates are not the same bug." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "Trainer token counter not updating vs multimodal iterable-dataset finetuning failure; different subsystems." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 breakages around model initialization, but one is buffer persistence/state handling and the other is custom-model init in notebooks; not the same concrete issue." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "pip extra/backend packaging omission vs backend-mapping validation error; related theme but distinct bugs." + }, + { + "left": "issue:40444", + "right": "issue:46326", + "accept": false, + "reason": "IterableDataset multi-image finetuning failure and mistral_common continue_final_message behavior are unrelated." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError and hf-initialization flag regression are separate problems." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer add_special_tokens conflict vs custom model initialization in Jupyter; different failure modes." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": true, + "reason": "Both point to the same audio feature-extraction path; the titles suggest a shared `_torch_extract_fbank_features` regression, likely triggered by the unsupported `center` argument." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "GGUF loading failures and v5 tokenizer-loading issues are separate model-format/code-path problems." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are version-gated loading regressions, but they affect different models and likely different tokenizer/config logic." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Template compilation failure vs GraniteMoEHybrid invalid method call are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Dequantized model offload loading failure and Qwen2.5-GGUF v5 loading failure are distinct." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet and stays conservative. The only accepted soft edge, 44206 \u2194 44220, is plausibly the same audio feature-extraction regression based on the titles and shared codepath. The rest of the pair verdicts are appropriately cautious and do not overclaim duplication across different subsystems." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 11, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28530", + "issue:34567", + "issue:36032", + "issue:41720", + "issue:42915", + "issue:43089", + "issue:43208", + "issue:43262", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43531", + "issue:43540", + "issue:43632", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43746", + "issue:43792", + "issue:43819", + "issue:43883", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44849", + "issue:44877", + "issue:44987", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45987", + "issue:46326", + "issue:46661", + "issue:46821", + "issue:46869", + "issue:46899", + "issue:47013" + ], + "soft_pairs": [ + "issue:43746|issue:46326", + "issue:43262|issue:43377", + "issue:43208|issue:47013", + "issue:44403|issue:45081", + "issue:43792|issue:46869", + "issue:43646|issue:44534", + "issue:42915|issue:45084", + "issue:43645|issue:46661", + "issue:43377|issue:43819", + "issue:44849|issue:45325", + "issue:43883|issue:44488", + "issue:44987|issue:45081", + "issue:43404|issue:44466", + "issue:28530|issue:34567", + "issue:43994|issue:44610", + "issue:42915|issue:44291", + "issue:36032|issue:43742", + "issue:44451|issue:45081", + "issue:44464|issue:45987", + "issue:44062|issue:44987", + "issue:44008|issue:46899", + "issue:42915|issue:44589", + "issue:44265|issue:44610", + "issue:43322|issue:43540", + "issue:43632|issue:46821", + "issue:43531|issue:44877", + "issue:36032|issue:41720", + "issue:44479|issue:45325", + "issue:43089|issue:45987" + ], + "prepared_review_unit_hash": "f949ffcdf74203ba29d9a66ea152ac758735fce17b3f1b51596944ff43f0ed3c", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36613, + "estimated_input_tokens": 9154, + "estimated_eval_tokens": 18564 + }, + "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 subsystem keywords, but they do not appear to be the same concrete bug/change. No soft edge should be merged; the closest mini-cluster is the v5 custom-model initialization regressions, with issue:43646 as the best representative.", + "confidence": 0.72, + "canonical_issue_reason": "issue:43646 is the broadest representative of the only coherent mini-cluster here (Transformers v5 custom model initialization regressions), and it is more canonical than the narrower variants around `_is_hf_initialized` or `init_empty_weights`.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best overall issue representative in this set because it captures the broad v5 custom-model initialization breakage, while most other issues are model-specific or unrelated. It is still not a duplicate of the rest of the cluster, just the strongest anchor among the issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43746", + "right": "issue:46326", + "accept": false, + "reason": "PEFT local-checkpoint loading and `continue_final_message` for mistral_common are different code paths and different failure modes." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio processor sampling-rate default vs MIMI batched-vs-single padding-mask mismatch are unrelated bugs." + }, + { + "left": "issue:43208", + "right": "issue:47013", + "accept": false, + "reason": "Both are xLSTM-related, but one is a bundle of training bugs for small models and the other is a specific unpacking crash in train mode; not the same bug." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Loading-noise behavior and the Mistral regex patch crash are unrelated issues." + }, + { + "left": "issue:43792", + "right": "issue:46869", + "accept": false, + "reason": "Whisper inference failure and YOLO multi-GPU device-map failure are different models and different code paths." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Custom model initialization regressions and non-persistent buffer junk are different v5 problems, not one concrete bug." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and template-node compilation failure are unrelated." + }, + { + "left": "issue:43645", + "right": "issue:46661", + "accept": false, + "reason": "Both concern custom models, but one is notebook re-initialization and the other is remote-code `config_class` handling; different underlying failures." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "MIMI padding-mask batching bug and DAC `from_latents`/forward mismatch are different audio-model bugs." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Hidden-states output bug in Qwen3.5 and Qwen2.5-VL rope/temporal scaling bug are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Missing `all_tied_weights_keys` on Molmo and loading `cjvt/sleng-bert` are unrelated model-loading problems." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "General load failure for one model and Mistral tokenizer regex patch crash are different root causes." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Tied-weight absence in Mistral3 and device-dependent serialization inconsistency both mention `lm_head.weight`, but the failure modes are not the same concrete bug." + }, + { + "left": "issue:28530", + "right": "issue:34567", + "accept": false, + "reason": "Early-stopping metric selection and `num_input_tokens_seen` not updating are unrelated Trainer issues." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Siglip2 nonsense outputs and OmDet-Turbo processor size mismatch are different model/processor bugs." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "FP8 config failure for Qwen3Moe and unexpected `_is_hf_initialized` arg in `init_empty_weights` are unrelated." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "T5 tokenizer method-name conflict and MobileLLM key error are unrelated loading issues." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "ScandiBERT loading failure and Mistral regex patch crash are different models and different causes." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Compiled chunked generation inconsistency and `StaticCache.get_seq_length()` return-type bug are related to generation APIs but not the same code-path defect." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "`AddedToken` argument collision and loading `physical-intelligence/fast` are unrelated." + }, + { + "left": "issue:44008", + "right": "issue:46899", + "accept": false, + "reason": "Gemma 3n variable-name collision and Gemma4 quantization dtype-cast bug are different multimodal model issues." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and missing Float8 storage object are unrelated precision/storage issues." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "`torch.export` failing on `torch_compilable_check` and a processor/model input-size mismatch are unrelated." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Llava-Next segfault on load and Qwen3OmniMoe video-processing `ValueError` are different multimodal failures." + }, + { + "left": "issue:43632", + "right": "issue:46821", + "accept": false, + "reason": "`_is_hf_initialized` breakage and TimesFM `window_size` attribute error are unrelated regressions." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and granite_speech strict config loading are different bugs." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer special-token conflict and Qwen3 auto device-map CUDA assert are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Both touch Qwen2.5-VL/video-position logic, but one is a broad v5.3.0 video-input regression and the other is a specific still-image temporal-position scaling bug; not enough to treat as the same bug." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation GPU-sync/dispatch overhead and `StaticCache.get_seq_length()` shape bug 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": 26, + "review_unit_index": 12, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42796", + "issue:43296", + "issue:43317", + "issue:43381", + "issue:43421", + "issue:43425", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43611", + "issue:43632", + "issue:43716", + "issue:43746", + "issue:43792", + "issue:43856", + "issue:43883", + "issue:44079", + "issue:44220", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44683", + "issue:44756", + "issue:44811", + "issue:44991", + "issue:45081", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45812", + "issue:45923", + "issue:46315", + "issue:46491", + "issue:46710", + "issue:46821", + "issue:46899" + ], + "soft_pairs": [ + "issue:43425|issue:44403", + "issue:45216|issue:45357", + "issue:43296|issue:43531", + "issue:40444|issue:45325", + "issue:43883|issue:44991", + "issue:43611|issue:44534", + "issue:44991|issue:46315", + "issue:43716|issue:46899", + "issue:44220|issue:44683", + "issue:44403|issue:44991", + "issue:43381|issue:45923", + "issue:42796|issue:43746", + "issue:45812|issue:46491", + "issue:43632|issue:44756", + "issue:33666|issue:40444", + "issue:43792|issue:44589", + "issue:44291|issue:44589", + "issue:40990|issue:44811", + "issue:42796|issue:46710", + "issue:43482|issue:43525", + "issue:35707|issue:40444", + "issue:44079|issue:45663", + "issue:43531|issue:44291", + "issue:43421|issue:44568", + "issue:41720|issue:45081", + "issue:42175|issue:45542", + "issue:43482|issue:45538", + "issue:44079|issue:44534", + "issue:43296|issue:46821", + "issue:44462|issue:45812", + "issue:43317|issue:43856", + "issue:42175|issue:43296" + ], + "prepared_review_unit_hash": "d5f3a8ff6a1e0c2203870224ad0f7561ba32efbba8ed84deb6cdb60d93be379d", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37287, + "estimated_input_tokens": 9322, + "estimated_eval_tokens": 18900 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated transformer/v5, tokenizer, model-loading, and training regressions. I only see one plausible duplicate pair: the two Qwen3.5 save_pretrained issues. No PRs are present.", + "confidence": 0.84, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:45812 is the broadest representative of the tokenizer-regression subset and has clear impact across all Granite models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch-version incompatibility and generic loading noise are different failure modes." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Both describe the same Qwen3.5 save_pretrained checkpoint regression; the second is a narrower symptom of the same bug." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL loading failure and Qwen3-MoE sliding_window behavior are unrelated." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Multi-image IterableDataset finetuning failure is not the same as the Qwen2.5-VL rope_index scaling bug." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Missing all_tied_weights_keys and tokenizer loading failure are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "base_model_prefix loading breakage and non-persistent buffer junk are different code paths." + }, + { + "left": "issue:44991", + "right": "issue:46315", + "accept": false, + "reason": "Tokenizer load failure and slow convert_tokens_to_ids regression are distinct bugs." + }, + { + "left": "issue:43716", + "right": "issue:46899", + "accept": false, + "reason": "Both are multimodal dtype issues, but they involve different models and triggers, so they are not clearly the same bug." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction and compiled flex_attention on torch>=2.9 are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic loading noise is not the same as tokenizer repository-resolution failure." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Eval-mode gradient checkpointing and missing supports_gradient_checkpointing are different issues." + }, + { + "left": "issue:42796", + "right": "issue:43746", + "accept": false, + "reason": "Tokenizer BPE-marker handling and GraniteSpeech PEFT checkpoint loading are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but Granite token IDs and CodeLlama leading-space round-trip are different bugs." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "_is_hf_initialized breakage and mmap OOM on Strix Halo are unrelated." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Multi-GPU training and multi-image iterable finetuning are not the same failure." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper run failure and Float8 storage lookup error are different problems." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage object are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "GPT-OSS perplexity regression and Whisper batch_decode skip_special_tokens bug are unrelated." + }, + { + "left": "issue:42796", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer BPE-marker handling and DeepSeek accuracy regression are different issues." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Qwen2.5-GGUF loading under v5 and Llama4Config missing pad_token_id are unrelated." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is not the same as the multi-image finetuning failure." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment and Gemma4 shared_kv_states/FSDP2 are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and init_empty_weights argument handling are different bugs." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Runtime post-processor updates and add_special_tokens BOS/EOS omission are different tokenizer issues." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3 auto device mapping cuda assert and Mistral regex patch crash are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "These are both backend-packaging complaints, but they are not clearly the same underlying defect." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "GGUF loading failure and CLIPTokenizer model_max_length are unrelated." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment and non-persistent buffer initialization are different bugs." + }, + { + "left": "issue:43296", + "right": "issue:46821", + "accept": false, + "reason": "PaddleOCR-VL loading failure and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "Ignoring tokenizer.json and wrong Granite token IDs are both tokenizer regressions, but not the same code-path bug." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "device_map=auto offload failure and Qwen3 MoE memory usage are different issues." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "TensorFlow backend packaging and PaddleOCR-VL model loading are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, issue:45216 vs issue:45357, is plausibly the same Qwen3.5 save_pretrained regression with a narrower symptom. The other pairwise rejections are cautious and do not overclaim equivalence. No PR pairs are present." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 13, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42796", + "issue:42915", + "issue:43262", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43493", + "issue:43582", + "issue:43606", + "issue:43611", + "issue:43644", + "issue:43646", + "issue:43701", + "issue:43746", + "issue:43818", + "issue:43819", + "issue:43828", + "issue:43873", + "issue:43994", + "issue:44079", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44466", + "issue:44479", + "issue:44509", + "issue:44829", + "issue:44877", + "issue:45072", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45584", + "issue:45701", + "issue:45823", + "issue:45834", + "issue:46402", + "issue:46821", + "issue:46909" + ], + "soft_pairs": [ + "issue:44509|issue:46909", + "issue:43611|issue:43646", + "issue:43377|issue:45491", + "issue:43644|issue:44079", + "issue:43818|issue:46402", + "issue:40444|issue:45381", + "issue:43818|issue:45357", + "issue:44265|issue:45072", + "issue:43493|issue:43994", + "issue:45325|issue:45834", + "issue:45325|issue:45381", + "issue:44829|issue:45479", + "issue:43454|issue:44466", + "issue:44291|issue:44877", + "issue:42915|issue:43482", + "issue:43701|issue:45325", + "issue:43873|issue:45701", + "issue:42796|issue:45381", + "issue:41553|issue:43582", + "issue:40990|issue:45584", + "issue:43606|issue:44265", + "issue:44479|issue:45823", + "issue:43262|issue:43746", + "issue:43828|issue:46821", + "issue:35707|issue:41720", + "issue:39692|issue:40444", + "issue:34634|issue:35707", + "issue:41720|issue:45823", + "issue:44186|issue:44265", + "issue:43606|issue:43819" + ], + "prepared_review_unit_hash": "428bbfe2a7334cea04387570a789843afc9545df5ceda275cd06861f2ef960a1", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36916, + "estimated_input_tokens": 9229, + "estimated_eval_tokens": 18714 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are a heterogeneous mix of unrelated bugs and docs regressions across different models and subsystems. I do not see any soft pair that is clearly the same underlying change/bug, so all candidate duplicate edges should be rejected.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the broadest and most active open bug in the set (offloading/quantization), so it is the best single representative issue, but it is not a true duplicate anchor for the other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44509", + "right": "issue:46909", + "accept": false, + "reason": "Both are docs-related, but one is about removed pipeline tasks in v5 while the other is a broken DIA example; different documentation problems." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both mention Transformers 5.0.0 regressions, but one is about `base_model_prefix` loading and the other about custom model initialization; different failure modes/code paths." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Different models and bugs: missing padding mask support in MIMI versus NaNs from Gemma3 sliding-window attention on mixed-length batches." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "One is about non-persistent buffer serialization in v5, the other about `ModelOutput` key assignment when a field was previously `None`; unrelated code paths." + }, + { + "left": "issue:43818", + "right": "issue:46402", + "accept": false, + "reason": "Different models and subsystems: Video-LLaVA tower architecture issues versus Sam3 text encoder weights not loading." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both involve Qwen2.5-VL video/image handling, but one is finetuning with multi-image iterable datasets and the other is wrong video `vision_position_ids`; not the same concrete bug." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Video-LLaVA tower/temporal-attention problem is unrelated to Qwen3.5 `save_pretrained` saving wrong visual encoder keys." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "`torch.export.export` failing on `torch_compilable_check` is a different issue from bf16 dtype mismatches in inference." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": false, + "reason": "Same model family, but one is a JAX-vs-HF implementation discrepancy and the other is a bad pipeline/AutoModel result; insufficient evidence they are the same code-path bug." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both touch position/index handling, but they are different models and different inputs: Qwen2.5-VL still-image temporal positions versus Kosmos2.5 long OCR indexing." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Both concern Qwen2.5-VL video/position IDs, but one is still-image temporal scaling and the other is video `vision_position_ids`; related area, not the same bug." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "One is a training regression with `flash_attention_3`, the other is degenerate zero loss from `problem_type=\"single_label_classification\"` with `num_labels=1`; different causes." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both involve tied `lm_head` weights, but one is failing to tie weights at init and the other is inconsistent serialization depending on device; different problems." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "`init_empty_weights` rejecting `_is_hf_initialized` is a different compatibility issue from strict config blocking `granite_speech` config loading." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Qwen2.5-GGUF loading failure are unrelated." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint key mismatch during resume is unrelated to Qwen2.5-VL rope/position index scaling." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to tokenizer output changes across versions." + }, + { + "left": "issue:42796", + "right": "issue:45381", + "accept": false, + "reason": "Tokenizer BPE-marker handling and Qwen2.5-VL video position IDs are different bug classes." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Bad AutoTokenizer error reporting for Voxtral is unrelated to the Apple Silicon `caching_allocator_warmup` TypeError." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "Extremely high perplexity on one model and Whisper empty-transcription failure are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch and `torch.export` failure with `torch_compilable_check` are different runtime paths." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Different models and symptoms: video input regression versus Gemma4 PLE device mismatch." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting is unrelated to GraniteSpeech PEFT checkpoint loading." + }, + { + "left": "issue:43828", + "right": "issue:46821", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to TimesFM `window_size` AttributeError." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation with `inputs_embeds`/`past_key_values` and Qwen3 auto-device-map CUDA asserts are different issues." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 docs example mistakes and Qwen2.5-VL finetuning with multiple images are unrelated." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "Bark `voice_preset` not working and progressive generation with embeddings/cache are unrelated model APIs." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert and Gemma4 PLE device mismatch are different models and different failure modes." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes and `torch.export` failures are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "CPU offload device mismatch and DAC `from_latents` vs forward mismatch 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": 26, + "review_unit_index": 14, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:26413", + "issue:28530", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42796", + "issue:42915", + "issue:42947", + "issue:43065", + "issue:43257", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43450", + "issue:43454", + "issue:43550", + "issue:43572", + "issue:43720", + "issue:43749", + "issue:43873", + "issue:43881", + "issue:44186", + "issue:44387", + "issue:44466", + "issue:44488", + "issue:44534", + "issue:44568", + "issue:44610", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45357", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45920", + "issue:45923", + "issue:46050", + "issue:46209", + "issue:46305", + "issue:46315", + "issue:46402", + "issue:46821" + ], + "soft_pairs": [ + "issue:43454|issue:43550", + "issue:43257|issue:45923", + "issue:42796|issue:46209", + "issue:44568|issue:45920", + "issue:45823|issue:46821", + "issue:43377|issue:44610", + "issue:43454|issue:45072", + "issue:42617|issue:42915", + "issue:40990|issue:42947", + "issue:46305|issue:46315", + "issue:43572|issue:45759", + "issue:43454|issue:43720", + "issue:43377|issue:44898", + "issue:43749|issue:44534", + "issue:43317|issue:43873", + "issue:43322|issue:46050", + "issue:36331|issue:42617", + "issue:41720|issue:43322", + "issue:44387|issue:45704", + "issue:41720|issue:44977", + "issue:36331|issue:42915", + "issue:44877|issue:44991", + "issue:43881|issue:44488", + "issue:40990|issue:45701", + "issue:43450|issue:45701", + "issue:41720|issue:42617", + "issue:44466|issue:45357", + "issue:43065|issue:46402", + "issue:26413|issue:28530", + "issue:43873|issue:44756", + "issue:44186|issue:45072" + ], + "prepared_review_unit_hash": "e36926b1ce3216439fddceeef0641ec8c5d48efb846aa362c7c780307200cc0f", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36920, + "estimated_input_tokens": 9230, + "estimated_eval_tokens": 18716 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs share broad themes like tokenization, loading, offloading, or quantization, but they do not look like the same concrete bug/change. No PRs are present.", + "confidence": 0.93, + "canonical_issue_reason": "No single true canonical duplicate stands out because the cluster spans many unrelated failure modes. If one issue must be used as an umbrella representative, #43873 is the broadest/highest-activity offloading/quantization report, but it is not a duplicate of the others.", + "canonical_pr_reason": null, + "best_issue_reason": "#43873 is the strongest representative only in the sense that it is broad, active, and central to the offloading/quantization subset; however, the cluster is too diverse for a real canonical duplicate issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "Different models and failures: weight tying/text generation corruption vs torch.compile + SDPA failure." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "Qwen MoE weight-loading with accelerate/deepspeed is unrelated to a missing gradient_checkpointing flag on a trust_remote_code model." + }, + { + "left": "issue:42796", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer BPE-marker handling is unrelated to FP8 inference failure on B200." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "Both are tokenizer regressions, but one is specific special-token insertion behavior while the other is wrong token IDs across multiple models." + }, + { + "left": "issue:45823", + "right": "issue:46821", + "accept": false, + "reason": "Device-mismatch during forward is unrelated to an AttributeError on window_size in TimesFM." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI batching/padding-mask inconsistency is a different code path from OmDet processor output shape mismatch." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "lm_head tying/serialization corruption is unrelated to bfloat16 dtype mismatches in CI." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "A script execution problem is unrelated to a Qwen3 MoE FineGrainedFP8Config failure." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "High perplexity on a model/dataset is not the same bug as gradient checkpointing being ineffective with PEFT LoRA." + }, + { + "left": "issue:46305", + "right": "issue:46315", + "accept": false, + "reason": "Sliding-window masking off-by-one is unrelated to slow convert_tokens_to_ids performance." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "A missing config field is unrelated to AutoModelForCausalLM.from_config failing to unwrap text_config." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Weight tying/text generation corruption is unrelated to BitNet packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "Batch-vs-single output mismatch in MIMI is unrelated to Perceiver image classification with non-default resolution." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "FSDP CPU RAM efficient loading is a different loading path from non-persistent buffer initialization junk." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both mention offloading/quantization, but one is a specific device_map auto load failure while the other is a broader offloading regression; not the same concrete bug." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Llava Next loading segfault and sinq save/load corruption are different failure modes and code paths." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "CustomTrainer compute_loss signature breakage is unrelated to not being able to run a 3d_parallel script." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Llava Next segfault are unrelated load/runtime failures." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 CUDA reserved-memory/OOM regression is unrelated to a T5 apex fused RMSNorm memory leak." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Both involve Qwen models, but one is a device-mapping assert and the other is flash-attention generation corruption." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading for granite_speech is unrelated to tokenizer loading failure for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Two separate model-loading failures with no evidence of the same root cause." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "Perplexity regression on one model/dataset is not the same as a general tokenization-change report." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor batch-shape bug is unrelated to tokenization behavior changing across versions." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "lm_head serialization inconsistency and wrong visual-encoder keys in Qwen3.5 are different save_pretrained issues." + }, + { + "left": "issue:43065", + "right": "issue:46402", + "accept": false, + "reason": "A dummy Conv2d in Sam3PixelDecoder is unrelated to missing text encoder weight loading in Sam3Model." + }, + { + "left": "issue:26413", + "right": "issue:28530", + "accept": false, + "reason": "resume_from_checkpoint epoch-iterator failure is unrelated to early stopping disabling from missing eval metric." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantization/offloading regression and disabling mmap on Strix Halo to avoid OOM are different problems." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer crashes on NER/padding inputs are unrelated to bfloat16 dtype mismatches." + } + ] + }, + "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": 26, + "review_unit_index": 15, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28530", + "issue:33666", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43317", + "issue:43404", + "issue:43526", + "issue:43531", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43708", + "issue:43720", + "issue:43881", + "issue:43883", + "issue:44008", + "issue:44186", + "issue:44361", + "issue:44423", + "issue:44451", + "issue:44464", + "issue:44610", + "issue:44671", + "issue:44734", + "issue:44743", + "issue:44898", + "issue:44991", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45685", + "issue:45701", + "issue:45823", + "issue:46291", + "issue:46612", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:46869", + "issue:47020" + ], + "soft_pairs": [ + "issue:46612|issue:46858", + "issue:43295|issue:45701", + "issue:45685|issue:46821", + "issue:43577|issue:43708", + "issue:43550|issue:43720", + "issue:43646|issue:46665", + "issue:36010|issue:42617", + "issue:43550|issue:43606", + "issue:44186|issue:44610", + "issue:44451|issue:47020", + "issue:42915|issue:43317", + "issue:43404|issue:45357", + "issue:44610|issue:44898", + "issue:46821|issue:46869", + "issue:44423|issue:45406", + "issue:43881|issue:44991", + "issue:44734|issue:46858", + "issue:43526|issue:43883", + "issue:28530|issue:33666", + "issue:41720|issue:45161", + "issue:44008|issue:45823", + "issue:43531|issue:43572", + "issue:42915|issue:46291", + "issue:44464|issue:46858", + "issue:36331|issue:41720", + "issue:45632|issue:45684", + "issue:44671|issue:44743", + "issue:44186|issue:44361" + ], + "prepared_review_unit_hash": "2c33de9ecf4e2467b6865c1da0e36eb65939e03ec2843c488951b35837a88644", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36269, + "estimated_input_tokens": 9068, + "estimated_eval_tokens": 18392 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft-edge pair looks like the same concrete bug/change; the set is a collection of unrelated Transformers regressions across loading, generation, tokenization, multimodal serving, and device-mapping paths.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43646 is the broadest umbrella-style issue in the set: a Transformers 5.0.0 regression breaking custom model initialization. It is more central and general than the more model-specific or symptom-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best single issue anchor for this cluster because it describes a foundational regression in model initialization/loading rather than a narrow model-specific failure.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46612", + "right": "issue:46858", + "accept": false, + "reason": "Both involve generation/cache behavior, but one is about beam-search cache reorder being skipped for several models and the other is a static-cache seq-length/device-tensor arithmetic bug on non-CUDA backends." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Both are version-related regressions, but one is about processor.tokenizer / image-to-tokenizer compatibility and the other is about tokenization changes; different code paths and symptoms." + }, + { + "left": "issue:45685", + "right": "issue:46821", + "accept": false, + "reason": "Unrelated issues: an MPS MoE histogram/int dtype failure versus a TimesFM window_size AttributeError." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Different subsystems: BLIP2 dtype propagation during load versus Trainer checkpoint-resume step calculation." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Different failures in different model paths: torch.compile with SDPA on Bamba versus packed-weight handling during accelerate loading for BitNet." + }, + { + "left": "issue:43646", + "right": "issue:46665", + "accept": false, + "reason": "Both are custom-model/loading related, but one is a broad initialization regression and the other is inconsistent from_pretrained behavior inside Python classes; not the same bug." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "Completely different problems: GenerationMixin import failure versus inability to run a specific 3d_parallel.py script." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Both are device/runtime bugs, but the concrete failures differ: SDPA+torch.compile on Bamba versus CPU offload device mismatch on suno/bark-small." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Both touch vision preprocessing, but one is a tokenizer crash on NER/padding while the other is an image processor size mismatch for OmDet-Turbo." + }, + { + "left": "issue:44451", + "right": "issue:47020", + "accept": false, + "reason": "Both are tokenizer-loading failures, but they concern different tokenizer implementations and distinct failure modes." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Different quantization/device-map problems: FineGrainedFP8Config on Qwen3MoE versus dequantized model loading with gpu+cpu offload." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Both concern model save/load correctness, but one is lm_head tying for Mistral3 while the other is incorrect visual encoder keys for Qwen3.5." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are image-size/preprocessing regressions, but they are not the same concrete bug: one is processor output size mismatch, the other is Perceiver interpolation failure." + }, + { + "left": "issue:46821", + "right": "issue:46869", + "accept": false, + "reason": "Different model bugs: TimesFM window_size attribute access versus YOLO multi-GPU device-map loading failure." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both affect transformers serve with multimodal processors, but one is a .to() call on a string during continuous batching and the other is a missing _tokenizer attribute on Gemma4Processor." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading issues, but they are unrelated: glm-4v-9b load failure versus tokenizer loading failure for EMBEDDIA/est-roberta." + }, + { + "left": "issue:44734", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache handling, but one is wrong tensor indexing in serve /v1/responses and the other is a device-tensor seq-length issue in _prefill with static cache and torch.compile." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "Different components entirely: BeitImageProcessorFast label reduction versus MolmoForCausalLM missing all_tied_weights_keys." + }, + { + "left": "issue:28530", + "right": "issue:33666", + "accept": false, + "reason": "No real overlap: early stopping metric selection versus Qwen2-VL multi-GPU training." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Both are distributed/device-related, but they are different concrete bugs: cudaErrorAssert on A800 versus only-TP not working for GPT-OSS MoE." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma multimodal forward issues, but the specific failures differ: variable-name collision in Gemma3n versus device mismatch in Gemma4 PLE." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Different model/config regressions: Qwen3-MoE sliding_window behavior versus StableLmConfig missing pad_token_idx." + }, + { + "left": "issue:42915", + "right": "issue:46291", + "accept": false, + "reason": "Different areas: MoE FP8 failure versus importing transformers.pipelines with a kernels dependency version/revision requirement." + }, + { + "left": "issue:44464", + "right": "issue:46858", + "accept": false, + "reason": "Both involve generation/compile/cache, but one is inconsistent outputs from compiled forward and the other is a seq-length tensor arithmetic bug in prefill." + }, + { + "left": "issue:36331", + "right": "issue:41720", + "accept": false, + "reason": "Unrelated trainer API compatibility issue versus Qwen3 auto device-mapping crash." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Both concern custom-model save/load workflows, but one is a cache-path collision and the other is permissions propagation from source files." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "Different model-specific bugs: CamemBERT masked-LM predictions versus Qwen3.5 recurrent state reset with cache." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Both are tokenizer crashes, but they affect different tokenizers and failure points; 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": 26, + "review_unit_index": 16, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28530", + "issue:29159", + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36032", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43316", + "issue:43322", + "issue:43329", + "issue:43388", + "issue:43425", + "issue:43452", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43606", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:44077", + "issue:44387", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44534", + "issue:44589", + "issue:44625", + "issue:44913", + "issue:44991", + "issue:45059", + "issue:45406", + "issue:45464", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:46665", + "issue:46821", + "issue:46858", + "issue:47020" + ], + "soft_pairs": [ + "issue:43425|issue:46821", + "issue:28530|issue:30064", + "issue:43316|issue:44913", + "issue:42915|issue:43322", + "issue:33357|issue:35707", + "issue:29159|issue:36032", + "issue:30064|issue:34634", + "issue:43873|issue:45704", + "issue:44387|issue:45701", + "issue:43540|issue:43782", + "issue:43531|issue:45759", + "issue:43550|issue:43653", + "issue:44534|issue:46665", + "issue:44492|issue:44509", + "issue:41720|issue:43746", + "issue:43452|issue:47020", + "issue:42490|issue:43329", + "issue:43606|issue:43720", + "issue:34567|issue:35707", + "issue:45406|issue:45464", + "issue:43531|issue:46821", + "issue:43388|issue:45059", + "issue:34634|issue:41720", + "issue:43606|issue:43653", + "issue:43531|issue:44589", + "issue:45759|issue:45859", + "issue:43540|issue:46858", + "issue:44451|issue:44991", + "issue:44077|issue:44625" + ], + "prepared_review_unit_hash": "65c946df86779c240e3a0a52959b531ad396153a064e3d44a4fe5a98c51921dd", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35772, + "estimated_input_tokens": 8943, + "estimated_eval_tokens": 18142 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly superficially similar issues across Transformers, but the pairwise evidence points to different root causes and code paths. I do not see a true duplicate cluster here.", + "confidence": 0.88, + "canonical_issue_reason": "No single canonical issue: the linked items span unrelated bugs in tokenization, config loading, multimodal processing, serving, quantization, and docs.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43452 is the broadest actionable loading regression in the set, but it still does not represent the others as a duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43425", + "right": "issue:46821", + "accept": false, + "reason": "Different problems: Torch compatibility vs a TimesFM `window_size` attribute error." + }, + { + "left": "issue:28530", + "right": "issue:30064", + "accept": false, + "reason": "Early-stopping metric lookup and segmentation-map processing are unrelated bugs." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Both are config-related, but one is a Gemma3 API mismatch and the other is GPTNeoX rotary_pct persistence." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Llava Next segfault are different models and failure modes." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "MacOS CLIP bus error and progressive generation with `inputs_embeds`/`past_key_values` are unrelated." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Tokenizer inconsistency vs a T5 `add_special_tokens` name conflict; same area, different bug." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Void segmentation-map handling is unrelated to BarkProcessor voice preset loading." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization offload behavior and an Apex RMSNorm memory leak are distinct issues." + }, + { + "left": "issue:44387", + "right": "issue:45701", + "accept": false, + "reason": "CUDA OOM under int4 quantization is not the same as tokenization changing across versions." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "One is video input processing in Qwen3OmniMoe; the other is a `from_pretrained` weight_only loading error in Qwen3VL." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE and unwrapping `text_config` for composite configs are different code paths." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is unrelated to BigBirdTokenizer mask token registration." + }, + { + "left": "issue:44534", + "right": "issue:46665", + "accept": false, + "reason": "Non-persistent buffer initialization and `from_pretrained` inside Python classes are separate regressions." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A cache-strategy typo does not match a docs update about removed pipeline tasks." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3 device-mapping CUDA assert and GraniteSpeech PEFT checkpoint loading are different failures." + }, + { + "left": "issue:43452", + "right": "issue:47020", + "accept": false, + "reason": "Both involve loading/tokenizers, but one is gguf/from_pretrained routing and the other is an XLMRobertaTokenizer/Unigram init bug." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Trailing-slash local model import warning is unrelated to the undefined video-branch helpers in multimodal token counting." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU-offload device mismatch in Bark-small and BitNet packed-weight unpacking are distinct loading/offload problems." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "Trainer token-count tracking and progressive generation behavior are unrelated." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing `_tokenizer` and Qwen3.5 streaming chat/completions failures are different serving bugs." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3-MoE sliding-window logic is unrelated to the Float8 storage lookup error." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Composite-config unwrapping and CPU-offload compatibility in Qwen3.5 MoE embedding are different issues." + }, + { + "left": "issue:43540", + "right": "issue:46858", + "accept": false, + "reason": "Video-input validation in Qwen3OmniMoe is unrelated to the static-cache slice arithmetic bug." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are load failures, but the model-specific regression reports do not indicate the same tokenizer/root cause." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "A `post_init` allowance policy for patchtsmixer is unrelated to Qwen3.5 `num_labels` propagation." + } + ] + }, + "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": 26, + "review_unit_index": 17, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:41720", + "issue:42886", + "issue:43381", + "issue:43421", + "issue:43493", + "issue:43526", + "issue:43644", + "issue:43653", + "issue:43746", + "issue:43782", + "issue:43844", + "issue:43873", + "issue:43881", + "issue:43927", + "issue:43994", + "issue:44062", + "issue:44368", + "issue:44451", + "issue:44464", + "issue:44514", + "issue:44545", + "issue:44610", + "issue:44625", + "issue:44671", + "issue:44843", + "issue:44898", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:45042", + "issue:45072", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45704", + "issue:45759", + "issue:45923", + "issue:46032", + "issue:46050", + "issue:46402", + "issue:46463", + "issue:46655", + "issue:46665", + "issue:46693", + "issue:47050" + ], + "soft_pairs": [ + "issue:45357|issue:46402", + "issue:43644|issue:46665", + "issue:43526|issue:43881", + "issue:44987|issue:45042", + "issue:44898|issue:45072", + "issue:44062|issue:44451", + "issue:44625|issue:45479", + "issue:43381|issue:43844", + "issue:39692|issue:44610", + "issue:41720|issue:45704", + "issue:43746|issue:46050", + "issue:44368|issue:45357", + "issue:43381|issue:45704", + "issue:46655|issue:46693", + "issue:44918|issue:44936", + "issue:43421|issue:43927", + "issue:45704|issue:45923", + "issue:45759|issue:45923", + "issue:43746|issue:43873", + "issue:42886|issue:44843", + "issue:44514|issue:44545", + "issue:30064|issue:36010", + "issue:39692|issue:43493", + "issue:43994|issue:44671", + "issue:43421|issue:43653", + "issue:44464|issue:47050", + "issue:43782|issue:45381", + "issue:46032|issue:46463" + ], + "prepared_review_unit_hash": "b345fda8be81ea60e77529db05256b7944d8cefbf4dbbbbffe1cc37a1057c679", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36371, + "estimated_input_tokens": 9093, + "estimated_eval_tokens": 18442 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous. The only clear duplicate issue pair is 44514/44545, which share the same Qwen2_5_VLProcessor batched-input crash with padding=False; 42886/44843 is thematically related around HF_HUB_OFFLINE tokenizer loading, but not strong enough to treat as the same bug from the visible evidence. No pull requests are present.", + "confidence": 0.89, + "canonical_issue_reason": "44514 is the best canonical issue anchor because it is the clearest and earliest report of the Qwen2_5_VLProcessor batched-input padding=False crash, with 44545 appearing to be a near-identical follow-up.", + "canonical_pr_reason": null, + "best_issue_reason": "44514 is the strongest representative issue in the set: it is specific, has the closest matching twin in 44545, and best captures the concrete batched-input crash path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45357", + "right": "issue:46402", + "accept": false, + "reason": "Different model families and failure modes: Qwen3.5 visual-encoder key saving vs Sam3 text-encoder weights not loading." + }, + { + "left": "issue:43644", + "right": "issue:46665", + "accept": false, + "reason": "Both involve loading behavior, but one is junk-filled non-persistent buffers and the other is inconsistent from_pretrained behavior inside Python classes." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "Unrelated bugs: BeitImageProcessorFast label reduction vs glm-4v-9b loading failure." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Different concrete problems: loading a specific repo on newer transformers vs PIL image processors requiring torchvision." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Different subsystems and symptoms: Perceiver image-size handling vs bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "Both are tokenizer-related failures, but one is an AddedToken argument conflict and the other is a model loading failure for ScandiBERT." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Both involve labels, but one is config propagation in Qwen3.5 and the other is a classification-loss degeneracy across multiple models." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Different issues: eval-mode gradient checkpointing restriction vs gradient growth with ZeRO-3 training." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "One is a SigLIP2 documentation/example problem; the other is an OmDet-Turbo processor image-size mismatch." + }, + { + "left": "issue:41720", + "right": "issue:45704", + "accept": false, + "reason": "Completely different failures: Qwen3 auto device mapping cuda assert vs a T5 RMSNorm memory leak." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Different code paths: PEFT adapter loading from local checkpoints vs quantized model save/load breakage." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "A warning about tie_word_embeddings in LoRA fine-tuning is not the same as incorrect visual encoder keys being saved." + }, + { + "left": "issue:43381", + "right": "issue:45704", + "accept": false, + "reason": "No shared underlying bug: gradient checkpointing eval-mode restriction vs apex RMSNorm memory leak." + }, + { + "left": "issue:46655", + "right": "issue:46693", + "accept": false, + "reason": "Different problems: video padding calculation in SmolVLM vs performance regression with FA2 long-context generation." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Different trainer failures; one is embedding unpacking with TRL SFT, the other is evaluate() failing after train()." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Both touch token/config handling, but one is runtime special-token post-processing and the other is saved custom token IDs causing IndexError." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "Unrelated: T5 FusedRMSNorm memory leak vs missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:45759", + "right": "issue:45923", + "accept": false, + "reason": "Different problems: unwrapping text_config in multimodal Qwen configs vs a trust_remote_code gradient-checkpointing flag omission." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Different areas: local checkpoint loading with PEFT adapters vs offloading failure under quantization." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both are about HF_HUB_OFFLINE tokenization paths, but one reports a cache-load failure and the other a specific unconditional model_info call; related, not clearly the same bug." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same concrete Qwen2_5_VLProcessor apply_chat_template crash on batched input with padding=False; identical titles and matching failure path." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Different domains and failures: image processor segmentation-map handling vs GenerationMixin import error." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Same model family, but not the same bug: documentation example errors vs HF/JAX implementation discrepancy." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different model families and symptoms: SigLIP2 nonsensical outputs vs CamemBERT masked-LM prediction regression." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Both concern special tokens, but one is about runtime post-processor updates and the other about a tokenizer-specific special-token registration bug." + }, + { + "left": "issue:44464", + "right": "issue:47050", + "accept": false, + "reason": "Different failures: compiled-forward chunked generation inconsistency vs ONNX export dtype mismatch in CanineModel attention." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen multimodal issues: weight_only loading error vs incorrect video vision_position_ids." + }, + { + "left": "issue:46032", + "right": "issue:46463", + "accept": false, + "reason": "Unrelated: Mamba2Mixer use_cache correctness bug vs T5 decoder validation error-message bug." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Mostly grounded, but the cluster summary overstates certainty. The packet supports 44514/44545 as a strong duplicate candidate because the titles are identical, yet it does not substantiate \"exact duplicate\" or \"identical reproduction conditions\" from the visible evidence. Likewise 42886/44843 looks plausibly related, but the summary should stay more conservative about claiming the same underlying bug. The soft-edge rejections are generally cautious, but the canonical/summary wording is a bit too strong for the packet." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 18, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:30064", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43295", + "issue:43335", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43611", + "issue:43618", + "issue:43643", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43756", + "issue:43819", + "issue:43873", + "issue:43927", + "issue:44062", + "issue:44186", + "issue:44220", + "issue:44361", + "issue:44488", + "issue:44625", + "issue:44661", + "issue:44857", + "issue:44877", + "issue:44898", + "issue:44991", + "issue:44995", + "issue:45081", + "issue:45356", + "issue:45381", + "issue:45834", + "issue:45907", + "issue:45987", + "issue:46294", + "issue:46665", + "issue:46693", + "issue:46829", + "issue:46832", + "issue:46858", + "issue:46932", + "issue:47020", + "issue:47050" + ], + "soft_pairs": [ + "issue:44857|issue:47050", + "issue:45081|issue:45356", + "issue:43421|issue:44186", + "issue:44995|issue:46858", + "issue:45356|issue:45907", + "issue:39692|issue:43873", + "issue:39692|issue:47050", + "issue:43720|issue:43819", + "issue:43295|issue:45834", + "issue:43653|issue:43720", + "issue:46829|issue:46832", + "issue:43550|issue:43927", + "issue:43550|issue:44361", + "issue:44625|issue:47050", + "issue:43596|issue:44220", + "issue:43531|issue:43643", + "issue:43540|issue:46294", + "issue:45381|issue:46693", + "issue:44062|issue:44488", + "issue:45987|issue:46829", + "issue:43611|issue:46665", + "issue:41720|issue:44991", + "issue:44220|issue:44661", + "issue:30064|issue:36331", + "issue:44186|issue:44898", + "issue:43653|issue:43819", + "issue:45987|issue:46932", + "issue:43335|issue:43756", + "issue:43550|issue:43819", + "issue:44488|issue:47020", + "issue:43618|issue:43697", + "issue:43643|issue:44877" + ], + "prepared_review_unit_hash": "d8f3c5e6060d0038c3e2817ff9c5056911a46f1a9c082e4e86283a8a9798d1b0", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37590, + "estimated_input_tokens": 9398, + "estimated_eval_tokens": 19052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated issues across different models, tokenizers, config loading, caching, and export/runtime paths. The soft links are mostly superficial topic overlap (especially around tokenizers/configs), but none look like the same underlying bug or change, so all soft edges should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44857", + "right": "issue:47050", + "accept": false, + "reason": "Different failures and code paths: LwDetrImageLoss AMP crash vs CanineModel fp16 ONNX dtype mismatch." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both involve tokenizer regex/loading, but the concrete bugs differ: missing backend_tokenizer vs codec handling regression." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Different tokenizer subsystems and symptoms; runtime post-processor updates are unrelated to LayoutLMv2 NER/padding crashes." + }, + { + "left": "issue:44995", + "right": "issue:46858", + "accept": false, + "reason": "Both mention cache behavior, but one is stale indexer state and the other is tensor-return shape in StaticCache." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer regression vs repo-file listing API bug; no shared underlying issue." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "Both mention quantization, but one is a docs example mismatch and the other is offloading behavior." + }, + { + "left": "issue:39692", + "right": "issue:47050", + "accept": false, + "reason": "Different model families and failures; docs/quantization example issues do not match ONNX export dtype errors." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "Separate model-specific bugs: packed-weight loading vs missing STE in DAC.from_latents." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal regressions, but the concrete problems are different: processor/tokenizer API break vs long-OCR index error." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Tokenizer special-token registration bug is unrelated to BitNet packed-weight loading." + }, + { + "left": "issue:46829", + "right": "issue:46832", + "accept": false, + "reason": "Different RT/vision model bugs: wrong output field vs feature-level mismatch TypeError." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "Torch.compile SDPA failure and DiaConfig save/load token ID loss are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "No shared code path: Bamba compile/SDPA vs MLukeTokenizer task AttributeError." + }, + { + "left": "issue:44625", + "right": "issue:47050", + "accept": false, + "reason": "Different components and failure modes: Qwen3.5 num_labels propagation vs Canine fp16 ONNX export." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "BertModel zero3 init IndexError and fbank feature extraction issue are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Qwen3-MoE sliding_window bug is unrelated to AutoConfig missing fields with trust_remote_code." + }, + { + "left": "issue:43540", + "right": "issue:46294", + "accept": false, + "reason": "Both involve Qwen video inputs, but one is a direct video-processing ValueError and the other is a chat-template token/features mismatch." + }, + { + "left": "issue:45381", + "right": "issue:46693", + "accept": false, + "reason": "Qwen2.5-VL video position IDs bug is unrelated to FA2 long-context performance regression." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "AddedToken constructor argument error is not the same as loading cjvt/sleng-bert failing." + }, + { + "left": "issue:45987", + "right": "issue:46829", + "accept": false, + "reason": "StaticCache seq-length shape bug and VideoPrism hidden_states assignment are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:46665", + "accept": false, + "reason": "Both are from_pretrained/loading issues, but the concrete failure modes and call paths differ." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Different subsystems entirely: Qwen3 auto device mapping crash vs tokenizer loading failure for a specific model." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio feature extraction bug and tokenizer-mapping add-new-model-like failure are unrelated." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image segmentation-map processing and CustomTrainer compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer NER/padding crash is unrelated to Perceiver image classification interpolation failure." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Tokenizer special-token registration bug is unrelated to DAC.from_latents STE behavior." + }, + { + "left": "issue:45987", + "right": "issue:46932", + "accept": false, + "reason": "StaticCache sequence-length typing and Gemma4 quantized embedder dtype casting are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug is unrelated to Smollm3 RoPE-layer dropping." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Different models and mechanisms: SDPA/torch.compile failure vs DAC latent conversion bug." + }, + { + "left": "issue:44488", + "right": "issue:47020", + "accept": false, + "reason": "Both are tokenizer loading issues, but one is a specific model load failure and the other is an XLMRoberta Unigram init type mismatch." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "CLIPOutput attentions regression and RTDetrV2 differing outputs are distinct vision model bugs." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both concern config loading strictness, but they are different code paths and symptoms, not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 19, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:34634", + "issue:36010", + "issue:39692", + "issue:43322", + "issue:43388", + "issue:43421", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43644", + "issue:43749", + "issue:43782", + "issue:43818", + "issue:43825", + "issue:43994", + "issue:44008", + "issue:44186", + "issue:44442", + "issue:44448", + "issue:44479", + "issue:44561", + "issue:44671", + "issue:44805", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45183", + "issue:45479", + "issue:45584", + "issue:45685", + "issue:45834", + "issue:46032", + "issue:46291", + "issue:46294", + "issue:46463", + "issue:46512", + "issue:46858", + "issue:47020" + ], + "soft_pairs": [ + "issue:44008|issue:46858", + "issue:44821|issue:45084", + "issue:34634|issue:36010", + "issue:44843|issue:45081", + "issue:39692|issue:45834", + "issue:44186|issue:44442", + "issue:43540|issue:44008", + "issue:44448|issue:44671", + "issue:45584|issue:45834", + "issue:45183|issue:45479", + "issue:43322|issue:45685", + "issue:43644|issue:43749", + "issue:44991|issue:47020", + "issue:43606|issue:45072", + "issue:44991|issue:46291", + "issue:44561|issue:45020", + "issue:43421|issue:44811", + "issue:43577|issue:43818", + "issue:43550|issue:45072", + "issue:43582|issue:46512", + "issue:44479|issue:45834", + "issue:43782|issue:44977", + "issue:43825|issue:43994", + "issue:43388|issue:46032", + "issue:44805|issue:44936", + "issue:46294|issue:46463" + ], + "prepared_review_unit_hash": "43b58086ed355997b7538f919012529837503d6104b02a7217f7ebb0587e9325", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35345, + "estimated_input_tokens": 8837, + "estimated_eval_tokens": 17930 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous Transformers issues spanning unrelated models, tokenizers, multimodal/video paths, training, and runtime config. I found one plausible duplicate pair around the XLM-Roberta / est-roberta tokenizer-loading regression; the rest should stay separate.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue for the full cluster; issue:47020 is the clearest root-cause report for the only plausible duplicate pair, while the other items are unrelated.", + "canonical_pr_reason": "No pull requests are present in this cluster.", + "best_issue_reason": "issue:47020 is the most specific and actionable report; it best represents the tokenizer-loading regression seen in issue:44991.", + "best_pr_reason": "No pull requests are present, so there is no best PR candidate.", + "soft_edge_verdicts": [ + { + "left": "issue:44008", + "right": "issue:46858", + "accept": false, + "reason": "Different failures in different code paths: Gemma3n forward variable collision vs static-cache slice arithmetic/device tensor issue." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "Unrelated: loading AutoImageProcessor from URL vs template-node compilation error." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "Different subsystems and symptoms: Bark voice_preset vs GenerationMixin import breakage." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both touch mistral regex handling, but one is offline model_info access and the other is a backend_tokenizer attribute crash; distinct bugs." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 example issues and Kosmos2.5 long-ocr index error are unrelated model-specific bugs." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Tokenizer crash on NER/padding vs AutoTokenizer failing to load a different tokenizer class." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different multimodal paths: Qwen video processing ValueError vs Gemma3n audio-mask attribute collision." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are v5 output regressions, but in different models and code paths; not the same bug." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription generation failure is unrelated to Kosmos2.5 OCR indexing." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Validation/error-message bug versus classification-loss bug; different underlying issues." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Segfault on Llava Next loading is unrelated to MPS histogram type support." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Non-persistent buffer junk fill and FSDP CPU RAM-efficient loading are different loading/runtime problems." + }, + { + "left": "issue:44991", + "right": "issue:47020", + "accept": true, + "reason": "Likely the same tokenizer-loading regression: est-roberta fails to load because XLMRobertaTokenizer passes a dict to Unigram expecting a sequence." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "Both involve dtype/device mismatches, but for different models, backends, and concrete failures." + }, + { + "left": "issue:44991", + "right": "issue:46291", + "accept": false, + "reason": "Tokenizer load failure for one model is not the same as pipelines import breaking on kernels v0.15." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Both concern trust-remote-code / recent-version breakage, but the specific removed helper and the broader remote-code regression are not the same bug." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Special-token post-processor update bug is unrelated to Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Blip2 dtype propagation and Video-LLaVA temporal-attention/model-weight issues are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Both are compile/dtype-related, but the failing models and code paths differ." + }, + { + "left": "issue:43582", + "right": "issue:46512", + "accept": false, + "reason": "AppleSilicon warmup TypeError and stale config.dtype after from_config are separate dtype/config bugs." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Qwen video regression and Kosmos2.5 OCR long-input indexing are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "From_pretrained weight_only error in Qwen3VL is unrelated to Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "Pipeline translation-message regression and SigLIP2 output nonsense are different issues." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Metrics batch-label truncation and Mamba2 use_cache incorrect results are unrelated code paths." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Mask-shape IndexError and trainer.evaluate-after-train failure are not the same bug." + }, + { + "left": "issue:46294", + "right": "issue:46463", + "accept": false, + "reason": "Qwen video chat-template mismatch and T5 decoder-input error-message bug are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: it correctly treats the set as mostly heterogeneous and only tentatively groups 44991 with 47020 as a plausible tokenizer-loading duplicate. The soft-edge verdicts are conservative, and there are no pull requests to evaluate. No obvious overstatement detected." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 20, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:42175", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43645", + "issue:43653", + "issue:43716", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43931", + "issue:44008", + "issue:44291", + "issue:44589", + "issue:44610", + "issue:44671", + "issue:44857", + "issue:44912", + "issue:45003", + "issue:45020", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45372", + "issue:45381", + "issue:45701", + "issue:46175", + "issue:46285", + "issue:46326", + "issue:46665", + "issue:46697", + "issue:46728", + "issue:46846", + "issue:46932" + ], + "soft_pairs": [ + "issue:43931|issue:46846", + "issue:43716|issue:44008", + "issue:43540|issue:45381", + "issue:28218|issue:29159", + "issue:44610|issue:45072", + "issue:43749|issue:44291", + "issue:43720|issue:45072", + "issue:43645|issue:46665", + "issue:43825|issue:43827", + "issue:34634|issue:36331", + "issue:43577|issue:43582", + "issue:42175|issue:45081", + "issue:44671|issue:45701", + "issue:45003|issue:46285", + "issue:46175|issue:46932", + "issue:46697|issue:46728", + "issue:43262|issue:43819", + "issue:43582|issue:44857", + "issue:43782|issue:45325", + "issue:43653|issue:44912", + "issue:44589|issue:45042", + "issue:43742|issue:45020", + "issue:45372|issue:46326", + "issue:43550|issue:44912", + "issue:36010|issue:43645", + "issue:36010|issue:36331", + "issue:43257|issue:43329" + ], + "prepared_review_unit_hash": "1e8589e7632c878c9d1e6ad2348b350064bcf1ea02918ca7269101d5695930a8", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36248, + "estimated_input_tokens": 9062, + "estimated_eval_tokens": 18380 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly heterogeneous. The only clear duplicate-style pair is the tokenizer spacing bug versus the slower/faster tokenizer inconsistency report; the rest are distinct model, loading, dtype, or docs issues and should stay separate.", + "confidence": 0.74, + "canonical_issue_reason": "Issue #29159 is the broader formulation of the tokenizer spacing/inconsistency bug; #28218 reads like a narrower symptom of the same underlying behavior.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #29159 is the best duplicate target among the issues because it captures the shared tokenizer behavior more generally than #28218.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43931", + "right": "issue:46846", + "accept": false, + "reason": "Different failures: Qwen3-VL weight-shape mismatch vs Qwen3.5 TP planning/linear_attn OOM-shape errors." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Both are multimodal loading bugs, but the concrete code paths differ: image preprocessor/model dtype mismatch vs Gemma 3n audio tensor name collision." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Both mention video inputs, but they target different models and likely different bugs; not the same concrete change." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": true, + "reason": "Both describe the same tokenizer added-token spacing inconsistency; #29159 is the broader duplicate target." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Unrelated subsystems: OmDet image-size mismatch vs bf16 dtype mismatches in different models." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "Both involve loading infrastructure, but one is broken CPU RAM-efficient loading and the other is an unexpected init_empty_weights argument." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Different issues: packed-weight unpacking during accelerate loading vs dtype mismatch inference." + }, + { + "left": "issue:43645", + "right": "issue:46665", + "accept": false, + "reason": "Both concern loading custom models, but the reported environments and failure modes differ too much to treat as the same bug." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "One is an error-message regression, the other is a documentation cleanup; related topic but not the same artifact or bug." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "Completely different areas: BarkProcessor voice_preset vs Trainer compute_loss signature change." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Different failure modes and platforms: BLIP2 dtype loading vs Apple Silicon allocator warmup TypeError." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "No common code path: extras/install backend packaging vs Mistral regex patching during tokenizer load." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "The titles are related to tokenization, but one is a specific CamemBERT MLM regression and the other is a broad version-to-version tokenization change report." + }, + { + "left": "issue:45003", + "right": "issue:46285", + "accept": false, + "reason": "Different problems: unsafe sys.modules access in modeling_utils vs CLIPTextModel structure divergence." + }, + { + "left": "issue:46175", + "right": "issue:46932", + "accept": false, + "reason": "Both touch Gemma4 vision/audio embedders, but one is a huge one-hot memory issue and the other is quantized dtype casting." + }, + { + "left": "issue:46697", + "right": "issue:46728", + "accept": false, + "reason": "Both are image-shape/count calculations, but for different processors and distinct bugs." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting is unrelated to DAC latent/STE behavior." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon allocator warmup and CUDA AMP image-loss crashes are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen-VL issues: weight_only loading error vs rope-index / temporal position scaling bug." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration is unrelated to MXFP4 quantization fallback." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Float8 storage lookup error vs PIL/torchvision dependency regression are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "A specific key error while loading one model is not the same as the broader remote_code breakage report." + }, + { + "left": "issue:45372", + "right": "issue:46326", + "accept": false, + "reason": "Both involve mistral_common, but one is an import error for Gemma 4 loading and the other is a tokenizer continue_final_message behavior bug." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile/SDPA failure is unrelated to quantization loading fallback." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "Import path regression for GenerationMixin vs Jupyter custom-model initialization failure are different bugs." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "One is an import regression, the other is a Trainer API signature change." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE loading/conversion and multimodal token counting are separate bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is mostly conservative and grounded in the packet. The only accepted pair, #28218 and #29159, is a plausible tokenizer-duplicate match based on the titles, and the rest of the soft-edge pairs are reasonably rejected as distinct bugs/subsystems. No mergeability issues for PRs arise here since these are issue pairs only." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 21, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:34567", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42175", + "issue:42490", + "issue:42915", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43638", + "issue:43643", + "issue:43696", + "issue:43697", + "issue:43742", + "issue:43746", + "issue:43819", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:43931", + "issue:44060", + "issue:44368", + "issue:44387", + "issue:44479", + "issue:44554", + "issue:44610", + "issue:44617", + "issue:44811", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45042", + "issue:45081", + "issue:45084", + "issue:45406", + "issue:45538", + "issue:45600", + "issue:45759", + "issue:45859", + "issue:46209", + "issue:46291", + "issue:47050" + ], + "soft_pairs": [ + "issue:43906|issue:46209", + "issue:43526|issue:43697", + "issue:43531|issue:45042", + "issue:43452|issue:44991", + "issue:42175|issue:46291", + "issue:43638|issue:44936", + "issue:43819|issue:43927", + "issue:45081|issue:45406", + "issue:44060|issue:44368", + "issue:43329|issue:45600", + "issue:45084|issue:45538", + "issue:43526|issue:44554", + "issue:42490|issue:43257", + "issue:43746|issue:44811", + "issue:45859|issue:47050", + "issue:43643|issue:45759", + "issue:43854|issue:44977", + "issue:38175|issue:44610", + "issue:43262|issue:44811", + "issue:39692|issue:44912", + "issue:43257|issue:43696", + "issue:43329|issue:43450", + "issue:43257|issue:43931", + "issue:43329|issue:44479", + "issue:44387|issue:44617", + "issue:42915|issue:43742", + "issue:34567|issue:40990" + ], + "prepared_review_unit_hash": "f0b90a491c32a4721a35bc54e417bc00b5f9a5c46357ab44e05aaa56c7938b40", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36100, + "estimated_input_tokens": 9025, + "estimated_eval_tokens": 18306 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the issues cover unrelated models, loaders, processors, training, quantization, and backend/dependency bugs. None of the soft pairs look like the same underlying bug/change.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue: the items are not duplicates of one another, so there is no single issue to preserve as the cluster representative.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44479 is the clearest self-contained regression report in the set, with a specific symptom and affected-model list, but it is still unrelated to the other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "Different problems: an isolated reproduction of an older issue versus an FP8 inference failure on B200." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "Unrelated models and symptoms: BEiT Fast label reduction versus RTDetrV2 output differences in v5." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Different code paths: Qwen3-MoE sliding-window behavior versus PIL image processors requiring torchvision." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading/tokenizer issues, but they affect different APIs, models, and failure modes." + }, + { + "left": "issue:42175", + "right": "issue:46291", + "accept": false, + "reason": "Different dependency regressions: missing TensorFlow backend on install versus kernels import requiring version/revision." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Training-related bugs, but one is a DeepSpeed ZeRO3 Bert index error and the other is an evaluate-after-train failure." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "Different models and bugs: DAC forward/from_latents mismatch versus DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Both involve tokenizer/processor loading, but one is a Mistral regex patch crash and the other is a Gemma4Processor attribute error." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Related to tied-weights warnings, but the underlying model/config paths are different and not the same bug." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Different subsystems: multimodal video token counting versus stale auto_mappings references breaking CI." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated bugs: template compilation error versus CLIPTokenizer model_max_length handling." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "Different components entirely: label reduction in BEiT versus an MPS attention correctness issue." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Different loading failures: trailing-slash import warning versus Qwen3 MoE weight conversion under accelerate+deepspeed." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "Different functionality: local PEFT checkpoint loading versus Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:45859", + "right": "issue:47050", + "accept": false, + "reason": "Different backends and failures: CPU offload incompatibility versus fp16 ONNX export dtype mismatch." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both are config/loading related, but one is trust_remote_code missing fields and the other is composite multimodal config unwrapping." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Different models and symptoms: GLM-4.7-Flash unit-test loading failure versus Qwen3.5 flash-attention generation issues." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Both involve model/processor mismatch-like symptoms, but they are for different models and different concrete failures." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Different audio-processing bugs: apply_chat_template sample-rate default versus Whisper batch_decode token skipping." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "Different quantization/documentation issues: SigLIP2 doc errors versus GPT-oss MXFP4 loading fallback behavior." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Different failure classes: Qwen3 MoE weight conversion versus GPT-oss-20b CUDA out-of-memory." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Both are video-related, but one is multimodal token counting and the other is batched video processor shape handling." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Different model-loading problems: MoE weight conversion under distributed loading versus Qwen3-VL weight shape mismatch." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Related to video paths, but 43329 is a specific undefined-helper bug in token counting while 44479 is a broader video-input regression." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Different OOM causes: quantization-related CUDA reserved memory growth versus Sam3Video model memory exhaustion." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Different loading errors: FineGrainedFP8Config failure in Qwen3Moe versus a key error loading MobileLLM-125M." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "Completely different areas: TrainerState token-count tracking versus an inference perplexity report on GPT-oss-20b." + } + ] + }, + "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": 26, + "review_unit_index": 22, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28282", + "issue:29127", + "issue:29159", + "issue:29870", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:39379", + "issue:42175", + "issue:42915", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43525", + "issue:43540", + "issue:43577", + "issue:43650", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43824", + "issue:43906", + "issue:43927", + "issue:44291", + "issue:44554", + "issue:44568", + "issue:44756", + "issue:44821", + "issue:44857", + "issue:44912", + "issue:44987", + "issue:45042", + "issue:45081", + "issue:45127", + "issue:45137", + "issue:45356", + "issue:45357", + "issue:45405", + "issue:45459", + "issue:45538", + "issue:45803", + "issue:46291", + "issue:46421", + "issue:46424", + "issue:46701", + "issue:46832", + "issue:47020" + ], + "soft_pairs": [ + "issue:44291|issue:45042", + "issue:42915|issue:44821", + "issue:29159|issue:34567", + "issue:45459|issue:45803", + "issue:46421|issue:46424", + "issue:43650|issue:43906", + "issue:28282|issue:29870", + "issue:43749|issue:44756", + "issue:43525|issue:46291", + "issue:28282|issue:34689", + "issue:43454|issue:43577", + "issue:44987|issue:45538", + "issue:43540|issue:46701", + "issue:43425|issue:46291", + "issue:43927|issue:44912", + "issue:44568|issue:45356", + "issue:33357|issue:34567", + "issue:39379|issue:44554", + "issue:42175|issue:43824", + "issue:44857|issue:46832", + "issue:43697|issue:45357", + "issue:45137|issue:45405", + "issue:43404|issue:45127", + "issue:28282|issue:29127", + "issue:45081|issue:47020", + "issue:43742|issue:44291" + ], + "prepared_review_unit_hash": "dcc228d062d03cf0fbbdd56f4bc9bce0218d67ad90ecb14e2e188bb336ceb5f8", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35117, + "estimated_input_tokens": 8780, + "estimated_eval_tokens": 17816 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a broad semantic false-positive: the items span unrelated bugs in model loading, tokenization, cache behavior, vision/text models, training utilities, and dependency packaging. I found no pair that looks like the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue stands out; the set is heterogeneous and the apparent similarities are mostly broad subsystem overlap rather than the same defect.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue is a strong global representative because the cluster does not share one concrete code path or user-facing failure mode.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Different failures: init_empty_weights argument handling vs PIL image processor dependency requirements." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Unrelated areas: FineGrainedFP8Config failure vs loading AutoImageProcessor from URL." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer inconsistency and TrainerState token counting are separate bugs." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve exception handling, but they are different code paths and different failure modes." + }, + { + "left": "issue:46421", + "right": "issue:46424", + "accept": false, + "reason": "Both concern cache behavior, but one is recompilation with chunked prefill and the other is max_cache_length semantics." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "A placeholder title vs an isolated reproduction of another issue; no shared bug." + }, + { + "left": "issue:28282", + "right": "issue:29870", + "accept": false, + "reason": "PyTorch import error and BART generate() UnboundLocalError are unrelated." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different problems." + }, + { + "left": "issue:43525", + "right": "issue:46291", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config is unrelated to pipelines import failing on kernels version/revision." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "Different model-loading bugs: lm_head tying vs dtype propagation in BLIP-2." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Loading a specific model and CLIPTokenizer max_length behavior are not the same defect." + }, + { + "left": "issue:43540", + "right": "issue:46701", + "accept": false, + "reason": "Video input validation in Qwen3OmniMoe is unrelated to DiffusionGemma pipeline output handling." + }, + { + "left": "issue:43425", + "right": "issue:46291", + "accept": false, + "reason": "Torch compatibility and kernels import requirements are separate dependency issues." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "Save/load token ID loss in DiaConfig is unrelated to MXFP4 quantization fallback behavior." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different underlying logic." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "CLIP bus error on macOS and TrainerState token counting are not the same issue." + }, + { + "left": "issue:39379", + "right": "issue:44554", + "accept": false, + "reason": "Different correctness bugs in different attention/multimodal paths." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Packaging backend composition issue vs missing exported model class import error." + }, + { + "left": "issue:44857", + "right": "issue:46832", + "accept": false, + "reason": "Both are RT-DETR-family, but one is an AMP loss crash and the other is a forward-shape TypeError." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetrV2 output drift and Qwen3.5 visual encoder key save_pretrained regression are unrelated." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and PEFT version pinning are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied weights, but one is a missing weight tying bug and the other is LoRA merge collapse with extended vocab." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "ImportError for PyTorch and a LayoutLMv3 error-message clarity issue are unrelated." + }, + { + "left": "issue:45081", + "right": "issue:47020", + "accept": false, + "reason": "Mistral regex patch crash and XLMRobertaTokenizer/Unigram constructor misuse are different tokenizer code paths." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "Key error loading MobileLLM and init_empty_weights argument handling are separate model-loading 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": 26, + "review_unit_index": 23, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:24643", + "issue:27957", + "issue:28282", + "issue:29159", + "issue:34567", + "issue:36331", + "issue:38175", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43257", + "issue:43452", + "issue:43531", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44568", + "issue:44589", + "issue:44792", + "issue:44821", + "issue:44849", + "issue:44898", + "issue:44936", + "issue:44964", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45081", + "issue:45325", + "issue:45381", + "issue:45488", + "issue:45538", + "issue:45701", + "issue:45907", + "issue:46399", + "issue:46512", + "issue:46693", + "issue:46821", + "issue:46869" + ], + "soft_pairs": [ + "issue:44936|issue:46821", + "issue:44479|issue:46693", + "issue:42915|issue:45538", + "issue:45538|issue:46869", + "issue:43452|issue:43582", + "issue:24643|issue:29159", + "issue:27957|issue:28282", + "issue:44062|issue:44991", + "issue:42915|issue:43749", + "issue:43577|issue:46512", + "issue:44849|issue:45701", + "issue:41720|issue:43540", + "issue:42175|issue:44488", + "issue:43540|issue:44560", + "issue:44987|issue:45907", + "issue:43540|issue:44479", + "issue:38175|issue:44898", + "issue:44479|issue:45381", + "issue:44291|issue:44821", + "issue:34567|issue:36331", + "issue:44792|issue:44964", + "issue:44568|issue:45488", + "issue:43257|issue:43756", + "issue:44062|issue:45081", + "issue:45042|issue:45081", + "issue:43582|issue:46821", + "issue:43742|issue:44589", + "issue:43531|issue:44821", + "issue:43756|issue:46399", + "issue:43756|issue:45325" + ], + "prepared_review_unit_hash": "0ba116de75251b9e2c97f0b582a01b1371991cd92389278f965e5a4d271ec9f4", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36058, + "estimated_input_tokens": 9015, + "estimated_eval_tokens": 18286 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous and does not collapse cleanly into a single bug/theme. All soft edges look like false positives: some are broadly related areas (Qwen video regressions, tokenizer regressions, trainer API changes), but they describe different concrete failures and fixes.", + "confidence": 0.93, + "canonical_issue_reason": "No canonical issue: the items span unrelated bugs across training, tokenization, multimodal models, packaging, and backend/runtime errors.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the broadest representative of the Qwen video-regression subset, but it still does not represent the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44936", + "right": "issue:46821", + "accept": false, + "reason": "Different failures and code paths: trainer evaluate-after-train behavior vs TimesFM window_size attribute handling." + }, + { + "left": "issue:44479", + "right": "issue:46693", + "accept": false, + "reason": "Both mention performance/regression, but one is a Qwen video-input bug and the other is an FA2 long-context slowdown." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated: Qwen3 MoE FP8 loading vs CLIP tokenizer max_length." + }, + { + "left": "issue:45538", + "right": "issue:46869", + "accept": false, + "reason": "Different subsystems: tokenizer max_length vs YOLO multi-GPU device-map failure." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "Unrelated runtime issues: gguf/AutoModel loading vs Apple Silicon allocator warmup TypeError." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "Different areas: DeepSpeed weight shape error vs tokenizer slow/fast inconsistency." + }, + { + "left": "issue:27957", + "right": "issue:28282", + "accept": false, + "reason": "FA2/XLM-RoBERTa model behavior is unrelated to missing PyTorch ImportError." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-related, but one is an AddedToken argument conflict and the other is a model-specific tokenizer loading failure." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Different bugs: Qwen3 MoE FP8 config vs FSDP CPU RAM efficient loading." + }, + { + "left": "issue:43577", + "right": "issue:46512", + "accept": false, + "reason": "Not the same bug: one leaves model/qformer weights in float32, the other leaves config.dtype stale after from_config(dtype=...)." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Different topics: Qwen3.5 hidden_states behavior vs version-dependent tokenization changes." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Different failure modes and models: auto device-map CUDA assert vs Qwen3OmniMoe video input ValueError." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Packaging/backend omission is unrelated to a model loading failure." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-input Qwen issues, but they affect different model families and report different concrete errors." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Unrelated: model loading regression vs repo file listing not forwarding revision." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Broadly similar Qwen video regressions, but not the same concrete bug or fix target." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Different vision-model bugs: zero probabilities vs Perceiver non-default resolution failure." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": false, + "reason": "Related Qwen2.5-VL video area, but one is a broad regression report and the other is a specific vision_position_ids symptom; not enough to treat as one bug." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "Unrelated: init_empty_weights argument breakage vs AutoImageProcessor URL loading." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "Both are trainer-related, but they are different API regressions with different impacted call sites." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Different multimodal model issues: Janus image-generation test failure vs Phi-4 multimodal loading failure." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different tokenizer classes and different concrete behaviors." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Unrelated model-specific bugs: Qwen3 MoE weight conversion vs Smollm3 RoPE layer dropping." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer bugs and different failure points." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Image-processor torchvision dependency issue is unrelated to Mistral tokenizer regex patching." + }, + { + "left": "issue:43582", + "right": "issue:46821", + "accept": false, + "reason": "Different runtime areas: allocator warmup on Apple Silicon vs TimesFM window_size handling." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "Different loading errors: MobileLLM key error vs missing Float8 storage object." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding_window bug is unrelated to AutoImageProcessor URL loading." + }, + { + "left": "issue:43756", + "right": "issue:46399", + "accept": false, + "reason": "Different model-specific regressions: Smollm3 RoPE layers vs Qwen3.5 kernel registration crash." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Both touch rotary/position logic, but they are different models and different erroneous 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": 26, + "review_unit_index": 24, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29870", + "issue:32090", + "issue:41762", + "issue:43452", + "issue:43526", + "issue:43540", + "issue:43577", + "issue:43697", + "issue:43716", + "issue:43720", + "issue:43906", + "issue:44466", + "issue:44488", + "issue:44493", + "issue:44568", + "issue:44625", + "issue:44737", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:44936", + "issue:45005", + "issue:45072", + "issue:45200", + "issue:45216", + "issue:45357", + "issue:45405", + "issue:45479", + "issue:45588", + "issue:45600", + "issue:45663", + "issue:45672", + "issue:45674", + "issue:45823", + "issue:45920", + "issue:46050", + "issue:46121", + "issue:46463", + "issue:46468", + "issue:46491", + "issue:46693", + "issue:46701", + "issue:46829", + "issue:46846", + "issue:46899" + ], + "soft_pairs": [ + "issue:45405|issue:45600", + "issue:44737|issue:46463", + "issue:43577|issue:45072", + "issue:43716|issue:46701", + "issue:45823|issue:46463", + "issue:43906|issue:46693", + "issue:44568|issue:46491", + "issue:24643|issue:29870", + "issue:45672|issue:45674", + "issue:44625|issue:44877", + "issue:43540|issue:46693", + "issue:43452|issue:44488", + "issue:43906|issue:46468", + "issue:28282|issue:32090", + "issue:43697|issue:46829", + "issue:45663|issue:46846", + "issue:44913|issue:46121", + "issue:44737|issue:45588", + "issue:43526|issue:45479", + "issue:46463|issue:46899", + "issue:43697|issue:45216", + "issue:45357|issue:45920", + "issue:44493|issue:46050", + "issue:44466|issue:45005", + "issue:43720|issue:44898", + "issue:45200|issue:46463", + "issue:41762|issue:44936" + ], + "prepared_review_unit_hash": "ce7b32a528eab764b44894cbb7031fbb69c6b1c5910d6ac7a303dae19038055c", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35891, + "estimated_input_tokens": 8973, + "estimated_eval_tokens": 18202 + }, + "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 bugs; the soft-similarity links do not look like true duplicates, so I would not merge any of them.", + "confidence": 0.87, + "canonical_issue_reason": "Issue 44466 is the best representative only because it is a concrete, well-scoped v5 save/load regression around tied weights; the rest of the cluster spans unrelated tokenizer, config, quantization, multimodal, and Trainer issues.", + "canonical_pr_reason": null, + "best_issue_reason": "44466 is the clearest standalone issue in the set: it has a specific symptom, an explicit v5 regression angle, and a likely actionable serialization bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI-related, but one is a missing PyPI release and the other is stale auto-mapping references; different failures and fixes." + }, + { + "left": "issue:44737", + "right": "issue:46463", + "accept": false, + "reason": "XLNet CPU placement bug and T5 decoder-input error message are unrelated code paths and symptoms." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype issues, but they affect different models and different inference paths, so they are not the same bug." + }, + { + "left": "issue:43716", + "right": "issue:46701", + "accept": false, + "reason": "One is a dtype mismatch in Mistral-3 preprocessing, the other is a broken image-text-to-text pipeline for DiffusionGemmaGenerationOutput." + }, + { + "left": "issue:45823", + "right": "issue:46463", + "accept": false, + "reason": "Gemma4 device-map forward mismatch and T5 decoder validation messaging are different problems." + }, + { + "left": "issue:43906", + "right": "issue:46693", + "accept": false, + "reason": "The reproduction of issue 38071 and the FA2 long-context performance regression are not the same concrete bug." + }, + { + "left": "issue:44568", + "right": "issue:46491", + "accept": false, + "reason": "Both are tokenizer regressions, but one is BOS/EOS insertion and the other is leading-space round-trip behavior; distinct issues." + }, + { + "left": "issue:24643", + "right": "issue:29870", + "accept": false, + "reason": "DeepSpeed training weight-shape error and BART decoder_attention_mask initialization error are unrelated failures." + }, + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Both involve 4-bit inference, but one is a uint8/torch.finfo crash and the other is lm_head exclusion handling in BitsAndBytesConfig." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and granite_speech strict-config loading are different configuration bugs." + }, + { + "left": "issue:43540", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3OmniMoe video processing failure and FA2 long-context performance regression do not share the same underlying code path." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "Both mention loading failures, but one is gguf_file API handling and the other is a specific model not loading; no clear duplicate relation." + }, + { + "left": "issue:43906", + "right": "issue:46468", + "accept": false, + "reason": "Isolated reproduction of an older issue and an official Nemotron example breakage are not the same bug." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for missing PyTorch and Trainer/_gpu_broadcast_one NoneType error are entirely different issues." + }, + { + "left": "issue:43697", + "right": "issue:46829", + "accept": false, + "reason": "RTDetrV2 output differences and VideoPrism hidden_states assignment are unrelated model-specific bugs." + }, + { + "left": "issue:45663", + "right": "issue:46846", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Qwen3.5 TP plan omissions are different distributed-training problems." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "GPTNeoX rotary_pct reload regression and rope-parameter JSON list TypeError are both rope-related, but they are distinct failure modes." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU arange bug and flash_attention s_aux=None crash are separate low-level runtime errors." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "A BeitImageProcessor label-reduction bug is not the same as a sequence-classification zero-loss problem." + }, + { + "left": "issue:46463", + "right": "issue:46899", + "accept": false, + "reason": "T5 decoder validation messaging and Gemma4 4-bit multimodal embedding dtype casts are unrelated." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output regression and Qwen3.5 save_pretrained checkpoint corruption are different areas and symptoms." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Incorrect Qwen3.5 visual-encoder keys on save and wrong token IDs across tokenizers are not the same bug." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Unexpected position-id keys on load and sinq quantization save/load breakage are different serialization problems." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both concern tied weights in v5, but one is device-dependent lm_head serialization and the other is translation-model tied-weight issues; not clearly the same fix." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "Packed-weight loading for BitNet AutoBitLinear and Perceiver interpolate_pos_encoding failure are unrelated." + }, + { + "left": "issue:45200", + "right": "issue:46463", + "accept": false, + "reason": "Gemma 4 mm_token_type_ids defaults and T5 decoder validation are different model-family bugs." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Gemma3 DeepSpeed ZeRO-3 loading failure and trainer.evaluate-after-train failure are not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 25, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:29159", + "issue:33357", + "issue:38175", + "issue:42898", + "issue:43295", + "issue:43452", + "issue:43525", + "issue:43531", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43761", + "issue:43819", + "issue:43927", + "issue:43994", + "issue:44361", + "issue:44442", + "issue:44464", + "issue:44568", + "issue:44743", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44857", + "issue:44898", + "issue:44991", + "issue:45356", + "issue:45372", + "issue:45375", + "issue:45542", + "issue:45701", + "issue:45704", + "issue:46170", + "issue:46178", + "issue:46402", + "issue:46491", + "issue:46701", + "issue:46821", + "issue:46829", + "issue:46832", + "issue:46932", + "issue:47000" + ], + "soft_pairs": [ + "issue:43643|issue:45375", + "issue:38175|issue:43295", + "issue:43688|issue:43749", + "issue:44361|issue:44442", + "issue:43653|issue:44361", + "issue:43452|issue:44843", + "issue:43819|issue:44811", + "issue:43994|issue:44898", + "issue:43531|issue:43742", + "issue:44857|issue:47000", + "issue:45372|issue:46170", + "issue:43761|issue:46829", + "issue:29159|issue:33357", + "issue:43525|issue:46701", + "issue:45542|issue:46178", + "issue:46832|issue:46932", + "issue:43688|issue:45704", + "issue:43295|issue:45356", + "issue:43688|issue:45701", + "issue:43653|issue:44568", + "issue:43646|issue:43749", + "issue:43756|issue:46402", + "issue:43452|issue:43577", + "issue:44464|issue:44743", + "issue:43653|issue:43927", + "issue:44821|issue:46821", + "issue:42898|issue:46491", + "issue:44843|issue:44991" + ], + "prepared_review_unit_hash": "ce42da09ac859b7a1cb1420687b3ac27371344c086dbefad23c0a26cb9e0f6dd", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35725, + "estimated_input_tokens": 8932, + "estimated_eval_tokens": 18120 + }, + "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 bug reports across tokenizers, vision/audio models, config loading, and backend edge cases. The soft pairs share broad themes at best, but none look like the same concrete bug or change.", + "confidence": 0.47, + "canonical_issue_reason": "Issue #46491 is the clearest, most concrete tokenizer regression here: it has a specific model, a versioned regression vs v4, and a reproducible encode\u2192decode symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #46491 is the best representative issue because it is open, narrowly scoped, and describes a user-facing tokenizer behavior regression with a clear comparison point.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both involve config fields being dropped, but one is trust_remote_code returning incomplete configs and the other is a specific Qwen3_5MoeVisionConfig field being stripped; different root causes." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Siglip2 zero probabilities and processor.tokenizer/image-passing regression are unrelated model-path failures." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Aux-loss normalization in MoE models is unrelated to the FSDP_CPU_RAM_EFFICIENT_LOADING breakage." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer load failures affecting different models and failure modes." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird special-token registration and MLukeTokenizer AttributeError are different tokenizer bugs." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both touch AutoTokenizer loading, but one is gguf_file support and the other is an offline-mode failure in _patch_mistral_regex; not the same bug." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC.from_latents/STE mismatch is unrelated to Whisper batch_decode skip_special_tokens behavior." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-model issues, but they are different models and different output/runtime problems." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding_window and MobileLLM key error are distinct model-loading/runtime bugs." + }, + { + "left": "issue:44857", + "right": "issue:47000", + "accept": false, + "reason": "Both are detection-loss crashes, but they fail in different code paths and with different conditions." + }, + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Gemma processor import breakage and Windows ROCm torch.distributed dependency issues are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:46829", + "accept": false, + "reason": "Both mention hidden_states, but they concern different models and different incorrect outputs." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer inconsistency vs MacOS bus error are not the same underlying bug." + }, + { + "left": "issue:43525", + "right": "issue:46701", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config is unrelated to DiffusionGemmaGenerationOutput pipeline failure." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Different backend-registration problems with different offending packages and symptoms." + }, + { + "left": "issue:46832", + "right": "issue:46932", + "accept": false, + "reason": "RT-DETR backbone/feature-level crash and Gemma4 dtype-casting fix are unrelated." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "MoE auxiliary-loss normalization and T5 apex RMSNorm memory leak are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Custom processor.tokenizer regression and Kimi-K2.5 codec/mistral-regex regression are different bugs." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "A specific aux-loss normalization bug is not the same as a broad tokenization-change report." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Different tokenizers and different special-token failures; not one concrete defect." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Custom model initialization breakage and FSDP CPU RAM efficient loading failure are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "gguf_file loading failure and BLIP2 dtype mismatch affect different code paths." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Both concern generation/cache behavior, but the models and mechanics differ; not enough to call them the same bug." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "Special-token registration in a tokenizer and DiaConfig losing custom token IDs after save/load are different layers and symptoms." + }, + { + "left": "issue:44821", + "right": "issue:46821", + "accept": false, + "reason": "AutoImageProcessor URL loading and TimesFM window_size AttributeError are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:46491", + "accept": false, + "reason": "Both involve tokenizer behavior changes, but one is clean_up_tokenization_spaces semantics and the other is CodeLlama leading-space round-trip regression; different defects." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline model_info() call in mistral regex patch and EMBEDDIA/est-roberta tokenizer loading failure do not appear to be the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 26, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:32090", + "issue:33666", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43611", + "issue:43643", + "issue:43688", + "issue:43720", + "issue:43761", + "issue:43782", + "issue:43825", + "issue:43927", + "issue:44361", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44484", + "issue:44514", + "issue:44554", + "issue:44561", + "issue:44610", + "issue:44625", + "issue:44704", + "issue:44734", + "issue:44964", + "issue:44977", + "issue:45290", + "issue:45405", + "issue:45440", + "issue:45491", + "issue:45584", + "issue:46002", + "issue:46050", + "issue:46612", + "issue:46693", + "issue:46846", + "issue:47013" + ], + "soft_pairs": [ + "issue:44479|issue:44625", + "issue:33666|issue:36010", + "issue:43761|issue:47013", + "issue:44466|issue:46050", + "issue:32090|issue:33666", + "issue:44484|issue:44704", + "issue:41720|issue:44464", + "issue:43688|issue:45440", + "issue:44977|issue:46693", + "issue:44464|issue:44977", + "issue:44734|issue:46612", + "issue:43504|issue:43550", + "issue:33666|issue:36331", + "issue:43262|issue:43295", + "issue:43278|issue:43688", + "issue:43504|issue:43720", + "issue:43782|issue:46693", + "issue:45491|issue:46002", + "issue:43611|issue:46846", + "issue:43577|issue:44554", + "issue:43720|issue:43927", + "issue:43606|issue:43927", + "issue:43504|issue:44610", + "issue:44514|issue:45290", + "issue:44964|issue:45405", + "issue:43525|issue:43572", + "issue:43643|issue:44561", + "issue:43720|issue:44361", + "issue:44625|issue:45584", + "issue:43493|issue:43825" + ], + "prepared_review_unit_hash": "e6256b7cecee1fbdacd78036a3da22186db4a0af51a61c5b44deb7f5d4606bc2", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36968, + "estimated_input_tokens": 9242, + "estimated_eval_tokens": 18740 + }, + "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/regressions across different models and subsystems. The soft similarities are mostly superficial (shared keywords like Qwen, processor, save/load, FA2, or NaNs) rather than the same underlying defect.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 46693 is the most central active report in the set: it has the most discussion, inbound references, and describes a broad generation regression with clear ongoing impact.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46693 is the best representative issue here because it is active, well-discussed, and broadly scoped compared with the mostly isolated model-specific bugs in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is a video-input regression and the other is a num_labels propagation bug; different code paths and symptoms." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Multi-GPU training for Qwen2-VL and a GenerationMixin import error are unrelated failures." + }, + { + "left": "issue:43761", + "right": "issue:47013", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and xLSTM train-mode unpacking crash are different model paths and bugs." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Both involve save/load behavior, but one is tied-weight serialization and the other is quantized sinq state restoration." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer GPU broadcast TypeError and Qwen2-VL multi-GPU training are not the same defect." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "A max_shard_size question and an AutoProcessor kwarg-passing bug are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Auto device mapping cudaErrorAssert and compiled chunked generation inconsistency are different failures." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Both are model-fidelity bugs, but they affect different architectures and different internal computations." + }, + { + "left": "issue:44977", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3.5 flash-attention correctness bug and FA2 long-context performance regression are not the same issue." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Compiled-forward chunked generation mismatch is distinct from the Qwen3.5 flash-attention bug." + }, + { + "left": "issue:44734", + "right": "issue:46612", + "accept": false, + "reason": "KV-cache continuation tensor indexing in serve and skipped beam-search cache reorder in generation are different code paths." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Beit legacy-field loading failure and Bamba torch.compile SDPA failure are unrelated." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and CustomTrainer compute_loss signature breakage are different training issues." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio processor sample-rate defaulting and processor.tokenizer/image-passing regression are separate processor bugs." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype mismatch during eval and auxiliary-loss normalization are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Legacy field loading in Beit and packed-weight unpacking in BitNet are different loading bugs." + }, + { + "left": "issue:43782", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3VL weight_only loading error and FA2 performance regression do not share the same underlying cause." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Both mention NaNs, but one is a Gemma3 sliding-window attention bug and the other is a classification-head init/checkpoint issue." + }, + { + "left": "issue:43611", + "right": "issue:46846", + "accept": false, + "reason": "base_model_prefix loading regression and Qwen3.5 TP-plan OOM/shape errors are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "Blip2 dtype loading mismatch and MPS attention correctness with mismatched head dims are different bugs." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "BitNet accelerate loading issue and DiaConfig token-ID loss after save/load are unrelated serialization/config bugs." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU offload device mismatch and DiaConfig field loss after save/load are not the same failure." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Beit legacy field loading and OmDet-Turbo processor size mismatch are distinct model/processor problems." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both are apply_chat_template crashes, but one is batched padding=False handling and the other is assistant tool-call content handling." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Phi-4 multimodal loading failure and an unreleased PEFT version pin are unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config and missing pad_token_idx in StableLmConfig are different model/config regressions." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both involve trust_remote_code, but one is missing config fields and the other is removal of is_torch_fx_available." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet packed-weight loading and MLukeTokenizer task AttributeError are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription generation are unrelated bugs." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task messaging 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": 26, + "review_unit_index": 27, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28218", + "issue:33357", + "issue:33666", + "issue:39692", + "issue:42796", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43452", + "issue:43493", + "issue:43504", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43701", + "issue:43824", + "issue:43844", + "issue:43927", + "issue:43994", + "issue:44291", + "issue:44361", + "issue:44451", + "issue:44466", + "issue:44488", + "issue:44554", + "issue:44568", + "issue:44589", + "issue:44792", + "issue:44843", + "issue:44849", + "issue:44977", + "issue:45245", + "issue:45357", + "issue:45405", + "issue:45520", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46050", + "issue:46192", + "issue:46209", + "issue:46365", + "issue:46489", + "issue:46710", + "issue:46821", + "issue:46832" + ], + "soft_pairs": [ + "issue:44977|issue:46209", + "issue:44554|issue:46365", + "issue:45823|issue:46050", + "issue:28218|issue:33357", + "issue:43452|issue:44451", + "issue:43994|issue:46192", + "issue:39692|issue:43295", + "issue:42796|issue:44977", + "issue:43493|issue:43844", + "issue:43278|issue:43844", + "issue:43646|issue:44466", + "issue:43278|issue:43701", + "issue:46209|issue:46710", + "issue:44568|issue:46489", + "issue:33357|issue:33666", + "issue:44849|issue:46209", + "issue:44792|issue:45405", + "issue:43927|issue:44361", + "issue:43575|issue:45405", + "issue:44568|issue:45812", + "issue:43824|issue:44589", + "issue:44488|issue:44843", + "issue:44849|issue:45357", + "issue:43572|issue:44291", + "issue:44589|issue:45520", + "issue:46821|issue:46832", + "issue:43504|issue:43606", + "issue:43334|issue:45759", + "issue:45245|issue:46710" + ], + "prepared_review_unit_hash": "235acbfdb2590e4eb7b2fb81aa37eaff3b5c6a5acd590d1e32b35af6addf3061", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35930, + "estimated_input_tokens": 8983, + "estimated_eval_tokens": 18222 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly heterogeneous; a few pairs share broad areas like tokenization or model loading, but none are close enough to be true duplicates or the same fix scope.", + "confidence": 0.93, + "canonical_issue_reason": "No single canonical issue is justified across this cluster because the items span unrelated bugs (tokenizers, multimodal loading, training/runtime failures, and model-specific regressions).", + "canonical_pr_reason": null, + "best_issue_reason": "If one issue must represent the most reusable triage anchor, issue 45812 is the strongest candidate because it is a broad, high-impact tokenizer regression with multiple references and nearby tokenizer-related reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Both are inference regressions, but one is Qwen3.5 flash-attention behavior and the other is FP8/B200 support; different code paths and fixes." + }, + { + "left": "issue:44554", + "right": "issue:46365", + "accept": false, + "reason": "MPS attention correctness and Trainer import failure are unrelated subsystems and failure modes." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Device-map forward mismatch and quantized save/load breakage are different bugs with different triggers." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer spacing bug and MacOS CLIP bus error are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "Both concern model loading, but one is gguf/from_pretrained plumbing and the other is a specific ScandiBERT load failure." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 wrong outputs and ignored return_tensors are different API bugs." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Documentation example errors and custom processor/tokenizer regression are not the same underlying issue." + }, + { + "left": "issue:42796", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer BPE-marker handling and Qwen3.5 flash-attention generation are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 implementation mismatch and DeepSpeed ZeRO-3 gradient growth are distinct model/training problems." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype change and gradient explosion are different training behaviors with different likely fixes." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage and tied-weight serialization inconsistency are separate regressions." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Dtype drift and resume_from_checkpoint key mismatch are unrelated." + }, + { + "left": "issue:46209", + "right": "issue:46710", + "accept": false, + "reason": "FP8 inference on B200 and DeepSeek accuracy regression after upgrade are not the same bug." + }, + { + "left": "issue:44568", + "right": "issue:46489", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different tokenization paths; not the same fix." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS bus error with CLIP and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Hidden-state output bug and FP8 inference failure are unrelated." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "A failing Janus image-generation test and an unreleased PEFT version bump do not share an underlying bug." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig save/load token ID loss and MLukeTokenizer AttributeError are different tokenizer/config issues." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM and dependency version bump are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both involve tokenizers, but one is missing special tokens for mdeberta and the other is wrong IDs for Granite; different concrete defects." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "ImportError for Qwen2.5-VL and missing Float8 storage are unrelated." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Specific model load failure and offline-mode tokenizer patch bug are not the same root cause." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Hidden-state behavior and incorrect visual-encoder key serialization are distinct Qwen3.5 issues." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Missing StableLmConfig field and unexpected init_empty_weights argument are different config/init regressions." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure and flash_attn import mapping error are unrelated." + }, + { + "left": "issue:46821", + "right": "issue:46832", + "accept": false, + "reason": "TimesFM window_size AttributeError and RT-DETR feature-level mismatch are different model-specific failures." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Legacy preset loading in Beit and CPU-offload device mismatch in bark-small are unrelated." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_id in Qwen3-VL config and text_config unwrapping in composite configs are different loading bugs." + }, + { + "left": "issue:45245", + "right": "issue:46710", + "accept": false, + "reason": "Category-limit runtime error and DeepSeek accuracy regression are not the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 28, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:29942", + "issue:30333", + "issue:32090", + "issue:34567", + "issue:38175", + "issue:42490", + "issue:43065", + "issue:43257", + "issue:43299", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43572", + "issue:43606", + "issue:43653", + "issue:43696", + "issue:43701", + "issue:43844", + "issue:43873", + "issue:43937", + "issue:43994", + "issue:44112", + "issue:44464", + "issue:44493", + "issue:44554", + "issue:44869", + "issue:44964", + "issue:45072", + "issue:45084", + "issue:45200", + "issue:45229", + "issue:45446", + "issue:45520", + "issue:45584", + "issue:45588", + "issue:45823", + "issue:46050", + "issue:46285", + "issue:46369", + "issue:46399", + "issue:46701", + "issue:46752", + "issue:46832", + "issue:46846" + ], + "soft_pairs": [ + "issue:44493|issue:46285", + "issue:43257|issue:43299", + "issue:43334|issue:43572", + "issue:43504|issue:43653", + "issue:43606|issue:44112", + "issue:46399|issue:46846", + "issue:45072|issue:46832", + "issue:43873|issue:46050", + "issue:44464|issue:45584", + "issue:43994|issue:44493", + "issue:43844|issue:44964", + "issue:29159|issue:32090", + "issue:45588|issue:46050", + "issue:44869|issue:46752", + "issue:45229|issue:45823", + "issue:42490|issue:43696", + "issue:45200|issue:45229", + "issue:29942|issue:30333", + "issue:43937|issue:46701", + "issue:45084|issue:45520", + "issue:28218|issue:34567", + "issue:43065|issue:46369", + "issue:15354|issue:38175", + "issue:43493|issue:43701", + "issue:44554|issue:45446" + ], + "prepared_review_unit_hash": "e1276c5024fc040ff737e45d9263d903ad07c515066eceef4246574156db2b07", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 25, + "serialized_chars": 34579, + "estimated_input_tokens": 8645, + "estimated_eval_tokens": 17546 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The proposed pairs are mostly near-matches only at a broad subsystem level (tokenizers, loading, attention, quantization, training), but they describe different concrete failures and fixes. I would not merge any of these as duplicates; the cluster is heterogeneous rather than one bug/change.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44493", + "right": "issue:46285", + "accept": false, + "reason": "Both mention token-related behavior, but one is about unexpected position-id keys across models while the other is a CLIP text-model structural regression. Different symptoms and code paths." + }, + { + "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 transformers dev config breakage (`pad_token_id`). Same family, different bug." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "These are separate missing-config-field issues in different models (`pad_token_id` vs `pad_token_idx`). Similar shape, not the same underlying defect." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Beit pretrained load failure from a legacy field vs BigBirdTokenizer special-token registration causing empty decode. Different components and fixes." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "CPU offload device mismatch in inference is not the same as a stale CI test failure in GraniteSpeech. Same general area, unrelated bugs." + }, + { + "left": "issue:46399", + "right": "issue:46846", + "accept": false, + "reason": "One is a missing decorator for `apply_rotary_pos_emb` registration; the other is a missing tensor-parallel plan entry for `linear_attn`. Distinct issues." + }, + { + "left": "issue:45072", + "right": "issue:46832", + "accept": false, + "reason": "bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated to RTDetr feature-level shape/type errors." + }, + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Both touch quantization/offloading, but one is offloading not working as expected and the other is save/load corruption after `sinq` quantization. Not the same code-path bug." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Compiled-forward chunked-generation inconsistency is unrelated to Whisper empty-transcription failure after `align_special_tokens`." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "Nonsensical SigLIP2 outputs and unexpected `position_id` keys are different model behaviors with different causes." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient growth during training is unrelated to a model-loading failure for Phi-4 multimodal." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer slow/fast inconsistency and `_gpu_broadcast_one` NoneType errors are entirely different subsystems and failures." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "Flash-attention `s_aux=None` crash and sinq quantized model save/load breakage are not the same bug." + }, + { + "left": "issue:44869", + "right": "issue:46752", + "accept": false, + "reason": "Whisper timestamp decode crash and empty chat-template input `IndexError` are unrelated edge-case crashes." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4, but one is multi-GPU OOM and the other is a device-map mismatch during forward. Different fixes." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Local-path import warning with trailing slash is unrelated to GPT-oss-20b CUDA OOM." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Text-only `mm_token_type_ids` defaulting to zeros and Gemma4 multi-GPU OOM are distinct problems." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 test failures and MLFlow job-status reporting after training failure are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:46701", + "accept": false, + "reason": "Invalid `GenerationConfig` for GLM-5 and DiffusionGemma pipeline failure are different model/integration issues." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Compilation of non-template nodes and Python 3.13 `flash_attn` import handling are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing after added tokens and `num_input_tokens_seen` not updating in TrainerState are different bugs." + }, + { + "left": "issue:43065", + "right": "issue:46369", + "accept": false, + "reason": "A dummy Conv2d in Sam3PixelDecoder is not the same as memory overflow in sam3-litetext; one is architecture scaffolding, the other is runtime memory pressure." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "torch.jit.script GeneratorExp unsupported during export and SigLIP2 zero-probability outputs are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 HF-vs-JAX discrepancy and checkpoint-resume key mismatch are different classes of problems." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "MPS attention correctness when value head dim differs from query is unrelated to the PyTorch version check bug for AuxRequest 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": 26, + "review_unit_index": 29, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28218", + "issue:29870", + "issue:29942", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43232", + "issue:43295", + "issue:43450", + "issue:43493", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43957", + "issue:44186", + "issue:44423", + "issue:44466", + "issue:44521", + "issue:44610", + "issue:44734", + "issue:44849", + "issue:44977", + "issue:44987", + "issue:45072", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45381", + "issue:45406", + "issue:45520", + "issue:45588", + "issue:45759", + "issue:45823", + "issue:46121", + "issue:46178", + "issue:46192", + "issue:46209", + "issue:46701", + "issue:46710", + "issue:46821" + ], + "soft_pairs": [ + "issue:41720|issue:46710", + "issue:44423|issue:44734", + "issue:44977|issue:46701", + "issue:37428|issue:43957", + "issue:43295|issue:43450", + "issue:44977|issue:46710", + "issue:45072|issue:45198", + "issue:45406|issue:46121", + "issue:45588|issue:45759", + "issue:43540|issue:45200", + "issue:43295|issue:43873", + "issue:43572|issue:46821", + "issue:44186|issue:46192", + "issue:43295|issue:44610", + "issue:45823|issue:46192", + "issue:43450|issue:43575", + "issue:43550|issue:43844", + "issue:29870|issue:29942", + "issue:45520|issue:46178", + "issue:43572|issue:44987", + "issue:37428|issue:43824", + "issue:45200|issue:45823", + "issue:42490|issue:43493", + "issue:39692|issue:46192", + "issue:43232|issue:44849", + "issue:43295|issue:43493", + "issue:44521|issue:45290", + "issue:42915|issue:43824", + "issue:45381|issue:46209", + "issue:44466|issue:45127", + "issue:28218|issue:35141" + ], + "prepared_review_unit_hash": "d9ea94e91d2f64327877bcbdf5f4f2ea116f3b2a68d28e31f338ca24c3415396", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37312, + "estimated_input_tokens": 9328, + "estimated_eval_tokens": 18912 + }, + "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 similarities are usually at the subsystem/model-family level, but the underlying failures differ, so none of the proposed edges should be merged.", + "confidence": 0.97, + "canonical_issue_reason": "issue:43295 is the best representative because it is the most connected soft hub and describes a broad regression in multimodal processor behavior, but the cluster is not a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43295 is the strongest global representative: broad impact, clear regression framing, and multiple soft links to nearby multimodal-processing reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41720", + "right": "issue:46710", + "accept": false, + "reason": "Different failures: auto device-mapping CUDA assert vs an accuracy regression after upgrade; same broad model usage, not the same bug." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both touch transformers serve, but one is a multimodal continuous-batching crash and the other is a KV-cache continuation indexing bug." + }, + { + "left": "issue:44977", + "right": "issue:46701", + "accept": false, + "reason": "Both involve generation/output issues, but they are on different model families and different code paths." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "ImportError for a missing flash-attention helper is unrelated to meta-device model loading breakage." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "One is a processor.tokenizer regression in custom multimodal code; the other is batched video processor shape handling." + }, + { + "left": "issue:44977", + "right": "issue:46710", + "accept": false, + "reason": "Same model family mention is not enough; one is flash-attention generation breakage, the other is an accuracy regression after version upgrade." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "Completely different models and symptoms: bfloat16 dtype mismatch vs save/tokenization failure." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Processor missing _tokenizer is unrelated to rope-validation type handling for a JSON-loaded list." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Flash-attention sink handling and unwrapping text_config are separate code-path issues." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Both are multimodal-adjacent, but one is a video-processing ValueError and the other is a missing default for mm_token_type_ids." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "A processor/tokenizer regression and quantization offloading behavior are not the same underlying bug." + }, + { + "left": "issue:43572", + "right": "issue:46821", + "accept": false, + "reason": "Unrelated config/API issues in different model families: StableLm pad_token_idx vs TimesFM window_size." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "Tokenizer crash on NER/batched padding is distinct from return_tensors being ignored when text_kwargs is explicitly passed." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Processor/tokenizer regression and OmDet-Turbo image-size mismatch are different bugs in different models." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Gemma4 device-map mismatch is unrelated to return_tensors being dropped in kwargs handling." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Video input shape handling and tensor-parallel OOM are different failure modes." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "torch.compile + SDPA failure is unrelated to ZeRO-3 gradient abnormality." + }, + { + "left": "issue:29870", + "right": "issue:29942", + "accept": false, + "reason": "Bart generate mask bug and Flash Attention 2 test failures do not share the same code path." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Both are import/backend errors, but they involve different missing backends and different failure mechanisms." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "StableLm config regression and a transformers>=5.1.0 loading failure are not the same bug." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Different import problems: missing flash_attention helper vs missing Qwen2_5_VL class export." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Both concern multimodal Gemma4, but one is a missing default tensor-type field and the other is a forward-device mismatch." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "A path-trailing-slash import warning is unrelated to a SigLIP2 implementation discrepancy." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "Documentation example errors are not the same as kwargs/return_tensors handling in preprocessing." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation helper state update after sync_gpus is a different issue from a Qwen3.5 hidden-states bug." + }, + { + "left": "issue:43295", + "right": "issue:43493", + "accept": false, + "reason": "Processor/tokenizer regression and SigLIP2 HF-vs-JAX discrepancy are separate multimodal problems." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Same apply_chat_template API, but different edge cases: multimodal assistant masks vs tool-call messages with no content." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Qwen2.5-VL import error are unrelated." + }, + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Qwen2.5-VL video position ids and FP8 inference on B200 are different model/runtime problems." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied weights, but one is inconsistent serialization and the other is post-LoRA model collapse after vocab extension." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer added-space behavior and post_init embedding reinitialization are distinct 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": 26, + "review_unit_index": 30, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:38175", + "issue:41720", + "issue:43262", + "issue:43278", + "issue:43334", + "issue:43335", + "issue:43388", + "issue:43450", + "issue:43452", + "issue:43525", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43827", + "issue:43856", + "issue:43901", + "issue:43950", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44442", + "issue:44561", + "issue:44610", + "issue:44737", + "issue:44792", + "issue:44857", + "issue:44898", + "issue:45083", + "issue:45198", + "issue:45200", + "issue:45325", + "issue:45412", + "issue:45563", + "issue:45584", + "issue:45593", + "issue:46192", + "issue:46209", + "issue:46234", + "issue:46285", + "issue:46512", + "issue:46821", + "issue:46894", + "issue:46899", + "issue:46932" + ], + "soft_pairs": [ + "issue:43278|issue:43856", + "issue:43824|issue:43975", + "issue:44737|issue:46894", + "issue:43452|issue:46512", + "issue:43653|issue:44112", + "issue:43262|issue:43450", + "issue:43950|issue:46285", + "issue:44442|issue:45198", + "issue:43334|issue:43525", + "issue:45200|issue:46899", + "issue:43526|issue:46192", + "issue:44898|issue:45412", + "issue:45200|issue:46932", + "issue:44792|issue:45563", + "issue:43819|issue:44112", + "issue:45200|issue:45325", + "issue:43335|issue:43550", + "issue:44561|issue:46234", + "issue:41720|issue:46209", + "issue:44610|issue:44857", + "issue:43827|issue:43901", + "issue:43819|issue:45083", + "issue:44857|issue:45593", + "issue:43388|issue:43526", + "issue:46512|issue:46821", + "issue:43606|issue:45083", + "issue:44206|issue:45584", + "issue:38175|issue:45412" + ], + "prepared_review_unit_hash": "f3b3fae81bcf8e44948ee7e17505b443c5aa77afc2560d6bcc0b58703fc8a178", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35831, + "estimated_input_tokens": 8958, + "estimated_eval_tokens": 18172 + }, + "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. The soft links are superficial theme matches (same model family, same subsystem, or both docs/runtime regressions) rather than true duplicates, so all soft edges should be rejected.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 45200 is the closest thing to a hub here: it is a concrete, reproducible Gemma4 multimodal bug and appears in multiple soft-similarity links. That said, it is not a true representative of the whole cluster because the rest of the items span unrelated models and code paths.", + "canonical_pr_reason": null, + "best_issue_reason": "45200 is the best single issue to anchor triage because it is specific and well-scoped, but the cluster is too heterogeneous for it to serve as a meaningful global canonical beyond its own Gemma4 area.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Both mention model internals, but one is a dtype regression in eval and the other is Qwen3 MoE training memory usage; different bugs and fixes." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "Import/export symbol resolution for Qwen2.5-VL is unrelated to a tokenizer detokenization regression in deepseek-coder." + }, + { + "left": "issue:44737", + "right": "issue:46894", + "accept": false, + "reason": "XLNet CPU placement in relative positions and CohereASR double-shift training loss are unrelated code paths." + }, + { + "left": "issue:43452", + "right": "issue:46512", + "accept": false, + "reason": "gguf_file loading breaks model/tokenizer factories; the other issue is stale config.dtype after from_config. Not the same bug." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and a stale CI device-override test in GraniteSpeech are different problems." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Both are processor bugs, but one is audio sampling-rate defaults and the other is batched video shape handling; separate fixes." + }, + { + "left": "issue:43950", + "right": "issue:46285", + "accept": false, + "reason": "Non-persistent buffer loading corruption and CLIP text-model structural inconsistency are not the same underlying issue." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "Tokenizer loading for FastSpeech2Conformer is different from Wav2Vec2 save_pretrained/tokenization failure." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both involve missing pad_token_id, but they are separate model/config classes and likely distinct fixes." + }, + { + "left": "issue:45200", + "right": "issue:46899", + "accept": false, + "reason": "Both are Gemma4 multimodal issues, but one is token-type-id defaulting and the other is unsafe dtype casting under 4-bit quantization; different code paths." + }, + { + "left": "issue:43526", + "right": "issue:46192", + "accept": false, + "reason": "reduce_labels returning one label and return_tensors being ignored are unrelated preprocessing bugs." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver resolution/interpolation failure and RT-DETR memory not being released are unrelated." + }, + { + "left": "issue:45200", + "right": "issue:46932", + "accept": false, + "reason": "Both are Gemma4-related, but this is about mm_token_type_ids defaults while the other is about dtype casting in quantized embedders." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "A failed image-generation test and a stale num_return_sequences warning in paged generate are different symptoms and fixes." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC forward/latent mismatch is unrelated to a GraniteSpeech CI stale test." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Gemma4 token-type defaults and Qwen2.5-VL rope temporal position scaling are unrelated multimodal bugs." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction and Bamba torch.compile SDPA failure are unrelated." + }, + { + "left": "issue:44561", + "right": "issue:46234", + "accept": false, + "reason": "Both concern utility availability, but one is a removed helper breaking trust_remote_code and the other is a missing import_utils symbol; not the same bug." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3 auto device mapping CUDA assert and FP8 inference on B200 are different hardware/runtime failures." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output resolution mismatch and LwDetr AMP/CUDA crash are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs regressions, but they affect different pages and different deprecations; not a duplicate change." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC latent mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "AMP/CUDA crash in LwDetr loss and D-FINE auxiliary-loss omission are different bugs." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Metric gathering drops tuple labels, while BeitImageProcessorFast reduces labels incorrectly; similar domain, different functions and fixes." + }, + { + "left": "issue:46512", + "right": "issue:46821", + "accept": false, + "reason": "Stale model config dtype and a missing window_size attribute in TimesFM are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU offload device mismatch in suno/bark-small is unrelated to qwen3_omni_moe helper behavior." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Unsupported center argument crash in LasrFeatureExtractor and empty-transcription Whisper generation failure are different issues." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "Siglip2 zero probabilities and RT-DETR memory release 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": 26, + "review_unit_index": 31, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28218", + "issue:28282", + "issue:29127", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:42490", + "issue:43278", + "issue:43335", + "issue:43425", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43720", + "issue:43819", + "issue:43825", + "issue:43854", + "issue:44112", + "issue:44351", + "issue:44448", + "issue:44466", + "issue:44493", + "issue:44514", + "issue:44521", + "issue:44743", + "issue:44855", + "issue:44857", + "issue:45083", + "issue:45290", + "issue:45520", + "issue:45588", + "issue:45593", + "issue:45759", + "issue:45859", + "issue:46170", + "issue:46285", + "issue:46326", + "issue:46424", + "issue:46431", + "issue:46566", + "issue:46858" + ], + "soft_pairs": [ + "issue:37428|issue:43425", + "issue:45759|issue:46285", + "issue:43335|issue:43653", + "issue:46424|issue:46858", + "issue:28282|issue:33357", + "issue:29942|issue:32090", + "issue:43611|issue:44466", + "issue:43335|issue:43504", + "issue:43550|issue:44112", + "issue:44855|issue:45083", + "issue:45520|issue:46170", + "issue:44448|issue:44743", + "issue:37428|issue:44351", + "issue:43335|issue:45083", + "issue:43335|issue:44112", + "issue:43335|issue:45593", + "issue:45290|issue:46326", + "issue:43278|issue:43825", + "issue:44514|issue:44521", + "issue:34689|issue:35141", + "issue:45588|issue:45859", + "issue:42490|issue:43854", + "issue:29127|issue:32090", + "issue:43819|issue:44493", + "issue:39692|issue:43526", + "issue:46431|issue:46566", + "issue:43606|issue:44857", + "issue:43720|issue:44112", + "issue:39692|issue:44493", + "issue:28218|issue:32090" + ], + "prepared_review_unit_hash": "1faecc408d582c50e0134c3539de917fa374018341508743406c7d34f5abd7a4", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37137, + "estimated_input_tokens": 9285, + "estimated_eval_tokens": 18826 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This looks like a noisy semantic cluster rather than a duplicate set: the items span many unrelated Transformers subsystems, so there is no single underlying bug/change to collapse onto.", + "confidence": 0.97, + "canonical_issue_reason": "No clear canonical duplicate: the issues are only loosely related by being bug reports, but they cover different code paths (tokenization, model loading, cache, chat templates, CI, etc.).", + "canonical_pr_reason": null, + "best_issue_reason": "No strong best issue for deduping; none of these is a reliable representative of the others because the cluster is not cohesive.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "Different failures: missing flash-attention helper import vs general Torch 2.10 incompatibility." + }, + { + "left": "issue:45759", + "right": "issue:46285", + "accept": false, + "reason": "Different model paths: Qwen multimodal config unwrapping vs CLIP text-model structure inconsistency." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Unrelated bugs: SwitchTransformers sparse-layer config vs BigBird mask-token registration." + }, + { + "left": "issue:46424", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache, but one is max_cache_length handling and the other is device-tensor seq_length arithmetic under static cache/compile." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "ImportError for missing PyTorch vs MacOS bus error loading a community CLIP model." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures vs Trainer GPU broadcast NoneType; not the same code path." + }, + { + "left": "issue:43611", + "right": "issue:44466", + "accept": false, + "reason": "Different loading/serialization regressions: base_model_prefix handling vs device-dependent lm_head.weight serialization." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Different causes: sparse layer creation with zero layers vs legacy field blocking BEiT pretrained loading." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "torch.compile SDPA failure in Bamba vs stale device-override CI test in GraniteSpeech." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "TorchScript parsing/import error vs helper returning unexpected feature lengths in qwen3_omni_moe." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "Different environment issues: flash_attn import key error on Python 3.13 vs Windows ROCm/torch.distributed dependency problems." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Pegasus v4/v5 output difference vs recurrent states resetting in modular_qwen3_5; not the same bug." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports: _flash_supports_window_size vs HybridCache." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Sparse-layer config bug and qwen3_omni_moe length helper bug are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "No shared concrete failure: config construction bug vs CI test flake." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers config bug vs D-FINE auxiliary-loss behavior; different subsystems." + }, + { + "left": "issue:45290", + "right": "issue:46326", + "accept": false, + "reason": "Both are chat-template related, but one crashes on tool-call messages and the other is continue_final_message on mistral_common tokenizers." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype drift during eval vs pipeline error message about unsupported translation tasks in v5." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both touch apply_chat_template, but one is a batched padding=False crash and the other is all-zero assistant masks for multimodal inputs." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Model loading regression for Llama 3.2 vision vs embedding reinitialization after resize; different mechanisms." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Flash-attention sink-less AttributeError vs Qwen3 MoE rotary embedding CPU-offload incompatibility." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Local-path trailing-slash warning vs inability to load a specific model in unit tests; not the same bug." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMv3 error-message clarity issue vs Trainer broadcast NoneType failure." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC from_latents / STE mismatch vs unexpected position-id keys; distinct code paths." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "SigLIP2 documentation example problems vs BeitImageProcessorFast reduce_labels bug." + }, + { + "left": "issue:46431", + "right": "issue:46566", + "accept": false, + "reason": "Processor missing chat template vs DiffusionGemma staying on meta with device_map=auto." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU-offload device mismatch in bark-small vs float16 AMP crash in LwDetrImageLoss." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "BitNet packed-weights loading issue vs GraniteSpeech CI stale override test." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "Documentation/example failures vs position-id key mismatch; unrelated." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer extra-space bug vs GPU broadcast NoneType error in Trainer." + } + ] + }, + "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": 26, + "review_unit_index": 32, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:26413", + "issue:27957", + "issue:29870", + "issue:29942", + "issue:31501", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:42175", + "issue:42490", + "issue:43278", + "issue:43329", + "issue:43381", + "issue:43550", + "issue:43575", + "issue:43701", + "issue:44206", + "issue:44360", + "issue:44448", + "issue:44464", + "issue:44493", + "issue:44561", + "issue:44617", + "issue:44779", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45127", + "issue:45245", + "issue:45290", + "issue:45310", + "issue:45381", + "issue:45459", + "issue:45478", + "issue:45593", + "issue:45823", + "issue:46234", + "issue:46424", + "issue:46655", + "issue:46697", + "issue:47000" + ], + "soft_pairs": [ + "issue:45310|issue:45478", + "issue:44448|issue:44779", + "issue:44206|issue:45823", + "issue:43701|issue:45083", + "issue:44617|issue:46697", + "issue:26413|issue:31501", + "issue:45127|issue:45245", + "issue:45459|issue:46234", + "issue:42490|issue:43575", + "issue:32090|issue:33357", + "issue:44857|issue:45072", + "issue:45593|issue:47000", + "issue:38175|issue:44493", + "issue:43550|issue:45083", + "issue:27957|issue:34689", + "issue:44464|issue:46424", + "issue:29942|issue:35141", + "issue:29870|issue:32090", + "issue:45290|issue:45381", + "issue:43278|issue:43329", + "issue:10000|issue:15354", + "issue:29870|issue:33290", + "issue:44448|issue:44977", + "issue:45290|issue:46655", + "issue:37428|issue:42175", + "issue:43278|issue:43381", + "issue:44360|issue:44561" + ], + "prepared_review_unit_hash": "03e336c7da0804860e56454f6f87302d208a65ca0388c3a63b861ab3431001d0", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35692, + "estimated_input_tokens": 8923, + "estimated_eval_tokens": 18102 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a grab-bag of unrelated Transformers issues; only the two Qwen3.5 MoE `from_pretrained` reports look like true duplicates. All other soft pairs are false positives based on broad subsystem overlap or similar error wording.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45310 is the earlier and more complete report of the Qwen3.5 MoE `from_pretrained` failure; issue:45478 is effectively the same bug with a near-identical title.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the best issue anchor in this set because it most clearly identifies the concrete failing model and version-specific regression.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same model, same `transformers>=5.4.0` `from_pretrained` failure, near-identical titles; these look like duplicates." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Both mention output changes across versions, but they concern different models and different failure modes." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "One is a `LasrFeatureExtractor` crash from an unsupported arg; the other is a Gemma4 device-map mismatch." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Both are bug reports, but one is about checkpoint key mismatch and the other about a Qwen3 Omni feature-extraction helper." + }, + { + "left": "issue:44617", + "right": "issue:46697", + "accept": false, + "reason": "CUDA OOM for Sam3Video is unrelated to Swin2SR image over-padding." + }, + { + "left": "issue:26413", + "right": "issue:31501", + "accept": false, + "reason": "Both affect training startup/resume, but the concrete errors and code paths differ." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/vocab-collapse is unrelated to the 2^24 category limit runtime error." + }, + { + "left": "issue:45459", + "right": "issue:46234", + "accept": false, + "reason": "Tokenizer protobuf error handling and a missing `import_utils` symbol are different bugs." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Local-path import warning and tensor-parallel OOM on Qwen2 are unrelated." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "A `NoneType` broadcast failure in Trainer is not the same as a MacOS bus error loading CLIP." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Both involve dtype/precision, but they hit different models and different execution paths." + }, + { + "left": "issue:45593", + "right": "issue:47000", + "accept": false, + "reason": "Different DETR-family losses and different concrete crashes; not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Zero probabilities in SigLIP2 is unrelated to unexpected position-id keys across models." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is unrelated to the Qwen3 Omni feature-length helper bug." + }, + { + "left": "issue:27957", + "right": "issue:34689", + "accept": false, + "reason": "Flash Attention 2 support in XLMRoberta and Llama 3.2 vision model loading are different problems." + }, + { + "left": "issue:44464", + "right": "issue:46424", + "accept": false, + "reason": "Compiled chunked generation inconsistency and cache-length handling are only broadly cache/generation related, not the same defect." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention test failures and embedding reinitialization on resize are unrelated." + }, + { + "left": "issue:29870", + "right": "issue:32090", + "accept": false, + "reason": "Bart generate mask initialization and Trainer GPU broadcast type handling are separate code paths." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat-template tool-call crash and Qwen2.5-VL video position ids are different multimodal bugs." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Eval dtype drift and undefined video-branch variables are unrelated." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBertModel issue and GeneratorExp/JIT export failure are completely different." + }, + { + "left": "issue:29870", + "right": "issue:33290", + "accept": false, + "reason": "Bart generation mask bug and DeepSpeed Adafactor OOM do not share the same underlying defect." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Pegasus v4/v5 output drift and Qwen3.5 flash-attention generation failure are different issues." + }, + { + "left": "issue:45290", + "right": "issue:46655", + "accept": false, + "reason": "Chat template tool-call handling and SmolVLM video padding are unrelated." + }, + { + "left": "issue:37428", + "right": "issue:42175", + "accept": false, + "reason": "Missing flash-attention utility import and TensorFlow not being installed with `.[torch]` are unrelated packaging/import problems." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Eval-mode checkpointing restriction is not the same as BF16-to-FP32 embedding dtype drift." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "A missing ReLU in the DSA indexer and removal of `is_torch_fx_available` are unrelated changes." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The duplicate call on issue:45310/issue:45478 is well supported by the packet: they have the same title and very similar issue bodies, with 45310 earlier in time. The other soft-edge rejections are conservative and consistent with the titles. No clear overstatement beyond a mild inference on completeness." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 33, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29870", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:39692", + "issue:40444", + "issue:41720", + "issue:42490", + "issue:43278", + "issue:43381", + "issue:43526", + "issue:43575", + "issue:43611", + "issue:43701", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:44186", + "issue:44361", + "issue:44448", + "issue:44466", + "issue:44561", + "issue:44568", + "issue:44779", + "issue:44855", + "issue:44871", + "issue:44977", + "issue:45198", + "issue:45229", + "issue:45440", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45803", + "issue:45823", + "issue:45920", + "issue:46121", + "issue:46192", + "issue:46399", + "issue:46421", + "issue:46858" + ], + "soft_pairs": [ + "issue:45920|issue:46192", + "issue:40444|issue:43526", + "issue:44855|issue:46399", + "issue:32090|issue:33290", + "issue:44977|issue:45229", + "issue:43927|issue:45698", + "issue:42490|issue:43873", + "issue:46421|issue:46858", + "issue:43873|issue:45229", + "issue:25251|issue:32090", + "issue:44448|issue:44466", + "issue:43381|issue:43701", + "issue:43976|issue:45440", + "issue:44871|issue:44977", + "issue:29127|issue:34567", + "issue:33357|issue:39692", + "issue:43278|issue:43575", + "issue:45588|issue:45823", + "issue:44561|issue:44779", + "issue:44568|issue:45491", + "issue:41720|issue:45229", + "issue:44186|issue:45198", + "issue:44361|issue:45198", + "issue:45198|issue:45803", + "issue:44977|issue:45563", + "issue:43611|issue:46121", + "issue:29870|issue:34689", + "issue:29942|issue:30064" + ], + "prepared_review_unit_hash": "303e627fb920a70631847f12120ede073148ed9fee7f3400f0f2abc9b109da95", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36377, + "estimated_input_tokens": 9095, + "estimated_eval_tokens": 18446 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs share broad themes like tokenizers, save/load, cache, or device mapping, but they describe different concrete bugs and code paths. None look like safe duplicates to merge.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the strongest representative here: it is active, has multiple inbound references, and describes a concrete quantization/offloading bug. The rest of the items are mostly unrelated regressions or model-specific failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "Both involve tokenizer/processor behavior, but one is wrong token IDs across models and the other is a return_tensors handling bug; different code paths." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "Both are vision/image-processing related, but one is dataset collation for multi-image prompts and the other is Beit label reduction; not the same bug." + }, + { + "left": "issue:44855", + "right": "issue:46399", + "accept": false, + "reason": "Both are import/runtime failures, but one is a Python 3.13 IndentationError in DebertaV2 and the other is a missing decorator for Qwen3.5 kernels; unrelated fixes." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer broadcast NoneType error vs Deepspeed Adafactor OOM are distinct failures with different root causes." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Flash-attention generation failure and multi-GPU CUDA OOM are different problems, not one underlying bug." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both are save/load related, but one loses custom token IDs and the other loads the wrong custom module; different persistence bugs." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Trailing-slash import warning and quantization offloading failure are unrelated." + }, + { + "left": "issue:46421", + "right": "issue:46858", + "accept": false, + "reason": "Both concern cache behavior, but one is spurious recompilation and the other is a device-tensor slice arithmetic issue in static cache; not mergeable as one fix." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Both can surface as memory-related, but offloading/quantization misbehavior is not the same as a model-specific CUDA OOM." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting bug and Trainer GPU broadcast NoneType error are entirely separate subsystems." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Both are v5 regressions, but one is output-difference on a specific model and the other is serialization of tied weights; different issues." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Eval-mode gradient checkpointing and resume_from_checkpoint key mismatch are unrelated trainer behaviors." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility and DeepseekV3MoE divergence are different bugs with no shared code path." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "EOS token configuration inconsistency and flash-attention generation failure are not the same defect." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "One is a clearer error message for LayoutLMv3 box inputs; the other is TrainerState token counting not updating." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "A MacOS bus error in a CLIP model and documentation example errors for SigLIP2 are separate problems." + }, + { + "left": "issue:43278", + "right": "issue:43575", + "accept": false, + "reason": "BF16-to-FP32 dtype drift during eval and tensor-parallel load OOM are not the same bug." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "One is an AttributeError on s_aux=None in flash_attention integration; the other is a PLE device mismatch in Gemma4 forward pass." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "A removed helper breaking trust_remote_code and a tokenizer regression in DeepSeek are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer special-token insertion and Gemma3 NaN embeddings from sliding-window attention are distinct code paths." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Auto device mapping CUDA assert and multi-GPU OOM are different model/runtime failures." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2 tokenizer padding/truncation crash and Wav2Vec2 save/tokenization failure are unrelated tokenizer bugs." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLuke tokenizer AttributeError and Wav2Vec2 save/tokenization failure do not share the same underlying fix." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Saving/tokenization failure in Wav2Vec2 and a bare-except swallowing errors in Fuyu are different problems." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Flash-attention generation failure and a stale warning in paged generate are not the same bug." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "Base-model-prefix loading breakage and rope-parameter list handling are unrelated." + }, + { + "left": "issue:29870", + "right": "issue:34689", + "accept": false, + "reason": "BART generate UnboundLocalError and Llama 3.2 vision model loading breakage are distinct regressions." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures and void segmentation map processing 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": 26, + "review_unit_index": 34, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:27957", + "issue:29127", + "issue:33290", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43381", + "issue:43412", + "issue:43421", + "issue:43526", + "issue:43575", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43824", + "issue:43931", + "issue:43994", + "issue:44206", + "issue:44265", + "issue:44279", + "issue:44387", + "issue:44442", + "issue:44462", + "issue:44479", + "issue:44743", + "issue:44857", + "issue:45072", + "issue:45198", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45712", + "issue:45834", + "issue:46369", + "issue:46693", + "issue:46701" + ], + "soft_pairs": [ + "issue:43653|issue:44743", + "issue:44206|issue:44479", + "issue:42915|issue:42994", + "issue:43756|issue:45440", + "issue:43653|issue:44442", + "issue:43065|issue:43994", + "issue:43065|issue:45412", + "issue:43575|issue:43701", + "issue:44265|issue:45198", + "issue:29127|issue:33357", + "issue:45372|issue:45588", + "issue:43412|issue:45588", + "issue:45491|issue:46693", + "issue:42490|issue:43381", + "issue:29127|issue:35141", + "issue:45245|issue:45563", + "issue:45459|issue:45712", + "issue:43994|issue:45412", + "issue:45440|issue:45593", + "issue:45372|issue:45406", + "issue:43575|issue:44387", + "issue:45406|issue:46701", + "issue:43421|issue:45491", + "issue:45412|issue:46369", + "issue:43066|issue:44462", + "issue:45372|issue:45459", + "issue:44442|issue:45072", + "issue:43526|issue:45834", + "issue:27957|issue:33290", + "issue:27957|issue:29127", + "issue:43824|issue:44279", + "issue:33290|issue:34689", + "issue:43526|issue:46693", + "issue:33290|issue:35141", + "issue:44857|issue:45588", + "issue:43994|issue:46369", + "issue:42915|issue:43931" + ], + "prepared_review_unit_hash": "5d8b240c4c33638534a2834cbb602b5c0d2d6880be176f532a8b02422e25cfd1", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 37, + "serialized_chars": 38699, + "estimated_input_tokens": 9675, + "estimated_eval_tokens": 19606 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The candidate pairs are mostly unrelated issues that only share broad subsystem names or vague keyword similarity. None should be merged as duplicates/aliases. No PRs are present. The cleanest representative issue in this set is 44462 because it describes a concrete, reproducible AutoTokenizer loading bug with a narrow fix surface.", + "confidence": 0.9, + "canonical_issue_reason": "No single canonical issue emerges from the set overall, but issue:44462 is the strongest representative because it is a concrete AutoTokenizer regression with a specific failure mode and clear code-path.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44462 is the most self-contained and actionable bug report here: precise repro, specific subsystem, and a focused fix path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "Different tokenization/output-decoding bug vs Qwen recurrent-state/cache behavior; only superficial similarity in model/inference context." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "LasrFeatureExtractor center-arg crash and Qwen VL video regression are different failure modes in different components." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "FP8 config failure during model use is not the same as quantized model saving failure; separate code paths." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Different models and different defects: RoPE layer selection mismatch vs native/remote implementation divergence." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBirdTokenizer special-token decode bug is unrelated to FastSpeech2Conformer tokenizer loading failure." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and Siglip2 nonsensical outputs are unrelated model-specific problems." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "A dummy layer implementation bug is not the same as an RT-DETR memory-release leak." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Tensor-parallel OOM on Qwen2 load vs resume checkpoint key mismatch are different bugs and code paths." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and MacOS CLIP bus error are not the same underlying bug." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 processor import failure is unrelated to flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executable export failure and flash-attention attribute error are different concrete failures." + }, + { + "left": "issue:45491", + "right": "issue:46693", + "accept": false, + "reason": "Gemma3 NaNs from sliding-window padding are not the same as a performance regression in FA2 long-context generation." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Local-path import warning with trailing slash is unrelated to gradient-checkpointing eval-mode restriction." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "Clarifying an error message is not the same as embedding reinitialization during post_init." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-count runtime error and stale num_return_sequences warning are unrelated." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer error masking from protobuf handling is unrelated to leftover dummy-class cleanup." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "Siglip2 inference output corruption and RT-DETR memory not releasing are different bugs." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepSeekV3MoE divergence from remote impl is unrelated to D-FINE auxiliary-loss behavior." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Gemma 4 processor loading import error and Gemma4Processor missing _tokenizer in serve are related by model family but not the same concrete failure." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Tensor-parallel OOM on one model is not the same as increased CUDA reserved memory under int4 quantization." + }, + { + "left": "issue:45406", + "right": "issue:46701", + "accept": false, + "reason": "Gemma4Processor serve crash is unrelated to DiffusionGemmaGenerationOutput pipeline failure." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Runtime special-token post-processor update issue is unrelated to Gemma3 NaN embeddings." + }, + { + "left": "issue:45412", + "right": "issue:46369", + "accept": false, + "reason": "RT-DETR memory leak and sam3-litetext memory overflow are different model-specific memory issues." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 is not the same as AutoTokenizer ignoring tokenizer.json from the repo." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma 4 import failure and protobuf-related tokenizer exception masking are unrelated." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure and dtype mismatch in bfloat16 inference are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "Incorrect label reduction in BeitImageProcessorFast is unrelated to Kosmos2.5 long-ocr index errors." + }, + { + "left": "issue:27957", + "right": "issue:33290", + "accept": false, + "reason": "Flash Attention 2 support for XLMRoberta and Adafactor/deepspeed OOM are unrelated." + }, + { + "left": "issue:27957", + "right": "issue:29127", + "accept": false, + "reason": "FA2 model support and LayoutLMv3 error-message clarity are different issues." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Missing Qwen2_5_VL import and a generic dependency issue are not the same underlying bug." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Optimizer-related OOM and model-loading shape mismatch are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:46693", + "accept": false, + "reason": "BeitImageProcessorFast label handling and FA2 long-context performance regression are unrelated." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Deepspeed Adafactor OOM and embedding reinit during post_init are different defects." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "AMP/CUDA crash in LwDetrImageLoss is unrelated to flash_attention.py s_aux=None handling." + }, + { + "left": "issue:43994", + "right": "issue:46369", + "accept": false, + "reason": "Siglip2 inference corruption and sam3-litetext memory overflow are different bugs." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are different model-loading/training 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": 26, + "review_unit_index": 35, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:27957", + "issue:29870", + "issue:42175", + "issue:42490", + "issue:43262", + "issue:43316", + "issue:43334", + "issue:43381", + "issue:43493", + "issue:43525", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43646", + "issue:43720", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:44062", + "issue:44077", + "issue:44186", + "issue:44206", + "issue:44442", + "issue:44479", + "issue:44514", + "issue:44743", + "issue:44811", + "issue:44855", + "issue:44871", + "issue:44898", + "issue:44913", + "issue:44977", + "issue:45003", + "issue:45137", + "issue:45245", + "issue:45250", + "issue:45325", + "issue:45381", + "issue:45440", + "issue:45588", + "issue:45600", + "issue:45859", + "issue:45907", + "issue:46121", + "issue:46234", + "issue:46399", + "issue:46640", + "issue:46697" + ], + "soft_pairs": [ + "issue:43526|issue:44186", + "issue:45250|issue:46640", + "issue:43334|issue:43531", + "issue:43381|issue:43873", + "issue:45588|issue:45600", + "issue:43646|issue:44077", + "issue:46121|issue:46399", + "issue:44514|issue:45381", + "issue:44977|issue:45245", + "issue:43761|issue:45859", + "issue:44743|issue:44811", + "issue:43262|issue:44206", + "issue:44442|issue:44898", + "issue:42490|issue:44743", + "issue:43720|issue:44442", + "issue:27957|issue:29870", + "issue:42175|issue:46234", + "issue:44077|issue:44479", + "issue:45003|issue:45440", + "issue:43316|issue:43525", + "issue:43493|issue:43827", + "issue:43873|issue:44871", + "issue:45325|issue:46697", + "issue:44062|issue:45907", + "issue:43643|issue:44913", + "issue:44855|issue:45137" + ], + "prepared_review_unit_hash": "4411fd35e1b7ad1988d6016668b95a539a605c079affa7e7120b21e753de0487", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35630, + "estimated_input_tokens": 8908, + "estimated_eval_tokens": 18072 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is a mixed bag of unrelated issues across model configs, tokenizers, vision/audio processors, flash-attention, and runtime regressions. None of the soft pairs look like the same concrete bug/change, so all proposed merges should be rejected.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue is appropriate: the set spans many unrelated bugs and feature regressions, so there is no clear cluster anchor.", + "canonical_pr_reason": "No PRs are present in this cluster.", + "best_issue_reason": "If one issue must be highlighted, issue 43873 is the strongest representative by activity and inbound references, but it is not a duplicate of the others.", + "best_pr_reason": "No PRs are present, so there is no best PR candidate.", + "soft_edge_verdicts": [ + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Different processor/tokenizer bugs: one is BeitImageProcessorFast label reduction, the other is LayoutLMv2 NER/padding crashes." + }, + { + "left": "issue:45250", + "right": "issue:46640", + "accept": false, + "reason": "Both mention Flash Attention 2, but the reports are too generic and not the same concrete failure mode." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Qwen3-VL pad_token_id loading failure is unrelated to Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing in eval mode is unrelated to quantization/offloading behavior." + }, + { + "left": "issue:45588", + "right": "issue:45600", + "accept": false, + "reason": "Different code paths: flash_attention sink-less crash vs removed Sam3LiteText auto_mapping references." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Both concern model initialization changes, but they describe different breakages and model-specific behavior." + }, + { + "left": "issue:46121", + "right": "issue:46399", + "accept": false, + "reason": "Rope-validation TypeError and Qwen3.5 kernel registration ValueError are distinct failures." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both involve Qwen2.5-VL, but one is chat-template batching and the other is video position-id scaling." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention generation issues are unrelated to the category-count RuntimeError." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression is unrelated to Qwen3_5Moe rotary embedding CPU offload compatibility." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Recurrent cache-state reset and Whisper batch_decode skip_special_tokens are different bugs in different components." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting is unrelated to LasrFeatureExtractor passing an unsupported center argument." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "FastSpeech2Conformer tokenizer loading and Perceiver image-size handling are unrelated issues." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local-path import warnings and Qwen3.5 recurrent-state resets do not share the same underlying bug." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "BitNet accelerate-loading unpacking bug is unrelated to AutoTokenizer loading FastSpeech2ConformerTokenizer." + }, + { + "left": "issue:27957", + "right": "issue:29870", + "accept": false, + "reason": "XLMRoberta Flash Attention 2 support and BART generate() decoder_attention_mask error are different problems." + }, + { + "left": "issue:42175", + "right": "issue:46234", + "accept": false, + "reason": "TensorFlow extra packaging and a missing import_utils helper are unrelated packaging/API issues." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "PatchTSMixer post_init policy and Qwen video-input regression are separate model-specific regressions." + }, + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "Unsafe sys.modules access in modeling_utils is unrelated to DeepseekV3MoE divergence from remote implementation." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3TextConfig API mismatch and Llama4Config missing pad_token_id are different config-schema issues." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and pipeline() documentation drift are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to Gemma-3 eos_token_id configuration inconsistency." + }, + { + "left": "issue:45325", + "right": "issue:46697", + "accept": false, + "reason": "Qwen2.5-VL rope index scaling and Swin2SR over-padding are unrelated image/video preprocessing bugs." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken special-argument duplication and list_repo_files revision forwarding are unrelated API bugs." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Missing fields from trust_remote_code config loading and GPTNeoX rotary_pct reload loss are different config deserialization bugs." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 TorchScript parsing failure is unrelated to the DeepSpeed ZeRO3 deque underflow." + } + ] + }, + "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": 26, + "review_unit_index": 36, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:25251", + "issue:27957", + "issue:28282", + "issue:32090", + "issue:42617", + "issue:42907", + "issue:43232", + "issue:43295", + "issue:43316", + "issue:43388", + "issue:43452", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43701", + "issue:43716", + "issue:43742", + "issue:43749", + "issue:43761", + "issue:43825", + "issue:43931", + "issue:44077", + "issue:44079", + "issue:44188", + "issue:44190", + "issue:44442", + "issue:44464", + "issue:44496", + "issue:44534", + "issue:44871", + "issue:44913", + "issue:44987", + "issue:45216", + "issue:45584", + "issue:45698", + "issue:45759", + "issue:45803", + "issue:46178", + "issue:46234", + "issue:46291", + "issue:46421", + "issue:46506", + "issue:46821", + "issue:46829", + "issue:46932" + ], + "soft_pairs": [ + "issue:43749|issue:43931", + "issue:43316|issue:45759", + "issue:44987|issue:46178", + "issue:44190|issue:44496", + "issue:43931|issue:45216", + "issue:42617|issue:43575", + "issue:43701|issue:46506", + "issue:43232|issue:43388", + "issue:44442|issue:45803", + "issue:43550|issue:43701", + "issue:43575|issue:43742", + "issue:27957|issue:32090", + "issue:43761|issue:44079", + "issue:43761|issue:44188", + "issue:43606|issue:43701", + "issue:25251|issue:28282", + "issue:46829|issue:46932", + "issue:42907|issue:45698", + "issue:43295|issue:43825", + "issue:16998|issue:22355", + "issue:43295|issue:44871", + "issue:44871|issue:45584", + "issue:46234|issue:46291", + "issue:44077|issue:44534", + "issue:43316|issue:43716", + "issue:44464|issue:46421", + "issue:43316|issue:46821", + "issue:43646|issue:44913", + "issue:43452|issue:44496" + ], + "prepared_review_unit_hash": "171d405fe182bf3143cf265faf4cde767881ac2da860116c6578ab36b410d5d3", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36315, + "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 collection of unrelated Transformers issues. The soft pairs mostly share broad keywords like loading, config, generation, or compilation, but they do not appear to describe the same concrete bug or fix.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "Both are model-loading failures, but one is about FSDP CPU RAM efficient loading and the other is a Qwen3-VL weight-shape mismatch; different code paths and symptoms." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both involve config APIs for multimodal models, but they target different model families and different failures; not the same bug." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "One is a specific model load regression, the other is an import-utils backend registration error; unrelated causes." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Different failure modes: dataset loading in a training script vs unrecognized model/config loading." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Both mention loading problems, but one is a checkpoint weight-shape mismatch and the other is a save_pretrained regression for a different model family." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "Both are distributed/model loading issues, but one is a script execution problem and the other is an OOM during tensor-parallel loading; not the same defect." + }, + { + "left": "issue:43701", + "right": "issue:46506", + "accept": false, + "reason": "Resume-from-checkpoint key mismatch is unrelated to CPU/CUDA output divergence in a vision model." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation state-update logic and gather_for_metrics label truncation are separate training/runtime bugs." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading failure and a bare-except in tensor conversion are different subsystems and symptoms." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "torch.compile/SDPA failure is unrelated to checkpoint key mismatch." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "OOM when loading a large Qwen model is not the same as a key error for a small MobileLLM model." + }, + { + "left": "issue:27957", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 support and a Trainer broadcast type error are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "One is missing hidden_states from CLIPVisionModel.forward; the other is generic ModelOutput key assignment behavior." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "Different issues in model outputs vs attention kernel branching under torch.compile." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to resume_from_checkpoint key mismatches." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline output nesting and missing PyTorch import are unrelated." + }, + { + "left": "issue:46829", + "right": "issue:46932", + "accept": false, + "reason": "Video classification output field misassignment is not the same as Gemma4 quantized-embedder dtype casting." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Both touch save/load flows, but one is saving dequantized models and the other is loading the wrong custom module after save_pretrained; different bugs." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "A regression in processor/tokenizer image handling is unrelated to pipeline error-message wording." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Model_max_length behavior for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Both involve tokenizer/config regressions, but one is processor.tokenizer/image passing and the other is eos_token_id mismatch; not the same fix." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Both mention special-token handling, but one is an eos_token_id config inconsistency and the other is empty-transcription generation failure after align_special_tokens." + }, + { + "left": "issue:46234", + "right": "issue:46291", + "accept": false, + "reason": "Missing availability helper and kernel dependency validation are distinct import-time issues." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "Optional post_init handling in one model class is unrelated to non-persistent buffer corruption in v5." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "Config API discrepancy in Gemma3TextConfig is unrelated to a vision dtype mismatch in Mistral-3." + }, + { + "left": "issue:44464", + "right": "issue:46421", + "accept": false, + "reason": "Both relate to chunked/compiled generation, but one is incorrect outputs and the other is spurious recompilation from cache lazy init." + }, + { + "left": "issue:43316", + "right": "issue:46821", + "accept": false, + "reason": "Gemma3 config API mismatch and TimesFM window_size AttributeError are unrelated." + }, + { + "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:43452", + "right": "issue:44496", + "accept": false, + "reason": "gguf_file handling for tokenizer/model loading is a different problem from unrecognized model config loading." + } + ] + }, + "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": 26, + "review_unit_index": 37, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:35141", + "issue:41950", + "issue:42490", + "issue:43065", + "issue:43352", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43475", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43540", + "issue:43572", + "issue:43650", + "issue:43673", + "issue:43688", + "issue:44162", + "issue:44190", + "issue:44242", + "issue:44514", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44743", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45542", + "issue:45600", + "issue:45632", + "issue:45698", + "issue:45803", + "issue:46050", + "issue:46170", + "issue:46178", + "issue:46202", + "issue:46234", + "issue:46365", + "issue:46369", + "issue:46424", + "issue:46640", + "issue:46894" + ], + "soft_pairs": [ + "issue:44871|issue:46050", + "issue:45803|issue:46365", + "issue:44162|issue:44792", + "issue:43404|issue:43441", + "issue:43065|issue:44617", + "issue:44242|issue:46894", + "issue:44610|issue:46369", + "issue:45632|issue:46424", + "issue:43540|issue:44514", + "issue:45542|issue:46234", + "issue:43572|issue:46178", + "issue:44792|issue:45600", + "issue:45803|issue:46170", + "issue:44792|issue:44977", + "issue:43504|issue:45698", + "issue:43650|issue:46202", + "issue:45127|issue:45200", + "issue:43352|issue:46640", + "issue:41950|issue:43475", + "issue:44534|issue:44913", + "issue:43525|issue:43531", + "issue:35141|issue:42490", + "issue:43673|issue:44743", + "issue:43388|issue:43688", + "issue:44190|issue:44964" + ], + "prepared_review_unit_hash": "ef50b5e5ad29e3779882d1b77b56205e4a08b31d4995f61ee55737d38c7dfa85", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 25, + "serialized_chars": 34475, + "estimated_input_tokens": 8619, + "estimated_eval_tokens": 17494 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose similarity bucket of unrelated issues; none of the soft pairs look like true duplicates or the same concrete fix. The strongest standalone issue is the save/load custom-module regression.", + "confidence": 0.34, + "canonical_issue_reason": "issue:45698 is the strongest representative: it has broad user impact, clear save_pretrained/from_pretrained reproduction, and the highest discussion/reference activity among the set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45698 is the best overall issue candidate because it is concrete, widely impactful, and more central than the other mostly model-specific edge cases.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Both involve config/state roundtrips, but one is eos_token_id inconsistency and the other is quantized model reload failure; different bugs and fixes." + }, + { + "left": "issue:45803", + "right": "issue:46365", + "accept": false, + "reason": "One is a bare-except bug in tensor conversion, the other is an import failure tied to datasets/Trainer loading; not the same code path." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 model breakage and a janus image-generation test failure are unrelated model-specific issues." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Both mention newer model generation problems, but they target different models and likely different FlashAttention/config paths." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Both are SAM 3 video-related, but one is a dummy Conv2d placeholder and the other is an OOM symptom; not the same fix." + }, + { + "left": "issue:44242", + "right": "issue:46894", + "accept": false, + "reason": "Both concern training loss, but the models and mechanics differ; no evidence they share the same underlying bug." + }, + { + "left": "issue:44610", + "right": "issue:46369", + "accept": false, + "reason": "Processor/model input-size mismatch is unrelated to a memory overflow when calling sam3-litetext." + }, + { + "left": "issue:45632", + "right": "issue:46424", + "accept": false, + "reason": "Both touch cache behavior, but one is a path collision for trust_remote_code and the other is max_cache_length semantics; different bugs." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Both are Qwen multimodal input issues, but one is video processing and the other is batched chat-template padding behavior." + }, + { + "left": "issue:45542", + "right": "issue:46234", + "accept": false, + "reason": "Both involve backend availability, but tensorboard/TensorFlow setup is not the same as missing falcon_mamba backend export." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "Missing pad_token_idx and a tensorflow_text backend mapping error are unrelated configuration/import failures." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "A janus test failure and stale auto_mappings references are different problems in different code paths." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Bare except in Fuyu tensor conversion is unrelated to a Windows/ROCm torch.distributed import dependency issue." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Both are generation-related, but janus image generation and Qwen3.5 flash-attention behavior are not the same bug." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both concern loading/saving models, but BEiT legacy preset loading is not the same as custom-module resolution after save_pretrained." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "The titles are too generic to imply the same bug; no shared code path or concrete change is evident." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "One is LoRA merge collapse with tied embeddings, the other is missing mm_token_type_ids defaults; different failure modes and fixes." + }, + { + "left": "issue:43352", + "right": "issue:46640", + "accept": false, + "reason": "Both mention Flash Attention 2, but model/trigger details are missing and there is no evidence they are the same concrete compatibility fix." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "A pipeline using image processors and a missing Sam3VisionEncoderOutput attribute are different SAM/video bugs." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer junk filling and GPTNeoX rotary_pct reloading are unrelated state/config regressions." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config and a Qwen3-MoE sliding_window issue are distinct model/config bugs." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Embedding reinitialization after resize and a local-path trailing-slash import warning are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both involve cache semantics, but chunked_prefill cache handling and recurrent-state reset in modular_qwen3_5.py are different issues." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Label gathering in metrics and auxiliary-loss normalization are unrelated training/runtime bugs." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "A local dataset loading failure and inability to load Phi-4 multimodal instruct are separate loading 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": 26, + "review_unit_index": 38, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:29870", + "issue:35141", + "issue:41720", + "issue:43065", + "issue:43335", + "issue:43550", + "issue:43644", + "issue:43673", + "issue:43827", + "issue:43854", + "issue:43931", + "issue:43994", + "issue:44060", + "issue:44077", + "issue:44190", + "issue:44242", + "issue:44479", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44792", + "issue:44855", + "issue:44869", + "issue:44877", + "issue:44913", + "issue:44933", + "issue:45072", + "issue:45245", + "issue:45362", + "issue:45381", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45632", + "issue:46082", + "issue:46369", + "issue:46421", + "issue:46617", + "issue:46655", + "issue:46697", + "issue:46832", + "issue:46858", + "issue:46897", + "issue:46932", + "issue:47013" + ], + "soft_pairs": [ + "issue:44734|issue:45406", + "issue:44855|issue:45588", + "issue:29870|issue:35141", + "issue:45072|issue:46932", + "issue:44869|issue:46832", + "issue:45632|issue:46617", + "issue:45245|issue:46858", + "issue:46082|issue:46697", + "issue:43827|issue:43994", + "issue:44617|issue:45584", + "issue:43644|issue:44913", + "issue:43931|issue:44060", + "issue:25251|issue:27957", + "issue:43335|issue:44855", + "issue:44077|issue:44877", + "issue:44242|issue:46897", + "issue:43673|issue:46421", + "issue:45381|issue:46369", + "issue:45563|issue:45584", + "issue:44933|issue:45362", + "issue:41720|issue:45245", + "issue:43550|issue:44190", + "issue:43065|issue:44610", + "issue:43335|issue:44190", + "issue:44617|issue:46369", + "issue:43854|issue:44792", + "issue:44479|issue:44734", + "issue:45588|issue:47013", + "issue:45381|issue:46655" + ], + "prepared_review_unit_hash": "5871efaee22c50a914093502cdddd4c65234f62ed6608b8b2be8790e740c8f83", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36535, + "estimated_input_tokens": 9134, + "estimated_eval_tokens": 18524 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is a heterogeneous mix of unrelated issues across different models, processors, generation paths, and config behaviors. The soft edges mostly reflect superficial wording similarity (e.g. cache, padding, dtype, generation) rather than the same concrete bug, so none should be merged as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "Both are runtime crashes in serving/generation, but the concrete failures differ: KV-cache continuation tensor indexing vs missing _tokenizer on Gemma4Processor." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Different failure modes: Python 3.13 IndentationError in a scripted model file vs flash_attention crash on s_aux=None for sink-less models." + }, + { + "left": "issue:29870", + "right": "issue:35141", + "accept": false, + "reason": "Both involve generation/model internals, but one is a BART decoder_attention_mask bug at generate() and the other is embedding reinitialization after resize_token_embeddings." + }, + { + "left": "issue:45072", + "right": "issue:46932", + "accept": false, + "reason": "Both mention dtype/precision issues, but one is a CI bfloat16 mismatch in SwitchTransformers/TimmWrapperModel and the other is Gemma4 quantized vision/audio embedder casting." + }, + { + "left": "issue:44869", + "right": "issue:46832", + "accept": false, + "reason": "Whisper timestamp decoding crashes on a trailing replacement character, while RT-DETR fails because num_feature_levels exceeds available backbone outputs." + }, + { + "left": "issue:45632", + "right": "issue:46617", + "accept": false, + "reason": "Both concern trust_remote_code cache/loading, but one is a cache path collision for same leaf directory names and the other is FileNotFoundError when loading from cache." + }, + { + "left": "issue:45245", + "right": "issue:46858", + "accept": false, + "reason": "Both are numeric/runtime edge cases, but one is a categories cardinality limit and the other is a cache.get_seq_length() device-tensor bug affecting slice arithmetic." + }, + { + "left": "issue:46082", + "right": "issue:46697", + "accept": false, + "reason": "Both are shape/config constraints, but LlamaConfig head_dim validation is unrelated to Swin2SR over-padding when size_divisor is already satisfied." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "The first is stale docs referencing removed pipeline() APIs; the second is a model behavior bug where AutoModel/pipeline gives nonsensical SigLIP2 results." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Sam3Video OOM is a memory-pressure issue, while Whisper empty transcription failure after align_special_tokens is a generation edge-case crash." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Both involve configuration/state persistence, but one is transformers v5 non-persistent buffer initialization junk and the other is GPTNeoXConfig rotary_pct not surviving reload." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Both are Qwen model loading issues, but one is Qwen3-VL weight shape mismatch and the other is Qwen3-Next incorrect tied-weights warning across dt_bias." + }, + { + "left": "issue:25251", + "right": "issue:27957", + "accept": false, + "reason": "Pipeline top_k nested-list output bug is unrelated to XLMRoberta Flash Attention 2 support." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation with zero sparse layers is unrelated to the DebertaV2 Python 3.13 IndentationError." + }, + { + "left": "issue:44077", + "right": "issue:44877", + "accept": false, + "reason": "PatchTSMixer optional post_init policy and strict config blocking granite_speech are different config-validation concerns with no shared code-path bug." + }, + { + "left": "issue:44242", + "right": "issue:46897", + "accept": false, + "reason": "Load-balancing loss not added when output_router_logits=False is an MoE loss-computation bug; Florence2 double-shift is a training-target alignment bug." + }, + { + "left": "issue:43673", + "right": "issue:46421", + "accept": false, + "reason": "Both mention cache, but one is missing GenerationMixin cache in v5.0.0 during chunked_prefill and the other is lazy cache init causing spurious recompilation." + }, + { + "left": "issue:45381", + "right": "issue:46369", + "accept": false, + "reason": "Both mention model regressions and video/image-like workloads, but qwen2.5-vl vision_position_ids is unrelated to Sam3-litetext memory overflow." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Paged generate stale warning for num_return_sequences is not the same bug as Whisper empty transcription failure after align_special_tokens." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Nonexistent image_utils import is a broken reference; Qwen3.5-35B crashes with transformers chat is a model/chat integration runtime failure." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Auto device mapping cudaErrorAssert on Qwen3 is unrelated to the max-categories RuntimeError." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Bamba-9B-v2 torch.compile+SDPA crash and local dataset loading in a training script are different subsystems and failure modes." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and OmDet-Turbo processor producing 640px instead of 224px are different model/processor bugs." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug and local dataset loading script failure are unrelated." + }, + { + "left": "issue:44617", + "right": "issue:46369", + "accept": false, + "reason": "Both are Sam3-related, but one is OOM in Sam3Video and the other is memory overflow for sam3-litetext; different models and likely different causes." + }, + { + "left": "issue:43854", + "right": "issue:44792", + "accept": false, + "reason": "GLM-4.7-Flash unit-test loading failure is unrelated to janus test_model_generate_images failure." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Video input regression across Qwen VL variants is unrelated to KV-cache continuation tensor indexing in serve responses." + }, + { + "left": "issue:45588", + "right": "issue:47013", + "accept": false, + "reason": "Both are runtime crashes, but one is a flash_attention s_aux=None AttributeError and the other is xLSTM train-mode unpacking mismatch with bare tensors." + }, + { + "left": "issue:45381", + "right": "issue:46655", + "accept": false, + "reason": "Both involve video/image preprocessing, but qwen2.5-vl vision_position_ids and SmolVLM padding calculation are distinct 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": 26, + "review_unit_index": 39, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:25251", + "issue:27957", + "issue:29870", + "issue:32090", + "issue:35141", + "issue:36246", + "issue:38617", + "issue:39692", + "issue:41628", + "issue:42175", + "issue:42796", + "issue:43295", + "issue:43526", + "issue:43540", + "issue:43643", + "issue:43673", + "issue:43854", + "issue:44112", + "issue:44190", + "issue:44246", + "issue:44279", + "issue:44462", + "issue:44464", + "issue:44496", + "issue:44521", + "issue:44734", + "issue:44737", + "issue:44779", + "issue:44938", + "issue:44964", + "issue:45003", + "issue:45127", + "issue:45276", + "issue:45290", + "issue:45357", + "issue:45381", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45600", + "issue:45632", + "issue:45803", + "issue:45907", + "issue:46234", + "issue:46489", + "issue:46697", + "issue:46752", + "issue:46858" + ], + "soft_pairs": [ + "issue:39692|issue:44190", + "issue:45127|issue:45357", + "issue:43295|issue:45276", + "issue:44464|issue:45563", + "issue:44779|issue:46489", + "issue:44112|issue:45600", + "issue:45276|issue:45584", + "issue:44521|issue:45381", + "issue:45290|issue:46752", + "issue:43540|issue:44734", + "issue:25251|issue:29870", + "issue:43526|issue:46697", + "issue:43643|issue:44496", + "issue:44462|issue:45907", + "issue:44938|issue:45542", + "issue:43854|issue:44964", + "issue:44246|issue:45003", + "issue:44737|issue:45803", + "issue:43673|issue:46858", + "issue:27957|issue:35141", + "issue:44964|issue:45600", + "issue:44112|issue:44964", + "issue:45632|issue:46234", + "issue:42175|issue:44279", + "issue:42796|issue:44521", + "issue:32090|issue:35141", + "issue:36246|issue:41628", + "issue:36246|issue:38617" + ], + "prepared_review_unit_hash": "6a4a5bf9356b9e5aca947823bbe1ce0c05be70d1c046083f49a3cd5a7e5adc98", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36492, + "estimated_input_tokens": 9123, + "estimated_eval_tokens": 18502 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose semantic grouping of unrelated Transformer issues; only one pair looks plausibly like the same underlying tokenizer regression. No PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 44521 is the best anchor if a single issue must be chosen: it is open, concrete, and has the strongest discussion/inbound activity. That said, the overall cluster is heterogeneous, so it is not a true umbrella duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "44521 is the most actionable and representative issue in the set, with the clearest bug report and strongest follow-on interest.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "Different problems: a SigLIP2 doc example/model-processor mismatch versus a local dataset loading failure in `run_image_classification_no_trainer.py`." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "Both touch model saving/resizing, but the bugs differ: LoRA merge with extended vocab collapse vs Qwen3.5 saving wrong visual encoder keys." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Different code paths: processor/tokenizer image-handling regression versus `resize_token_embeddings` not updating Gemma4 embedding/output layers." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Different symptoms: inconsistent outputs from compiled chunked generation vs a stale warning about `num_return_sequences` in paged generate." + }, + { + "left": "issue:44779", + "right": "issue:46489", + "accept": true, + "reason": "Both report the same DeepSeek tokenizer/BPE-marker regression in v5; 46489 reads like a model-specific manifestation or missing edge case from the same fix path." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Unrelated CI failures: GraniteSpeech device-override test flakiness vs auto-mapping references to removed Sam3LiteText configs." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Different bugs in different models: Gemma4 embedding resize state propagation vs Whisper failing on empty transcription after special-token alignment." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both are multimodal template issues, but one is assistant-mask generation and the other is video vision-position IDs; not the same bug." + }, + { + "left": "issue:45290", + "right": "issue:46752", + "accept": false, + "reason": "Same function, but different edge cases: tool-call assistant messages with no content versus an entirely empty message list." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Different subsystems and failures: Qwen3OmniMoe video-input validation versus server-side KV-cache continuation tensor indexing." + }, + { + "left": "issue:25251", + "right": "issue:29870", + "accept": false, + "reason": "Different areas: pipeline `top_k` output nesting regression versus BART generate() hitting `decoder_attention_mask` initialization error." + }, + { + "left": "issue:43526", + "right": "issue:46697", + "accept": false, + "reason": "Both are image-processor bugs, but one is `reduce_labels` returning too little data and the other is Swin2SR over-padding images." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Different loading/config issues: missing fields when `trust_remote_code=True` versus an unrecognized model due to missing `model_type`." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Different concerns: AutoTokenizer ignoring `tokenizer.json` versus `list_repo_files` not forwarding `revision` for CI." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Different environment/backend problems: Python 3.14 import failure versus TensorBoard installed without TensorFlow causing backend errors." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are model-loading complaints, but they concern different models and likely different loader paths." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "One is slow import performance; the other is an unsafe `sys.modules` access bug in `modeling_utils`." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "Different bugs: XLNet device placement regression versus Fuyu swallowing exceptions with a bare `except:`." + }, + { + "left": "issue:43673", + "right": "issue:46858", + "accept": false, + "reason": "Both involve cache/prefill, but one is missing GenerationMixin cache in v5.0.0 and the other is a device-tensor `seq_length` arithmetic issue on non-CUDA backends." + }, + { + "left": "issue:27957", + "right": "issue:35141", + "accept": false, + "reason": "Different topics: XLMRoberta Flash Attention 2 support versus token embedding resize reinitialization with untied output embeddings." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Different models and failures: Phi-4 multimodal loading versus stale auto-mapping references to removed configs." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "GraniteSpeech CI test failure and Phi-4 multimodal model loading are unrelated." + }, + { + "left": "issue:45632", + "right": "issue:46234", + "accept": false, + "reason": "Different hub/import issues: local cache-path collision for `trust_remote_code` versus a missing utility export in `import_utils`." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "Both are dependency-related, but one is a specific packaging extra omission and the other is a generic transformers dependency complaint." + }, + { + "left": "issue:42796", + "right": "issue:44521", + "accept": false, + "reason": "Different features: DeepSeek tokenizer BPE-marker handling versus multimodal assistant-mask generation in `apply_chat_template`." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Different runtime errors: trainer GPU broadcast rejecting `NoneType` versus embedding reinitialization after vocab resize." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Different import failures: a missing model-specific processor export versus missing top-level `AutoImageProcessor`." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Unrelated missing imports in different modules; no shared code path." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative about cluster cohesion. The chosen anchor issue, 44521, is supported by its open state, discussion activity, and inbound references. The note that the cluster is heterogeneous is also consistent with the mostly dissimilar issue titles and the soft pairs. The only accepted soft edge, 44779/46489, is plausibly the same DeepSeek tokenizer regression path and is not overstated. No PR-pair mergeability concerns apply here since the packet contains only issues." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 40, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:41950", + "issue:42907", + "issue:43295", + "issue:43299", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43479", + "issue:43526", + "issue:43531", + "issue:43582", + "issue:43650", + "issue:43673", + "issue:43824", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44220", + "issue:44279", + "issue:44295", + "issue:44336", + "issue:44496", + "issue:44521", + "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:45600", + "issue:45759", + "issue:46082", + "issue:46121", + "issue:46123", + "issue:46369", + "issue:46424", + "issue:46506", + "issue:46515", + "issue:46693", + "issue:46932" + ], + "soft_pairs": [ + "issue:43404|issue:45216", + "issue:43531|issue:45600", + "issue:46082|issue:46121", + "issue:44295|issue:46506", + "issue:41950|issue:44016", + "issue:43673|issue:46424", + "issue:43295|issue:45335", + "issue:43582|issue:46932", + "issue:45083|issue:46506", + "issue:44496|issue:45042", + "issue:43976|issue:44987", + "issue:43526|issue:43901", + "issue:43901|issue:44496", + "issue:43673|issue:44336", + "issue:44496|issue:45759", + "issue:44683|issue:45446", + "issue:43381|issue:45593", + "issue:44016|issue:44220", + "issue:42907|issue:43441", + "issue:43650|issue:44220", + "issue:44016|issue:44062", + "issue:43976|issue:44279", + "issue:44792|issue:46424", + "issue:43441|issue:45588", + "issue:43824|issue:44933", + "issue:44521|issue:46693", + "issue:44279|issue:46515", + "issue:43479|issue:45375", + "issue:45335|issue:45584", + "issue:43299|issue:45216", + "issue:46123|issue:46369" + ], + "prepared_review_unit_hash": "b3b16ef7aa0d8bcbae1ac0fce1a48cf41cf9346cc49795e0f389d1c19c65917c", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36847, + "estimated_input_tokens": 9212, + "estimated_eval_tokens": 18680 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose cluster of mostly unrelated Transformers issues, with repeated themes around multimodal model/config loading and processor regressions but no true duplicate pair. The best anchor is a Qwen multimodal loading regression; the soft edges are mostly same-subsystem, not same-bug matches.", + "confidence": 0.43, + "canonical_issue_reason": "issue:43299 is the clearest central bug in the set: a concrete multimodal model-loading regression with a well-defined failure mode and release window, and it best represents the Qwen/config-loading slice of the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43299 is the most suitable representative issue overall because it is high-signal, concrete, and closest to the recurring multimodal loading/config failures seen across the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Related multimodal family, but one is untied lm_head weights and the other is save_pretrained checkpoint correctness; different bugs/code paths." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior vs stale auto_mappings references to removed configs; unrelated root causes." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "Both are config validation issues, but head_dim handling and rope ignore-key parsing are distinct failures." + }, + { + "left": "issue:44295", + "right": "issue:46506", + "accept": false, + "reason": "Position_ids buffer read error vs CPU/CUDA output divergence; not the same defect." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "Video-classification processor lookup bug vs a notebook syntax error; no shared underlying issue." + }, + { + "left": "issue:43673", + "right": "issue:46424", + "accept": false, + "reason": "Both mention cache, but chunked_prefill cache missing and max_cache_length semantics for quantized cache are different." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Processor.tokenizer/image regression vs resize_token_embeddings not updating decoder.embed_tokens; different code paths." + }, + { + "left": "issue:43582", + "right": "issue:46932", + "accept": false, + "reason": "Apple Silicon TypeError in warmup vs Gemma4 quantized-embedder dtype casting; unrelated." + }, + { + "left": "issue:45083", + "right": "issue:46506", + "accept": false, + "reason": "Helper length computation bug vs CPU/CUDA result mismatch; not the same failure." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrecognized model config vs PIL image processors needing torchvision; different problems." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility issue vs a specific model-loading failure; different root cause." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug vs documentation mismatch in TextClassificationPipeline." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Docs mentioning return_all_scores vs model config import failure; unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Cache-missing regression vs ANSI codes in loading_report; no overlap beyond general utility code." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model loading vs failing to unwrap text_config in composite multimodal configs; different code paths." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Compiled flex_attention failure on torch>=2.9 vs incorrect version gate for AuxRequest import; different issues." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode vs D-FINE auxiliary-loss behavior; unrelated." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error vs audio feature extraction helper bug; no shared underlying defect." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Dequantized model saving issue vs flash-attention runtime failure on Ministral-3; same model family but different bug." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "Placeholder/invalid issue vs audio feature extraction bug; no relation." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error vs AddedToken special-argument TypeError; unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Version-support problem vs generic dependency issue; not the same bug." + }, + { + "left": "issue:44792", + "right": "issue:46424", + "accept": false, + "reason": "Janus image-generation test failure vs cache max_length semantics; different areas." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both are flash-attention themed, but one is Ministral-3-specific and the other is an s_aux=None AttributeError for sink-less models; not mergeable as one fix." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Missing Qwen2_5_VL export vs nonexistent image_utils import; both import errors, but different symbols and causes." + }, + { + "left": "issue:44521", + "right": "issue:46693", + "accept": false, + "reason": "All-zero assistant masks in multimodal chat templates vs FA2 long-context performance regression; not the same issue." + }, + { + "left": "issue:44279", + "right": "issue:46515", + "accept": false, + "reason": "Dependency issue vs torch 2.4.1 incompatibility; environment problems, but different ones." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Phi4MultimodalConfig default initialization bug vs missing Qwen3_5MoeVisionConfig field; different config objects and failures." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "Token embedding resize bug vs Whisper empty-transcription failure after align_special_tokens; unrelated." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Both are Qwen multimodal regressions, but one is model loading and the other is save_pretrained checkpoint corruption; different fixes." + }, + { + "left": "issue:46123", + "right": "issue:46369", + "accept": false, + "reason": "Final partial batch dropping vs sam3-litetext memory overflow; unrelated performance/behavior 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": 26, + "review_unit_index": 41, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:39692", + "issue:41950", + "issue:42796", + "issue:43295", + "issue:43299", + "issue:43381", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43653", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44075", + "issue:44242", + "issue:44279", + "issue:44291", + "issue:44327", + "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:46209", + "issue:46399", + "issue:46512", + "issue:46614", + "issue:46752", + "issue:47000", + "issue:47085" + ], + "soft_pairs": [ + "issue:44327|issue:46752", + "issue:46614|issue:47085", + "issue:44279|issue:45405", + "issue:43295|issue:44779", + "issue:42796|issue:43650", + "issue:39692|issue:43901", + "issue:44075|issue:44368", + "issue:43577|issue:45216", + "issue:44351|issue:45278", + "issue:41950|issue:45020", + "issue:43976|issue:45405", + "issue:43381|issue:47000", + "issue:44466|issue:45276", + "issue:44779|issue:45812", + "issue:46082|issue:46399", + "issue:43299|issue:45357", + "issue:46082|issue:46512", + "issue:43653|issue:45593", + "issue:43650|issue:46209", + "issue:41950|issue:45823", + "issue:44242|issue:45200", + "issue:43299|issue:43854", + "issue:44462|issue:45081", + "issue:45588|issue:47000", + "issue:44016|issue:44291", + "issue:44683|issue:45542", + "issue:43479|issue:43525" + ], + "prepared_review_unit_hash": "a95247e3a6657f9a110aaaa8022592e4e54ff08a44d28bf1dabedf7b920feb16", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35414, + "estimated_input_tokens": 8854, + "estimated_eval_tokens": 17964 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs mostly share only broad themes (tokenizers, multimodal models, docs, regressions) but not the same underlying bug or change, so none should be merged as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out; the items span unrelated failures across different models, pipelines, configs, and documentation.", + "canonical_pr_reason": null, + "best_issue_reason": "No strong global best issue: 45020 is the broadest recent regression report, but it does not meaningfully represent the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44327", + "right": "issue:46752", + "accept": false, + "reason": "Different code paths and failure modes: QA decode_spans top-k bounds vs apply_chat_template on an empty list." + }, + { + "left": "issue:46614", + "right": "issue:47085", + "accept": false, + "reason": "Tokenizer backend state bug vs Rust tokenizer thread-safety regression; unrelated underlying issues." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A generic dependency problem is not the same as bumping MIN_PEFT_VERSION to an unreleased release." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Both involve tokenizers, but one is processor/tokenizer API regression in multimodal code and the other is DeepSeek tokenization output corruption." + }, + { + "left": "issue:42796", + "right": "issue:43650", + "accept": false, + "reason": "Ministral BPE handling bug is unrelated to the data request issue." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs-related, but they concern different examples and different behavior, not the same mistake." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "SGD args being ignored is unrelated to the tie_word_embeddings warning." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Different regressions: loading dtype stays float32 vs save_pretrained serialization of Qwen3.5 checkpoints." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "A specific missing import/export is not the same as broad import errors after upgrade." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Video-classification processor lookup bug and remote_code breakage are different problems." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version compatibility issue vs dependency pin/version-release issue." + }, + { + "left": "issue:43381", + "right": "issue:47000", + "accept": false, + "reason": "Gradient-checkpointing eval-mode guard is unrelated to RTDetrLoss cost-matrix infeasibility." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Serialization of tied lm_head weights is different from resize_token_embeddings not propagating to embeddings." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and different token-ID behavior." + }, + { + "left": "issue:46082", + "right": "issue:46399", + "accept": false, + "reason": "Config validation for head_dim is unrelated to a missing kernel decorator causing rotary-pos-emb registration failure." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Model loading failure and save_pretrained visual-key serialization bug are different code paths." + }, + { + "left": "issue:46082", + "right": "issue:46512", + "accept": false, + "reason": "Hidden-size/head-dim validation is unrelated to from_config leaving config.dtype stale." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "BigBirdTokenizer special-token registration is unrelated to D-FINE auxiliary-loss behavior." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "The titles are unrelated; 'ADD THE DATA' is not the same as FP8 inference failing on B200." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Processor lookup in a video pipeline is unrelated to Gemma4 device-placement mismatch." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Load-balancing loss gating and mm_token_type_ids defaults are different multimodal training issues." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Qwen3VL loading regression and GLM unit-test loading issue are separate model-specific failures." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "Ignoring tokenizer.json and crashing in _patch_mistral_regex are distinct tokenizer-loading bugs." + }, + { + "left": "issue:45588", + "right": "issue:47000", + "accept": false, + "reason": "Flash-attention s_aux=None crash and RTDetrLoss cost-matrix infeasibility are unrelated." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "A notebook syntax error is not the same as an init_empty_weights TypeError." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on new torch versions is unrelated to a TensorFlow backend install/import issue." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Multimodal default config initialization and missing pad_token_id are separate config 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": 26, + "review_unit_index": 42, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:33290", + "issue:42175", + "issue:42796", + "issue:42994", + "issue:43317", + "issue:43441", + "issue:43525", + "issue:43643", + "issue:43673", + "issue:43723", + "issue:43761", + "issue:43824", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44162", + "issue:44186", + "issue:44188", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44336", + "issue:44373", + "issue:44496", + "issue:44589", + "issue:44623", + "issue:44671", + "issue:44749", + "issue:44779", + "issue:44987", + "issue:45081", + "issue:45103", + "issue:45362", + "issue:45372", + "issue:45397", + "issue:45542", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45823", + "issue:45834", + "issue:45859", + "issue:45907", + "issue:46655", + "issue:46726", + "issue:46851", + "issue:47000" + ], + "soft_pairs": [ + "issue:33290|issue:45656", + "issue:45081|issue:45907", + "issue:42994|issue:43317", + "issue:43441|issue:45834", + "issue:44336|issue:46851", + "issue:44623|issue:44749", + "issue:45103|issue:45372", + "issue:45859|issue:46851", + "issue:44261|issue:44263", + "issue:44623|issue:45397", + "issue:42175|issue:43976", + "issue:43901|issue:44246", + "issue:45600|issue:45712", + "issue:44671|issue:44779", + "issue:43525|issue:43881", + "issue:43673|issue:43761", + "issue:44016|issue:45823", + "issue:44186|issue:47000", + "issue:43824|issue:43976", + "issue:44987|issue:45542", + "issue:44188|issue:44336", + "issue:42796|issue:43441", + "issue:43525|issue:43643", + "issue:43723|issue:44162", + "issue:44589|issue:45362", + "issue:44336|issue:44373", + "issue:46655|issue:46726", + "issue:44246|issue:44496" + ], + "prepared_review_unit_hash": "c79ca437ad2fe0c901a479f28b15bcb0921f0903965251547e6819915da2cd3e", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36017, + "estimated_input_tokens": 9005, + "estimated_eval_tokens": 18266 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items are largely heterogeneous bug reports across unrelated models, loading paths, processors, and docs/performance regressions. None of the soft pairs look like the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Both mention Deepspeed/optimizer behavior, but one is Adafactor OOM and the other is double-stepping; different failure modes and fixes." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral tokenizer regex crash vs hub API revision forwarding; unrelated code paths." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "One is quantized model saving, the other is loading a dequantized model with device offload; different lifecycle stages." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Different models and bugs: Mistral tokenizer/BPE markers vs Kosmos2.5 OCR index error." + }, + { + "left": "issue:44336", + "right": "issue:46851", + "accept": false, + "reason": "ANSI color logging issue vs Qwen3.5 variable-length position_ids bug; unrelated." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Processor save_pretrained missing files vs a reported slowdown after upgrading Transformers; not the same problem." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "Auto-docstring annotation crash vs Gemma 4 processor import failure; distinct subsystems." + }, + { + "left": "issue:45859", + "right": "issue:46851", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is CPU offload incompatibility and the other is variable-length input handling." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Different model internals and symptoms: MLA layer norm precision/config issue vs GLM-MoE split-return behavior." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Processor file-saving bug vs Gemma-4 ZeRO-3 loading failure; unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Package backend inclusion problem vs Python version compatibility; different environment issues." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs mismatch for pipeline arguments vs intermittent import slowdown; unrelated." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both are repo maintenance/CI issues, but one is stale auto_mappings references and the other is leftover dummy classes; different fixes." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "CamemBERT masked-LM regression vs DeepSeek tokenizer regression; different models and code paths." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Llama4Config missing pad_token_id vs glm-4v-9b loading failure; different model-specific errors." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Generation cache regression in chunked_prefill vs CLIPVision hidden_states regression; not the same bug." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error vs Gemma4 device-map mismatch during forward; completely different." + }, + { + "left": "issue:44186", + "right": "issue:47000", + "accept": false, + "reason": "Tokenizer padding/truncation crash vs RTDetr loss infeasible cost matrix; unrelated model failures." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Missing Qwen2_5_VL export vs Python support/version issue; unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45542", + "accept": false, + "reason": "Model loading failure vs missing TensorFlow backend installation; different problems." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile vs terminal ANSI output issue; unrelated." + }, + { + "left": "issue:42796", + "right": "issue:43441", + "accept": false, + "reason": "Same model family, but tokenizer BPE-marker handling and FlashAttention failure are different concrete bugs." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Both involve missing config fields, but one is a model config attribute bug and the other is trust_remote_code returning incomplete config objects." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Generic tokenizer loading issue vs ESM2 breakage; different models and failure modes." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage type lookup failure vs Qwen chat crash; unrelated." + }, + { + "left": "issue:46655", + "right": "issue:46726", + "accept": false, + "reason": "Video processor padding bug vs missing vision backend/image processor error; different causes." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import slowdown vs unrecognized model/config.json failure; not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 43, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:36246", + "issue:41628", + "issue:42491", + "issue:42757", + "issue:42886", + "issue:43257", + "issue:43299", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43784", + "issue:43901", + "issue:43976", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44351", + "issue:44485", + "issue:44496", + "issue:44779", + "issue:44871", + "issue:45003", + "issue:45072", + "issue:45083", + "issue:45103", + "issue:45137", + "issue:45292", + "issue:45325", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45561", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:46132", + "issue:46291", + "issue:46399", + "issue:46506", + "issue:46614", + "issue:46665", + "issue:46752", + "issue:46821", + "issue:46829", + "issue:46851" + ], + "soft_pairs": [ + "issue:45325|issue:46851", + "issue:46291|issue:46399", + "issue:44297|issue:45397", + "issue:45083|issue:46851", + "issue:43784|issue:45542", + "issue:41628|issue:43976", + "issue:46399|issue:46851", + "issue:43526|issue:45003", + "issue:45561|issue:45563", + "issue:43540|issue:46752", + "issue:45103|issue:46399", + "issue:45072|issue:46829", + "issue:44246|issue:44871", + "issue:44295|issue:45397", + "issue:43531|issue:45362", + "issue:46614|issue:46665", + "issue:45588|issue:46851", + "issue:42491|issue:43299", + "issue:45003|issue:46821", + "issue:44336|issue:44485", + "issue:44242|issue:45593", + "issue:43550|issue:46506", + "issue:42886|issue:46132", + "issue:44779|issue:45292", + "issue:36246|issue:42757", + "issue:43901|issue:45003", + "issue:36246|issue:44351", + "issue:43257|issue:45137", + "issue:44496|issue:45003" + ], + "prepared_review_unit_hash": "8bb6f3e3217beb590a54f105001bdca9504540e44efd228ce1f107e0101c17c5", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36472, + "estimated_input_tokens": 9118, + "estimated_eval_tokens": 18492 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated single-issue reports. The soft pairs share broad keywords or model families, but none look like the same underlying bug or change, so I would not merge any of them as duplicates.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 46851 is the strongest standalone bug report in this set: it describes a concrete, reproducible runtime failure in a specific code path (Qwen3.5 variable-length `position_ids`) with a clear model surface and likely fix location.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46851 is the best issue candidate overall because it is specific, actionable, and centered on a clear runtime regression rather than a vague import/docs/config mismatch.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45325", + "right": "issue:46851", + "accept": false, + "reason": "Both mention Qwen and `position_ids`, but one is a Qwen2.5-VL temporal rope-scaling bug and the other is a Qwen3.5 variable-length input crash; different model families and code paths." + }, + { + "left": "issue:46291", + "right": "issue:46399", + "accept": false, + "reason": "Both involve kernels-related failures, but one is a `transformers.pipelines` import error from kernel version requirements and the other is a Qwen3.5 kernel registration/decorator crash." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save metadata mismatch and a Gemma-4 ZeRO3 `from_pretrained` failure are unrelated bugs with different surfaces and fixes." + }, + { + "left": "issue:45083", + "right": "issue:46851", + "accept": false, + "reason": "Both are Qwen3-related, but one is an omni-MoE feature-length helper issue and the other is a text/chat `position_ids` regression; not the same bug." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Both are import/dependency problems, but the affected packages and failure modes differ (`nn` NameError in sentence-transformers vs missing TensorFlow backend handling for tensorboard)." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "Missing `AutoImageProcessor` export and a Python-version compatibility regression are different issues, even though both are import-time failures." + }, + { + "left": "issue:46399", + "right": "issue:46851", + "accept": false, + "reason": "One is a kernel registration decorator error for Qwen3.5, the other is a variable-length `position_ids` failure; same model family but not the same concrete defect." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BEiT Fast `reduce_labels` behavior and unsafe `sys.modules` access are unrelated code-path bugs." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "A pytest-xdist file race in testing utils is unrelated to a stale `num_return_sequences` warning in generation." + }, + { + "left": "issue:43540", + "right": "issue:46752", + "accept": false, + "reason": "A Qwen3OmniMoe video-input ValueError and an empty-list chat-template IndexError are different runtime failures." + }, + { + "left": "issue:45103", + "right": "issue:46399", + "accept": false, + "reason": "Auto-docstring future-annotations crash and Qwen3.5 kernel registration failure are unrelated." + }, + { + "left": "issue:45072", + "right": "issue:46829", + "accept": false, + "reason": "One is a dtype mismatch under `torch.compile`/SDPA, the other is a wrong field assignment in `ImageClassifierOutput`; not the same bug." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import slowness and Gemma-3 `eos_token_id` inconsistency are unrelated issues." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Reading `position_ids` after buffer registration and a model-loading failure in Gemma-4 are different defects." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Both are Qwen-related, but one is a `sliding_window` issue in Qwen3-MoE and the other is a chat crash in Qwen3.5; different symptoms and likely fixes." + }, + { + "left": "issue:46614", + "right": "issue:46665", + "accept": false, + "reason": "A stale phonemizer backend after language switching is unrelated to inconsistent `from_pretrained` behavior inside Python classes." + }, + { + "left": "issue:45588", + "right": "issue:46851", + "accept": false, + "reason": "A flash-attention `s_aux=None` crash and a Qwen3.5 `position_ids` bug do not share the same code path." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both mention Qwen3 MoE loading regressions, but one is LoRA compatibility across hf4.x/hf5.x and the other is Qwen3VL MoE loading on dev0; not the same concrete issue." + }, + { + "left": "issue:45003", + "right": "issue:46821", + "accept": false, + "reason": "Unsafe `sys.modules` access and a TimesFM `window_size` AttributeError are unrelated." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI code output in loading report is unrelated to a GLM-5 RoPE implementation discussion/bug." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Both concern conditional losses, but they affect different models and flags and do not look like one underlying bug." + }, + { + "left": "issue:43550", + "right": "issue:46506", + "accept": false, + "reason": "A Bamba SDPA/`torch.compile` failure and CPU-vs-CUDA result divergence in PPDocLayoutV3 are different issues." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline tokenizer cache loading and `AttentionInterface.register` behavior changes are unrelated." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "DeepSeek tokenizer regression and `resize_token_embeddings` not updating output embeddings are different code paths." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different packages and likely different fixes." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mentioning `return_all_scores` and unsafe `sys.modules` access are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Missing `Qwen2_5_VLImageProcessor` and missing `HybridCache` are separate import/export problems." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Qwen3 MoE weight conversion with accelerate+deepspeed and a DeepSpeed ZeRO3 deque error are different failures." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "An unrecognized model/config.json problem is unrelated to unsafe `sys.modules` access." + } + ] + }, + "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": 26, + "review_unit_index": 44, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:36246", + "issue:41950", + "issue:43262", + "issue:43295", + "issue:43352", + "issue:43475", + "issue:43526", + "issue:43540", + "issue:43596", + "issue:44038", + "issue:44162", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44451", + "issue:44466", + "issue:44479", + "issue:44623", + "issue:44749", + "issue:44857", + "issue:44877", + "issue:44987", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45250", + "issue:45292", + "issue:45335", + "issue:45375", + "issue:45459", + "issue:45561", + "issue:45600", + "issue:45759", + "issue:45859", + "issue:46170", + "issue:46369", + "issue:46399", + "issue:46506", + "issue:46655", + "issue:46726", + "issue:46752", + "issue:46894", + "issue:46897", + "issue:47000" + ], + "soft_pairs": [ + "issue:41950|issue:44220", + "issue:43540|issue:46655", + "issue:43262|issue:44246", + "issue:44451|issue:46752", + "issue:44479|issue:46655", + "issue:45375|issue:45759", + "issue:45003|issue:45561", + "issue:45137|issue:45600", + "issue:45292|issue:47000", + "issue:44295|issue:44749", + "issue:44162|issue:45200", + "issue:45859|issue:46399", + "issue:43295|issue:45292", + "issue:36246|issue:43596", + "issue:43475|issue:46369", + "issue:44466|issue:45335", + "issue:44623|issue:46726", + "issue:44242|issue:44857", + "issue:43526|issue:46897", + "issue:44877|issue:45375", + "issue:44038|issue:46506", + "issue:29870|issue:30333", + "issue:44987|issue:45459", + "issue:43526|issue:46894", + "issue:43352|issue:45250", + "issue:44479|issue:46170" + ], + "prepared_review_unit_hash": "83f75e21697a7401378a2be7dca1f59472bcb7ea59da77819874689828fd0335", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34999, + "estimated_input_tokens": 8750, + "estimated_eval_tokens": 17756 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are not a true duplicate cluster; they mostly share broad subsystem keywords (video/Qwen/config/modeling) but describe distinct bugs. No soft edge pair should be merged.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44479 is the closest umbrella candidate because it is the broadest, most recent video-regression report and the nearest thematic match to several other items; however, it still does not actually cover the other issues.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the best representative issue for this set, but only as a loose thematic anchor, not as a real duplicate canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Different bugs: video-classification image-processor lookup vs audio feature extraction." + }, + { + "left": "issue:43540", + "right": "issue:46655", + "accept": false, + "reason": "Both are video-related, but they hit different model/code paths and have different root causes." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug is unrelated to import slowness." + }, + { + "left": "issue:44451", + "right": "issue:46752", + "accept": false, + "reason": "Model load failure vs empty-list chat template IndexError; not the same issue." + }, + { + "left": "issue:44479", + "right": "issue:46655", + "accept": false, + "reason": "Same broad video area, but different models and distinct failures; not one concrete bug." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Both touch Qwen multimodal config handling, but one is a missing field and the other is from_config text_config unwrapping." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unrelated: sys.modules access bug vs xdist race in testing utils." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and removed Sam3LiteText config references are different problems." + }, + { + "left": "issue:45292", + "right": "issue:47000", + "accept": false, + "reason": "Token embedding resize sync bug is unrelated to RTDetrLoss infeasible cost matrix." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer error vs data-filtering slowdown after upgrade; different code paths." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "Different models and failures: ESM2 breakage vs Gemma 4 mm_token_type_ids defaulting." + }, + { + "left": "issue:45859", + "right": "issue:46399", + "accept": false, + "reason": "Both mention Qwen3.5, but one is CPU offload compatibility and the other is kernel registration/decorator failure." + }, + { + "left": "issue:43295", + "right": "issue:45292", + "accept": false, + "reason": "Processor/tokenizer regression is unrelated to resize_token_embeddings not updating outputs." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Missing Qwen2_5_VLImageProcessor import is unrelated to DeepSpeed/BertModel zero3 indexing." + }, + { + "left": "issue:43475", + "right": "issue:46369", + "accept": false, + "reason": "SAM3 video encoder missing attribute vs SAM3 LiteText memory overflow are different issues." + }, + { + "left": "issue:44623", + "right": "issue:46726", + "accept": false, + "reason": "save_pretrained missing files is not the same as unrecognized image processor due to missing vision backend." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Load-balancing loss omission and AMP image-loss crash are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:46897", + "accept": false, + "reason": "BEiT fast label reduction bug vs Florence2 training double-shift bug; different models and fixes." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Strict config loading granite_speech is unrelated to Qwen config field handling." + }, + { + "left": "issue:44038", + "right": "issue:46506", + "accept": false, + "reason": "Qwen3-VL-Moe bug and PPDocLayoutV3 CPU/CUDA divergence are not the same change." + }, + { + "left": "issue:29870", + "right": "issue:30333", + "accept": false, + "reason": "BART generation UnboundLocalError vs MLflow job status reporting; unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Loading a model and masking tokenizer protobuf errors are different issues." + }, + { + "left": "issue:43526", + "right": "issue:46894", + "accept": false, + "reason": "BEiT label reduction bug is unrelated to CohereASR training-loss double-shift." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "Specific model not supporting Flash Attention 2.0 is not the same as the generic FA2 topic." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "Video-input regression vs Windows ROCm/torch.distributed import dependency 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": 26, + "review_unit_index": 45, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:42199", + "issue:42886", + "issue:43065", + "issue:43408", + "issue:43475", + "issue:43540", + "issue:43653", + "issue:43701", + "issue:43723", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44361", + "issue:44368", + "issue:44483", + "issue:44623", + "issue:44749", + "issue:44877", + "issue:44977", + "issue:44995", + "issue:45103", + "issue:45200", + "issue:45229", + "issue:45397", + "issue:45412", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45656", + "issue:45685", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46170", + "issue:46202", + "issue:46421", + "issue:46431", + "issue:46710", + "issue:46726", + "issue:46832", + "issue:46851", + "issue:46858", + "issue:47068" + ], + "soft_pairs": [ + "issue:42199|issue:46851", + "issue:44060|issue:44077", + "issue:44368|issue:44977", + "issue:45412|issue:46123", + "issue:43653|issue:44162", + "issue:45397|issue:45656", + "issue:43065|issue:43408", + "issue:44877|issue:46170", + "issue:43475|issue:43540", + "issue:42886|issue:45229", + "issue:46170|issue:46832", + "issue:43723|issue:46710", + "issue:43931|issue:44368", + "issue:44188|issue:46202", + "issue:44368|issue:45563", + "issue:44995|issue:45200", + "issue:44483|issue:46202", + "issue:43906|issue:46202", + "issue:44188|issue:46421", + "issue:44995|issue:45588", + "issue:45593|issue:47068", + "issue:45685|issue:46431", + "issue:45987|issue:46082", + "issue:45103|issue:46858", + "issue:44361|issue:44749", + "issue:43065|issue:46123", + "issue:44297|issue:44623", + "issue:43994|issue:46726", + "issue:44749|issue:45656", + "issue:43701|issue:44749" + ], + "prepared_review_unit_hash": "38eae7d74dd354014f77d97152afe00124991125f57a06306404e0e689ac1f6d", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36875, + "estimated_input_tokens": 9219, + "estimated_eval_tokens": 18694 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft pairs are mostly broad topical similarities, not true duplicates; I rejected all of them because they describe different bugs, code paths, or model families.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42199", + "right": "issue:46851", + "accept": false, + "reason": "Different models and failures: DETR cardinality/background-class logic vs Qwen3.5 variable-length position_ids." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Unrelated issues: tied-weights warning in Qwen3-Next vs PatchTSMixer post_init policy." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "Both mention Qwen3.5, but one is a LoRA warning and the other is a flash-attention generation bug." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "Different subsystems: RT-DETR GC/memory release vs MaskGenerationPipeline batching/drop logic." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer special-token registration bug vs ESM2 breakage; no shared code path." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Both involve deepspeed, but loading with ZeRO-3 and double optimizer stepping are distinct bugs." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Both are SAM3-related, but one is a dummy layer implementation and the other is a model-type mismatch warning." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Strict config loading vs Windows/ROCm import dependency issue; unrelated failures." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "Different model families and symptoms: SAM3 vision encoder attribute error vs Qwen3OmniMoe video processing ValueError." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Tokenizer cache/offline loading bug vs Gemma4 multi-GPU CUDA OOM; unrelated." + }, + { + "left": "issue:46170", + "right": "issue:46832", + "accept": false, + "reason": "Import/dependency failure vs RT-DETR forward shape mismatch; no common underlying defect." + }, + { + "left": "issue:43723", + "right": "issue:46710", + "accept": false, + "reason": "Tokenizer-loading regression vs accuracy regression after upgrade; different problem classes." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape load error vs Qwen3.5 warning about tie_word_embeddings; unrelated." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile vs Gemma pipeline roundtrip token issue; different code paths." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Qwen3.5 tie_word_embeddings warning vs stale num_return_sequences warning in paged generate." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa stale indexer cache on second forward vs Gemma4 mm_token_type_ids training requirement." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "Chat-completions API request rejection vs Gemma pipeline roundtrip failure; unrelated." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "One is an isolated reproduction of another issue; the Gemma roundtrip bug is a different concrete failure." + }, + { + "left": "issue:44188", + "right": "issue:46421", + "accept": false, + "reason": "Both touch cache/compile behavior, but one is divergent kernels and the other is spurious recompilation from lazy init." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Different bugs: stale cache in GlmMoeDsa vs flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:45593", + "right": "issue:47068", + "accept": false, + "reason": "D-FINE auxiliary-loss behavior vs a deepspeed loss host-sync concern; not the same defect." + }, + { + "left": "issue:45685", + "right": "issue:46431", + "accept": false, + "reason": "MPS int histogram failure vs processor chat-template absence; no overlap." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "Cache return-type contract bug vs LlamaConfig head_dim validation; different issues." + }, + { + "left": "issue:45103", + "right": "issue:46858", + "accept": false, + "reason": "Docstring helper crash with future annotations vs static-cache seq-length tensor bug; unrelated." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "Tokenizer task AttributeError vs performance regression after version upgrade; different symptoms and code paths." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "SAM3 pixel-decoder implementation issue vs pipeline batching logic; unrelated." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both concern save_pretrained, but tokenizer_config class mismatch and processor missing files are distinct serialization bugs." + }, + { + "left": "issue:43994", + "right": "issue:46726", + "accept": false, + "reason": "Siglip2 wrong outputs vs missing vision-backend error in AutoProcessor; different failure modes." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Data filtering slowdown vs double optimizer step under deepspeed; unrelated." + }, + { + "left": "issue:43701", + "right": "issue:44749", + "accept": false, + "reason": "Checkpoint key mismatch vs data filtering slowdown; no shared 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": 26, + "review_unit_index": 46, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:39692", + "issue:41628", + "issue:42371", + "issue:42994", + "issue:43232", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43723", + "issue:43784", + "issue:43873", + "issue:43986", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44373", + "issue:44623", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44945", + "issue:44964", + "issue:45092", + "issue:45137", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45397", + "issue:45478", + "issue:45656", + "issue:46132", + "issue:46326", + "issue:46399", + "issue:46463", + "issue:46726", + "issue:46752", + "issue:46821", + "issue:46851", + "issue:46877", + "issue:47000", + "issue:47020" + ], + "soft_pairs": [ + "issue:43646|issue:44162", + "issue:44623|issue:45656", + "issue:44871|issue:46132", + "issue:46752|issue:47020", + "issue:44964|issue:45137", + "issue:45375|issue:45478", + "issue:42994|issue:43482", + "issue:39692|issue:46726", + "issue:43475|issue:43643", + "issue:43475|issue:46821", + "issue:43986|issue:46726", + "issue:44188|issue:46851", + "issue:28282|issue:30990", + "issue:45230|issue:45397", + "issue:43232|issue:43673", + "issue:46877|issue:47000", + "issue:45310|issue:45375", + "issue:44038|issue:44623", + "issue:46463|issue:46877", + "issue:41628|issue:44933", + "issue:44373|issue:46851", + "issue:42994|issue:43525", + "issue:44869|issue:45356", + "issue:43723|issue:46326", + "issue:42371|issue:45092", + "issue:43873|issue:44945", + "issue:43784|issue:46399" + ], + "prepared_review_unit_hash": "15dc14f8634b7e849cd2c8666a7153e78dc8852c5d28491af817e6a4c63f6df8", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 34860, + "estimated_input_tokens": 8715, + "estimated_eval_tokens": 17686 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly a grab-bag of unrelated Transformers bugs, with one clear duplicate chain around Qwen3.5-MoE `from_pretrained` failures. The most representative issue is the one that names the concrete missing config field behind that regression.", + "confidence": 0.69, + "canonical_issue_reason": "Issue 45375 is the best canonical issue because it identifies the concrete root cause (`deepstack_visual_indexes` being dropped by `@strict`) behind the Qwen3.5-MoE loading regression.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45375 is the strongest global representative: it is specific enough to explain the failure mode, while the other Qwen3.5-MoE reports are mostly generic symptoms of the same load problem.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Both are v5-related regressions, but one is generic custom-model initialization and the other is an ESM2-specific failure path; they are not the same bug." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "`processor.save_pretrained(...)` missing files and a double optimizer step under DeepSpeed are different code paths and different failure modes." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Gemma EOS config mismatch and `AttentionInterface.register` behavior change are unrelated issues in different subsystems." + }, + { + "left": "issue:46752", + "right": "issue:47020", + "accept": false, + "reason": "An empty chat list crashing `apply_chat_template` is unrelated to `XLMRobertaTokenizer` passing the wrong type to `Unigram`." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Loading a multimodal Phi-4 model and a DeepSpeed ZeRO3 deque pop are separate bugs with no clear shared code path." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": true, + "reason": "Both reports point to the same Qwen3.5-MoE `from_pretrained` regression under Transformers v5.4.x; 45375 gives the concrete missing-field root cause." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different quantization/loading problems, not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:46726", + "accept": false, + "reason": "A broken documentation example is not the same as AutoProcessor's missing-vision-backend error handling." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM 3 video attribute loss and `trust_remote_code=True` returning incomplete config objects are different model/config issues." + }, + { + "left": "issue:43475", + "right": "issue:46821", + "accept": false, + "reason": "These are distinct `AttributeError`s in different models (`Sam3VisionEncoderOutput` vs TimesFM `window_size`)." + }, + { + "left": "issue:43986", + "right": "issue:46726", + "accept": false, + "reason": "Both involve vision backend availability, but one is a video-model import crash without torchvision and the other is an unrecognized image processor error; not the same concrete bug." + }, + { + "left": "issue:44188", + "right": "issue:46851", + "accept": false, + "reason": "The first is a torch.compile attention-kernel divergence; the second is a Qwen3.5 variable-length `position_ids` bug." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "Missing PyTorch import support and Sentence Transformers getting stuck loading are different symptoms and likely different fixes." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "A placeholder bug report is not a duplicate of the Gemma-4 ZeRO3 `from_pretrained` failure." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Both are generation-related, but one is about `_update_model_kwargs_for_generation` after `sync_gpus` and the other is a v5.0 cache regression during chunked prefill." + }, + { + "left": "issue:46877", + "right": "issue:47000", + "accept": false, + "reason": "Import-time docstring lint output and an RTDetrLoss infeasible cost-matrix crash are unrelated." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": true, + "reason": "Both describe the same Qwen3.5-MoE `from_pretrained` breakage in v5.4.x; 45375 is the more specific root-cause report." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe loading and processor save-pretrained file emission are different problems despite both touching processor/model code." + }, + { + "left": "issue:46463", + "right": "issue:46877", + "accept": false, + "reason": "A decoder-input validation message bug and stdout doc-lint noise at import time are not the same issue." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "A missing `AutoImageProcessor` import and a non-existent `image_utils` import are separate API surface problems." + }, + { + "left": "issue:44373", + "right": "issue:46851", + "accept": false, + "reason": "A docstring typo for `position_ids` is unrelated to the Qwen3.5 variable-length input failure." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized model saving and a missing `pad_token_id` on `Llama4Config` are different failures." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper timestamp decoding and Kimi-K2.5 tokenizer codec handling are separate tokenizer/decoding regressions." + }, + { + "left": "issue:43723", + "right": "issue:46326", + "accept": false, + "reason": "Tokenizer loading in v5 and `continue_final_message` with `mistral_common` are different paths and different bugs." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 API guidance and InternVL2 meta-initialization incompatibility are unrelated issues." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior and incorrect pipeline-parallel LLM output are different runtime bugs." + }, + { + "left": "issue:43784", + "right": "issue:46399", + "accept": false, + "reason": "Sentence-transformers import failure and Qwen3.5 kernel-registration crash are unrelated code paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: 45310 and 45478 are clearly the same Qwen3.5-MoE from_pretrained regression, and 45375 is the more specific root-cause report naming the missing `deepstack_visual_indexes` field. The soft-edge rejections are conservative and the accepted pairs look mergeable as a single bugfix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 47, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:33290", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:42994", + "issue:43232", + "issue:43479", + "issue:43704", + "issue:43824", + "issue:43976", + "issue:44038", + "issue:44222", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44661", + "issue:44743", + "issue:44792", + "issue:44857", + "issue:44933", + "issue:44945", + "issue:45081", + "issue:45103", + "issue:45137", + "issue:45310", + "issue:45405", + "issue:45406", + "issue:45561", + "issue:45593", + "issue:45600", + "issue:45712", + "issue:45753", + "issue:46202", + "issue:46693", + "issue:46701", + "issue:46829", + "issue:46877", + "issue:47000", + "issue:47059", + "issue:47068", + "issue:47085" + ], + "soft_pairs": [ + "issue:45561|issue:47059", + "issue:45137|issue:47000", + "issue:41628|issue:44351", + "issue:36010|issue:41628", + "issue:44945|issue:45753", + "issue:44857|issue:46877", + "issue:44661|issue:47059", + "issue:45406|issue:46202", + "issue:42994|issue:43824", + "issue:44038|issue:45310", + "issue:45081|issue:45103", + "issue:45593|issue:46877", + "issue:44222|issue:45310", + "issue:46202|issue:46701", + "issue:42673|issue:43704", + "issue:43232|issue:44792", + "issue:42831|issue:44222", + "issue:42175|issue:44933", + "issue:44496|issue:45561", + "issue:45405|issue:46877", + "issue:30333|issue:33290", + "issue:44743|issue:47068", + "issue:45561|issue:45600", + "issue:43976|issue:44933", + "issue:30990|issue:35141", + "issue:42898|issue:44568", + "issue:43479|issue:46829", + "issue:42175|issue:45712", + "issue:44568|issue:47085", + "issue:44623|issue:45310", + "issue:45600|issue:47059", + "issue:42673|issue:46693" + ], + "prepared_review_unit_hash": "7d7fd15c23d22216fe62d25c95d2be0def45162def84013ed6e50baeace052cf", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37174, + "estimated_input_tokens": 9294, + "estimated_eval_tokens": 18844 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is mostly a heterogeneous similarity cluster with one real duplicate pair and many false-positive matches across unrelated bugs/features. The only strong duplicate is the Qwen3ForCausalLM VRAM leak reports.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue for the whole cluster: most items are unrelated. For the one true duplicate pair, issue 42673 is the better anchor because it is earlier and more discussed.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 42673 is the best representative only for the genuine duplicate subcluster (paired with 43704), but it does not represent the rest of the mixed cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45561", + "right": "issue:47059", + "accept": false, + "reason": "pytest-xdist file race vs a huggingface_hub CLI completion breakage; different code paths." + }, + { + "left": "issue:45137", + "right": "issue:47000", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error vs RTDetr loss infeasible cost-matrix crash; unrelated failures." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports: AutoImageProcessor vs HybridCache." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "GenerationMixin import failure is a different symbol/module than AutoImageProcessor." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallelism output bug vs Qwen3_5 image infinite loop; not the same bug." + }, + { + "left": "issue:44857", + "right": "issue:46877", + "accept": false, + "reason": "CUDA AMP crash in LwDetrImageLoss vs auto_docstring stderr at import time." + }, + { + "left": "issue:44661", + "right": "issue:47059", + "accept": false, + "reason": "Tokenizer-mapping model registration issue vs CLI completion attribute error." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer during serve vs Gemma pipeline roundtrip tokenization issue." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized save_pretrained failure vs missing Qwen2_5_VLForConditionalGeneration import; different functionality." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen3 MoE, but the titles point to different model/version-specific loading problems without clear same code path." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Mistral tokenizer regex crash vs auto_docstring annotation crash; unrelated." + }, + { + "left": "issue:45593", + "right": "issue:46877", + "accept": false, + "reason": "D-FINE auxiliary-loss behavior vs auto_docstring import-time diagnostics; unrelated." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained failure and Qwen3.5 MoE from_pretrained failure are different code paths." + }, + { + "left": "issue:46202", + "right": "issue:46701", + "accept": false, + "reason": "Gemma pipeline roundtrip issue vs image-text-to-text pipeline output class issue; related model family but different bug." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same Qwen3ForCausalLM VRAM leak in multiple dataloader threads; these are duplicate reports of the same bug." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs/update bug after sync_gpus vs janus image-generation test failure; different paths." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "FP8 accuracy regression vs FP8 save_pretrained failure; same broad feature area but different issues." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Packaging backend omission vs missing image_utils import; unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Model config loading error vs testing-utils xdist race; unrelated." + }, + { + "left": "issue:45405", + "right": "issue:46877", + "accept": false, + "reason": "Dependency-version bump complaint vs docstring diagnostics printing; unrelated." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting on failed training vs deepspeed OOM; different failure modes." + }, + { + "left": "issue:44743", + "right": "issue:47068", + "accept": false, + "reason": "Qwen3 recurrent-state reset bug vs host synchronization in deepspeed loss; unrelated." + }, + { + "left": "issue:45561", + "right": "issue:45600", + "accept": false, + "reason": "pytest-xdist race vs stale auto_mappings references; unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python-version compatibility issue vs missing import from image_utils; unrelated." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Loading hang for Sentence Transformers vs embedding resize/post_init reinitialization; unrelated." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Token cleanup behavior change vs BOS/EOS insertion regression; both tokenizer-related but not the same bug." + }, + { + "left": "issue:43479", + "right": "issue:46829", + "accept": false, + "reason": "Multimodal config default initialization vs VideoPrism output field mismatch; unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "TensorFlow extras packaging vs leftover dummy PT objects; unrelated." + }, + { + "left": "issue:44568", + "right": "issue:47085", + "accept": false, + "reason": "Special-tokens behavior regression vs tokenizer thread-safety regression; different defects." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "processor.save_pretrained missing files vs Qwen3.5 MoE loading error; different code paths." + }, + { + "left": "issue:45600", + "right": "issue:47059", + "accept": false, + "reason": "auto_mappings CI breakage vs CLI completion attribute error; unrelated." + }, + { + "left": "issue:42673", + "right": "issue:46693", + "accept": false, + "reason": "Qwen3ForCausalLM VRAM leak vs FA2 long-context performance regression; different issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The summary stays conservative: it identifies the identical Qwen3ForCausalLM VRAM-leak pair (42673/43704) as the only strong duplicate and treats the rest as heterogeneous false positives. The soft-edge verdicts are cautious and do not overclaim sameness where the titles point to different bugs or code paths." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 48, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28282", + "issue:30333", + "issue:34689", + "issue:35141", + "issue:39692", + "issue:42222", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:43116", + "issue:43295", + "issue:43479", + "issue:43723", + "issue:43784", + "issue:43824", + "issue:43874", + "issue:44016", + "issue:44060", + "issue:44075", + "issue:44079", + "issue:44186", + "issue:44190", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44843", + "issue:44869", + "issue:44933", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45200", + "issue:45356", + "issue:45405", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:46077", + "issue:46123", + "issue:46231", + "issue:46909", + "issue:47085" + ], + "soft_pairs": [ + "issue:44186|issue:47085", + "issue:46077|issue:46123", + "issue:42898|issue:45812", + "issue:43479|issue:44869", + "issue:43723|issue:44568", + "issue:44016|issue:46231", + "issue:39692|issue:46909", + "issue:43784|issue:43824", + "issue:43479|issue:44079", + "issue:42898|issue:44779", + "issue:44843|issue:47085", + "issue:27957|issue:30333", + "issue:42673|issue:47085", + "issue:30333|issue:34689", + "issue:42898|issue:43295", + "issue:42222|issue:45200", + "issue:43874|issue:45092", + "issue:44295|issue:45656", + "issue:44734|issue:45103", + "issue:42898|issue:45356", + "issue:30333|issue:35141", + "issue:22355|issue:28282", + "issue:44222|issue:44297", + "issue:44075|issue:46231", + "issue:43116|issue:44190", + "issue:45405|issue:45712", + "issue:44933|issue:45712", + "issue:44297|issue:45656", + "issue:42831|issue:44623", + "issue:44060|issue:45125" + ], + "prepared_review_unit_hash": "e740abd294763e2fb60855dc235cf71f1b00605461e5bdf057726a67afa2b6be", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36908, + "estimated_input_tokens": 9227, + "estimated_eval_tokens": 18710 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a true duplicate set: the items span many unrelated bugs and docs regressions across tokenizers, model loading, multimodal configs, training, and infrastructure. All soft pairs should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No clear canonical issue: the cluster is a grab bag of unrelated bugs, regressions, docs issues, and infrastructure problems.", + "canonical_pr_reason": null, + "best_issue_reason": "No single issue is a good global representative because the topics are heterogeneous and do not share one underlying code path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44186", + "right": "issue:47085", + "accept": false, + "reason": "Different problems: LayoutLMv2 NER/padding crash vs fast tokenizer thread-safety regression." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Gemma4 attention-mask bug and MaskGenerationPipeline partial-batch dropping are unrelated code paths." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Tokenizer cleanup behavior change is a different regression from Granite token-ID corruption." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config default initialization and Whisper decode crash are unrelated." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Generic AutoTokenizer loading issue is not the same as mDeBERTa special-token insertion regression." + }, + { + "left": "issue:44016", + "right": "issue:46231", + "accept": false, + "reason": "Notebook syntax error and a docs parameter-name mismatch are separate documentation issues." + }, + { + "left": "issue:39692", + "right": "issue:46909", + "accept": false, + "reason": "Different broken examples for different models; no shared underlying bug." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Sentence-transformers import failure and missing Qwen2.5-VL export are distinct import issues." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config defaulting bug vs ModelOutput key assignment bug; unrelated internals." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both involve tokenizers, but one is a global cleanup regression and the other is a Deepseek codec-handling regression." + }, + { + "left": "issue:27957", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention model-loading issue and MLflow reporting issue have nothing in common." + }, + { + "left": "issue:42673", + "right": "issue:47085", + "accept": false, + "reason": "Qwen3 VRAM leak in dataloader threads is different from tokenizer thread-safety." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow failure reporting and Llama 3.2 Vision loading regression are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:43295", + "accept": false, + "reason": "Tokenizer cleanup regression is not the same as processor.tokenizer/image-passing regression." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Vitpose model breakage and Gemma4 multimodal token-type defaults are different model families and bugs." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Missing image-processor method and remote-code meta-init incompatibility are different failures." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Position_ids buffer access error and double optimizer stepping under deepspeed are unrelated." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "Serve KV-cache indexing crash and auto_docstring future-annotations crash are separate subsystems." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Kimi-K2.5 tokenizer codec regression is a different issue from clean_up_tokenization_spaces behavior changes." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "MLflow reporting and embedding reinitialization are unrelated bugs." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Missing transformers.onnx module and missing PyTorch dependency are both import errors but with different causes." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "FP8 MoE save_pretrained bug and tokenizer save_pretrained config mismatch are different save paths." + }, + { + "left": "issue:44075", + "right": "issue:46231", + "accept": false, + "reason": "Optimizer SGD args being ignored is unrelated to a docs parameter-name typo." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Classification example returning empty results and image-classification dataset loading are separate example-script bugs." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "PEFT version bump and leftover dummy classes are unrelated maintenance issues." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Missing image_utils import and dummy_pt_objects cleanup are different problems." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch has nothing to do with deepspeed optimizer stepping." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FineGrainedFP8 accuracy and processor.save_pretrained missing files are unrelated." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Incorrect tied-weights warning and missing tensor-parallel plan are distinct Qwen3-Next 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": 26, + "review_unit_index": 49, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:30990", + "issue:38175", + "issue:41628", + "issue:42757", + "issue:42796", + "issue:42831", + "issue:42907", + "issue:43122", + "issue:43232", + "issue:43475", + "issue:43479", + "issue:43504", + "issue:43644", + "issue:43723", + "issue:44190", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44360", + "issue:44373", + "issue:44448", + "issue:44484", + "issue:44485", + "issue:44623", + "issue:44977", + "issue:45071", + "issue:45092", + "issue:45245", + "issue:45310", + "issue:45834", + "issue:46050", + "issue:46082", + "issue:46132", + "issue:46209", + "issue:46305", + "issue:46369", + "issue:46424", + "issue:46431", + "issue:46614", + "issue:46655", + "issue:46710", + "issue:46762", + "issue:46909", + "issue:46932", + "issue:47059" + ], + "soft_pairs": [ + "issue:44448|issue:46132", + "issue:42796|issue:43723", + "issue:44222|issue:44623", + "issue:43479|issue:43644", + "issue:45245|issue:46132", + "issue:41628|issue:42757", + "issue:44360|issue:46762", + "issue:44373|issue:44485", + "issue:46132|issue:46710", + "issue:44297|issue:45310", + "issue:43504|issue:47059", + "issue:45071|issue:46082", + "issue:43723|issue:44977", + "issue:44222|issue:44484", + "issue:43122|issue:46209", + "issue:46614|issue:46932", + "issue:44263|issue:45092", + "issue:44295|issue:44623", + "issue:43232|issue:46424", + "issue:46209|issue:46431", + "issue:43232|issue:45834", + "issue:30990|issue:38175", + "issue:44263|issue:44373", + "issue:44190|issue:46082", + "issue:46369|issue:46909", + "issue:43475|issue:46305", + "issue:46305|issue:46655", + "issue:42907|issue:46050", + "issue:44263|issue:44360", + "issue:42831|issue:45310", + "issue:22355|issue:27957" + ], + "prepared_review_unit_hash": "4a77ca958ad2d80bf4393b7a6d80e89ef0c23a2dba30408b7bcfe95a56e2d15e", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36527, + "estimated_input_tokens": 9132, + "estimated_eval_tokens": 18520 + }, + "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 Transformers regressions and model-specific bugs (tokenization, config validation, quantization, attention, save/load, generation, and CLI/import issues).", + "confidence": 0.95, + "canonical_issue_reason": "Issue 45071 is the broadest representative of the cluster\u2019s dominant v5 regression theme, but there is no single underlying bug shared by most items.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45071 is the best global representative because it describes a general Transformers v5 compatibility regression rather than a narrowly model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "Different bugs: Pegasus v4/v5 output change vs AttentionInterface.register altering callback behavior." + }, + { + "left": "issue:42796", + "right": "issue:43723", + "accept": false, + "reason": "Both are tokenizer-related, but one is a BPE-marker handling bug and the other is an AutoTokenizer import/load issue." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Different save_pretrained paths: FP8 MoE checkpoint saving vs processor artifact omissions." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Different code paths: multimodal config default initialization vs non-persistent buffer initialization." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Unrelated bugs: category-count runtime limit vs attention registration behavior." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols and different root causes." + }, + { + "left": "issue:44360", + "right": "issue:46762", + "accept": false, + "reason": "Different issues in attention/math: missing ReLU in DSA indexer vs numerical mismatch in Minimax MSA attention." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Docstring mismatch is not the same as a RoPE implementation discussion." + }, + { + "left": "issue:46132", + "right": "issue:46710", + "accept": false, + "reason": "AttentionInterface behavior change is not the same as a model accuracy regression after upgrade." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Different tokenizer persistence/load failures with different symptoms and likely different fixes." + }, + { + "left": "issue:43504", + "right": "issue:47059", + "accept": false, + "reason": "Unrelated failures: legacy field loading vs CLI breakage from huggingface_hub API changes." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Generic PretrainedConfig type-checking breakage is not the same as LlamaConfig head_dim validation." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer load failure and flash-attention generation issue are different code paths." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "Checkpoint-saving bug vs a question about the default max_shard_size." + }, + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer behavior changed across versions, but the other is an FP8 inference failure on B200." + }, + { + "left": "issue:46614", + "right": "issue:46932", + "accept": false, + "reason": "Different multilingual phonemizer state bug vs Gemma4 quantized embedder dtype casting fix." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return handling in a MoE indexer is unrelated to remote-code meta-initialization incompatibility." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Buffer registration/read error is not the same as missing files in processor.save_pretrained." + }, + { + "left": "issue:43232", + "right": "issue:46424", + "accept": false, + "reason": "Both touch generation/cache logic, but they involve different behaviors and different cache settings." + }, + { + "left": "issue:46209", + "right": "issue:46431", + "accept": false, + "reason": "FP8 inference on B200 and missing chat template are unrelated problems." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwarg update after sync_gpus is unrelated to long-OCR index overflow in Kosmos2.5." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "A loading hang and zero-probability outputs are different issues." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "MoE indexing bug vs docstring wording issue are not duplicates." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Dataset loading in a script is unrelated to LlamaConfig head_dim validation." + }, + { + "left": "issue:46369", + "right": "issue:46909", + "accept": false, + "reason": "Model memory overflow and a broken documentation example are different classes of problems." + }, + { + "left": "issue:43475", + "right": "issue:46305", + "accept": false, + "reason": "SAM3 missing attribute bug and Moonshine sliding-window off-by-one are different model paths." + }, + { + "left": "issue:46305", + "right": "issue:46655", + "accept": false, + "reason": "Both are video-related, but the bugs are different: mask off-by-one vs padding calculation." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both involve quantization/save-load, but they are different quantization flows and failure modes." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Different DSA-related bugs: split return handling vs missing ReLU." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue is not the same as Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:22355", + "right": "issue:27957", + "accept": false, + "reason": "One is a missing transformers.onnx module import, the other is an XLMRoberta flash-attention question." + } + ] + }, + "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": 26, + "review_unit_index": 50, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:22355", + "issue:27957", + "issue:28218", + "issue:30990", + "issue:33357", + "issue:39379", + "issue:39401", + "issue:41628", + "issue:42757", + "issue:42796", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43540", + "issue:43644", + "issue:43646", + "issue:43824", + "issue:44038", + "issue:44188", + "issue:44261", + "issue:44279", + "issue:44327", + "issue:44373", + "issue:44483", + "issue:44568", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45092", + "issue:45278", + "issue:45357", + "issue:45381", + "issue:45406", + "issue:45593", + "issue:45753", + "issue:46002", + "issue:46172", + "issue:46305", + "issue:46328", + "issue:46369", + "issue:46402", + "issue:46909", + "issue:47059" + ], + "soft_pairs": [ + "issue:45092|issue:45357", + "issue:44279|issue:46328", + "issue:44483|issue:45406", + "issue:44964|issue:47059", + "issue:42831|issue:44038", + "issue:44829|issue:46002", + "issue:41628|issue:43824", + "issue:45381|issue:46305", + "issue:43408|issue:46402", + "issue:44261|issue:45092", + "issue:43644|issue:46305", + "issue:39379|issue:46172", + "issue:22355|issue:28218", + "issue:42831|issue:45753", + "issue:44327|issue:47059", + "issue:39401|issue:44568", + "issue:42757|issue:45278", + "issue:43408|issue:46369", + "issue:44188|issue:44373", + "issue:42886|issue:42994", + "issue:27957|issue:30990", + "issue:42886|issue:43066", + "issue:42796|issue:43122", + "issue:45381|issue:46909", + "issue:43232|issue:45593", + "issue:43540|issue:46909", + "issue:43646|issue:46002", + "issue:42907|issue:44945", + "issue:30990|issue:33357" + ], + "prepared_review_unit_hash": "81e67b80da83028743db2a9e6f062578d61974dc1fad01ef2d3909cbf98b6207", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 35923, + "estimated_input_tokens": 8981, + "estimated_eval_tokens": 18218 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous; the soft edges mostly connect superficially similar regression reports, but none are clearly the same underlying bug/change. I would not merge any of the proposed pairs.", + "confidence": 0.89, + "canonical_issue_reason": "If a representative issue is needed, 45278 is the broadest and most umbrella-like report, but the set is too mixed to treat it as a true canonical duplicate target.", + "canonical_pr_reason": null, + "best_issue_reason": "45278 is the best global representative because it captures the broad import-regression theme more than the others, even though it should not be used to collapse the whole cluster into one duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Both are multimodal/model-regression reports, but one is a meta-init incompatibility and the other is incorrect save_pretrained key handling." + }, + { + "left": "issue:44279", + "right": "issue:46328", + "accept": false, + "reason": "A generic dependency issue is unrelated to a tokenizer padding performance regression." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Serve request acceptance failure and a Gemma4Processor attribute crash are different code paths." + }, + { + "left": "issue:44964", + "right": "issue:47059", + "accept": false, + "reason": "Model loading failure and CLI completion breakage are unrelated failures." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "FP8 accuracy degradation and Qwen3-VL-Moe loading/behavior issues are not the same bug." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Flash-attention training degeneration and NaNs after checkout/init have different symptoms and likely different roots." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different model areas." + }, + { + "left": "issue:45381", + "right": "issue:46305", + "accept": false, + "reason": "A wrong video position-id computation and an off-by-one sliding window mask are different tensor bugs." + }, + { + "left": "issue:43408", + "right": "issue:46402", + "accept": false, + "reason": "A model-type warning and missing text-encoder weights are not the same failure." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Layernorm epsilon precision drift is unrelated to remote-code checkpoint meta-initialization incompatibility." + }, + { + "left": "issue:43644", + "right": "issue:46305", + "accept": false, + "reason": "Non-persistent buffer initialization junk and streaming mask off-by-one are unrelated." + }, + { + "left": "issue:39379", + "right": "issue:46172", + "accept": false, + "reason": "These affect different model internals: mllama mask handling vs a DeepSeek tensor transpose issue." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Missing an ONNX module and adding an extra space after added tokens are unrelated tokenizer/export problems." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FP8 accuracy loss and a model looping on a specific image are different underlying bugs." + }, + { + "left": "issue:44327", + "right": "issue:47059", + "accept": false, + "reason": "QA top-k bounds checking and a hub CLI completion attribute error do not share a concrete code path." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Wrong offset mapping and missing BOS/EOS insertion are distinct tokenizer regressions." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "Both mention import failures, but one is a specific missing symbol while the other is a broad post-update report; not enough evidence they are the same bug." + }, + { + "left": "issue:43408", + "right": "issue:46369", + "accept": false, + "reason": "A model-type warning and a memory overflow while calling the model are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel branching under torch.compile and a docstring typo are not duplicates." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model save failure are different subsystems." + }, + { + "left": "issue:27957", + "right": "issue:30990", + "accept": false, + "reason": "XLM-R Flash Attention 2 support and a sentence-transformers loading hang are not the same issue." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline cache loading and wrong tokenizer decoder type are separate tokenizer regressions." + }, + { + "left": "issue:42796", + "right": "issue:43122", + "accept": false, + "reason": "Tokenizer BPE-marker handling and cross-version tokenization differences are related only broadly, not the same concrete bug." + }, + { + "left": "issue:45381", + "right": "issue:46909", + "accept": false, + "reason": "A model video-input bug and a broken documentation example are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus and D-FINE auxiliary-loss behavior are different changes." + }, + { + "left": "issue:43540", + "right": "issue:46909", + "accept": false, + "reason": "Video-input ValueError in Qwen3OmniMoe is unrelated to a docs example bug." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage and NaNs in classification heads do not look like the same code-path problem." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Quantized-model save failures and pipeline-parallel wrong outputs are different failures." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "A loading hang and a macOS bus error on a different model/version 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": 26, + "review_unit_index": 51, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:22355", + "issue:29159", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42907", + "issue:43116", + "issue:43122", + "issue:43329", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43576", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44060", + "issue:44117", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44488", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45478", + "issue:45600", + "issue:46172", + "issue:46326", + "issue:46614", + "issue:46762", + "issue:47059" + ], + "soft_pairs": [ + "issue:44295|issue:44297", + "issue:42907|issue:44964", + "issue:43576|issue:43784", + "issue:43976|issue:44945", + "issue:43329|issue:47059", + "issue:41628|issue:45278", + "issue:43756|issue:44060", + "issue:42371|issue:44263", + "issue:44117|issue:44704", + "issue:42907|issue:43937", + "issue:43784|issue:45600", + "issue:39401|issue:44488", + "issue:39401|issue:44521", + "issue:44625|issue:44829", + "issue:44704|issue:45290", + "issue:46172|issue:46762", + "issue:43122|issue:46326", + "issue:43408|issue:43475", + "issue:44360|issue:44485", + "issue:43122|issue:44521", + "issue:44038|issue:44297", + "issue:42907|issue:43452", + "issue:22355|issue:29159", + "issue:44038|issue:44295", + "issue:45216|issue:45478", + "issue:44038|issue:44222", + "issue:43824|issue:44351", + "issue:43116|issue:45200", + "issue:39401|issue:46614", + "issue:43122|issue:44779", + "issue:42757|issue:44351" + ], + "prepared_review_unit_hash": "03151fc76614f2c04ee5de94b1fca07c73b619a506aa5dadc732be8d890d7093", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36827, + "estimated_input_tokens": 9207, + "estimated_eval_tokens": 18670 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is noisy: most items are unrelated tokenizer, multimodal, or model-specific bugs that only share vague wording. The closest common thread is a set of import/export regressions in Transformers v5, so the broad import-regression report is the best representative, but most soft links should be rejected.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45278 is the broadest representative of the import-regression subset and can subsume several specific \"cannot import\" reports, whereas the others are narrower symptom reports.", + "canonical_pr_reason": null, + "best_issue_reason": "It is the most general user-facing report in the cluster and best captures the shared failure mode of top-level import breakage after upgrade.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Different bugs: buffer access error vs tokenizer_config mismatch." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Different model/save/load paths and different failure modes." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Both are import/runtime regressions, but they concern different symbols and code paths." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Unrelated issues: Python compatibility vs pipeline-parallel output correctness." + }, + { + "left": "issue:43329", + "right": "issue:47059", + "accept": false, + "reason": "Multimodal token counting bug vs CLI completion breakage from huggingface_hub." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Related by broad import breakage, but 45278 is an umbrella report and not the same concrete missing symbol." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Different model-specific architectural bugs, not the same code-path." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 configuration guidance vs a DSA indexer tensor-splitting issue." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Both touch from_pretrained-related code, but the defects are different and not mergeable into one fix." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Different GLM-5 generation/configuration failures." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Importing sentence-transformers failing on nn vs stale auto_mappings references." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Different tokenizer/model issues for different models." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer offset mapping bug vs assistant mask generation for multimodal chat templates." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Config propagation bug vs training instability under flash_attention_3." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "cached_file kwargs forwarding vs chat template crash on tool-call messages." + }, + { + "left": "issue:46172", + "right": "issue:46762", + "accept": false, + "reason": "Different model implementations and different numerical discrepancy reports." + }, + { + "left": "issue:43122", + "right": "issue:46326", + "accept": false, + "reason": "Both tokenizer-related, but one is a version regression and the other is a mistral_common chat-template behavior bug." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Same model family, but a config warning is not the same bug as a missing output attribute." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Completely different model/algorithm bugs." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Tokenization regression vs multimodal assistant mask handling." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe load bug vs tokenizer_class metadata mismatch." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Dequantized model save failure vs gguf_file loading breakage." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "Different modules and different tokenizer behavior problems." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Different subsystems: model loading vs buffer read/write." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Related Qwen3.5 family, but one is save_pretrained regression and the other is from_pretrained loading error." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Different bugs: model load failure vs FP8 save_pretrained issue." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different fixes." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Example script multi-label output bug vs Gemma 4 mm_token_type_ids defaulting." + }, + { + "left": "issue:39401", + "right": "issue:46614", + "accept": false, + "reason": "Different tokenizer implementations and different statefulness bugs." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Different tokenization regressions affecting different models/paths." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Both are import failures, but one is huggingface_hub and the other is transformers; 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": 26, + "review_unit_index": 52, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:36296", + "issue:38617", + "issue:39379", + "issue:42175", + "issue:42371", + "issue:42673", + "issue:42898", + "issue:43066", + "issue:43116", + "issue:43232", + "issue:43278", + "issue:43493", + "issue:43611", + "issue:43618", + "issue:43825", + "issue:43874", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44075", + "issue:44188", + "issue:44355", + "issue:44485", + "issue:44610", + "issue:44625", + "issue:44779", + "issue:44936", + "issue:44938", + "issue:44945", + "issue:45072", + "issue:45083", + "issue:45278", + "issue:45357", + "issue:45446", + "issue:45478", + "issue:45753", + "issue:46002", + "issue:46231", + "issue:46328", + "issue:46614", + "issue:46726", + "issue:46829", + "issue:46877", + "issue:46909", + "issue:46981", + "issue:47116" + ], + "soft_pairs": [ + "issue:36296|issue:43116", + "issue:43278|issue:44945", + "issue:22355|issue:29127", + "issue:42175|issue:44938", + "issue:39379|issue:43232", + "issue:43493|issue:45753", + "issue:46231|issue:46877", + "issue:44188|issue:46328", + "issue:38617|issue:45278", + "issue:43937|issue:44485", + "issue:42898|issue:46614", + "issue:43976|issue:45446", + "issue:43611|issue:46002", + "issue:44038|issue:45478", + "issue:43066|issue:44779", + "issue:42673|issue:45072", + "issue:44355|issue:46231", + "issue:44625|issue:46002", + "issue:45357|issue:45478", + "issue:43116|issue:43493", + "issue:44610|issue:46909", + "issue:45083|issue:45446", + "issue:42371|issue:43874", + "issue:46726|issue:46981", + "issue:44075|issue:44936", + "issue:43618|issue:46829", + "issue:46328|issue:47116", + "issue:43116|issue:43825" + ], + "prepared_review_unit_hash": "bc57161992e5b4c4af5698861416fd88103d0004c50d46d5ee06f42c2a14a775", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35732, + "estimated_input_tokens": 8933, + "estimated_eval_tokens": 18122 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the paired items are mostly unrelated regressions, import errors, model-specific bugs, docs issues, and pipeline problems. I found no soft pair that looks like the same underlying bug/change, so no duplicate merge is warranted.", + "confidence": 0.94, + "canonical_issue_reason": "issue:42371 is the broadest and most discussed item in the set, but the cluster is not a true duplicate group; it is only a weak stand-in representative.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42371 has the highest discussion activity and the most general framing among the issues, making it the best representative if one must be chosen.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Different bugs: tensor-parallel training behavior vs. multi-label example script returning empty results." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Both involve training/eval behavior, but one is a dtype regression and the other is incorrect output under pipeline parallelism." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Unrelated: missing transformers.onnx module vs. LayoutLMv3 error-message clarity." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Different failures: pip extra backend composition vs. Python 3.14 import/load breakage." + }, + { + "left": "issue:39379", + "right": "issue:43232", + "accept": false, + "reason": "mllama mask handling bug and generation kwargs update after sync_gpus are different code paths." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Different model-specific regressions: SigLIP2 implementation mismatch vs. Qwen3_5 image-specific infinite loop." + }, + { + "left": "issue:46231", + "right": "issue:46877", + "accept": false, + "reason": "Docs-only mismatch versus import-time stdout logging from auto_docstring; not the same issue." + }, + { + "left": "issue:44188", + "right": "issue:46328", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to SequenceFeatureExtractor padding inefficiency." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-related, but one is a specific missing symbol and the other is a broad post-upgrade import breakage." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "GenerationConfig validation error and GLM-5 RoPE discussion do not describe the same bug." + }, + { + "left": "issue:42898", + "right": "issue:46614", + "accept": false, + "reason": "Tokenization cleanup behavior change and stale espeak backend state are separate tokenizer issues." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Version-support problem vs. incorrect PyTorch version gate for AuxRequest import; different fixes needed." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Model loading with base_model_prefix is unrelated to NaNs in classification heads after checkout/init." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3.5/Qwen3-VL-Moe, but they are different regressions in different versions and code paths." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Tokenizer regressions in v5, but for different models and different concrete failures." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in multithreaded Qwen3ForCausalLM use is unrelated to dtype mismatches in inference tests." + }, + { + "left": "issue:44355", + "right": "issue:46231", + "accept": false, + "reason": "Compiled Python file errors are unrelated to a documentation parameter-name typo." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and classification-head NaNs are separate bugs." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5-related, but one is save_pretrained key regression and the other is from_pretrained loading failure." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Different scopes: example script output handling versus model implementation discrepancy." + }, + { + "left": "issue:44610", + "right": "issue:46909", + "accept": false, + "reason": "Processor input-size mismatch is unrelated to a broken documentation example." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Feature-length helper behavior and PyTorch version checking are not the same underlying change." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 API guidance and missing image-patch method in GLM46V are unrelated." + }, + { + "left": "issue:46726", + "right": "issue:46981", + "accept": false, + "reason": "Both involve AutoProcessor, but one is missing vision backend messaging and the other is retry behavior on HTTP 429." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer argument handling is unrelated to trainer.evaluate failing after train." + }, + { + "left": "issue:43618", + "right": "issue:46829", + "accept": false, + "reason": "Both are output-field issues, but they affect different model classes and different attributes." + }, + { + "left": "issue:46328", + "right": "issue:47116", + "accept": false, + "reason": "Padding performance and streaming pipeline regression are separate concerns." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Example-script multi-label behavior is not the same as pipeline translation-task support messaging." + } + ] + }, + "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": 26, + "review_unit_index": 53, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28348", + "issue:33453", + "issue:39379", + "issue:39401", + "issue:42222", + "issue:42673", + "issue:42796", + "issue:43064", + "issue:43066", + "issue:43441", + "issue:43450", + "issue:43673", + "issue:43688", + "issue:43824", + "issue:43827", + "issue:43937", + "issue:43986", + "issue:43992", + "issue:44117", + "issue:44263", + "issue:44291", + "issue:44380", + "issue:44410", + "issue:44466", + "issue:44704", + "issue:44779", + "issue:44977", + "issue:44995", + "issue:45071", + "issue:45092", + "issue:45310", + "issue:45405", + "issue:45446", + "issue:45563", + "issue:45588", + "issue:46002", + "issue:46077", + "issue:46328", + "issue:46640", + "issue:46697", + "issue:46710", + "issue:46752", + "issue:46762", + "issue:46897", + "issue:46981", + "issue:47000", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:43992|issue:44704", + "issue:43066|issue:46710", + "issue:28348|issue:46640", + "issue:42673|issue:42796", + "issue:44995|issue:46697", + "issue:43827|issue:45446", + "issue:33453|issue:39401", + "issue:45092|issue:46762", + "issue:44263|issue:46762", + "issue:43824|issue:46981", + "issue:43064|issue:43673", + "issue:42222|issue:43441", + "issue:43673|issue:46328", + "issue:39379|issue:43688", + "issue:42222|issue:47000", + "issue:45405|issue:45446", + "issue:43937|issue:45563", + "issue:42222|issue:46752", + "issue:44380|issue:46077", + "issue:44291|issue:45071", + "issue:44117|issue:45588", + "issue:44466|issue:46002", + "issue:39379|issue:46897", + "issue:39401|issue:44779", + "issue:44410|issue:45310", + "issue:43450|issue:43986", + "issue:43937|issue:44977", + "issue:44704|issue:47110", + "issue:45071|issue:47059" + ], + "prepared_review_unit_hash": "435b082902a2d580ca484b02ad6f27f6d0794adf72ab05dce4702607f234ec93", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36168, + "estimated_input_tokens": 9042, + "estimated_eval_tokens": 18340 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft pair looks like a true duplicate; most are only loosely related by broad themes like tokenizer regressions, FlashAttention, or v5 behavior changes. I\u2019d use issue:44779 as the best representative issue because it has the clearest concrete regression report, but I would not merge any of these pairs.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44779 is the strongest standalone bug report here: it describes a concrete tokenizer regression in v5 with a specific model and clear behavioral change, making it the best representative issue for this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44779 has the most actionable, model-specific tokenizer regression description and is the best single issue to anchor related triage.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Different bugs: one is missing encoder weights during UMT5 loading, the other is AutoProcessor forwarding kwargs incorrectly." + }, + { + "left": "issue:43066", + "right": "issue:46710", + "accept": false, + "reason": "Both are v5-related, but one is a tokenizer decoder-type issue and the other is an accuracy regression in a model; not the same code-path bug." + }, + { + "left": "issue:28348", + "right": "issue:46640", + "accept": false, + "reason": "FlashAttention is the only overlap; one is a feature request for GPT2 support and the other is an unspecified FA2 issue." + }, + { + "left": "issue:42673", + "right": "issue:42796", + "accept": false, + "reason": "Unrelated model areas: VRAM leak in Qwen3ForCausalLM versus tokenizer BPE marker handling in Ministral." + }, + { + "left": "issue:44995", + "right": "issue:46697", + "accept": false, + "reason": "Different subsystems entirely: GLM MoE cache crash versus Swin2SR image processor padding." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still mentioning pipeline() is unrelated to a PyTorch version check bug in flex_attention." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Both are tokenizer regressions, but one is a generic loading regression and the other is a Qwen3 offset_mapping bug; not the same issue." + }, + { + "left": "issue:45092", + "right": "issue:46762", + "accept": false, + "reason": "Meta-init incompatibility for old InternVL2 checkpoints is not the same as numerical differences in Minimax MSA attention." + }, + { + "left": "issue:44263", + "right": "issue:46762", + "accept": false, + "reason": "A torch.split return-value bug in GlmMoeDsaIndexer is unrelated to Minimax attention numerical differences." + }, + { + "left": "issue:43824", + "right": "issue:46981", + "accept": false, + "reason": "ImportError for a missing class and repeated HTTP 429 retries are different failure modes in different layers." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Optimizer-state corruption in Trainer/FSDP2/PEFT is unrelated to missing GenerationMixin cache during chunked prefill." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "Vitpose breakage and Ministral FlashAttention failure are separate model-specific bugs." + }, + { + "left": "issue:43673", + "right": "issue:46328", + "accept": false, + "reason": "Generation cache behavior and SequenceFeatureExtractor padding performance are unrelated." + }, + { + "left": "issue:39379", + "right": "issue:43688", + "accept": false, + "reason": "mllama mask handling and OLMoE/GPT Oss auxiliary-loss normalization affect different code paths." + }, + { + "left": "issue:42222", + "right": "issue:47000", + "accept": false, + "reason": "Vitpose model breakage is unrelated to RTDetrLoss infeasible cost-matrix errors." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A dependency-version bump issue is unrelated to an incorrect PyTorch version check in flex_attention." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "Invalid GenerationConfig and a stale paged-generate warning are different generation issues." + }, + { + "left": "issue:42222", + "right": "issue:46752", + "accept": false, + "reason": "Vitpose breakage is unrelated to apply_chat_template crashing on an empty list." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "Both touch attention, but GPT2 attention scaling under SDPA/FlashAttention is not the same bug as Gemma4 building causal masks when bidirectional attention is requested." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "One is an unexpected init_empty_weights argument, the other is a PretrainedConfig type-checking break; related to v5, but not the same defect." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer-mapping None handling and a flash_attention crash on s_aux=None are unrelated." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "Serialization of tied weights versus NaNs after checkout/init are different problems." + }, + { + "left": "issue:39379", + "right": "issue:46897", + "accept": false, + "reason": "Both mention training-loss style bugs, but the affected models and failure mechanisms differ." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but one is about offset_mapping and the other about tokenization correctness on DeepSeek; they do not look like the same concrete bug." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing qkv/gate projections in qwen3next and a Qwen3.5 MoE from_pretrained error are not the same code-path." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Video processor output-shape bugs and AutoProcessor import/crash behavior without torchvision are distinct." + }, + { + "left": "issue:44704", + "right": "issue:47110", + "accept": false, + "reason": "Keyword forwarding in AutoProcessor and a special_tokens list-vs-dict type error are different bugs." + }, + { + "left": "issue:45071", + "right": "issue:47059", + "accept": false, + "reason": "PretrainedConfig type checking and CLI completion breakage from huggingface_hub are unrelated." + }, + { + "left": "issue:43066", + "right": "issue:46762", + "accept": false, + "reason": "A tokenizer/decoder-type issue is unrelated to Minimax attention numerical discrepancies." + }, + { + "left": "issue:44977", + "right": "issue:46697", + "accept": false, + "reason": "DeepSeek tokenizer regression and Swin2SR padding behavior 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": 26, + "review_unit_index": 54, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42617", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43208", + "issue:43408", + "issue:43742", + "issue:43937", + "issue:44075", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44496", + "issue:44829", + "issue:44877", + "issue:44933", + "issue:45071", + "issue:45081", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45478", + "issue:45561", + "issue:45632", + "issue:45656", + "issue:46178", + "issue:46209", + "issue:46231", + "issue:46298", + "issue:46328", + "issue:46512", + "issue:46617", + "issue:46752", + "issue:46762", + "issue:46894", + "issue:46981", + "issue:47116" + ], + "soft_pairs": [ + "issue:45341|issue:46328", + "issue:45071|issue:46512", + "issue:44829|issue:45216", + "issue:45341|issue:45561", + "issue:45632|issue:46981", + "issue:43408|issue:43742", + "issue:42617|issue:45341", + "issue:44117|issue:45081", + "issue:38617|issue:41628", + "issue:43122|issue:46298", + "issue:42222|issue:46178", + "issue:39401|issue:46209", + "issue:44877|issue:46231", + "issue:42371|issue:44261", + "issue:44484|issue:45310", + "issue:43937|issue:44496", + "issue:44933|issue:45278", + "issue:44261|issue:46762", + "issue:38617|issue:44220", + "issue:42371|issue:46617", + "issue:44410|issue:45478", + "issue:44075|issue:45656", + "issue:43097|issue:44493", + "issue:43066|issue:43122", + "issue:44263|issue:44485", + "issue:43208|issue:46894", + "issue:46752|issue:47116" + ], + "prepared_review_unit_hash": "7b6eb736882329f1441df19f0437f645a7f5997d89fcbc4943943bc74e6102ae", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35435, + "estimated_input_tokens": 8859, + "estimated_eval_tokens": 17974 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is highly heterogeneous: most items are unrelated bugs/regressions across tokenizer, model loading, config, optimizer, and testing utilities. I found no soft-edge pair that is clearly the same underlying issue; all candidate pairs are too broad or target different code paths.", + "confidence": 0.93, + "canonical_issue_reason": "No single issue cleanly represents the whole set because it mixes many unrelated bugs. If one must be chosen as the best canonical anchor, issue 43066 is the most concrete and self-contained regression report.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43066 is the strongest triage anchor: it describes a specific, user-visible tokenizer regression with a clear symptom and likely narrow code path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45341", + "right": "issue:46328", + "accept": false, + "reason": "One is a testing_utils bug; the other is a performance issue in SequenceFeatureExtractor.pad. Different modules and failure modes." + }, + { + "left": "issue:45071", + "right": "issue:46512", + "accept": false, + "reason": "PretrainedConfig type-checking regression vs stale config.dtype after from_config; related to config handling but not the same bug." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Flash-attention training degeneration and Qwen3.5 save_pretrained regression are unrelated code paths." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "A testing_utils bug versus a pytest-xdist race condition; different symptoms and fixes." + }, + { + "left": "issue:45632", + "right": "issue:46981", + "accept": false, + "reason": "Cache-path collision for trust_remote_code is unrelated to repeated HTTP 429 retries in AutoProcessor.from_pretrained." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "SAM3 model-type warning and MobileLLM key error are different model-loading problems." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "3d_parallel.py execution failure and a small testing_utils bug do not share the same underlying defect." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer mapping None-handling and Mistral regex patch backend_tokenizer access are distinct tokenizer issues." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different compatibility breaks." + }, + { + "left": "issue:43122", + "right": "issue:46298", + "accept": false, + "reason": "Tokenizer output regression and noisy logging are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:46178", + "accept": false, + "reason": "Broken vitpose models and unsupported tensorflow_text backend mapping are separate failures." + }, + { + "left": "issue:39401", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3 offset_mapping bug and FP8 inference failure on B200 are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:46231", + "accept": false, + "reason": "Strict config loading for granite_speech and a docs/parameter-name mismatch are different issues." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API-settings guidance and MLA rms_norm_eps precision behavior are unrelated." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "save_pretrained shard-size question and Qwen3.5 Moe from_pretrained error are different concerns." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GenerationConfig validation and unrecognized model/config.json loading are separate code paths." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "A missing image_utils import and broad import errors after upgrade are too different to treat as the same bug." + }, + { + "left": "issue:44261", + "right": "issue:46762", + "accept": false, + "reason": "Both mention numerical behavior, but they affect different attention implementations and likely different fixes." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "ImportError for layer_type_validation is unrelated to _torch_extract_fbank_features()." + }, + { + "left": "issue:42371", + "right": "issue:46617", + "accept": false, + "reason": "TF32 behavior settings and trust_remote_code cache FileNotFoundError are unrelated." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Missing qwen3next projections and Qwen3.5 Moe from_pretrained error are different model-definition/loading bugs." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args ignored and optimizer.step() called twice with deepspeed are optimizer-related but not the same concrete defect." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Removed tie_embeddings_and_encoder_decoder and unexpected position-id keys are distinct regression reports." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Both are tokenizer v5 regressions, but one is decoder-type mismatch and the other is changed tokenization output; not clearly the same code-path bug." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "torch.split return-value issue and GLM-5 RoPE implementation problem are different bugs." + }, + { + "left": "issue:43208", + "right": "issue:46894", + "accept": false, + "reason": "xLSTM training blockers and CohereASR double-shift loss bug are unrelated model-specific issues." + }, + { + "left": "issue:46752", + "right": "issue:47116", + "accept": false, + "reason": "Empty-list apply_chat_template crash and streaming pipeline regression 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": 26, + "review_unit_index": 55, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:27957", + "issue:38617", + "issue:41628", + "issue:42757", + "issue:42886", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43441", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43673", + "issue:43874", + "issue:43931", + "issue:43986", + "issue:44038", + "issue:44077", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44315", + "issue:44360", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44829", + "issue:44987", + "issue:45003", + "issue:45071", + "issue:45341", + "issue:45372", + "issue:45375", + "issue:45440", + "issue:45464", + "issue:45478", + "issue:45588", + "issue:45753", + "issue:46050", + "issue:46123", + "issue:46172", + "issue:46369", + "issue:46897", + "issue:46981" + ], + "soft_pairs": [ + "issue:44077|issue:46981", + "issue:44483|issue:45464", + "issue:43931|issue:44410", + "issue:43673|issue:45341", + "issue:41628|issue:45071", + "issue:27957|issue:43352", + "issue:44484|issue:45478", + "issue:43874|issue:44261", + "issue:43012|issue:43577", + "issue:44263|issue:46172", + "issue:43441|issue:44829", + "issue:44360|issue:46172", + "issue:43931|issue:44038", + "issue:43097|issue:43452", + "issue:43986|issue:45341", + "issue:45003|issue:45341", + "issue:42886|issue:46981", + "issue:44261|issue:44485", + "issue:45753|issue:46369", + "issue:43531|issue:45341", + "issue:43097|issue:45588", + "issue:44117|issue:44987", + "issue:43874|issue:44263", + "issue:43673|issue:46123", + "issue:44410|issue:45375", + "issue:43097|issue:46050", + "issue:45440|issue:46172", + "issue:38617|issue:42757", + "issue:44038|issue:44410", + "issue:43986|issue:45372", + "issue:43064|issue:44315", + "issue:43208|issue:46897" + ], + "prepared_review_unit_hash": "8d375073e9e479f0e90eb1fde7e1d31788a5950b3493d5bb054d6e3c567eb607", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36552, + "estimated_input_tokens": 9138, + "estimated_eval_tokens": 18532 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is not a duplicate set; it mixes unrelated regressions across imports, model loading, Flash Attention, config/schema issues, and pipeline/runtime behavior.", + "confidence": 0.98, + "canonical_issue_reason": "No single issue is a clear canonical duplicate target; the items describe different failure modes and code paths.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue fits the cluster, because there is no one underlying bug shared by the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44077", + "right": "issue:46981", + "accept": false, + "reason": "Different bugs: a patchtsmixer API restriction vs AutoProcessor retry behavior on 429/HF_ENDPOINT." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Both involve chat/completions, but one is request validation and the other is streaming inference on a specific Qwen model." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Different Qwen model failures: weight-shape mismatch vs missing attention projections in qwen3next." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: GenerationMixin chunked_prefill cache issue vs a testing_utils bug." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "Different regressions: AutoImageProcessor import failure vs PreTrainedConfig type-checking breakage." + }, + { + "left": "issue:27957", + "right": "issue:43352", + "accept": false, + "reason": "Both mention Flash Attention 2, but they affect different models and different code paths." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "Unrelated: save_pretrained shard-size question vs Qwen3.5 MoE loading error." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different model internals: missing image-patch helper vs missing rms_norm_eps in MLA q_a_layernorm." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "Both involve dtype/precision, but one is a PyTorch warning during compile and the other is a Blip2 dtype mismatch on load." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "Different code paths: GlmMoeDsaIndexer split handling vs deepseek-v4 tensor transpose/cast." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both mention FlashAttention, but one is a model support failure and the other is training degeneracy with flash_attention_3." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "Different bugs in different models: DSA indexer activation issue vs deepseek-v4 tensor handling." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both are Qwen3-VL-related, but one is a specific shape mismatch and the other is a separate Qwen3-VL-Moe v5 issue." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Different problems: removed tie_embeddings API vs gguf_file/from_pretrained loading breakage." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: AutoProcessor/torchvision dependency crash vs testing_utils bug." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "Different areas: sys.modules access in modeling_utils vs testing_utils." + }, + { + "left": "issue:42886", + "right": "issue:46981", + "accept": false, + "reason": "Both are hub/network-related, but offline cache loading and HTTP 429 retry behavior are distinct bugs." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "Different model math issues: MLA layernorm epsilon bug vs GLM-5 RoPE implementation question." + }, + { + "left": "issue:45753", + "right": "issue:46369", + "accept": false, + "reason": "Different symptoms and models: Qwen3_5 infinite-loop on one image vs SAM3 memory overflow." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: Qwen3-MoE sliding_window behavior vs testing_utils bug." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Different issues: API removal in config handling vs flash_attention.py crashing on sink-less models." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Both are load-time failures, but TOKENIZER_MAPPING_NAMES=None and a specific model load regression are not the same bug." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Different GLM subsystems: image-patch count helper vs DSA torch.split return handling." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Different generation bugs: missing cache during chunked prefill vs MaskGenerationPipeline dropping the final partial batch." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen-family issues, but one is missing projections and the other is a config field silently dropped by @strict." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "Different behaviors: removed tie_embeddings API vs quantized model save/load breakage." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Different DeepSeek issues: divergence from remote implementation vs a tensor cast/transpose bug." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are ImportError reports, but they involve different missing symbols from different dependencies." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Related model family, but the reported failures are different and not the same concrete bug." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Different import/dependency problems affecting different model processors." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Different training issues: FSDP2/PEFT optimizer-state corruption vs Liger Kernel not applied with model_init." + }, + { + "left": "issue:43208", + "right": "issue:46897", + "accept": false, + "reason": "Different model-specific training bugs; only loosely similar as loss/training regressions." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 56, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:38617", + "issue:42890", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43502", + "issue:43550", + "issue:43825", + "issue:43827", + "issue:43873", + "issue:43906", + "issue:43976", + "issue:43986", + "issue:43992", + "issue:44016", + "issue:44062", + "issue:44117", + "issue:44351", + "issue:44368", + "issue:44380", + "issue:44451", + "issue:44484", + "issue:44655", + "issue:44704", + "issue:44734", + "issue:44936", + "issue:44938", + "issue:45103", + "issue:45584", + "issue:45712", + "issue:45753", + "issue:46132", + "issue:46172", + "issue:46178", + "issue:46298", + "issue:46365", + "issue:46421", + "issue:46655", + "issue:46664", + "issue:46701", + "issue:46710", + "issue:46752", + "issue:46821", + "issue:47116" + ], + "soft_pairs": [ + "issue:46655|issue:46664", + "issue:43986|issue:44484", + "issue:46298|issue:46752", + "issue:43097|issue:43873", + "issue:38617|issue:44351", + "issue:42890|issue:46298", + "issue:44062|issue:44117", + "issue:43012|issue:46821", + "issue:44117|issue:46701", + "issue:46172|issue:46710", + "issue:43502|issue:43992", + "issue:44704|issue:45103", + "issue:44936|issue:46365", + "issue:44451|issue:47116", + "issue:44380|issue:46421", + "issue:44938|issue:46178", + "issue:43986|issue:44016", + "issue:43064|issue:47116", + "issue:42907|issue:45753", + "issue:42907|issue:44655", + "issue:43012|issue:43550", + "issue:45712|issue:46298", + "issue:43012|issue:44368", + "issue:43827|issue:47116", + "issue:43992|issue:44484", + "issue:43352|issue:43976", + "issue:43208|issue:43550", + "issue:43825|issue:44655", + "issue:43066|issue:43906", + "issue:43352|issue:44380", + "issue:44734|issue:47116", + "issue:43097|issue:46298", + "issue:45584|issue:46365", + "issue:44380|issue:46132" + ], + "prepared_review_unit_hash": "74df4f8dec0eb806afcb3869b60f521cf0ef52065130a6751051ef4de50108ad", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37730, + "estimated_input_tokens": 9433, + "estimated_eval_tokens": 19122 + }, + "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. A few pairs share a subsystem or version theme, but none look like the same underlying bug/change closely enough to merge; no soft edge is accepted.", + "confidence": 0.71, + "canonical_issue_reason": "issue:47116 is the broadest open v5/pipeline regression item and the closest thing to an umbrella representative here, though the overall set is still mostly unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:47116 is the most general current issue in the set, but it should not be treated as a duplicate of the other reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46655", + "right": "issue:46664", + "accept": false, + "reason": "Same video processor area, but one is about padding math and the other about the meaning of num_frames; different bugs." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "Different problems: a crash loading a video model via AutoProcessor vs a question about save_pretrained shard sizing." + }, + { + "left": "issue:46298", + "right": "issue:46752", + "accept": false, + "reason": "No overlap in bug shape: noisy logging vs an IndexError on empty chat-template input." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "v5 embedding-flag removal is unrelated to quantization offloading behavior." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different code paths." + }, + { + "left": "issue:42890", + "right": "issue:46298", + "accept": false, + "reason": "Seed-related test flakiness is unrelated to a noisy logging message." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizer construction argument collision vs tokenizer mapping returning None are distinct failure modes." + }, + { + "left": "issue:43012", + "right": "issue:46821", + "accept": false, + "reason": "PyTorch warning during bfloat16 compilation is unrelated to a TimesFM AttributeError on window_size." + }, + { + "left": "issue:44117", + "right": "issue:46701", + "accept": false, + "reason": "Tokenizer mapping fallback bug vs a pipeline failure for DiffusionGemma are not the same change." + }, + { + "left": "issue:46172", + "right": "issue:46710", + "accept": false, + "reason": "A specific tensor transpose bug is not the same as a broader accuracy regression after upgrading versions." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "Local-files-only network leakage and a missing embed_tokens.weight are unrelated loader issues." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "Passing kwargs to cached_file and auto_docstring crashes from future annotations are unrelated." + }, + { + "left": "issue:44936", + "right": "issue:46365", + "accept": false, + "reason": "Trainer.evaluate failing after train is unrelated to an import conflict when importing Trainer." + }, + { + "left": "issue:44451", + "right": "issue:47116", + "accept": false, + "reason": "Loading one specific model and a streaming pipeline regression are different concrete failures." + }, + { + "left": "issue:44380", + "right": "issue:46421", + "accept": false, + "reason": "GPT2 attention scaling being ignored is not the same as cache lazy-init causing recompilation." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 import/load failure and an invalid backend mapping error are different root causes." + }, + { + "left": "issue:43986", + "right": "issue:44016", + "accept": false, + "reason": "Video-model import/runtime crash vs a notebook syntax error are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:47116", + "accept": false, + "reason": "FSDP2 optimizer-state corruption and streaming pipeline regression do not describe the same bug." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Saving dequantized models and an infinite loop on a specific image are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both mention saving, but one is dequantized models and the other is Pipeline.save_pretrained; different artifacts and symptoms." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A compiler warning in bfloat16 code is not the same as a torch.compile failure on Bamba with SDPA." + }, + { + "left": "issue:45712", + "right": "issue:46298", + "accept": false, + "reason": "Left is about leftover dummy classes and repo checks; right is a noisy logging message." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "Both are warnings, but they concern different config topics and different code paths." + }, + { + "left": "issue:43827", + "right": "issue:47116", + "accept": false, + "reason": "Docs still referencing pipeline() after removals is a documentation issue, not the streaming regression itself." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing encoder weights vs a question about max_shard_size are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "FlashAttention support in one model is unrelated to the Python-version compatibility issue." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training bugs and a Bamba torch.compile SDPA failure are different model-specific defects." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Incorrect v5 translation messaging and saving Pipeline objects are separate issues." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Wrong tokenizer decoder type and a reproduction of another issue are not the same bug." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "FlashAttention support gating is unrelated to GPT2 attention scaling being ignored under SDPA/FlashAttention." + }, + { + "left": "issue:44734", + "right": "issue:47116", + "accept": false, + "reason": "KV-cache continuation crash in serve is a different code path from a streaming pipeline regression." + }, + { + "left": "issue:43097", + "right": "issue:46298", + "accept": false, + "reason": "API removal around tie_embeddings_and_encoder_decoder is unrelated to a noisy logging message." + }, + { + "left": "issue:45584", + "right": "issue:46365", + "accept": false, + "reason": "Whisper empty-transcription failure and Trainer import problems are unrelated." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "GPT2 backend-specific attention scaling bug and AttentionInterface.register behavior change 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": 26, + "review_unit_index": 57, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:38617", + "issue:42175", + "issue:42490", + "issue:42757", + "issue:42890", + "issue:42994", + "issue:43012", + "issue:43064", + "issue:43208", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43526", + "issue:43582", + "issue:43611", + "issue:43618", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43867", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44077", + "issue:44162", + "issue:44222", + "issue:44230", + "issue:44355", + "issue:44380", + "issue:44483", + "issue:44485", + "issue:44514", + "issue:44625", + "issue:44655", + "issue:44938", + "issue:45276", + "issue:45290", + "issue:45478", + "issue:45479", + "issue:46050", + "issue:46077", + "issue:46172", + "issue:46365", + "issue:46877", + "issue:47059" + ], + "soft_pairs": [ + "issue:44162|issue:46172", + "issue:44077|issue:46365", + "issue:42890|issue:47059", + "issue:44222|issue:45478", + "issue:36683|issue:43582", + "issue:42757|issue:43824", + "issue:44380|issue:45479", + "issue:43064|issue:44380", + "issue:42175|issue:46365", + "issue:42490|issue:43012", + "issue:44483|issue:44514", + "issue:38617|issue:43502", + "issue:43618|issue:44380", + "issue:43756|issue:44485", + "issue:45276|issue:46077", + "issue:43618|issue:43761", + "issue:43208|issue:43526", + "issue:36010|issue:42757", + "issue:43299|issue:43867", + "issue:42994|issue:44655", + "issue:44038|issue:44230", + "issue:44483|issue:45290", + "issue:44625|issue:46365", + "issue:43352|issue:43937", + "issue:44625|issue:46077", + "issue:44355|issue:46877", + "issue:43611|issue:44655", + "issue:36683|issue:43475", + "issue:44655|issue:46050", + "issue:43976|issue:44938", + "issue:44938|issue:47059" + ], + "prepared_review_unit_hash": "79a46ed3d9a14caa822688aa18bd1dbf024f6cc74eb3cc4cd557242d228a799b", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36748, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a grab bag of unrelated regressions and import/runtime errors. I only see one plausible duplicate pair: the two CLIP vision output regression reports. Everything else is too far apart in model, API, or failure mode to merge safely.", + "confidence": 0.89, + "canonical_issue_reason": "Issue 43761 is the best representative because it describes the concrete CLIPVisionModel.forward v5 regression, and it is the closest match to the only plausible duplicate in the set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43761 is the clearest and most specific issue in the only likely duplicate pair, so it is the best global issue representative for this cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "Different models and failures: ESM2 training breakage vs a DeepSeek-V4 tensor operation bug." + }, + { + "left": "issue:44077", + "right": "issue:46365", + "accept": false, + "reason": "Unrelated issues: patchtsmixer post_init validation vs Trainer import failure tied to datasets." + }, + { + "left": "issue:42890", + "right": "issue:47059", + "accept": false, + "reason": "Both are test/runtime problems, but one is a seed-related SAM HQ test flake and the other is a CLI incompatibility with huggingface_hub 1.22.0." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "Both involve MoE/quantization, but one is save_pretrained FP8 failure and the other is a from_pretrained loading error for Qwen3.5 Moe." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Completely different subsystems: Gemma3 config attribute error vs Apple Silicon allocator warmup TypeError." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different packages and code paths." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Different bugs: GPT2 attention scaling is ignored in backend attention implementations, while the other is a zero-loss classification issue." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Different areas and symptoms: distributed optimizer state corruption vs attention scaling configuration handling." + }, + { + "left": "issue:42175", + "right": "issue:46365", + "accept": false, + "reason": "Package dependency/import problems, but one is backend extras installation and the other is Trainer import shadowing by datasets." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Both mention warnings, but one is a local model path warning and the other is a PyTorch dtype precision warning." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Different chat-path failures: an HTTP API request rejection vs a processor crash on batched template application." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "Different code paths: missing configuration_utils export vs local_files_only still allowing network requests." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "CLIP output assignment regression vs GPT2 attention scaling backend behavior; not the same bug." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Same broad RoPE/model family theme, but one is a Smollm3 layer drop bug and the other is GLM-5 RoPE implementation discussion." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Different model behaviors: token embedding resize propagation vs causal-vs-bidirectional attention masking." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": true, + "reason": "Both look like the same v5 CLIP vision regression where forward output fields are no longer populated as expected." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "Unrelated: xLSTM training blockers vs a BeitImageProcessorFast label reduction bug." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different removed/moved symbols from different modules." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Different loading failures: Qwen3VL-Moe model loading regression vs a state_dict sorting-related load error." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Both involve save_pretrained, but one is quantized model saving and the other is Pipeline object serialization; too different to unify." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Same model family, but one is a general v5 Qwen3-VL-Moe bug and the other is fp8 inference support; not the same concrete defect." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Both touch chat/template flows, but one is API request validation and the other is assistant tool-call handling during tokenization." + }, + { + "left": "issue:44625", + "right": "issue:46365", + "accept": false, + "reason": "Different issues: Qwen3.5 num_labels propagation vs a Trainer import conflict with datasets." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Nemotron Flash Attention support and GLM-5 GenerationConfig validation are separate model-specific problems." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Different model/config bugs: label propagation vs Gemma4 attention mask construction." + }, + { + "left": "issue:44355", + "right": "issue:46877", + "accept": false, + "reason": "Different import-time behaviors: compiled Python file errors vs docstring lint output at import time." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Different regressions: base_model_prefix loading breakage vs Pipeline save_pretrained failure." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Different attribute errors on different models: Gemma3Config vocab_size vs Sam3VisionEncoderOutput fpn_position_embeddings." + }, + { + "left": "issue:44655", + "right": "issue:46050", + "accept": false, + "reason": "Both involve save/load, but one is Pipeline serialization and the other is sinq quantized model round-tripping." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility issues, but they are about different interpreter versions and different failure modes." + }, + { + "left": "issue:44938", + "right": "issue:47059", + "accept": false, + "reason": "Both are environment compatibility problems, but one is Python 3.14 import failure and the other is a CLI break with huggingface_hub 1.22.0." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and the conservative rejections are well supported by the titles. The lone accepted soft pair, issue:43618 and issue:43761, is plausibly the same CLIP v5 regression and could be fixed in one PR. No obvious overstatement beyond that." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 58, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28282", + "issue:29870", + "issue:33453", + "issue:36296", + "issue:36683", + "issue:41669", + "issue:42890", + "issue:42898", + "issue:43010", + "issue:43064", + "issue:43502", + "issue:43525", + "issue:43575", + "issue:43618", + "issue:43867", + "issue:43976", + "issue:43992", + "issue:44062", + "issue:44075", + "issue:44077", + "issue:44112", + "issue:44246", + "issue:44355", + "issue:44360", + "issue:44393", + "issue:44492", + "issue:44855", + "issue:44908", + "issue:44964", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45561", + "issue:45850", + "issue:46077", + "issue:46202", + "issue:46285", + "issue:46365", + "issue:46512", + "issue:46664", + "issue:46851", + "issue:47000", + "issue:47110" + ], + "soft_pairs": [ + "issue:43867|issue:45216", + "issue:36683|issue:44077", + "issue:46077|issue:46512", + "issue:36683|issue:47110", + "issue:43618|issue:45216", + "issue:36683|issue:43525", + "issue:29870|issue:36296", + "issue:43064|issue:44492", + "issue:43010|issue:45278", + "issue:43992|issue:45310", + "issue:44075|issue:46365", + "issue:44908|issue:46851", + "issue:29870|issue:33453", + "issue:44393|issue:47000", + "issue:43867|issue:46202", + "issue:44355|issue:44855", + "issue:43867|issue:45375", + "issue:42890|issue:45561", + "issue:43010|issue:45850", + "issue:36296|issue:43575", + "issue:41669|issue:44246", + "issue:43618|issue:44077", + "issue:43010|issue:43976", + "issue:43867|issue:45357", + "issue:43992|issue:45478", + "issue:28282|issue:33453", + "issue:43618|issue:46285", + "issue:44360|issue:45440", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:43502|issue:44062", + "issue:42898|issue:46664" + ], + "prepared_review_unit_hash": "c1d2b7c52143cf672d8c258cf8cfe10a5a84c3217b515ae9fe31c74ee25f2f0c", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36406, + "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 mostly a loose collection of unrelated Transformers issues (import/load regressions, model-config attribute errors, training bugs, tokenizer behavior, and performance complaints). The closest common thread is upgrade-related import/load breakage, but the individual reports are not the same underlying bug.", + "confidence": 0.81, + "canonical_issue_reason": "issue:45278 is the broadest and most representative report in the set: a version-upgrade-induced spike in import failures. It best captures the cluster\u2019s loose import/regression theme without overfitting to one model-specific failure.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best single issue to anchor this cluster because it is broad, user-facing, and closest to the shared import/regression complaints seen elsewhere in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Both are load/save-related, but one is a generic state_dict loading error and the other is a Qwen3.5 save_pretrained regression; not the same concrete bug." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Different models and different failures: missing vocab_size vs patchtsmixer post_init validation. No shared bug path." + }, + { + "left": "issue:46077", + "right": "issue:46512", + "accept": false, + "reason": "One is about Gemma4 attention-mask semantics, the other about stale config dtype after from_config; unrelated behavior." + }, + { + "left": "issue:36683", + "right": "issue:47110", + "accept": false, + "reason": "Gemma3Config missing vocab_size and special_tokens list/type handling are unrelated issues." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIPOutput attentions assignment and Qwen3.5 save_pretrained checkpoint correctness are different code paths." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Both mention missing config attributes, but they are different models with different missing fields and likely different fixes." + }, + { + "left": "issue:29870", + "right": "issue:36296", + "accept": false, + "reason": "BART generate decoder mask bug and tensor-parallel training bug are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "FSDP/PEFT optimizer state corruption and a typo in cache strategy text are not the same problem." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "One is a no_grad decorator request on cache updates; the other is a broad import-error regression report." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "UMT5Encoder missing embed_tokens.weight and Qwen3.5 Moe from_pretrained failure are different model-loading defects." + }, + { + "left": "issue:44075", + "right": "issue:46365", + "accept": false, + "reason": "Optimizer argument handling and Trainer import depending on datasets are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:46851", + "accept": false, + "reason": "Scheduler kwargs propagation and Qwen3.5 variable-length position_ids are different subsystems." + }, + { + "left": "issue:29870", + "right": "issue:33453", + "accept": false, + "reason": "BART generate unbound variable and tokenizer-loading regression are not the same failure." + }, + { + "left": "issue:44393", + "right": "issue:47000", + "accept": false, + "reason": "Qwen3-VL bounding-box output errors and RTDetrLoss infeasible cost matrix are different model/task failures." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "State-dict loading issues and Gemma pipeline roundtrip failure on \u2581 are not the same bug." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Both involve import/runtime errors, but one is generic compiled-file failures and the other is a Python 3.13 torch.jit parsing issue." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "Generic load_model/state_dict sorting issue and Qwen3.5 vision-config field loss are different problems." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "Missing set_seed in tests and xdist race on captured_info.txt are both test-flakiness reports but not the same defect." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A specific no_grad/cache-update concern is not the same as a vague complaint that many models broke since v5.0." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor parallel training bug and Qwen2-57B-A14B-Instruct OOM are different runtime failures." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Both concern import performance, but one is a specific import-star cleanup/perf issue and the other is a vague intermittent slow-import report; not enough to treat as the same underlying bug." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIP attentions assignment and patchtsmixer post_init allowance are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache update no_grad and Python-version compatibility are unrelated." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "State_dict sorting/loading problems are not the same as Qwen3.5 visual-encoder key serialization." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "UMT5Encoder missing weights and a repeated Qwen3.5 Moe from_pretrained error are different model families and likely different code paths." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "Missing PyTorch environment import error and tokenizer-loading regression are unrelated." + }, + { + "left": "issue:43618", + "right": "issue:46285", + "accept": false, + "reason": "CLIPOutput attentions assignment and CLIP text-model structure inconsistency are different CLIP-side issues." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "DSA indexer ReLU discussion and DeepseekV3MoE divergence are unrelated model implementation concerns." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Test seeding flakiness and Phi-4 multimodal load failure are different issues." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "SAM HQ test flakiness and GraniteSpeech device-override CI failure are both test issues but not the same underlying bug." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "local_files_only network leakage and AddedToken keyword collision are different API bugs." + }, + { + "left": "issue:42898", + "right": "issue:46664", + "accept": false, + "reason": "Tokenizer cleanup-space behavior and SmolVLM num_frames semantics 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": 26, + "review_unit_index": 59, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:42548", + "issue:42757", + "issue:43010", + "issue:43065", + "issue:43232", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43576", + "issue:43723", + "issue:43824", + "issue:43827", + "issue:43874", + "issue:43901", + "issue:44222", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44734", + "issue:44749", + "issue:44961", + "issue:45071", + "issue:45125", + "issue:45375", + "issue:45468", + "issue:45478", + "issue:45850", + "issue:46132", + "issue:46143", + "issue:46144", + "issue:46617", + "issue:46664", + "issue:46726", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:44734|issue:46617", + "issue:43723|issue:44556", + "issue:44222|issue:45125", + "issue:43502|issue:44485", + "issue:43824|issue:46617", + "issue:43010|issue:43065", + "issue:45850|issue:47110", + "issue:43874|issue:46617", + "issue:43576|issue:46132", + "issue:44387|issue:44556", + "issue:30333|issue:33453", + "issue:45125|issue:45478", + "issue:42548|issue:44351", + "issue:30990|issue:31795", + "issue:44355|issue:44749", + "issue:45125|issue:45375", + "issue:43502|issue:44222", + "issue:43502|issue:44961", + "issue:43827|issue:44961", + "issue:46664|issue:46726", + "issue:43010|issue:43232", + "issue:30333|issue:36296", + "issue:43299|issue:44556", + "issue:29870|issue:31795", + "issue:33453|issue:34689", + "issue:44355|issue:45071", + "issue:42757|issue:43576", + "issue:43576|issue:45071", + "issue:43576|issue:47059", + "issue:31795|issue:38617", + "issue:43576|issue:43824", + "issue:44315|issue:44360", + "issue:44961|issue:45468", + "issue:44485|issue:44961", + "issue:44351|issue:46617", + "issue:36296|issue:43901", + "issue:43352|issue:44961", + "issue:43475|issue:46617", + "issue:46143|issue:46144" + ], + "prepared_review_unit_hash": "0dba2dfd84d8543ce9814863610c17028a96b71feae0fe0ee311a37702a9f600", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 39, + "serialized_chars": 38964, + "estimated_input_tokens": 9741, + "estimated_eval_tokens": 19738 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous Transformers issues (loading/import regressions, docs, multimodal/model-specific bugs). Only the two RoFormer reports appear to describe the same underlying bug in the model\u2019s attention path.", + "confidence": 0.78, + "canonical_issue_reason": "issue:46144 is the best anchor because it states the likely root cause more directly: RoFormer attention is not using the attention interface, which matches the kwargs-passthrough symptom in the paired issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:46144 is the most suitable representative of the only true duplicate pair and reads like the root-cause report rather than the downstream symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44734", + "right": "issue:46617", + "accept": false, + "reason": "Different failures: KV-cache continuation tensor indexing vs loading a trust_remote_code model from cache." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both are loading-related, but one is tokenizer loading in v5 and the other is a checkpoint reload/version-compatibility problem." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "Different model features: FP8 save_pretrained for MoE vs tensor-parallel plan missing for Qwen3.5 MoE." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated bugs: local_files_only still triggering network requests vs GLM-5 RoPE implementation." + }, + { + "left": "issue:43824", + "right": "issue:46617", + "accept": false, + "reason": "Different root causes: missing Qwen2_5_VL export vs cache FileNotFoundError when loading trust_remote_code models." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "Cache/layer no_grad decoration is unrelated to the Sam3PixelDecoder dummy Conv2d issue." + }, + { + "left": "issue:45850", + "right": "issue:47110", + "accept": false, + "reason": "One is a broad v5 regression complaint; the other is a specific special_tokens list/key error." + }, + { + "left": "issue:43874", + "right": "issue:46617", + "accept": false, + "reason": "Missing GLM46V image-processor method vs cache loading FileNotFoundError are different bugs." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Different areas and symptoms: transformers env CLI breakage vs AttentionInterface.register behavior change." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Memory/OOM under int4 quantization is unrelated to checkpoint reload failures after upgrade." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow job-status reporting vs tokenizer loading regression are unrelated." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but one is missing _tp_plan and the other is a from_pretrained error; not enough to treat as the same bug." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports: PreTrainedModel vs HybridCache." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers loading hang and forward-doc confusion are different kinds of issues." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled Python file errors are unrelated to a performance regression after version upgrade." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Tensor-parallel plan missing is unrelated to a config field being dropped by @strict." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "Network access despite local_files_only is unrelated to FP8 save_pretrained MoE behavior." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "No meaningful relation between local_files_only network requests and the racoon issue." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs still referencing pipeline() is unrelated to the racoon issue." + }, + { + "left": "issue:46664", + "right": "issue:46726", + "accept": false, + "reason": "Both are vision-related, but num_frames semantics and missing image-processor backend are different bugs." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Both mention generation/cache internals, but one is about no_grad decoration and the other about sync_gpus handling." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLflow reporting and tensor-parallel training are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both involve loading after version changes, but the affected models and failure modes differ." + }, + { + "left": "issue:29870", + "right": "issue:31795", + "accept": false, + "reason": "BART generation UnboundLocalError and forward-doc confusion are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer-loading regression and Llama 3.2 vision model-loading breakage are distinct issues." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file errors are unrelated to PretrainedConfig type checking in v5.4.0." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing huggingface_hub export and broken transformers env CLI are separate breakages." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "CLI breakage and PretrainedConfig type-checking regression are not the same bug." + }, + { + "left": "issue:43576", + "right": "issue:47059", + "accept": false, + "reason": "Different CLI failures with different external-version triggers and different missing attributes." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Docs confusion around model.forward is unrelated to a missing configuration_utils import." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "CLI env command breakage is unrelated to Qwen2.5-VL import failure." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel application with model_init is unrelated to a DSA indexer/ReLU discussion." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "No relation between the racoon issue and Gemma-4 audio positional encoding." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation and the racoon issue are unrelated." + }, + { + "left": "issue:44351", + "right": "issue:46617", + "accept": false, + "reason": "Missing HybridCache import and trust_remote_code cache FileNotFoundError are different failures." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor parallel training bug and TextClassificationPipeline docs mismatch are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44961", + "accept": false, + "reason": "Flash Attention support error for Nemotron and the racoon issue are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:46617", + "accept": false, + "reason": "Missing SAM3 vision encoder attribute and cache FileNotFoundError are different bugs." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": true, + "reason": "Both report the same RoFormer attention-path bug: kwargs are not being propagated because the attention implementation is not using the attention interface." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the only positive duplicate call (issue:46143/issue:46144) is stated conservatively and fits the titles. The other soft-edge rejections are also cautious and supported by the issue titles, so I don\u2019t see evidence of overstating similarity." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 60, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:41628", + "issue:41669", + "issue:42371", + "issue:42548", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43867", + "issue:43976", + "issue:44038", + "issue:44222", + "issue:44230", + "issue:44246", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44704", + "issue:44908", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45125", + "issue:45230", + "issue:45341", + "issue:45362", + "issue:45375", + "issue:45542", + "issue:45753", + "issue:45850", + "issue:46095", + "issue:46209", + "issue:46218", + "issue:46328", + "issue:46664", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:41628|issue:42548", + "issue:44222|issue:44230", + "issue:45375|issue:47110", + "issue:43867|issue:47110", + "issue:44315|issue:46209", + "issue:43525|issue:47110", + "issue:45362|issue:45850", + "issue:43673|issue:44492", + "issue:44485|issue:46664", + "issue:44273|issue:45542", + "issue:44623|issue:45230", + "issue:44038|issue:45125", + "issue:44273|issue:45003", + "issue:44360|issue:46664", + "issue:41669|issue:43673", + "issue:44393|issue:45071", + "issue:44393|issue:45753", + "issue:43576|issue:44556", + "issue:43525|issue:45850", + "issue:44556|issue:45092", + "issue:43572|issue:45850", + "issue:44230|issue:46095", + "issue:44492|issue:45850", + "issue:45850|issue:47059", + "issue:42371|issue:44492", + "issue:41669|issue:46218", + "issue:43408|issue:44315", + "issue:44246|issue:44273", + "issue:42890|issue:44908", + "issue:43576|issue:44908", + "issue:43352|issue:44393", + "issue:43976|issue:44393", + "issue:44295|issue:45230", + "issue:44908|issue:45071", + "issue:44297|issue:45230", + "issue:44492|issue:46328", + "issue:41669|issue:44351", + "issue:44704|issue:44908", + "issue:41669|issue:45341", + "issue:43976|issue:46218" + ], + "prepared_review_unit_hash": "3c1365398e1a7c671af5955b614546d417891f71bf421274e3e4894d363ea469", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 40, + "serialized_chars": 38963, + "estimated_input_tokens": 9741, + "estimated_eval_tokens": 19738 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is noisy and mostly groups broad import/lazy-loading/runtime-regression reports rather than one concrete defect. The best representative is the open lazy-loading issue, but none of the soft pairs are safe duplicate merges.", + "confidence": 0.63, + "canonical_issue_reason": "issue:44273 is the broadest and most central representative: it directly targets lazy loading behavior, which is the closest shared theme across many of the import-related reports in this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44273 is the best issue to anchor the cluster because it is open, clearly describes the core lazy-loading problem, and can serve as the umbrella for adjacent import-related symptoms.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import-name errors, but they concern different symbols (AutoImageProcessor vs PreTrainedModel) and likely different broken import paths." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "Both mention FP8/moe, but one is about saving pretrained checkpoints and the other about inference support; different concrete failures." + }, + { + "left": "issue:45375", + "right": "issue:47110", + "accept": false, + "reason": "One is a missing config field in Qwen3_5MoeVisionConfig; the other is a special_tokens type error. Not the same bug." + }, + { + "left": "issue:43867", + "right": "issue:47110", + "accept": false, + "reason": "State-dict loading/sorting error is unrelated to the special_tokens list/keys failure." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "Liger Kernel/model_init application is unrelated to FP8 inference on B200." + }, + { + "left": "issue:43525", + "right": "issue:47110", + "accept": false, + "reason": "Llama4Config missing pad_token_id is a config attribute issue, not the special_tokens list/dict mismatch." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Qwen3.5 chat crash and the generic v5 regression report are too broad and do not point to the same concrete code path." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Chunked prefill cache regression and a cache-strategy typo are different cache-related problems." + }, + { + "left": "issue:44485", + "right": "issue:46664", + "accept": false, + "reason": "GLM-5 RoPE implementation and SmolVLM num_frames semantics involve different model families and different functionality." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading problems are unrelated to the TensorBoard-without-TensorFlow backend error." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Missing processor save_pretrained files is not the same as a generic placeholder bug report." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL/Moe bug and missing _tp_plan are distinct model-specific issues." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Both touch import/loading internals, but one is a user-visible lazy-loading failure and the other is unsafe sys.modules access; not enough to call them the same bug." + }, + { + "left": "issue:44360", + "right": "issue:46664", + "accept": false, + "reason": "DSA indexer/ReLU discussion and video num_frames semantics are unrelated." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-star cleanup/performance and chunked-prefill cache missing are different changes and symptoms." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Qwen3-VL bounding-box hallucination is unrelated to PretrainedConfig type-checking regressions." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Both are Qwen3.5/VL issues, but one is bounding-box output and the other is an infinite-loop image case; different failure modes." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Both are version-upgrade regressions, but one is the env command and the other is checkpoint reload compatibility; not the same bug." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config is specific and unrelated to a broad 'v5 breaks too many models' complaint." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both concern v5 migration, but one is checkpoint reload compatibility and the other is remote-code/meta-init incompatibility; different code paths." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLmConfig pad_token_idx missing is a specific config regression, not the generic v5 breakage report." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "FP8 inference support for Qwen models is unrelated to a deepseekv4 weight-conversion question." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "Cache-strategy typo is a narrow naming/doc bug, not the broad v5 regression report." + }, + { + "left": "issue:45850", + "right": "issue:47059", + "accept": false, + "reason": "Generic v5 breakage and a cli completion failure with huggingface_hub 1.22.0 are different issues." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 API guidance and cache-strategy typo are unrelated." + }, + { + "left": "issue:41669", + "right": "issue:46218", + "accept": false, + "reason": "Import-star cleanup/performance does not match the Qwen3_5ForCausalLM import-module failure." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "sam3_video/sam3_tracker warning and Liger Kernel/model_init are different model-initialization problems." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Import taking long sometimes and lazy loading not working properly overlap on imports, but they describe different symptoms and likely different fixes." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Missing set_seed in a test is unrelated to inverse_sqrt scheduler kwargs handling." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "env command breakage and scheduler kwargs propagation are unrelated subsystems." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2.0 support gating for Nemotron and Qwen3-VL bounding-box errors are unrelated model-specific bugs." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version compatibility and Qwen3-VL bounding-box errors are different regressions." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Reading position_ids after buffer registration is a concrete tensor/buffer bug; the other item is just a generic bug report placeholder." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler keyword handling and PretrainedConfig type checking are unrelated." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save_pretrained metadata mismatch is unrelated to the generic bug report placeholder." + }, + { + "left": "issue:44492", + "right": "issue:46328", + "accept": false, + "reason": "Cache-strategy typo and SequenceFeatureExtractor.pad performance issue are different areas and different fixes." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star cleanup/performance and missing HybridCache import are different import-related bugs." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor kwargs forwarding and inverse_sqrt scheduler kwargs are unrelated code paths." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Import-star cleanup/performance is unrelated to the testing_utils bug." + }, + { + "left": "issue:43976", + "right": "issue:46218", + "accept": false, + "reason": "Python 3.9/3.10 compatibility and Qwen3_5ForCausalLM import failure are not the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 26, + "review_unit_index": 61, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:28348", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:38617", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:43066", + "issue:43298", + "issue:43352", + "issue:43502", + "issue:43519", + "issue:43576", + "issue:43618", + "issue:43704", + "issue:44077", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44273", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44655", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:44938", + "issue:44998", + "issue:45230", + "issue:45250", + "issue:45290", + "issue:45310", + "issue:45468", + "issue:45850", + "issue:46018", + "issue:46095", + "issue:46172", + "issue:46218", + "issue:46655", + "issue:46894", + "issue:47059", + "issue:47110" + ], + "soft_pairs": [ + "issue:44077|issue:44861", + "issue:31795|issue:33453", + "issue:43576|issue:44273", + "issue:44315|issue:44829", + "issue:43519|issue:46655", + "issue:44360|issue:45468", + "issue:44273|issue:44908", + "issue:45230|issue:45310", + "issue:43519|issue:44485", + "issue:45468|issue:46894", + "issue:43576|issue:46218", + "issue:44655|issue:44861", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:31795|issue:34689", + "issue:44485|issue:45468", + "issue:43704|issue:44485", + "issue:46018|issue:46172", + "issue:43502|issue:43519", + "issue:42371|issue:43704", + "issue:45850|issue:46218", + "issue:43519|issue:44623", + "issue:44263|issue:45468", + "issue:44261|issue:45468", + "issue:43298|issue:46095", + "issue:43519|issue:43704", + "issue:44351|issue:46218", + "issue:43352|issue:46218", + "issue:44861|issue:47110", + "issue:42913|issue:43066", + "issue:44861|issue:47059", + "issue:28348|issue:45250", + "issue:44908|issue:44998", + "issue:44492|issue:44998", + "issue:44998|issue:45850", + "issue:43618|issue:44998", + "issue:44938|issue:44998", + "issue:44861|issue:45290", + "issue:31515|issue:38617", + "issue:44230|issue:44623", + "issue:28348|issue:43352" + ], + "prepared_review_unit_hash": "72212d6bd36ddd2bf79af05d4eec842cd98efb590e6cf749bc64b8f328a144c1", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 41, + "serialized_chars": 39107, + "estimated_input_tokens": 9777, + "estimated_eval_tokens": 19810 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated issue reports that only share broad Transformers themes like loading, tokenizers, model internals, or Flash Attention. I don\u2019t see any true duplicate pair among the soft edges, so all of them should stay split.", + "confidence": 0.59, + "canonical_issue_reason": "issue:45850 is the closest thing to an umbrella report because it broadly describes v5 breaking many models, but it is still too vague to be a strong duplicate anchor.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the broadest representative of the mostly heterogeneous set, though it is not a precise bug report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Different bugs: optional post_init validation vs _get_tied_weight_keys crashing on a list." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Docs confusion about forward args vs a tokenizer loading regression." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "CLI env command breakage vs lazy loading behavior; different code paths." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger kernel application through model_init vs flash_attention_3 training instability." + }, + { + "left": "issue:43519", + "right": "issue:46655", + "accept": false, + "reason": "Different processor bugs in different models: timestamp math vs video padding math." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "Unrelated model internals: DSA indexer ReLU vs Gemma-4 audio positional encoding." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading vs inverse_sqrt scheduler kwargs; no shared underlying bug." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Generic placeholder report vs a concrete Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Processor timestamp calculation and GLM-5 RoPE are different subsystems." + }, + { + "left": "issue:45468", + "right": "issue:46894", + "accept": false, + "reason": "Different audio/model bugs: Gemma-4 positional encoding vs CohereASR double-shift training loss." + }, + { + "left": "issue:43576", + "right": "issue:46218", + "accept": false, + "reason": "CLI env command issue vs Qwen3_5 import failure; unrelated symptoms." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Pipeline save_pretrained serialization vs tied-weight-key lookup crash." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Processor timestamp math vs scheduler parameter handling." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Codebase import-star cleanup/perf discussion vs a runtime VRAM leak." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion vs a concrete model-loading regression." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Different model implementations: GLM-5 RoPE vs Gemma-4 audio positional encoding." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "Memory leak in Qwen3ForCausalLM vs positional-encoding math in GLM-5." + }, + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "DeepSeek-V4 expert gating vs a tensor dtype/transpose bug; not the same fix." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "local_files_only download enforcement vs processor timestamp calculation." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 configuration API change vs a VRAM leak; different concerns." + }, + { + "left": "issue:45850", + "right": "issue:46218", + "accept": false, + "reason": "Broad v5 breakage report vs a specific import error for Qwen3_5ForCausalLM." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Timestamp calculation bug vs missing files in processor.save_pretrained." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "torch.split/indexer logic vs Gemma-4 audio positional encoding." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "Missing rms_norm_eps in MLA q_a_layernorm vs an audio positional-encoding issue." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Continuous batching audio support vs a DeepSeek-V4 weight conversion question." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Processor timestamp math vs a model VRAM leak." + }, + { + "left": "issue:44351", + "right": "issue:46218", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different root causes." + }, + { + "left": "issue:43352", + "right": "issue:46218", + "accept": false, + "reason": "Nemotron flash-attention support gap vs Qwen3_5 import failure; different problems." + }, + { + "left": "issue:44861", + "right": "issue:47110", + "accept": false, + "reason": "Different attribute/type errors in unrelated utilities." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both concern tokenizer v5 behavior, but one is a broad regression and the other is a specific decoder-type bug." + }, + { + "left": "issue:44861", + "right": "issue:47059", + "accept": false, + "reason": "Tied-weight-key crash vs a CLI completion incompatibility with huggingface_hub." + }, + { + "left": "issue:28348", + "right": "issue:45250", + "accept": false, + "reason": "Specific GPT2LMHeadModel support request vs a generic Flash Attention 2.0 issue; too broad to merge." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwarg bug vs an unrelated placeholder issue." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "Cache-strategy typo vs an unrelated placeholder issue." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "Placeholder/non-technical report vs a broad v5 regression report." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression vs an unrelated placeholder issue." + }, + { + "left": "issue:44938", + "right": "issue:44998", + "accept": false, + "reason": "Python 3.14 import/load failure vs an unrelated placeholder issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weight-key bookkeeping crash vs chat template tool-call parsing crash." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow from_pretrained checkpoint loading vs an ImportError for layer_type_validation." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support request vs missing files on processor.save_pretrained." + }, + { + "left": "issue:28348", + "right": "issue:43352", + "accept": false, + "reason": "Same broad Flash Attention topic, but different models and different concrete fixes; not mergeable into one PR." + } + ] + }, + "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": 26, + "review_unit_index": 62, + "review_unit_count": 62, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-513", + "nodes": [ + "issue:29870", + "issue:30333", + "issue:31515", + "issue:31795", + "issue:36296", + "issue:37934", + "issue:42936", + "issue:43240", + "issue:43298", + "issue:43668", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:44230", + "issue:45310", + "issue:46018", + "issue:46144" + ], + "soft_pairs": [ + "issue:29870|issue:31515", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:44230|issue:45310", + "issue:43298|issue:44230", + "issue:43717|issue:46144", + "issue:42936|issue:46018", + "issue:43668|issue:44031", + "issue:42936|issue:43240", + "issue:43240|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:46144", + "issue:42936|issue:43813", + "issue:43668|issue:43813", + "issue:43240|issue:43668", + "issue:37934|issue:43668", + "issue:42936|issue:43668" + ], + "prepared_review_unit_hash": "a68e2429ace1029cc8a6aea5b6bc06c4a487c8aac0e298e1d95483345956034d", + "component_budget": { + "node_count": 513, + "item_count": 513, + "soft_pair_count": 1886, + "serialized_chars": 978511, + "estimated_input_tokens": 244628, + "estimated_eval_tokens": 489512 + }, + "budget": { + "node_count": 17, + "item_count": 17, + "soft_pair_count": 18, + "serialized_chars": 14946, + "estimated_input_tokens": 3737, + "estimated_eval_tokens": 7730 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes runtime bugs, performance complaints, documentation issues, config/type-hint problems, and model-specific feature requests. None of the soft edges look like the same underlying bug/change, so they should all be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No safe canonical issue stands out; the items span unrelated topics rather than one duplicate bug or change.", + "canonical_pr_reason": null, + "best_issue_reason": "No single issue is globally best suited as the representative; the cluster is too mixed and the soft links are only superficial similarity.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:29870", + "right": "issue:31515", + "accept": false, + "reason": "BART generate() UnboundLocalError vs slow from_pretrained checkpoint loading: different code paths and failure modes." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow run-state reporting after failed training is unrelated to checkpoint load performance." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Model loading slowdown and tensor-parallel training bug are distinct problems in different execution paths." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Checkpoint load slowness is not the same issue as confusing forward-API documentation." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "Both involve Qwen3.5/MoE, but one is fp8 support and the other is a from_pretrained regression; not the same concrete fix." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching audio-model support and fp8 Qwen inference support are unrelated feature requests." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba/Mamba-2 init_weights usage and RoFormer attention-interface implementation are separate model-specific issues." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Mask2Former ignore_value handling and DeepSeek-V4 shared-expert gating are unrelated code paths." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "Incorrect type hint on ModernBERTConfig.norm_eps is unrelated to tokenizer regex warning behavior." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Unused ignore_value in Mask2Former and missing kwargs in loss calculation are different bugs in different functions." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss-kwargs propagation and RoFormer attention-interface usage do not describe the same change." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2Former ignore_value handling is unrelated to tokenizer regex pattern warnings." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "Different model implementations and different failure modes; no shared underlying bug." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Ignore_value logic in Mask2Former is unrelated to a quantization-operation typo." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "A config type-hint issue and a typo in quantization code are not the same underlying change." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Missing loss kwargs and an incorrect config type hint are unrelated." + }, + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Question about Llama4TextL2Norm/RMSNorm is unrelated to ModernBERT norm_eps typing." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2Former ignore_value handling and ModernBERT config typing 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": 2, + "component_count": 26, + "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": "ecf5b3d41c47bf73ea7475bdc2ab72962db5cdb13682014c258abd10c67d6993", + "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 and do not share the same underlying bug or feature request: tokenizer truncation behavior, an optimizer/scheduler feature request, and Wav2Vec2 word-boundary timestamps. The cluster is mixed-topic, so no soft edges should be accepted.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 10519 is the most concrete and broadly scoped of the three, but it still does not represent the others; the cluster has no true canonical issue because the items are unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Different domains and problems: tokenizer truncation direction vs Wav2Vec2 word time offsets. No shared code-path or underlying bug." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "Unrelated feature requests: truncation behavior vs ReduceLROnPlateau-like scheduler functionality. Same library, but not the same issue." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "Completely different concerns: learning-rate scheduling vs speech model word boundary timestamps. No evidence of a common 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": 3, + "component_count": 26, + "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": "5a0bf21a6b1804fee52072e1fca44e5b84cbd1cdc076d5b649f1f66b7f1a9833", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues do not look like duplicates. One concerns Tapas tokenization behavior, one concerns a bad authentication error for private model downloads, and one concerns failing to load bert-base-uncased. They touch different failure modes and likely different code paths.", + "confidence": 0.93, + "canonical_issue_reason": "No strong duplicate set here; if one issue must be chosen as the representative, #16618 is the broadest and most discussed model-loading report, but it is still distinct from the other two.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the best standalone issue because it is the clearest user-facing model loading failure with the most general scope among the three.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tapas tokenization mismatch is a tokenizer/code-path problem, while #16618 is about loading bert-base-uncased; these are different underlying bugs." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "#15980 is specifically about an incorrect error message for private model access without login, not the same failure as being unable to load bert-base-uncased." + } + ] + }, + "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": 26, + "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": "22884d158073538dc85ac613e3555c0587c8df0a6567698eedff35c8aae7c259", + "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 feature requests about different architecture-support workstreams: Flash Attention 2 support vs GGUF support. They share a generic 'community contribution' framing but not the same underlying bug or change.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 26350 is a concrete, long-running request for Flash Attention 2 support across more architectures, which is self-contained and distinct from the GGUF request.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 26350 is the stronger canonical pick only because it is the earlier, more established thread with substantial discussion; it should not be treated as a duplicate of 33260.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different features and code paths: Flash Attention 2 support versus GGUF support. Same broad theme, but not the same underlying change or bug." + } + ] + }, + "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": 26, + "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-27088-9", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370", + "issue:46222", + "issue:46226", + "issue:46931" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:38435", + "issue:27088|issue:28936", + "issue:28936|issue:43834", + "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:46222", + "issue:38435|issue:46226", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:43834|issue:46222", + "issue:43834|issue:46226", + "issue:43834|issue:46931", + "issue:44365|issue:44370", + "issue:44365|issue:46222", + "issue:44365|issue:46226", + "issue:44365|issue:46931", + "issue:44370|issue:46222", + "issue:44370|issue:46226", + "issue:44370|issue:46931", + "issue:46222|issue:46226", + "issue:46222|issue:46931", + "issue:46226|issue:46931" + ], + "prepared_review_unit_hash": "7448c522d9ee16c722b40ea33b1d6ac87ea69b85b74a4083bfaff67491e5c655", + "component_budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 29, + "serialized_chars": 15984, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 + }, + "budget": { + "node_count": 9, + "item_count": 9, + "soft_pair_count": 29, + "serialized_chars": 15984, + "estimated_input_tokens": 3996, + "estimated_eval_tokens": 8248 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly separate docs-translation requests by language, not one shared issue. The only clear duplicates are the near-identical placeholder-template issues and the exact duplicate Thai-titled pair.", + "confidence": 0.83, + "canonical_issue_reason": "issue:43834 is the earliest generic template-style issue and is the best representative of the exact placeholder subgroup.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43834 is the most canonical representative because it uses the generic translation-request template and predates the other near-identical placeholder issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Different language request (Turkish vs generic placeholder), so not the same underlying issue." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Different language request; only broadly similar translation intent." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Turkish request does not match the generic placeholder issue closely enough to be a duplicate." + }, + { + "left": "issue:27088", + "right": "issue:38435", + "accept": false, + "reason": "Different target language (Turkish vs Romanian)." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Different target language (Turkish vs Spanish)." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish translation request is distinct from the generic placeholder issue." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Different language request; not the same issue." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Different target language (Spanish vs Romanian)." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Spanish request and generic placeholder issue are not clearly the same underlying request." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian translation request is distinct from the generic placeholder issue." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Romanian request and generic placeholder issue are not duplicates." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Romanian request differs from the generic template issue." + }, + { + "left": "issue:38435", + "right": "issue:46222", + "accept": false, + "reason": "Different titles and no evidence they are the same translation request." + }, + { + "left": "issue:38435", + "right": "issue:46226", + "accept": false, + "reason": "Different titles and no concrete sign of duplication." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": true, + "reason": "Near-identical generic placeholder issues with extremely high body similarity." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": true, + "reason": "Near-identical generic placeholder issues; looks like the same template issue duplicated." + }, + { + "left": "issue:43834", + "right": "issue:46222", + "accept": false, + "reason": "Titles differ materially; body similarity alone is not enough to prove the same issue." + }, + { + "left": "issue:43834", + "right": "issue:46226", + "accept": false, + "reason": "Different title/content signal; not safely the same issue." + }, + { + "left": "issue:43834", + "right": "issue:46931", + "accept": false, + "reason": "Different title and no clear shared request despite similar body template." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": true, + "reason": "Near-exact duplicate placeholder issues with matching translation-template wording." + }, + { + "left": "issue:44365", + "right": "issue:46222", + "accept": false, + "reason": "Different title and not enough evidence of the same underlying request." + }, + { + "left": "issue:44365", + "right": "issue:46226", + "accept": false, + "reason": "Different title and no explicit link to the same issue." + }, + { + "left": "issue:44365", + "right": "issue:46931", + "accept": false, + "reason": "Different title; similarity appears to be template-level only." + }, + { + "left": "issue:44370", + "right": "issue:46222", + "accept": false, + "reason": "Different title and no clear proof of duplication." + }, + { + "left": "issue:44370", + "right": "issue:46226", + "accept": false, + "reason": "Different title and no concrete same-request evidence." + }, + { + "left": "issue:44370", + "right": "issue:46931", + "accept": false, + "reason": "Different title and only loose body similarity." + }, + { + "left": "issue:46222", + "right": "issue:46226", + "accept": true, + "reason": "Exact same title and effectively identical issue text, so this is a duplicate." + }, + { + "left": "issue:46222", + "right": "issue:46931", + "accept": false, + "reason": "Both are Thai-titled, but the titles differ and duplication is not certain." + }, + { + "left": "issue:46226", + "right": "issue:46931", + "accept": false, + "reason": "Different title; similar body template is not enough to merge them." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: the cluster is dominated by language-specific docs-translation requests, while the placeholder-template issues and the Thai exact-match pair are the only clearly duplicate-looking groups. The accepted soft-edge verdicts are conservative overall, and the rejected edges avoid overclaiming across distinct language requests or materially different titles." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 26, + "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": "bdc2d7cb70aecfa353116e38c305d9591a20538cf2598297bb5c6c6d2172560e", + "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 cluster is noisy: most items are separate model-request issues across different model families and modalities, not true duplicates. The soft similarities are mostly template-level ('Add/Support X') rather than the same underlying change. No PRs are present.", + "confidence": 0.93, + "canonical_issue_reason": "No single canonical duplicate stands out; the issues span unrelated requests (BERT-like configs, audio ASR/TTS, vision-language encoders, keypoint models, sparse transformers).", + "canonical_pr_reason": null, + "best_issue_reason": "If a representative issue is needed, 46133 is the most detailed and discussion-heavy request, but it is still not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36718", + "right": "issue:47004", + "accept": false, + "reason": "Both are keypoint requests, but RoMa matcher and LoMa detector/matcher are different models and changes." + }, + { + "left": "issue:36718", + "right": "issue:41115", + "accept": false, + "reason": "RoMa keypoint matcher vs MiniCPM3 architecture; unrelated changes." + }, + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "EuroBert config addition and DEIMv2 are different model families." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert config vs Qwen3-TTS support; different modality and code path." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "EuroBert config and RF-DETR are unrelated model additions." + }, + { + "left": "issue:36817", + "right": "issue:42954", + "accept": false, + "reason": "EuroBert config vs XiaoMi MiMo; same request pattern, different model." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are distinct detection/vision model requests." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "RF-DETR vs LFM2.5 Audio; unrelated modality." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR vs weight-sparse transformer; different code paths." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are unrelated model-support requests." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR vs Qwen3-TTS; different model family and task." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are unrelated additions." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT vs sparse transformer; same broad request style, not same change." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT vs LFM2.5 Audio; different subsystem." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are different model requests." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT vs ModernVBERT models; both BERT-ish, but distinct requests." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture vs Qwen3-ASR support; unrelated." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 architecture and sparse transformer are different changes." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT target different model families." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 vs Voxtral TTS support; different tasks." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 and Qwen3-TTS are unrelated model-support requests." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT models vs Qwen3-ASR; different model and modality." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are not the same change." + }, + { + "left": "issue:42503", + "right": "issue:42954", + "accept": false, + "reason": "ModernVBERT models vs XiaoMi MiMo; unrelated." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like models with RoPE vs Voxtral TTS support; different code paths." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "Both are BERT-related, but ModernVBERT and RoPE support are separate requests." + }, + { + "left": "issue:42738", + "right": "issue:42954", + "accept": false, + "reason": "BERT-like RoPE support vs MiMo request; unrelated." + }, + { + "left": "issue:42738", + "right": "issue:46653", + "accept": false, + "reason": "BERT/RoPE feature request vs a person-named issue; no shared change." + }, + { + "left": "issue:42954", + "right": "issue:44205", + "accept": false, + "reason": "MiMo and SAM3-LiteText are different model additions." + }, + { + "left": "issue:42954", + "right": "issue:43671", + "accept": false, + "reason": "MiMo vs Qwen3-TTS; different model families." + }, + { + "left": "issue:42954", + "right": "issue:45177", + "accept": false, + "reason": "MiMo and DeepSeek-OCR2 are unrelated requests." + }, + { + "left": "issue:42954", + "right": "issue:46653", + "accept": false, + "reason": "MiMo vs person-named issue; clearly unrelated." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Both are speech-related, but TTS and ASR support are different changes." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Both are TTS-ish model requests, but Qwen3-TTS and Voxtral TTS are different models." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS vs DeepSeek-OCR2; unrelated." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-TTS and sparse transformer are different feature areas." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Both touch audio, but Qwen3-ASR and LFM2.5 Audio are different model-specific requests." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR vs EUPE; unrelated." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "ASR support vs sparse transformer; different code paths." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are different speech-model requests." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT vs Voxtral TTS; unrelated model families." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are unrelated." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "Both are vision-language/perception model requests, but EUPE and TIPSv2 are distinct models and changes." + }, + { + "left": "issue:42503", + "right": "issue:46653", + "accept": false, + "reason": "ModernVBERT models vs person-named issue; no shared change." + }, + { + "left": "issue:45027", + "right": "issue:46653", + "accept": false, + "reason": "Voxtral TTS support vs person-named 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": 7, + "component_count": 26, + "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": "6d1f0b8215a8bbdb298a07b5b4ed40bd8aa1b4929666e549d65c9168b044d331", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1505, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1505, + "estimated_input_tokens": 377, + "estimated_eval_tokens": 1010 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are unrelated: one is a closed docs request about adding `pruna` as an optimization framework, the other is an open technical issue about a Vision RoPE refactor. They should not be treated as duplicates or merged into one cluster.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the titles and contexts point to different problems in different areas (documentation request vs. vision model refactor).", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46443 is the more substantive active technical issue, while 38740 is a closed documentation request and not a close match.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38740", + "right": "issue:46443", + "accept": false, + "reason": "Reject: these do not describe the same underlying bug or change. One is a docs enhancement request; the other is a Vision RoPE refactor." + } + ] + }, + "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": 26, + "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": "7f0c79329c28c999d75fce71ef63435d87765e2c29415f54def2c0fc45846424", + "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": "These two issues are not duplicates: one is about block decoding support, while the other is about DeepSeek v3.2 model support. They appear to be unrelated feature requests.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different underlying requests with no shared target or code path: block decoding vs. DeepSeek v3.2 support." + } + ] + }, + "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": 26, + "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": "6fa336f35c2683d18964d54fd7f8e974e2eb51b98697d1d714634662133f280d", + "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 about different model components and different changes: one requests a `from_pretrained()` flag for TiledFusedLogitsLoss/FLCE, the other asks for TiledMLP integration to reduce memory usage. They are not duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "Issue 41306 is the more specific, self-contained request, with a clear API target (`from_pretrained()`) and a concrete feature flag change.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 41306 is the best representative issue because it names an exact integration point and a narrowly scoped user-facing change, making it easier to triage and implement.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41306", + "right": "issue:41826", + "accept": false, + "reason": "Different underlying requests: FLCE/TiledFusedLogitsLoss flag handling in `from_pretrained()` versus adding TiledMLP for lower memory footprint. No evidence they describe the same bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 10, + "component_count": 26, + "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": "7f5c737f958b8035fe8472e168a1a5c0d70488ac5704c25bf49b3b1b9383e701", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1582, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1582, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are related only at a broad Trainer/custom-trainer level, but they describe different problems: one is about `remove_unused_column` customizability, the other about FLOPS estimation for custom trainers. Not duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43598 is the broader, more general Trainer behavior discussion; issue 43600 is a separate FLOPS-estimation bug and should not be merged into it.", + "canonical_pr_reason": null, + "best_issue_reason": "43598 is the best single representative of the cluster because it centers on Trainer customizability, which is the broader theme, while 43600 is narrower and distinct.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Reject: they share a custom-trainer theme, but the concrete problems differ (`remove_unused_column` behavior vs FLOPS counting). They do not look like 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": 26, + "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": "efa2909250967205293f9a0046c42ab4985c12aa85e88035cf2de7845c1fdfa2", + "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 broad \u201cQwen model support\u201d level, but they target different model families and different API surfaces: one requests registration for Qwen3-Omni in base/generative auto classes, the other requests sequence-classification support for Qwen3.5. They should not be treated as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43684 is the broader, more central request: add AutoModel / AutoModelForConditionalGeneration registration for Qwen3-Omni. It represents the core model-registration gap more directly than the narrower task-head request.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43684 is the best representative of the cluster because it is the broader model-registration item and more likely to generalize as the main support request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying changes: Qwen3-Omni base/generative auto-registration versus Qwen3.5 sequence-classification support. They involve different configs, different auto classes, and 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": 12, + "component_count": 26, + "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": "12b0a4205b4ff9e4c8d7328fd85e0fe3f8dd12b63edbd4c7c7e709c1cdb1f616", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4267, + "estimated_input_tokens": 1067, + "estimated_eval_tokens": 2390 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4267, + "estimated_input_tokens": 1067, + "estimated_eval_tokens": 2390 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Three issues share the exact same title, \"huggingface \", and are likely duplicates of one another; the \"Fett\" issue appears unrelated.", + "confidence": 0.82, + "canonical_issue_reason": "Issue 43845 is the earliest of the matching \"huggingface \" reports, so it is the best canonical representative for that duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43845 is the strongest overall issue candidate because it is the earliest clear duplicate cluster member and has the same exact title as the other matching reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles and no other evidence of the same underlying problem." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Different titles and no shared issue-specific details tying them to the same bug." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "Different titles; this looks unrelated to the repeated \"huggingface \" reports." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": true, + "reason": "Same exact title and very close creation times, with no evidence suggesting different problems." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": true, + "reason": "Same exact title and likely the same duplicate report family." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": true, + "reason": "Same exact title and same apparent underlying issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: 43845, 43846, and 43859 all share the exact title \"huggingface \", while 43835 is clearly different. The canonical choice of 43845 is conservative and supported by it being the earliest of the matching set. The soft-edge verdicts are also appropriately cautious given only soft-similarity evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 26, + "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": "66516f9754ac9d7750d8e7a7ffe49da188fbf8b435fb32a32005169dd0ef6ebd", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 12, + "serialized_chars": 7349, + "estimated_input_tokens": 1838, + "estimated_eval_tokens": 3932 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 12, + "serialized_chars": 7349, + "estimated_input_tokens": 1838, + "estimated_eval_tokens": 3932 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues do not form a duplicate cluster; they look like unrelated spam/low-information posts mixed with a few generic suggestions.", + "confidence": 0.96, + "canonical_issue_reason": "No canonical issue: the set mixes spam placeholders, a docs suggestion, a benchmark tip, and a Google LLC mention with no shared concrete bug/request.", + "canonical_pr_reason": null, + "best_issue_reason": "44418 is the only minimally descriptive non-spam issue, but it still does not match the others closely.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are just \"\" placeholders; no concrete shared bug or change is described." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Both are just \"\" placeholders; no concrete shared bug or change is described." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "Spam placeholder vs. documentation enhancement suggestion; unrelated." + }, + { + "left": "issue:44371", + "right": "issue:47036", + "accept": false, + "reason": "Spam placeholder vs. community benchmark tip; unrelated." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder vs. \"Check out \\\"Google LLC\\\"\"; unrelated." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Documentation enhancement vs. spam placeholder; no shared underlying request." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Documentation enhancement vs. spam placeholder; no shared underlying request." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Both are spam placeholders, but that is not enough to establish the same underlying issue." + }, + { + "left": "issue:44572", + "right": "issue:47036", + "accept": false, + "reason": "Spam placeholder vs. community benchmark tip; unrelated." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder vs. \"Check out \\\"Google LLC\\\"\"; unrelated." + }, + { + "left": "issue:44573", + "right": "issue:47036", + "accept": false, + "reason": "Spam placeholder vs. community benchmark tip; unrelated." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder vs. \"Check out \\\"Google LLC\\\"\"; unrelated." + } + ] + }, + "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": 26, + "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": "3d492ccd4205fb5795844eb35d862144dc698003e7036abcfa29204ffdf1e7bd", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "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 issues are highly similar and likely report the same underlying problem, but the evidence is limited to identical titles, similar body lengths, and soft-similarity signals. #45145 is the better canonical thread because it already has an inbound reference.", + "confidence": 0.81, + "canonical_issue_reason": "#45145 is a likely duplicate pairmate to #45151 and has the slight edge as the more established thread, with one inbound reference already pointing to it.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is the best issue candidate because it appears to be the more central thread for this likely duplicate report and already has external attention.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": true, + "reason": "The titles are identical and the bodies are similarly sized, so they likely describe the same issue report; however, the match is based on limited evidence rather than a confirmed link." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet supports a strong similarity claim, but the analyst overstates certainty. We only have identical titles, similar body lengths, and a soft-similarity signal; there is no body text comparison, explicit link, or other hard evidence that these are definitively the same issue. The duplicate and canonical-thread conclusions should be phrased more conservatively as likely/possible rather than clear." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 15, + "component_count": 26, + "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": 26, + "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": "72f99de081f23f7720df49f59fbc5373cad39c85200061cc54974bd8411db1e4", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1475, + "estimated_input_tokens": 369, + "estimated_eval_tokens": 994 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1475, + "estimated_input_tokens": 369, + "estimated_eval_tokens": 994 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are only superficially similar by title format; they refer to different names and different dates, with no shared context or linkage. They should not be treated as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No canonical issue: the issues appear to be separate entries about different individuals, with no body text or evidence connecting them.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good canonical duplicate target; there is no substantive overlap beyond a similar naming pattern.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "Different names and dates with no contextual overlap; this is similarity in format only, not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 17, + "component_count": 26, + "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": "c6dd9cf117a2ba9100ebba25c336dee99b4065d67f34253e3e7aa44cfcb2f2b7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two closed issues with different personal-name/date titles and no body or linkage evidence; they do not appear to describe the same underlying problem.", + "confidence": 0.93, + "canonical_issue_reason": "No strong canonical choice: both issues are content-free and only contain distinct name/date strings, so neither is a better duplicate representative.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good global best representative because there is no bug report content, reproduction, or shared subject matter to establish a canonical topic.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "Reject: the titles refer to different people and different dates, with no shared description, 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": 26, + "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": "cadb012a1858009c1b65f0524eaa44ac3352a0af434b80e89a5f70fdb296595d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1541, + "estimated_input_tokens": 386, + "estimated_eval_tokens": 1028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one asks for a new Qwen3_5 token-classification/value-model class, while the other requests FP8 training support. They target different features and code paths.", + "confidence": 0.96, + "canonical_issue_reason": "Issue #45810 is the more concrete, narrowly scoped feature request and is the better representative of the cluster if one issue must be picked.", + "canonical_pr_reason": null, + "best_issue_reason": "#45810 is the most actionable and specific item; #46295 is a separate training-capability request with different implementation scope.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45810", + "right": "issue:46295", + "accept": false, + "reason": "Reject: these do not describe the same underlying bug or change. One is about adding a Qwen3_5 token classification/value-model class, the other is about FP8 training support." + } + ] + }, + "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": 26, + "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": "fe1aa7d9dea593e37968a33cebf8e1cdb8d460aee0146511a63f33c419466d2b", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5727, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5727, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster appears to contain two highly similar issue pairs\u201445870/45871 and 45872/45873\u2014plus 45869 as a related but separate item. The evidence supports treating the two pairs as likely duplicates, but not calling them exact duplicates from the packet alone. 45869 is not the same underlying request.", + "confidence": 0.85, + "canonical_issue_reason": "45870 is a strong canonical anchor for the \"Example: Financial time series prediction with DataForge\" subcluster because it has an identical-title counterpart and sits in the most clearly duplicate-looking pair, while 45869 is unrelated and 45872/45873 form a separate similar pair.", + "canonical_pr_reason": null, + "best_issue_reason": "45870 is the best overall issue representative because it anchors one of the two strongest duplicate-like issue pairs and is more representative of the cluster than the unrelated 45869.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Different artifacts: a dataset bundle promo versus a financial time-series example; related finance theme only, not the same issue." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Different artifacts: dataset bundle promo versus an example issue; the shared financial-data context is too broad to treat as a duplicate." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Identical title and matching body length make this a likely duplicate pair representing the same underlying issue." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "They share financial-time-series/DataForge wording, but one is an example and the other is a scraping-service topic; not the same underlying request." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Same broad domain, but the concrete topic differs from 45870's example/prediction request." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Related wording only; the concrete issue described is different." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Similar theme, but not the same specific issue or change request." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Identical title and matching body length make this a likely duplicate pair representing the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The clustering summary is directionally consistent, but it overstates the evidence by calling 45870/45871 and 45872/45873 \"exact duplicate\" pairs. The packet only provides soft-similarity signals, identical titles, and matching body lengths\u2014not the bodies themselves\u2014so a more conservative phrasing would be \"highly similar\" or \"likely duplicates.\" The 45869 exclusion looks reasonable, but the duplicate certainty is stronger than the packet supports." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 20, + "component_count": 26, + "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": "792c6b893d7675c0c12aa65cdee999e5ed68ed466906c18143822d64195cc456", + "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": "Both issues involve static-cache failures in linear-attention paths, but they describe different concrete symptoms: one is shape corruption during `early_initialization`, the other is a `KeyError` for `linear_attention` in Qwen3.5.", + "confidence": 0.87, + "canonical_issue_reason": "Issue 46439 is the better canonical representative because it describes the core static-cache/linear-attention failure in a more general, subsystem-level way and includes the concrete shape-corruption runtime error.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46439 is the strongest standalone issue because it is more broadly applicable and more diagnostic of the underlying bug than the product-specific Qwen3.5 `KeyError` report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46439", + "right": "issue:46441", + "accept": false, + "reason": "Reject: these look related to the same static-cache/linear-attention area, but the reported failures are different concrete bugs/symptoms (`RuntimeError` from wrong shapes vs `KeyError: 'linear_attention'`) and there is not enough evidence that one fix would plausibly cover both." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 21, + "component_count": 26, + "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": "d4ca78fd840ab4230e9e286d08e99f3d32d4202e942f346d1d1e658c65056f4d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1705, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1705, + "estimated_input_tokens": 427, + "estimated_eval_tokens": 1110 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are near-exact duplicates with the same title and body describing the same Wav2Vec2CTCTokenizer v5 special-token bug. The open issue with inbound references is the better canonical tracker.", + "confidence": 0.99, + "canonical_issue_reason": "Issue 46552 is the better canonical tracker because it is still open and already has inbound references, while 46551 is closed.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46552 is the strongest representative of the cluster: it is open, matches the same problem statement exactly, and appears to be the active report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46551", + "right": "issue:46552", + "accept": true, + "reason": "Same title and same body describing the same tokenizer bug; these are duplicates of the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The verdict is grounded in the packet: the two issues have identical titles, identical body length, and a soft_similarity score of 1.0, so treating them as duplicates is conservative. Choosing the open issue with inbound references as the canonical tracker is also supported." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 22, + "component_count": 26, + "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": "70edf92b0407f7f7eb31553f758d6a9e4be7dc72664abed57964ca07f8bc59e9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1568, + "estimated_input_tokens": 392, + "estimated_eval_tokens": 1040 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1568, + "estimated_input_tokens": 392, + "estimated_eval_tokens": 1040 + }, + "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 training. The symptoms, likely failure modes, and model context are different.", + "confidence": 0.91, + "canonical_issue_reason": "Issue 46657 is the better canonical issue because it describes a concrete training-time tensor shape mismatch, which is a specific bug report with clearer failure semantics than the broader OOM symptom in 46692.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46657 is the best issue anchor for this cluster because it is more specific and likely easier to map to a single underlying code-path problem.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46657", + "right": "issue:46692", + "accept": false, + "reason": "Reject: tensor size mismatch during object-detection training and post-epoch OOM in RT-DETR v2 are different failure modes; the titles do not indicate 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": 23, + "component_count": 26, + "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": "3459b776c9c289e70658a1ad5b6f4725eca1cd1c9429dc72c47e66d69e7393c9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1671, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1671, + "estimated_input_tokens": 418, + "estimated_eval_tokens": 1092 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two different bug reports in different model-loading paths: ZeRO-3 zero.Init partitioning vs 4-bit device_map/max_memory dispatch. I would not merge them as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46822", + "right": "issue:46823", + "accept": false, + "reason": "Different underlying problems and subsystems: 46822 is about ZeRO-3 failing to partition a composite submodule during multi-GPU load, while 46823 is about 4-bit loading with device_map='auto' and max_memory incorrectly placing modules on CPU. They do not look like the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 24, + "component_count": 26, + "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:47049", + "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": "3f93be8d8c99d759bd199fb97f05fe24ac5a0507681364d860000ec1f7227540", + "component_budget": { + "node_count": 11, + "item_count": 11, + "soft_pair_count": 34, + "serialized_chars": 18420, + "estimated_input_tokens": 4605, + "estimated_eval_tokens": 9466 + }, + "budget": { + "node_count": 11, + "item_count": 11, + "soft_pair_count": 34, + "serialized_chars": 18420, + "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 a dated series of separate [serge] integration failure triage issues, not duplicate reports of one underlying bug. The soft similarities reflect near-identical triage templates and recurring wording, but there\u2019s no evidence they should be merged.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 46906 is the earliest item in the dated triage series and is the cleanest anchor, though the set looks like separate daily reports rather than duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 47037 has the most discussion activity, so it is the strongest representative if one issue must stand in for the series.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46906", + "right": "issue:46940", + "accept": false, + "reason": "Same triage-series title, but these are different daily reports; no evidence of one shared bug." + }, + { + "left": "issue:46906", + "right": "issue:46948", + "accept": false, + "reason": "Near-identical template, but different dates and no concrete sign they describe the same failure instance." + }, + { + "left": "issue:46906", + "right": "issue:46969", + "accept": false, + "reason": "They are separate dated triage entries, not the same underlying issue." + }, + { + "left": "issue:46906", + "right": "issue:46985", + "accept": false, + "reason": "Same broad wording, but each issue is a distinct daily triage report." + }, + { + "left": "issue:46906", + "right": "issue:47037", + "accept": false, + "reason": "Recurring triage series with no evidence these are duplicate bug reports." + }, + { + "left": "issue:46940", + "right": "issue:46948", + "accept": false, + "reason": "Adjacent daily triage reports; similarity comes from the shared template, not a shared defect." + }, + { + "left": "issue:46940", + "right": "issue:46969", + "accept": false, + "reason": "Different dates in the same triage series; not enough to merge as duplicates." + }, + { + "left": "issue:46940", + "right": "issue:46985", + "accept": false, + "reason": "These look like separate daily check-ins on integration failures, not one issue." + }, + { + "left": "issue:46940", + "right": "issue:47049", + "accept": false, + "reason": "Same recurring title, but no evidence of the same concrete bug or change." + }, + { + "left": "issue:46948", + "right": "issue:46969", + "accept": false, + "reason": "Both are triage reports, but they are distinct dated entries." + }, + { + "left": "issue:46948", + "right": "issue:46985", + "accept": false, + "reason": "Shared subsystem wording is not enough to treat them as duplicates." + }, + { + "left": "issue:46948", + "right": "issue:46998", + "accept": false, + "reason": "Separate daily triage issues; no same-failure evidence." + }, + { + "left": "issue:46969", + "right": "issue:46985", + "accept": false, + "reason": "Different daily triage reports with similar wording only." + }, + { + "left": "issue:46969", + "right": "issue:47037", + "accept": false, + "reason": "These are distinct dated triage entries, not one bug report." + }, + { + "left": "issue:46969", + "right": "issue:47091", + "accept": false, + "reason": "Same report series, but no sign they reference the same underlying incident." + }, + { + "left": "issue:46985", + "right": "issue:46998", + "accept": false, + "reason": "Near-duplicate template text, but separate daily triage items." + }, + { + "left": "issue:46985", + "right": "issue:47037", + "accept": false, + "reason": "No concrete shared bug; just recurring triage wording." + }, + { + "left": "issue:46985", + "right": "issue:47049", + "accept": false, + "reason": "These appear to be independent daily integration-failure triage reports." + }, + { + "left": "issue:46998", + "right": "issue:47053", + "accept": false, + "reason": "Similar triage framing, but different dates and no shared failure details." + }, + { + "left": "issue:46998", + "right": "issue:47037", + "accept": false, + "reason": "Distinct daily reports; similarity is only superficial." + }, + { + "left": "issue:46998", + "right": "issue:47049", + "accept": false, + "reason": "Same template, but not the same underlying bug." + }, + { + "left": "issue:46998", + "right": "issue:47091", + "accept": false, + "reason": "These are separate triage records, not mergeable duplicates." + }, + { + "left": "issue:46985", + "right": "issue:47091", + "accept": false, + "reason": "Repeated triage-series wording, but each issue is a different day\u2019s report." + }, + { + "left": "issue:47037", + "right": "issue:47049", + "accept": false, + "reason": "Adjacent triage issues; no evidence of the same concrete change or bug." + }, + { + "left": "issue:47037", + "right": "issue:47091", + "accept": false, + "reason": "Same report family, but not the same incident." + }, + { + "left": "issue:47037", + "right": "issue:47053", + "accept": false, + "reason": "Different dated triage entries with shared boilerplate only." + }, + { + "left": "issue:47049", + "right": "issue:47053", + "accept": false, + "reason": "Similar wording is due to the daily triage template, not a duplicate issue." + }, + { + "left": "issue:47049", + "right": "issue:47091", + "accept": false, + "reason": "These are separate daily integration-failure reports." + }, + { + "left": "issue:47053", + "right": "issue:47091", + "accept": false, + "reason": "No concrete shared bug; just the same triage series." + }, + { + "left": "issue:47053", + "right": "issue:47102", + "accept": false, + "reason": "Different daily triage issues, not one underlying failure." + }, + { + "left": "issue:47091", + "right": "issue:47102", + "accept": false, + "reason": "Same series, but each issue is a distinct day\u2019s triage note." + }, + { + "left": "issue:46985", + "right": "issue:47091", + "accept": false, + "reason": "Repeated triage-series wording, but still separate daily reports." + }, + { + "left": "issue:47037", + "right": "issue:47102", + "accept": false, + "reason": "No evidence these two track the same bug; they are dated triage entries." + }, + { + "left": "issue:46985", + "right": "issue:47102", + "accept": false, + "reason": "Different daily reports with similar boilerplate only." + }, + { + "left": "issue:47049", + "right": "issue:47102", + "accept": false, + "reason": "Same recurring triage format, but not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 25, + "component_count": 26, + "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": "d7283f66b2f94c1a6a2229be03193ec9be2b7133b2a5ca52444459f718e9d5a5", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1490, + "estimated_input_tokens": 373, + "estimated_eval_tokens": 1002 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1490, + "estimated_input_tokens": 373, + "estimated_eval_tokens": 1002 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two issues share the exact title \"Add RWKV7 model support\" and have a strong soft-similarity signal, so they likely represent the same feature request; however, the packet does not expose the issue bodies, so the evidence is not enough to claim identical text. The open issue is the better cluster representative.", + "confidence": 0.86, + "canonical_issue_reason": "#46984 is the better canonical issue because it is the open, active tracker for the same titled request.", + "canonical_pr_reason": null, + "best_issue_reason": "#46984 best represents the cluster as the current open issue for this request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:46983", + "right": "issue:46984", + "accept": false, + "reason": "The pair looks likely related based on the identical title and high soft-similarity signal, but the packet does not show the underlying issue text, so there is not enough evidence here to confidently mark them as the same underlying request." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Reject: the cluster is plausibly the same feature request, but the analyst overstates the evidence. The packet only shows identical titles and a soft_similarity signal (with Jaccard), not the actual issue text, so \u201chighly similar issue text\u201d is unsupported. Also, \u201cno distinguishing metadata\u201d is a bit too strong because the issues differ in body length and activity. A more conservative verdict would be: likely duplicate/same request based on identical title and soft similarity." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 26, + "component_count": 26, + "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": "e0c9657be4f95e7418395212c87c98296a36c43218a3bb7a16c20877d43788dc", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1625, + "estimated_input_tokens": 407, + "estimated_eval_tokens": 1070 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1625, + "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 issues are topically similar (both about AI/LLM-based vulnerability discovery), but they read like different writeups rather than the same underlying bug or change. I would not merge them as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "No clear duplicate canonical emerges from the titles alone; if forced, 47015 is the more general/central phrasing, but the pair does not look like a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 47015 is the broader, more representative title for the shared AI vulnerability-discovery theme, so it is the better cluster anchor if one must be chosen.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:47007", + "right": "issue:47015", + "accept": false, + "reason": "They share a broad subject area, but the titles describe different angles: one is about lessons from using advanced LLMs, the other about a novel transformer-based cybersecurity discovery system. That is not enough to treat them as the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 33, + "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-40546-41", + "nodes": [ + "pull_request:43451", + "pull_request:44517", + "pull_request:45694", + "pull_request:46196", + "pull_request:46266", + "pull_request:46658", + "pull_request:46707", + "pull_request:46836", + "pull_request:46903", + "pull_request:46912", + "pull_request:46954", + "pull_request:46957", + "pull_request:46958", + "pull_request:46968", + "pull_request:46974", + "pull_request:46995", + "pull_request:47003", + "pull_request:47025", + "pull_request:47034", + "pull_request:47035", + "pull_request:47069", + "pull_request:47072", + "pull_request:47077", + "pull_request:47078", + "pull_request:47082", + "pull_request:47089", + "pull_request:47090", + "pull_request:47092", + "pull_request:47093", + "pull_request:47096", + "pull_request:47100", + "pull_request:47107", + "pull_request:47108", + "pull_request:47113", + "pull_request:47114", + "pull_request:47117", + "pull_request:47118", + "pull_request:47119" + ], + "soft_pairs": [ + "pull_request:46903|pull_request:47090", + "pull_request:46903|pull_request:46958", + "pull_request:46954|pull_request:46958", + "pull_request:46954|pull_request:46968", + "pull_request:46958|pull_request:46974", + "pull_request:47089|pull_request:47119", + "pull_request:46958|pull_request:46968", + "pull_request:46968|pull_request:46974", + "pull_request:47072|pull_request:47119", + "pull_request:47072|pull_request:47089", + "pull_request:47077|pull_request:47078", + "pull_request:46968|pull_request:47114", + "pull_request:46707|pull_request:46968", + "pull_request:46266|pull_request:46957", + "pull_request:44517|pull_request:47082", + "pull_request:47034|pull_request:47035", + "pull_request:47069|pull_request:47089", + "pull_request:46196|pull_request:46995", + "pull_request:47093|pull_request:47114", + "pull_request:46707|pull_request:47114", + "pull_request:46707|pull_request:47108", + "pull_request:47089|pull_request:47113", + "pull_request:46266|pull_request:46707", + "pull_request:47108|pull_request:47114", + "pull_request:47082|pull_request:47108", + "pull_request:46912|pull_request:47118", + "pull_request:47107|pull_request:47118", + "pull_request:47117|pull_request:47118", + "pull_request:47077|pull_request:47114", + "pull_request:47100|pull_request:47118", + "pull_request:46912|pull_request:47025", + "pull_request:46196|pull_request:47118", + "pull_request:47100|pull_request:47117", + "pull_request:47114|pull_request:47118", + "pull_request:46658|pull_request:47114", + "pull_request:46707|pull_request:47082", + "pull_request:46266|pull_request:47114", + "pull_request:47025|pull_request:47107", + "pull_request:46912|pull_request:47082", + "pull_request:43451|pull_request:47082", + "pull_request:46912|pull_request:47078", + "pull_request:47092|pull_request:47117", + "pull_request:47035|pull_request:47089", + "pull_request:43451|pull_request:46995", + "pull_request:46196|pull_request:47025", + "pull_request:46836|pull_request:46995", + "pull_request:46707|pull_request:47025", + "pull_request:45694|pull_request:47069", + "pull_request:46266|pull_request:47034", + "pull_request:47034|pull_request:47108", + "pull_request:45694|pull_request:47107", + "pull_request:46995|pull_request:47096", + "pull_request:47100|pull_request:47113", + "pull_request:47035|pull_request:47113", + "pull_request:47035|pull_request:47069", + "pull_request:46957|pull_request:47108", + "pull_request:46995|pull_request:47003", + "pull_request:47069|pull_request:47113", + "pull_request:46912|pull_request:47077", + "pull_request:47113|pull_request:47117", + "pull_request:46707|pull_request:47077", + "pull_request:47113|pull_request:47118", + "pull_request:43451|pull_request:46266", + "pull_request:47069|pull_request:47082", + "pull_request:45694|pull_request:47035", + "pull_request:43451|pull_request:47107", + "pull_request:43451|pull_request:47078", + "pull_request:47078|pull_request:47096", + "pull_request:43451|pull_request:47117", + "pull_request:46707|pull_request:46957", + "pull_request:47003|pull_request:47078", + "pull_request:47092|pull_request:47107" + ], + "prepared_review_unit_hash": "f50e831e038b96a317cf229f62ba926afa45c62c46524500aad585ab0a85dbf8", + "component_budget": { + "node_count": 41, + "item_count": 41, + "soft_pair_count": 119, + "serialized_chars": 146957, + "estimated_input_tokens": 36740, + "estimated_eval_tokens": 73736 + }, + "budget": { + "node_count": 38, + "item_count": 38, + "soft_pair_count": 72, + "serialized_chars": 113726, + "estimated_input_tokens": 28432, + "estimated_eval_tokens": 57120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Only the CohereASR training-loss PRs look like true duplicates of the same underlying bug. The rest mostly share broad subsystems or filenames but fix different issues, so they should stay separate.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46903 is the best canonical PR: it fixes the CohereASR double-shift loss bug and also carries the related PPFormulaNet change, with an explicit issue link and the most complete scope among the duplicates.", + "best_issue_reason": null, + "best_pr_reason": "PR 46903 is the strongest representative of the duplicate set because it is the broadest, has the issue reference, and addresses the same loss-shift bug on the shared CohereASR code path.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46903", + "right": "pull_request:47090", + "accept": true, + "reason": "Same CohereASR training-loss double-shift bug in the same code path; 47090 is just an alternate implementation of the same fix." + }, + { + "left": "pull_request:46903", + "right": "pull_request:46958", + "accept": true, + "reason": "Both fix the CohereASR double-shift loss bug in the same files; 46903 just bundles an extra PPFormulaNet fix too." + }, + { + "left": "pull_request:46954", + "right": "pull_request:46958", + "accept": false, + "reason": "Different bugs: Pytest-v9 fp32-module behavior in GlmAsr vs CohereASR training loss shifting." + }, + { + "left": "pull_request:46954", + "right": "pull_request:46968", + "accept": false, + "reason": "Unrelated changes: GLM-ASR pytest/fp32 handling vs docs autolink fixes." + }, + { + "left": "pull_request:46958", + "right": "pull_request:46974", + "accept": false, + "reason": "Same broad area only by name; one is CohereASR loss logic, the other is a Musicgen generate typo." + }, + { + "left": "pull_request:47089", + "right": "pull_request:47119", + "accept": false, + "reason": "Both are small validation fixes, but for different code paths: AqlmConfig type checks vs image padding handling." + }, + { + "left": "pull_request:46958", + "right": "pull_request:46968", + "accept": false, + "reason": "CohereASR loss bug vs documentation autolink cleanup; no shared underlying change." + }, + { + "left": "pull_request:46968", + "right": "pull_request:46974", + "accept": false, + "reason": "Docs autolink cleanup and Musicgen generate typo are unrelated." + }, + { + "left": "pull_request:47072", + "right": "pull_request:47119", + "accept": false, + "reason": "Chat-schema enum parsing and image padding are different utility bugs." + }, + { + "left": "pull_request:47072", + "right": "pull_request:47089", + "accept": false, + "reason": "Chat-template schema parsing vs AqlmConfig error-message/type checks are unrelated." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47078", + "accept": false, + "reason": "Both touch Japanese docs, but they change different content and wording, not the same change." + }, + { + "left": "pull_request:46968", + "right": "pull_request:47114", + "accept": false, + "reason": "Docs autolink cleanup vs weight-conversion docs update; same documentation area only." + }, + { + "left": "pull_request:46707", + "right": "pull_request:46968", + "accept": false, + "reason": "Different changes: native FSDP2 migration vs docs autolink fixes." + }, + { + "left": "pull_request:46266", + "right": "pull_request:46957", + "accept": false, + "reason": "Both touch generation/modeling internals, but one is DINO conversion mapping and the other is decode optimization for FP8/MoE." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47082", + "accept": false, + "reason": "Both add model registry entries, but for different models and different features." + }, + { + "left": "pull_request:47034", + "right": "pull_request:47035", + "accept": false, + "reason": "Neuron SFT scripts/training args vs torch.compile backend default; related platform area, but different fixes." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47089", + "accept": false, + "reason": "Different bugs: num_labels returning None vs AqlmConfig type-error wording." + }, + { + "left": "pull_request:46196", + "right": "pull_request:46995", + "accept": false, + "reason": "Different accelerator/model-support additions: SDPA dispatch docs/model changes vs MLU support for flash linear attention." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47114", + "accept": false, + "reason": "New model support for OpenPangu-V2 vs documentation for custom-code conversion mapping." + }, + { + "left": "pull_request:46707", + "right": "pull_request:47114", + "accept": false, + "reason": "FSDP2 migration work and conversion-mapping docs are unrelated." + }, + { + "left": "pull_request:47089", + "right": "pull_request:47113", + "accept": false, + "reason": "AqlmConfig validation messages vs audio feature/processor requirements are unrelated." + }, + { + "left": "pull_request:46266", + "right": "pull_request:46707", + "accept": false, + "reason": "Different core changes: DINO conversion mapping vs native FSDP2 support." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47114", + "accept": false, + "reason": "Doc-builder pin comment cleanup vs conversion-mapping docs content; not the same change." + }, + { + "left": "pull_request:47108", + "right": "pull_request:47082", + "accept": false, + "reason": "Doc-builder workflow comment cleanup vs RoMa model support are unrelated." + }, + { + "left": "pull_request:47082", + "right": "pull_request:47096", + "accept": false, + "reason": "Different model additions: RoMa keypoint matcher vs exportable kimi support." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47118", + "accept": false, + "reason": "RoPE test expansion across models vs cache dispatch simplification; not the same bug." + }, + { + "left": "pull_request:47107", + "right": "pull_request:47118", + "accept": false, + "reason": "MoE decode optimization for submodel experts vs cache-layer dispatch simplification are distinct changes." + }, + { + "left": "pull_request:47117", + "right": "pull_request:47118", + "accept": false, + "reason": "Output-record capture docs/code changes vs cache dispatch refactor are unrelated." + }, + { + "left": "pull_request:47100", + "right": "pull_request:47118", + "accept": false, + "reason": "kernels API compatibility fix vs cache dispatch simplification; different code paths." + }, + { + "left": "pull_request:47114", + "right": "pull_request:47118", + "accept": false, + "reason": "Weight-conversion docs update vs cache dispatch refactor are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47118", + "accept": false, + "reason": "Exportable kimi support and cache dispatch simplification are different features." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47118", + "accept": false, + "reason": "OpenVINO exporter support vs cache dispatch simplification are unrelated." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47003", + "accept": false, + "reason": "Different hardware/runtime support areas; flash-linear-attention availability is not the same as OpenVINO exporter work." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47070", + "accept": false, + "reason": "Config property typing change and chat-template enum parsing are unrelated utility fixes." + }, + { + "left": "pull_request:47092", + "right": "pull_request:47089", + "accept": false, + "reason": "Test expectations for MiniCPM3 on XPU vs AqlmConfig validation messages are unrelated." + }, + { + "left": "pull_request:47035", + "right": "pull_request:47089", + "accept": false, + "reason": "Torch-compile backend defaulting and quantization-config error text are different changes." + }, + { + "left": "pull_request:47034", + "right": "pull_request:47108", + "accept": false, + "reason": "Neuron training experiment scripts vs doc-builder workflow comment cleanup are unrelated." + }, + { + "left": "pull_request:47034", + "right": "pull_request:47069", + "accept": false, + "reason": "Neuron training scripts vs num_labels property typing are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47078", + "accept": false, + "reason": "OpenVINO exporter support and Japanese docs wording changes are unrelated." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47114", + "accept": false, + "reason": "Japanese DistilBERT translation vs English weight-converter docs update are different documentation tasks." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47108", + "accept": false, + "reason": "FP8/MoE decode optimization and doc-builder comment cleanup are unrelated." + }, + { + "left": "pull_request:47092", + "right": "pull_request:47107", + "accept": false, + "reason": "Test expectation updates for XPU vs MoE decode optimization are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47082", + "accept": false, + "reason": "Molmo2 model addition and RoMa model addition are different model integrations." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46995", + "accept": false, + "reason": "Molmo2 model support vs flash-linear-attention availability are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46266", + "accept": false, + "reason": "Different model addition vs Dino conversion-mapping modernization." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47107", + "accept": false, + "reason": "Molmo2 model support and MoE decode optimization are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47078", + "accept": false, + "reason": "Molmo2 model addition vs Japanese docs wording changes are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47117", + "accept": false, + "reason": "Molmo2 model support and output-recording docs/code changes are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47117", + "accept": false, + "reason": "OpenVINO exporter support vs output-capturing changes are unrelated." + }, + { + "left": "pull_request:47072", + "right": "pull_request:47114", + "accept": false, + "reason": "Schema parsing fix and conversion-mapping docs update are unrelated." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47025", + "accept": false, + "reason": "RoPE test expansion vs Diffusion Gemma test adjustments are different model test changes." + }, + { + "left": "pull_request:47100", + "right": "pull_request:47117", + "accept": false, + "reason": "kernels API compatibility fix vs output-capturing changes are unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47117", + "accept": false, + "reason": "Audio-feature/processor requirements vs output-capturing changes are unrelated." + }, + { + "left": "pull_request:47113", + "right": "pull_request:47118", + "accept": false, + "reason": "Audio requirements fixes and cache dispatch refactor are unrelated." + }, + { + "left": "pull_request:47089", + "right": "pull_request:47096", + "accept": false, + "reason": "Quantization-config type checking and exportable kimi support are unrelated." + }, + { + "left": "pull_request:47069", + "right": "pull_request:47082", + "accept": false, + "reason": "num_labels typing fix vs RoMa model addition are unrelated." + }, + { + "left": "pull_request:45694", + "right": "pull_request:47035", + "accept": false, + "reason": "TrainingArguments field formatting/respect-split-batches work and torch.compile backend defaulting are different fixes." + }, + { + "left": "pull_request:45694", + "right": "pull_request:47107", + "accept": false, + "reason": "TrainingArguments behavior change vs MoE decode optimization are unrelated." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47082", + "accept": false, + "reason": "RoPE tests and RoMa model addition do not address the same underlying issue." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47078", + "accept": false, + "reason": "RoPE tests vs Japanese docs wording changes are unrelated." + }, + { + "left": "pull_request:47092", + "right": "pull_request:47117", + "accept": false, + "reason": "XPU test expectations and output-recording changes are unrelated." + }, + { + "left": "pull_request:47035", + "right": "pull_request:47113", + "accept": false, + "reason": "Torch-compile backend defaulting vs audio feature/processor requirement guards are unrelated." + }, + { + "left": "pull_request:47035", + "right": "pull_request:47069", + "accept": false, + "reason": "Different TrainingArguments/configuration fixes with no shared bug." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47114", + "accept": false, + "reason": "FP8/MoE generation optimization and conversion-mapping docs are unrelated." + }, + { + "left": "pull_request:46707", + "right": "pull_request:47108", + "accept": false, + "reason": "FSDP2 migration and doc-builder comment cleanup are unrelated." + }, + { + "left": "pull_request:47077", + "right": "pull_request:47108", + "accept": false, + "reason": "Japanese translation changes and workflow comment cleanup are unrelated." + }, + { + "left": "pull_request:47078", + "right": "pull_request:47114", + "accept": false, + "reason": "Japanese wording edits and weight-converter docs update are unrelated." + }, + { + "left": "pull_request:47093", + "right": "pull_request:47114", + "accept": false, + "reason": "New OpenPangu-V2 model support vs conversion-mapping documentation are unrelated." + }, + { + "left": "pull_request:47096", + "right": "pull_request:47114", + "accept": false, + "reason": "Exportable kimi support vs conversion-mapping docs are unrelated." + }, + { + "left": "pull_request:47117", + "right": "pull_request:47118", + "accept": false, + "reason": "Output-capture mapping changes and cache dispatch simplification are unrelated." + }, + { + "left": "pull_request:46958", + "right": "pull_request:47114", + "accept": false, + "reason": "CohereASR loss bug fix vs conversion-mapping docs are unrelated." + }, + { + "left": "pull_request:46903", + "right": "pull_request:47114", + "accept": false, + "reason": "CohereASR/PPFormulaNet loss fix vs conversion-mapping docs are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47096", + "accept": false, + "reason": "OpenVINO exporter support and kimi exportability are different exporter features." + }, + { + "left": "pull_request:46954", + "right": "pull_request:47118", + "accept": false, + "reason": "Pytest-v9 GLM-ASR change vs cache dispatch simplification are unrelated." + }, + { + "left": "pull_request:47025", + "right": "pull_request:47107", + "accept": false, + "reason": "Diffusion Gemma test fixes vs MoE decode optimization are unrelated." + }, + { + "left": "pull_request:46196", + "right": "pull_request:47118", + "accept": false, + "reason": "Wav2Vec2 SDPA dispatch support and cache dispatch simplification are unrelated." + }, + { + "left": "pull_request:47089", + "right": "pull_request:47117", + "accept": false, + "reason": "AqlmConfig validation messages vs output-capturing changes are unrelated." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47108", + "accept": false, + "reason": "FP8/MoE optimization and doc-builder workflow comment cleanup are unrelated." + }, + { + "left": "pull_request:46995", + "right": "pull_request:47003", + "accept": false, + "reason": "Hardware availability check and exporter addition are different features." + }, + { + "left": "pull_request:46954", + "right": "pull_request:47025", + "accept": false, + "reason": "GLM-ASR pytest compatibility and Diffusion Gemma test fixes are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The conclusions are grounded in the packet and the duplicate judgments are conservative. The CohereASR PRs are the only clearly same-bug cluster, and the rest of the pairwise rejections stay within evidence shown by titles/diffs. No obvious overstatement or non-mergeable pairing issue stands out." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 33, + "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-40546-41", + "nodes": [ + "pull_request:40546", + "pull_request:43451", + "pull_request:44517", + "pull_request:45694", + "pull_request:46196", + "pull_request:46658", + "pull_request:46707", + "pull_request:46836", + "pull_request:46880", + "pull_request:46895", + "pull_request:46912", + "pull_request:46957", + "pull_request:46958", + "pull_request:47003", + "pull_request:47025", + "pull_request:47034", + "pull_request:47035", + "pull_request:47077", + "pull_request:47090", + "pull_request:47092", + "pull_request:47096", + "pull_request:47100", + "pull_request:47107", + "pull_request:47113" + ], + "soft_pairs": [ + "pull_request:43451|pull_request:44517", + "pull_request:46836|pull_request:47107", + "pull_request:46912|pull_request:47113", + "pull_request:46912|pull_request:47100", + "pull_request:45694|pull_request:46707", + "pull_request:44517|pull_request:46836", + "pull_request:44517|pull_request:47096", + "pull_request:44517|pull_request:47003", + "pull_request:43451|pull_request:46957", + "pull_request:46957|pull_request:47096", + "pull_request:46957|pull_request:47003", + "pull_request:46658|pull_request:46895", + "pull_request:43451|pull_request:47100", + "pull_request:43451|pull_request:45694", + "pull_request:43451|pull_request:46895", + "pull_request:46895|pull_request:47096", + "pull_request:46880|pull_request:47092", + "pull_request:46895|pull_request:47090", + "pull_request:46895|pull_request:46958", + "pull_request:46958|pull_request:47090", + "pull_request:47025|pull_request:47092", + "pull_request:46707|pull_request:47034", + "pull_request:46658|pull_request:46836", + "pull_request:43451|pull_request:46836", + "pull_request:46836|pull_request:47092", + "pull_request:46836|pull_request:47096", + "pull_request:43451|pull_request:47077", + "pull_request:46836|pull_request:47003", + "pull_request:40546|pull_request:43451", + "pull_request:43451|pull_request:46196", + "pull_request:46196|pull_request:47096", + "pull_request:46196|pull_request:47003", + "pull_request:43451|pull_request:46658", + "pull_request:46658|pull_request:47096", + "pull_request:46658|pull_request:47003", + "pull_request:43451|pull_request:47092", + "pull_request:43451|pull_request:47096", + "pull_request:43451|pull_request:47003", + "pull_request:47003|pull_request:47096", + "pull_request:43451|pull_request:47034", + "pull_request:47034|pull_request:47096", + "pull_request:47092|pull_request:47096", + "pull_request:43451|pull_request:47035", + "pull_request:47035|pull_request:47096", + "pull_request:43451|pull_request:46912", + "pull_request:47003|pull_request:47034", + "pull_request:47003|pull_request:47092" + ], + "prepared_review_unit_hash": "f714fabfac04e2a6b3650546ccc473ba85e40f85d050bba923f20e0dd01ec0cf", + "component_budget": { + "node_count": 41, + "item_count": 41, + "soft_pair_count": 119, + "serialized_chars": 146957, + "estimated_input_tokens": 36740, + "estimated_eval_tokens": 73736 + }, + "budget": { + "node_count": 24, + "item_count": 24, + "soft_pair_count": 47, + "serialized_chars": 76358, + "estimated_input_tokens": 19090, + "estimated_eval_tokens": 38436 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are not duplicates: they\u2019re unrelated model additions, exporter features, test tweaks, or doc-only changes. The only real duplicate cluster here is the CohereASR training-loss double-shift fixes, where all three PRs address the same loss-path bug.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "pull_request:47090 is the clearest expression of the shared CohereASR bug fix: it names the exact training-loss double-shift problem and matches the same code path as 46895 and 46958.", + "best_issue_reason": null, + "best_pr_reason": "pull_request:47090 is the best representative PR for the duplicate set because it is the most explicit about the underlying bug and is directly interchangeable with the other two CohereASR fixes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43451", + "right": "pull_request:44517", + "accept": false, + "reason": "Both add different models (Molmo2 vs Qwen3-TTS); same kind of integration work, but not the same change." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47107", + "accept": false, + "reason": "Unrelated changes: Unlimited OCR model integration vs MoE decode optimization in generation." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47113", + "accept": false, + "reason": "RoPE test/model coverage work is unrelated to audio feature/processor requirement fixes." + }, + { + "left": "pull_request:46912", + "right": "pull_request:47100", + "accept": false, + "reason": "Different problems: RoPE tests vs a kernels API/deprecation compatibility fix." + }, + { + "left": "pull_request:45694", + "right": "pull_request:46707", + "accept": false, + "reason": "TrainingArguments batch-size semantics and FSDP2 migration are unrelated changes." + }, + { + "left": "pull_request:44517", + "right": "pull_request:46836", + "accept": false, + "reason": "Different new model integrations; same broad auto-mapping/docs pattern, but not the same model or bug." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47096", + "accept": false, + "reason": "Qwen3-TTS integration and Kimi exportability are separate features." + }, + { + "left": "pull_request:44517", + "right": "pull_request:47003", + "accept": false, + "reason": "Qwen3-TTS model support is unrelated to adding an OpenVINO exporter." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46957", + "accept": false, + "reason": "Molmo2 model support and finegrained FP8 decode optimization are different changes." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47096", + "accept": false, + "reason": "Both touch generation/export code, but they solve different issues in different paths." + }, + { + "left": "pull_request:46957", + "right": "pull_request:47003", + "accept": false, + "reason": "Finegrained FP8 decode optimization is not the same as adding an OpenVINO exporter backend." + }, + { + "left": "pull_request:46658", + "right": "pull_request:46895", + "accept": false, + "reason": "Step-3.7-Flash conversion/model support is unrelated to CohereASR loss-shift fixing." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47100", + "accept": false, + "reason": "New model support and a kernels API compatibility fix are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45694", + "accept": false, + "reason": "Molmo2 integration and batch-size semantics in TrainingArguments are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46895", + "accept": false, + "reason": "Molmo2 support and CohereASR loss correction are unrelated." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47096", + "accept": false, + "reason": "CohereASR loss-shift fix and exportability work for Kimi are unrelated." + }, + { + "left": "pull_request:46880", + "right": "pull_request:47092", + "accept": false, + "reason": "Adding XPU expectations to tests and adding XPU expectations to different tests are separate test-only changes." + }, + { + "left": "pull_request:46895", + "right": "pull_request:47090", + "accept": true, + "reason": "Same underlying CohereASR training-loss double-shift bug and same code path." + }, + { + "left": "pull_request:46895", + "right": "pull_request:46958", + "accept": true, + "reason": "Both fix CohereASR by correcting the training loss computation against the same labels path." + }, + { + "left": "pull_request:46958", + "right": "pull_request:47090", + "accept": true, + "reason": "These are the same CohereASR double-shift fix expressed with different implementation text." + }, + { + "left": "pull_request:47025", + "right": "pull_request:47092", + "accept": false, + "reason": "Diffusion Gemma test adjustments and MiniCPM3 XPU expectations are unrelated test maintenance." + }, + { + "left": "pull_request:46707", + "right": "pull_request:47034", + "accept": false, + "reason": "FSDP2 migration and Neuron SFT experiment scaffolding are different workstreams." + }, + { + "left": "pull_request:46658", + "right": "pull_request:46836", + "accept": false, + "reason": "Step-3.7-Flash support and Unlimited OCR support are different model additions." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46836", + "accept": false, + "reason": "Different model integrations, not the same underlying bug or change." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47092", + "accept": false, + "reason": "Unlimited OCR integration and XPU test expectations for MiniCPM3 are unrelated." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47096", + "accept": false, + "reason": "Unlimited OCR model support and exportability fixes for Kimi are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47077", + "accept": false, + "reason": "Molmo2 support and Japanese translation updates are unrelated." + }, + { + "left": "pull_request:46836", + "right": "pull_request:47003", + "accept": false, + "reason": "Unlimited OCR integration and OpenVINO exporter work are unrelated." + }, + { + "left": "pull_request:40546", + "right": "pull_request:43451", + "accept": false, + "reason": "VibeVoice and Molmo2 are distinct model integrations." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46196", + "accept": false, + "reason": "Model integration work and SDPA dispatch support for wav2vec2 are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46658", + "accept": false, + "reason": "Molmo2 addition and Step-3.7-Flash conversion/model work are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47092", + "accept": false, + "reason": "Molmo2 model support and MiniCPM3 XPU test expectations are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47096", + "accept": false, + "reason": "Molmo2 model support and Kimi exportability fixes are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47003", + "accept": false, + "reason": "Molmo2 model support and OpenVINO exporter work are unrelated." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47034", + "accept": false, + "reason": "Molmo2 model integration and Neuron SFT experiment code are unrelated." + }, + { + "left": "pull_request:47034", + "right": "pull_request:47096", + "accept": false, + "reason": "Neuron SFT experiment scaffolding and Kimi exportability fixes are unrelated." + }, + { + "left": "pull_request:47092", + "right": "pull_request:47096", + "accept": false, + "reason": "Different concerns: test expectations for MiniCPM3 vs exporter compatibility for Kimi." + }, + { + "left": "pull_request:43451", + "right": "pull_request:47035", + "accept": false, + "reason": "Molmo2 integration and torch.compile backend defaults for Neuron are unrelated." + }, + { + "left": "pull_request:47035", + "right": "pull_request:47096", + "accept": false, + "reason": "Torch.compile backend defaulting and Kimi exportability are unrelated changes." + }, + { + "left": "pull_request:43451", + "right": "pull_request:46912", + "accept": false, + "reason": "Molmo2 support and RoPE testing changes are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47034", + "accept": false, + "reason": "OpenVINO exporter work and Neuron experiment scaffolding are unrelated." + }, + { + "left": "pull_request:47003", + "right": "pull_request:47092", + "accept": false, + "reason": "OpenVINO exporter changes and MiniCPM3 test expectations are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 2, + "component_count": 33, + "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": "ae0228b294007c1df897730f5e3e6da0a3439be44e8a43f240af761654a34edf", + "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": "These PRs are unrelated: one adds TimesFM covariate forecasting support, the other adds DINOv3 image classification support. They do not share a concrete bug or change target, so they should not be clustered as duplicates.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different models and code paths (TimesFM forecasting vs DINOv3 image classification); no shared underlying bug or 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": 3, + "component_count": 33, + "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": 33, + "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-41855-13", + "nodes": [ + "pull_request:41855", + "pull_request:42668", + "pull_request:43251", + "pull_request:43254", + "pull_request:43424", + "pull_request:43823", + "pull_request:43995", + "pull_request:44827", + "pull_request:45170", + "pull_request:45213", + "pull_request:45655", + "pull_request:46048", + "pull_request:46267" + ], + "soft_pairs": [ + "pull_request:41855|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:42668|pull_request:43823", + "pull_request:43251|pull_request:43823", + "pull_request:43251|pull_request:44827", + "pull_request:43424|pull_request:44827", + "pull_request:43424|pull_request:43823", + "pull_request:43823|pull_request:45655", + "pull_request:43823|pull_request:43995", + "pull_request:43995|pull_request:45213", + "pull_request:43995|pull_request:44827", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:46267", + "pull_request:44827|pull_request:46048", + "pull_request:44827|pull_request:45655", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "eed9b0f61172e747b7743f0c157d77938f29cc2bee49926b57ec2f1f72eb55b8", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 18, + "serialized_chars": 34786, + "estimated_input_tokens": 8697, + "estimated_eval_tokens": 17650 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 18, + "serialized_chars": 34786, + "estimated_input_tokens": 8697, + "estimated_eval_tokens": 17650 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly unrelated PRs that happen to touch Transformers internals; the only true duplicate pair is the two loss-utils PRs that both forward extra kwargs into fixed_cross_entropy. Everything else is just broad thematic overlap or different subsystems.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43251 is the best canonical representative: it is the more complete, issue-linked fix for fixed_cross_entropy forwarding kwargs to nn.functional.cross_entropy, and it cleanly subsumes the overlapping 43254 change.", + "best_issue_reason": null, + "best_pr_reason": "PR 43251 is the strongest standalone duplicate target in the cluster because it addresses the concrete cross-entropy kwargs bug directly and has the explicit issue link plus review activity.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Tokenizer missing-methods work and executorch dynamic-shape export tests are unrelated changes." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Mistral tokenizer methods vs adding MobileLLM are different models and different fixes." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor robustness for AudioFlamingo3 is unrelated to introducing MobileLLM." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs forwarding and MobileLLM model addition are unrelated code paths." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "Loss-function kwargs fix and Mistral4 test/model changes are not the same underlying bug." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch export test coverage and Mistral4 test fixes are different changes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Executorch export testing and MobileLLM model addition do not target the same issue." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "New model implementation and config-resolution ordering fix are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "MobileLLM addition and Falcon output-collection refactor are different model-level changes." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon forward/refactor work is unrelated to the model-creation skill PR." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-collection refactor and Mistral4 tests do not fix the same bug." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test fixes and a layrnorm-to-layernorm rename are different issues." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46267", + "accept": false, + "reason": "Mistral4 test fixes and chat-response parsing updates are unrelated subsystems." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 test cleanup and adding a TDT loss kernel are not the same change." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 tests and cls.config resolution order are unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 test fixes and the model-creation skill are entirely different artifacts." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "MobileLLM model addition and the model-creation skill are not the same underlying code change." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs patch the same concrete code path in fixed_cross_entropy by forwarding supported kwargs to nn.functional.cross_entropy, and they could plausibly be merged into one PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: the only clearly mergeable same-bugfix pair is PR 43251/43254 on fixed_cross_entropy, while the rest are distinct models/subsystems. The soft-edge rejections are conservative and the canonical/duplicate reasoning is supported by the issue link and overlapping diff." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 5, + "component_count": 33, + "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": "60d54e4042a086551d05822b7ca30a4dae8e7097966f10f0f0cf93dd48f6f36f", + "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": "Two open PRs that both bump GitHub Actions dependencies in workflow files, but they target different action families and different upgrade goals. They overlap in affected workflows, yet they are not the same concrete change and should not be merged as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "No issue is present. If one PR must represent the cluster, PR #42975 is the better anchor because it is the more specific compatibility-focused workflow action upgrade and better explains the shared automation maintenance theme.", + "best_issue_reason": null, + "best_pr_reason": "PR #42975 is the better representative because its title and diff clearly center on a single compatibility objective (Node 24 support) across workflow actions, whereas PR #42976 is a separate, narrower set of action version bumps.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Reject as duplicate: both are workflow-maintenance PRs, but #42975 updates checkout/cache/upload-artifact for Node 24 compatibility while #42976 updates docker/build-push-action and a few other actions to latest versions. They touch some of the same workflows, but they do not appear to 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": 6, + "component_count": 33, + "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": "dfda101fa3db3c4fcead9e2eef1e6e69535d8c685b2626494e6a57b021e05910", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 13047, + "estimated_input_tokens": 3262, + "estimated_eval_tokens": 6780 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 13047, + "estimated_input_tokens": 3262, + "estimated_eval_tokens": 6780 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster groups four independent model-addition PRs (ViT NEPA, Molmo, CTSM, Step3p7). They share some boilerplate docs/auto-registration files, but each introduces a different model family and codepath, so they are not duplicates.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45490 is the best representative only in a bookkeeping sense: it has the broadest review activity and touches the standard model-registration stack, but it is not a true duplicate of the others.", + "best_issue_reason": null, + "best_pr_reason": "None of the PRs is a good global duplicate target; they add different models and would not be mergeable into one fix/change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo); shared docs/auto files are generic scaffolding, not the same underlying change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "Different architectures and codepaths (ViT NEPA vs CTSM); overlap is only in registry/docs plumbing." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are unrelated model implementations; shared filenames do not indicate a duplicate bug or change." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46562", + "accept": false, + "reason": "CTSM and Step3p7 are separate model additions; despite some shared auto-mapping/timeseries files, they are not the same concrete change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 7, + "component_count": 33, + "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": "9a9ea4746a9369d4995f5c108adc338510e822d68a4310dfdac73aa714198a15", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19590, + "estimated_input_tokens": 4898, + "estimated_eval_tokens": 10052 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 15, + "serialized_chars": 19590, + "estimated_input_tokens": 4898, + "estimated_eval_tokens": 10052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster contains one clear duplicate pair in YOLOS (46886/46922). The remaining PRs are not duplicates: they cover separate fixes in tests, FP8/loading, generation validation, and CI/workflow infrastructure, with substantive code changes in several of them.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46922 is the best canonical representative for the YOLOS pair because it includes the concrete forward-path device fix and the matching no_split_modules update; 46886 only adds the metadata-side split guard.", + "best_issue_reason": null, + "best_pr_reason": "PR 46922 is the strongest standalone PR in the cluster because it addresses the actual YOLOS runtime mismatch directly and subsumes the smaller metadata-only change in 46886.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Both are test-related, but they target different models and different failures; no same underlying bug or change." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper tokenization tests and FP8/dequantization work are unrelated fixes." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46380", + "accept": false, + "reason": "Different subsystems and different issues; the overlap is only superficial test activity." + }, + { + "left": "pull_request:43291", + "right": "pull_request:46922", + "accept": false, + "reason": "Whisper test cleanup is unrelated to the YOLOS device-mapping fix." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46539", + "accept": false, + "reason": "Both touch tests or CI, but they address different models and different problems." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "CLIP/ViT test edits are unrelated to the FP8 loading/dequantization changes." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46380", + "accept": false, + "reason": "Test edits versus generation/watermarking and whisper-related code are not the same change." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46922", + "accept": false, + "reason": "Different model families and different fixes; the shared test framing is not enough." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46539", + "accept": false, + "reason": "FP8 quantization/loading changes are unrelated to the Phimoe CI/runtime fix." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46380", + "accept": false, + "reason": "No shared concrete bug: one is FP8/dequantization, the other is generation validation plus broader test updates." + }, + { + "left": "pull_request:44815", + "right": "pull_request:46922", + "accept": false, + "reason": "Different model code paths and different runtime issues; not the same fix." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46539", + "accept": false, + "reason": "Generation watermarking validation and CI memory/device-map infrastructure are separate problems." + }, + { + "left": "pull_request:46380", + "right": "pull_request:46922", + "accept": false, + "reason": "These are unrelated fixes in different subsystems." + }, + { + "left": "pull_request:46539", + "right": "pull_request:46922", + "accept": false, + "reason": "CI device/memory patching for Phimoe is unrelated to the YOLOS dispatch bug." + }, + { + "left": "pull_request:46886", + "right": "pull_request:46922", + "accept": true, + "reason": "Both target the same YOLOS device_map=\"auto\" mismatch in the same file and issue; 46922 adds the concrete forward-path fix on top of the metadata change." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The duplicate-pair call on 46886/46922 is grounded: they share the YOLOS file and issue target, and 46922 adds the forward-path fix on top of the metadata change. However, the cluster summary overstates the rest of the set as \"test-only\"; 44815, 46380, and 46539 all include substantive code and workflow changes beyond tests. Keep the unrelated-vs-duplicate distinction, but make the non-duplicate summary more conservative." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 33, + "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": "f3b5dfa29b1eeeec5d73304b01518884a99a4686aab5f00d1d82b9fd3b95f9b0", + "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": "All three PRs are separate \u201c[don\u2019t merge]\u201d experiments touching different areas: packaging/test deps, CI/comment-workflow plumbing, and a security-gate workflow change. The shared similarity is superficial, so there is no duplicate/canonical PR here.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the three pull requests address different concrete changes and do not look like the same underlying fix or feature.", + "best_issue_reason": null, + "best_pr_reason": "No single PR is a good duplicate target; the overlap is only broad subsystem/test-file similarity, not the same code-path problem.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:46747", + "accept": false, + "reason": "Different changes: 43488 edits setup/dependency versions and a test file, while 46747 only changes the PR security-gate workflow. They do not fix the same bug or change the same code path." + }, + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "Only a loose shared test filename appears. 43488 is about packaging/dependency/version tweaks, whereas 43532 rewires multiple CI/workflow files and test utilities. Not the same underlying change." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46747", + "accept": false, + "reason": "Both touch CI/workflow infrastructure, but the concrete edits differ: 43532 changes comment/scheduled/slack workflows plus testing utils, while 46747 only switches the security gate trigger. Too broad to treat 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": 33, + "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": "6638dac6ea438fa5d496c92b7b353a885e0e663c03dff08256cdb5f5a22d90d9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6083, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6083, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related segmentation pipeline additions, but they target different tasks and code paths: Promptable Concept Segmentation vs Promptable Visual Segmentation. They are not duplicates.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR here; the two PRs introduce distinct pipelines with different model/test/doc surfaces.", + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a duplicate of the other. If forced to pick a representative, 43613 is broader, but it is not a substitute for 43612 because it serves a different segmentation task.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Reject: the PRs add different pipelines for different tasks and model stacks. Sharing docs/registry updates and a broad segmentation theme is not enough to treat them as the same fix/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 10, + "component_count": 33, + "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": "4c51409bc5b0480e7f7c782520ab43f8c7a2fc9461299737c0a1509ef9b26fea", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated Trainer changes in the same file: one adjusts gradient-accumulation loss scaling, the other refactors signature-column filtering for custom trainers/PEFT. They are not the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the PRs target different code paths and different behaviors in `trainer.py`, so they should not be deduplicated.", + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a duplicate representative; if a single PR must be picked for triage, 43651 is the more concrete fix, but it is not the same change as 43663.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different features in the same module: 43651 changes gradient accumulation loss normalization, while 43663 adds an overridable signature-column helper for input filtering. They do not fix the same bug and would not be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 11, + "component_count": 33, + "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": "d6749780c17d4e9f43264adb638c2428b80258736dd04c4afc13bebc1fde2d2a", + "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": "Both PRs address the same TypeAdapter-related runtime NameError in `src/transformers/cli/serve.py`, but `43842` is the narrower and more direct fix; `43656` includes broader typing/annotation cleanup around the same file.", + "confidence": 0.81, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43842 is the best canonical PR: it directly fixes the `TypeAdapter` annotation/runtime import issue with a minimal change and a precise title/body tied to the pydantic-not-installed failure mode.", + "best_issue_reason": null, + "best_pr_reason": "PR 43842 is the best overall representative because it targets the exact crash condition with the smallest, most focused patch.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": true, + "reason": "Same underlying bug in the same code path: a `TypeAdapter` NameError in the CLI serve module. The implementations differ, but both are plausible fixes for the same runtime annotation/import failure." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The comparison is grounded: both PR titles point to a TypeAdapter NameError in `src/transformers/cli/serve.py`, and the narrower `43842` looks like a plausible canonical representative. The soft-edge acceptance is conservative enough given the shared file and closely related fix patterns." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 33, + "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": 13, + "component_count": 33, + "review_unit_index": 1, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43996", + "pull_request:43997", + "pull_request:44007", + "pull_request:44010", + "pull_request:44013", + "pull_request:44017", + "pull_request:44018", + "pull_request:44019", + "pull_request:44024", + "pull_request:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44066|pull_request:44071", + "pull_request:44018|pull_request:44068", + "pull_request:44066|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44018|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44068|pull_request:44116", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44085", + "pull_request:44018|pull_request:44086", + "pull_request:44071|pull_request:44085", + "pull_request:44068|pull_request:44086", + "pull_request:44085|pull_request:44086", + "pull_request:44098|pull_request:44101", + "pull_request:44059|pull_request:44086", + "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:43997|pull_request:44073", + "pull_request:44068|pull_request:44071", + "pull_request:44071|pull_request:44086", + "pull_request:44017|pull_request:44071", + "pull_request:44073|pull_request:44161", + "pull_request:44073|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44056|pull_request:44101", + "pull_request:44018|pull_request:44059", + "pull_request:44059|pull_request:44066", + "pull_request:44101|pull_request:44161", + "pull_request:43997|pull_request:44017", + "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: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:44071|pull_request:44129", + "pull_request:43997|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", + "pull_request:44073|pull_request:44074" + ], + "prepared_review_unit_hash": "3905ff5780a46127b5d318441f6a9c6cdcd33bfc1e28ae673f4a0e5f2aa23c50", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 76, + "serialized_chars": 84109, + "estimated_input_tokens": 21028, + "estimated_eval_tokens": 42312 + }, + "budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 72, + "serialized_chars": 76930, + "estimated_input_tokens": 19233, + "estimated_eval_tokens": 38722 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a broad output-tracing refactor theme spanning many different model implementations. I only accept soft edges when they are the same concrete model/code path (not just the same standardized-decorator pattern or the same tracking issue). The true duplicates here are the GPT-Neo slice, the GPT-J slices, and the Flaubert slice; most other pairs are model-specific refactors and should stay separate.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44722 is the clearest representative of the concrete GPT-J/CodeGen output-tracing change: it edits the same GPT-J/CodeGen code paths, explicitly targets the tracking issue, and is broader/more complete than the smaller follow-up slices.", + "best_issue_reason": null, + "best_pr_reason": "PR 44722 is the best single representative for the cluster because it directly captures the underlying GPT-J/CodeGen refactor rather than only the generic standardized-output theme.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Same high-level output-tracing theme, but different model families and code paths (CVT/FNet vs GPT-J)." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Both are refactors around output tracing, but they target unrelated models and implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Both touch GPT-J output tracing in the same code path; this looks like the same underlying refactor and could plausibly be merged." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "RegNet and EfficientNet are separate model implementations; same issue/theme is not enough." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet and GPT-J are unrelated code paths despite both being standardized output-tracing work." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "Different model families and different forward paths; only the refactor style overlaps." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa-v2 are unrelated implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen and MGP-STR are unrelated; same decorator theme is too broad." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen versus MPT is not the same concrete bug/change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both are GPT-Neo output-tracing refactors on the same model code path, so they look like the same underlying change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen and GPT-Neo are different model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "Different models: GPT-Neo versus GPT-J/CodeGen." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet and GPT-J/CodeGen are unrelated code paths." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are unrelated model implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet and GPT-J are unrelated; same issue label would not make them duplicates." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are completely different models/code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet versus MPT; only the refactor motif overlaps." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-Neo and SpeechT5 are unrelated implementations." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model families." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR are unrelated model code paths." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are different models; same theme is not enough." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert are unrelated model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 and MGP-STR are unrelated code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-2 and GPT-J/CodeGen are different model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are unrelated despite similar output-tracing cleanup." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "Segformer and LongT5 are different code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "Segformer and ViLT are unrelated model implementations." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and Segformer are separate models; same refactor theme is too broad." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are unrelated code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are unrelated implementations." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet and Flaubert are different model 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 unrelated code paths." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet and LongT5 are unrelated models." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44085", + "accept": false, + "reason": "MPT and GPT-J are different model implementations." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44161", + "accept": false, + "reason": "CVT/FNet and LongT5 are unrelated code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", + "accept": false, + "reason": "CVT/FNet and Segformer are different model implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBERT and DeBERTa-v2 are unrelated models." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa-v2 and TextNet are unrelated code paths." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44161", + "accept": false, + "reason": "ViLT and LongT5 are unrelated model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-2 and MPT are different models." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44101", + "accept": false, + "reason": "MPNet and Flaubert are unrelated code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44018", + "accept": false, + "reason": "Segformer and GPT-Neo are unrelated models." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44129", + "accept": false, + "reason": "MPT and SpeechT5 are unrelated implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-Neo and SpeechT5 are unrelated." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44074", + "accept": false, + "reason": "RegNet and TextNet are different model code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44073", + "accept": false, + "reason": "CVT/FNet and ViLT are unrelated implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44018", + "accept": false, + "reason": "SqueezeBERT and GPT-Neo are unrelated models." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBERT and ViLT are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model families." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44154", + "accept": false, + "reason": "EfficientNet and VITS are unrelated models." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44101", + "accept": false, + "reason": "SqueezeBERT and Flaubert are unrelated code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "Segformer and ViLT are unrelated model implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "Segformer and LongT5 are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated models." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT-2 are different model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are unrelated despite similar refactor style." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44071", + "accept": false, + "reason": "MPNet and MPT are unrelated model code paths." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model families." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are unrelated models." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen and GPT-Neo are different model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor on the same concrete code paths; this is a plausible duplicate/follow-up." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both are GPT-J output-tracing refactors on the same code path and look like the same underlying change." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44116", + "accept": true, + "reason": "Both modify the same Flaubert attention/output-tracing code path and appear to be the same refactor in different slices." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet and ImageGPT are unrelated model implementations." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "ViLT and TextNet are separate models; same standardized-output pattern is not enough." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays mostly conservative. The accepted soft edges are all supported by shared model/code-path overlap, and the representative PR choice is reasonable. I don\u2019t see an overstatement beyond the evidence provided." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 33, + "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: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": "3c914234c2212f60cbcc35f5023cfc1d5aa6dcc0ecd5b172e2349ff34bcf1e8c", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 76, + "serialized_chars": 84109, + "estimated_input_tokens": 21028, + "estimated_eval_tokens": 42312 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 4, + "serialized_chars": 12757, + "estimated_input_tokens": 3190, + "estimated_eval_tokens": 6636 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "All items are model-specific refactors to the new standardized output-tracing utilities, but they target different architectures and code paths, so they are not duplicates of the same underlying change.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "No issue artifact is present; among the PRs, pull_request:44056 is a strong representative because it cleanly shows the output-tracing refactor pattern with an explicit output-recording map in a single model file.", + "best_issue_reason": null, + "best_pr_reason": "pull_request:44056 is a good cluster representative because it captures the core migration pattern (adding capture_outputs/can_return_tuple and wiring _can_record_outputs) in a focused, model-local change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "Same refactor theme, but different models (MPNet vs VisualBert) and different forward paths/classes; too broad to treat as one concrete PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "Both migrate output tracing, but they apply to different architectures (MPNet vs TextNet) with distinct implementation details; not the same underlying change." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBERT are unrelated model code paths; the similarity is only the shared utility migration, not a single fix/change." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBERT and SpeechT5 touch different model internals and output contracts; they are separate refactors, not one mergeable 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": 33, + "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": "675ccaa035afdd47710e3af7b9c3717d7fa937325243b86e24ae74743aea89c9", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40165, + "estimated_input_tokens": 10042, + "estimated_eval_tokens": 20340 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40165, + "estimated_input_tokens": 10042, + "estimated_eval_tokens": 20340 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All artifacts are separate, model-specific output-tracing refactors; they share a pattern, but not the same underlying change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "pull_request:44025 is the broadest representative: it refactors both depth_anything and prompt_depth_anything with the new tracing decorator.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model implementations and forward paths; same refactor theme only." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different architectures; not the same concrete change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Different model code paths; only the tracing refactor pattern matches." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Unrelated model families; not mergeable as one fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Different backbones and output paths; too broad to be duplicate." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "Different models and output signatures; not the same bug/change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model targets; shared refactor style is insufficient." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "MobileNetV1 vs Depth Anything are separate implementations." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Different wrapper/model code paths; not a duplicate change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "Separate encoder-decoder family and model-specific refactor." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Vision-text dual encoder and UnivNet are unrelated code paths." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model classes; same output-tracing theme only." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and forward implementations." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Distinct models; no shared underlying fix beyond refactor style." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "UnivNet and UperNet are different model paths." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different architectures; not the same code-path problem." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "Different tasks and modules; not mergeable as one PR." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are unrelated implementations." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and output contracts." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Mamba/Falcon-Mamba and SuperPoint are unrelated code paths." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Different models and concrete changes; only generic tracing refactor overlaps." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Timm backbone and CodeGen fix different model code paths." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet and CodeGen are unrelated changes." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen do not share the same underlying bug." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Depth Anything and Vision Encoder-Decoder are separate model wrappers." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "Different encoder-decoder implementations; not the same fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Depth Anything and SuperPoint are unrelated model paths." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Different models and forward logic; too broad to merge together." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "Depth Anything and DPR are separate implementations." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Vision Encoder-Decoder and RWKV are unrelated code paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Vision and speech encoder-decoder wrappers are distinct changes." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different models; same tracing refactor theme only." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "Vision Encoder-Decoder and DPR do not fix the same problem." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "Speech encoder-decoder and SuperPoint are unrelated." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and code paths." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "Speech Encoder-Decoder and DPR are separate model-specific refactors." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV have different forward implementations." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "Different models and output shapes; not duplicate changes." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are unrelated model 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": 15, + "component_count": 33, + "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": "aed7a9dda648e230859867ef620c8e6b8a67853c8d2f37ce455cedbe54d14ff9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6551, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6551, + "estimated_input_tokens": 1638, + "estimated_eval_tokens": 3532 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate model-addition PRs: PenguinVL implementation vs Voxtral TTS support. They share some boilerplate auto-mapping/docs paths, but they target different models and codepaths, so they should not be deduplicated.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45401 is the clearer cluster representative because it has an explicit issue target and a focused implementation for Voxtral TTS, but it is still not a duplicate of #44662.", + "best_issue_reason": null, + "best_pr_reason": "PR #44662 is the more substantial standalone PR in this cluster because it is a non-draft, broader implementation with more review activity. It best represents the PenguinVL side, though the two PRs are unrelated and should remain separate.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different models and changes: PenguinVL implementation versus Voxtral TTS support. Shared docs/auto-registry plumbing is generic and not enough to indicate the same underlying bug or 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": 16, + "component_count": 33, + "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": "2adda2d8ad55b04474d3128861f0042600e18a90266982d6cc93f64f6fb583f8", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4407, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two PRs are related to fusion infrastructure, but they target different layers and goals: one introduces a generic Module Fusion API, while the other adds KernelConfig-based n-to-1 kernel fusion integration and an example. They are not the same underlying change.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44979 is the more foundational change: it introduces the core Module Fusion API and tests, which the other PR appears to build on rather than duplicate.", + "best_issue_reason": null, + "best_pr_reason": "PR #44979 is the best representative of this cluster because it defines the underlying fusion abstraction; PR #45363 is a more specific application/integration of that idea.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "They are related but not duplicates: #44979 adds a generic module fusion API, while #45363 wires KernelConfig-driven n-to-1 kernel fusion into model loading and includes an example. Different concrete code paths and purposes, so they would not 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": 17, + "component_count": 33, + "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": "5d5cec7046e6de36683a75dd189047000df17abf0e8622e8ebff5b1c9f82e38e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests patch the same AttributeError in `_patch_mistral_regex` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; the second PR adds test coverage but targets the same concrete fix.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same code fix as 45086 plus a test file, making it the more complete change.", + "best_issue_reason": null, + "best_pr_reason": "PR 45317 best represents the cluster since it fixes the same bug and includes validation in tests, while 45086 is the narrower variant.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying bug and same code-path fix in the same file; the extra test in 45317 does not change the core change, so they are 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": 18, + "component_count": 33, + "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": "3dda778432d4e2906eb503ea5233e2a3d4d3224f0299669e176d224d3395908e", + "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": "Two related NemotronH PRs, but they address different layers of the work: one is a refactor/conversion-mapping inheritance change, the other splits NemotronH into dense/sparse models with docs/tests. They are not duplicates.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45569 is the better canonical representative because it is the broader, more complete product change: it adds the new dense/sparse model split, updates auto-mapping/docs, and includes tests. PR #45115 is a narrower refactor focused on inheritance and conversion mapping.", + "best_issue_reason": null, + "best_pr_reason": "PR #45569 is the strongest standalone PR in the cluster because it covers the user-visible architecture/API change end to end and is backed by docs and tests. PR #45115 is more of a supporting refactor than the primary change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "Not the same concrete change. #45115 focuses on NemotronH refactoring and weight-conversion renames toward GraniteMoeHybrid compatibility, while #45569 introduces separate NemotronH dense/sparse model classes, docs, auto-mapping, and tests. They overlap in files but are not mergeable 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": 19, + "component_count": 33, + "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-45332-2", + "nodes": [ + "pull_request:45332", + "pull_request:45333" + ], + "soft_pairs": [ + "pull_request:45332|pull_request:45333" + ], + "prepared_review_unit_hash": "e1f538e3a8f67a93180df1f5d48a51400acb9a53ca7cd468ea8379c5e78ffdeb", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6081, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6081, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two pull requests are closely related and share config/docs/test files, but they are not duplicates: one adds broader heterogeneous model support and the other focuses on per-layer configuration utilities. They look like complementary or layered work, not the same mergeable change.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45332 is the broader, more complete change set: it includes heterogeneous modeling plus the config work, while #45333 is scoped to configuration utilities and docs/tests.", + "best_issue_reason": null, + "best_pr_reason": "PR #45332 is the best representative of the cluster because it covers the main code-path change and subsumes the narrower configuration-focused PR.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Shared files indicate overlap in the heterogeneous config subsystem, but the PRs are not the same concrete fix/change: #45332 adds full heterogeneous model support and touches modeling/cache/integration code, while #45333 is limited to per-layer configuration utilities and docs/tests. They would not cleanly collapse into one PR without mixing distinct scopes." + } + ] + }, + "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": 33, + "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": "15c15eb216bdab8b16a5b8c993f7404196922ea36a85b778f8327309da9b02ea", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are typing-related, but they are not the same change: #45415 is a broad type-checking sweep across many Transformers modules plus pyproject type-ignore config, while #45425 is a targeted typing refactor adding shared aliases/protocols for specific modeling/PEFT paths. Their overlap is incidental rather than a shared concrete bug/fix.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "#45415 is the broader umbrella typing PR and best represents the cluster\u2019s general theme, but it is still distinct from the targeted typing refactor in #45425.", + "best_issue_reason": null, + "best_pr_reason": "#45415 is the best representative PR here because it spans the largest set of affected files and includes the type-checking configuration change, making it the most canonical of the two typing cleanups.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "They share a typing/annotation theme and a few filenames, but #45415 is a broad type-checking sweep while #45425 adds specific protocols/aliases for modeling_utils/PEFT. They do not appear to fix the same concrete code-path problem and would not naturally merge 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": 21, + "component_count": 33, + "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": "5918793089f2df4fbb63e7ff2bf298f1862aaf6efaf3675f9b19321bb4bd5179", + "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": "These PRs are related at a high level (expert parallelism loading), but they fix different bugs on different code paths: one changes config serialization / EP detection for accelerate+ZeRO-3, the other prevents parameter overwrites during FSDP2 loading. They should not be treated as duplicates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: 45548 is about serializing distributed_config and exposing has_ep, while 45662 fixes shard-aware parameter assignment during loading; they do not resolve the same bug.", + "best_issue_reason": null, + "best_pr_reason": "PR 45662 is the stronger representative of a concrete runtime loading corruption fix, but it still does not subsume 45548's config/EP-detection changes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Different underlying failures: 45548 handles distributed_config serialization and EP detection for accelerate/ZeRO-3, while 45662 adds post-shard wrapping to stop experts being overwritten during FSDP2 loading." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 33, + "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": "c4e5eaf3d6ca960dc1b218ca43a2f1788ab766531dfbf4ce95252b1d4cafd778", + "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 address the same xdist-related captured_info collision problem under issue 45561; 45645 is the more complete implementation because it covers both the writer and CI log collection paths.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45645 is the broader, more complete fix: it makes the test debug output xdist-safe and also updates the workflow to collect multiple captured_info files without losing CI logs.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the best representative because it resolves the artifact-collision problem end-to-end, including the CI-side consumer of those logs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying xdist captured_info collision fix; 45645 appears to be a superset that adds the CI workflow handling needed for the new per-worker files." + } + ] + }, + "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": 33, + "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": 24, + "component_count": 33, + "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": "64658a3aea57a890b911d9936c831fe223d3f3113ff2dfc3154d10a9f843b7ba", + "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 in the sense that both touch distributed/MoE code, but they are not the same underlying change. PR 46126 is a focused EP/router-parallel integration, while PR 46269 is a broad distributed branch base with many unrelated updates, including CI and multiple distributed subsystems.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46126 is the more specific, concrete code-path change: it targets router/EP behavior in tensor parallel, trainer integration, and Qwen3 MoE config. PR 46269 is a broad umbrella branch base touching many areas and is not a precise duplicate candidate.", + "best_issue_reason": null, + "best_pr_reason": "PR 46126 is the best representative of the cluster because it is narrowly scoped and clearly describes a single functional change. PR 46269 is too broad and mixed to serve as the canonical change for duplicate triage.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46126", + "right": "pull_request:46269", + "accept": false, + "reason": "Do not merge: they overlap in distributed filenames, but 46269 is a broad branch-base PR spanning many unrelated changes, not the same concrete bug fix or feature as the focused EP/router integration in 46126." + } + ] + }, + "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": 33, + "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": "01cb246bab2d5f472887bb3d28fc5e4f236aea8bcfbe85983cafd968c10e4ef2", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 31, + "serialized_chars": 26663, + "estimated_input_tokens": 6666, + "estimated_eval_tokens": 13588 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 31, + "serialized_chars": 26663, + "estimated_input_tokens": 6666, + "estimated_eval_tokens": 13588 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is heterogeneous: most PRs are unrelated workflow tweaks, and the pr-ci-security-gate items are similar in file but address different concerns. I found no true duplicate PR pair to merge.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46554 is the most substantive and representative workflow change in the only tight subcluster: it adds concrete security-gate logic, while the nearby pr-ci-security-gate PRs make different, non-overlapping hardening edits.", + "best_issue_reason": null, + "best_pr_reason": "PR 46554 is the best overall representative because it contains the clearest underlying code-path change; the other PRs are separate workflow maintenance or security-hardening tweaks rather than duplicates.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46138", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflows and different behavior; only a broad CI automation theme is shared." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflows; one updates self-comment CI stdin handling, the other changes PR CI security-gate permissions." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46563", + "accept": false, + "reason": "Unrelated workflow files and different change goals." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflows; no shared underlying bug or code path." + }, + { + "left": "pull_request:46138", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files and different fixes; similarity is only superficial." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflows with unrelated behavior changes." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46501", + "accept": false, + "reason": "Different workflow files; one is SSH runner hardening, the other anti-slop permissions." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46502", + "accept": false, + "reason": "Different workflow files and unrelated changes." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow files and distinct security concerns." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow files; no same underlying bug." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46503", + "accept": false, + "reason": "Both touch workflows, but they change different jobs and unrelated logic." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46502", + "accept": false, + "reason": "Different workflow updates in separate files; not the same fix." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46504", + "accept": false, + "reason": "Unrelated workflow changes; only generic CI maintenance overlaps." + }, + { + "left": "pull_request:46501", + "right": "pull_request:46554", + "accept": false, + "reason": "Different files and unrelated security/CI behavior." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46503", + "accept": false, + "reason": "Different workflow files and separate behavior changes." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files; no shared concrete bug." + }, + { + "left": "pull_request:46502", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow files and distinct changes." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46504", + "accept": false, + "reason": "Different workflow files and unrelated edits." + }, + { + "left": "pull_request:46503", + "right": "pull_request:46554", + "accept": false, + "reason": "Separate workflow concerns; not mergeable as one duplicate fix." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46554", + "accept": false, + "reason": "Different workflow files and different operational behavior." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46558", + "accept": false, + "reason": "Different workflow files; one is commit-message handling, the other PR CI permissions." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46558", + "accept": false, + "reason": "Same file, but different changes: one adds security checks, the other adjusts permissions/approval gating." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46563", + "accept": false, + "reason": "Same workflow file, but the diffs address different concerns and are not the same code-path fix." + }, + { + "left": "pull_request:46554", + "right": "pull_request:46716", + "accept": false, + "reason": "Same workflow file, but one changes gate logic while the other changes reusable-workflow secrets handling." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46563", + "accept": false, + "reason": "Same file, but different modifications with different intent." + }, + { + "left": "pull_request:46558", + "right": "pull_request:46716", + "accept": false, + "reason": "Same file, but permission/approval tweaks are distinct from secrets hardening." + }, + { + "left": "pull_request:46563", + "right": "pull_request:46716", + "accept": false, + "reason": "Same file, but comment/documentation-only note versus secrets setting change." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow files and unrelated updates." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46563", + "accept": false, + "reason": "Different workflow files and no shared underlying issue." + }, + { + "left": "pull_request:46504", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow files and different behavior changes." + }, + { + "left": "pull_request:46432", + "right": "pull_request:46716", + "accept": false, + "reason": "Different workflow files; no concrete duplicate 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": 26, + "component_count": 33, + "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": 27, + "component_count": 33, + "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-46281-2", + "nodes": [ + "pull_request:46281", + "pull_request:46890" + ], + "soft_pairs": [ + "pull_request:46281|pull_request:46890" + ], + "prepared_review_unit_hash": "6ade0bd89f46e66aafee8beb11e392e67120030763c0e422ebf72ad9c77b9053", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4279, + "estimated_input_tokens": 1070, + "estimated_eval_tokens": 2396 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4279, + "estimated_input_tokens": 1070, + "estimated_eval_tokens": 2396 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate security-hardened PRs that both block path traversal, but in different code paths: sharded checkpoint loading from an index (`hub.py`) versus resolving `transformers_weights` in model loading (`modeling_utils.py`). They are related by theme, not duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "No single PR is the global duplicate-canonical for this cluster. PR 46281 fixes shard filename traversal in checkpoint index loading, while PR 46890 fixes traversal in `transformers_weights` resolution; they address different concrete bugs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46281", + "right": "pull_request:46890", + "accept": false, + "reason": "Both are path-traversal fixes, but they target different inputs and code paths (`weight_map` shard filenames vs `transformers_weights` config resolution), so 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": 28, + "component_count": 33, + "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": "cd5ed2136983f3635789ca040bd6f3390991d4f85d2823ab6d9e19c5d97cc7b5", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4410, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4410, + "estimated_input_tokens": 1103, + "estimated_eval_tokens": 2462 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two pull requests address different tokenizer problems: one fixes CodeLlama decode stripping real leading spaces, while the other fixes incorrect tokenizer-class selection for DeepSeek-Coder/DeepSeek-LLM checkpoints. They are not duplicates and should remain separate changes.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "No single canonical PR stands out because these are separate bugs in different code paths: CodeLlama decoding behavior versus AutoTokenizer checkpoint-class routing.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46574", + "right": "pull_request:46580", + "accept": false, + "reason": "Different underlying bugs and fixes. PR 46574 changes CodeLlama decoder setup to preserve leading spaces on decode; PR 46580 changes AutoTokenizer checkpoint-prefix routing for DeepSeek models that load the wrong tokenizer class." + } + ] + }, + "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": 33, + "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": "b469ff071725075b88f6d73728065ef16d1fadc8564978681dcb85080aaa5169", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5267, + "estimated_input_tokens": 1317, + "estimated_eval_tokens": 2890 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5267, + "estimated_input_tokens": 1317, + "estimated_eval_tokens": 2890 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related padding-mask fixes, but they address different code paths: one fixes batch-size-1 masking in several linear-attention model implementations, while the other changes the shared recurrent masking utility for continued forwards/chunked prefill/cache continuation. They should not be deduplicated or merged as the same PR.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47087 is the better canonical PR because it targets the shared masking utility, has an explicit issue link, and fixes a broader recurrent-forward masking bug with clearer scope.", + "best_issue_reason": null, + "best_pr_reason": "PR 47087 is the strongest standalone PR: it fixes a concrete shared bug in recurrent masking behavior during continued forwards and is backed by tests and an issue target. PR 46773 is narrower and model-specific.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46773", + "right": "pull_request:47087", + "accept": false, + "reason": "Both concern skipped padding masks, but the underlying bugs differ: 46773 fixes a batch-size-1 condition in linear-attention model code, while 47087 fixes recurrent mask handling for continued forwards in shared masking utilities. They are not the same concrete change 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": 30, + "component_count": 33, + "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-46781-3", + "nodes": [ + "pull_request:46781", + "pull_request:46993", + "pull_request:46994" + ], + "soft_pairs": [ + "pull_request:46781|pull_request:46993", + "pull_request:46993|pull_request:46994" + ], + "prepared_review_unit_hash": "bb3e7a2fbf81fc74f0c8295117fc2abd368826afbf0c44f1987643ec6460d3ce", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 6604, + "estimated_input_tokens": 1651, + "estimated_eval_tokens": 3558 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 6604, + "estimated_input_tokens": 1651, + "estimated_eval_tokens": 3558 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate pull requests touching different generation/cache validation paths. They share a broad Transformers-generation theme, but not the same underlying bug or mergeable change, so neither soft edge should be accepted.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:46781", + "right": "pull_request:46993", + "accept": false, + "reason": "Different code paths and failures: QuantizedCache layer-type validation vs. prompt lookup decoding max_length handling. They are unrelated bug fixes, not one mergeable PR." + }, + { + "left": "pull_request:46993", + "right": "pull_request:46994", + "accept": false, + "reason": "One fixes candidate generation length capping, the other validates empty lists in logits processors. Same subsystem, but distinct issues and 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": 31, + "component_count": 33, + "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-47026-2", + "nodes": [ + "pull_request:47026", + "pull_request:47046" + ], + "soft_pairs": [ + "pull_request:47026|pull_request:47046" + ], + "prepared_review_unit_hash": "7553ee12103cf181e5a8615c5dbfecdfeb53218819b003242476cf1f66cb514b", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4442, + "estimated_input_tokens": 1111, + "estimated_eval_tokens": 2478 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4442, + "estimated_input_tokens": 1111, + "estimated_eval_tokens": 2478 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two xLSTM PRs overlap on the same train-mode forward crash around `return_last_states=False`; PR 47046 appears to be the broader, more complete fix and likely subsumes PR 47026.", + "confidence": 0.84, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 47046 is the more complete xLSTM fix: it targets the same train-mode crash path, adds `use_cache` handling as well, and includes tests/logging changes, so it is the better canonical PR for this bug family.", + "best_issue_reason": null, + "best_pr_reason": "PR 47046 is the best overall PR because it covers the shared crash scenario more comprehensively and appears to include validation coverage, whereas PR 47026 is a narrower subset fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47026", + "right": "pull_request:47046", + "accept": true, + "reason": "Both PRs address the same underlying xLSTM train-mode crash path involving `return_last_states=False` in `modeling_xlstm.py`; PR 47046 looks like a superset that also handles `use_cache`, so they are close enough to treat as the same change cluster." + } + ] + }, + "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": 33, + "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-47032-6", + "nodes": [ + "pull_request:47032", + "pull_request:47045", + "pull_request:47051", + "pull_request:47060", + "pull_request:47074", + "pull_request:47076" + ], + "soft_pairs": [ + "pull_request:47032|pull_request:47051", + "pull_request:47032|pull_request:47045", + "pull_request:47045|pull_request:47076", + "pull_request:47045|pull_request:47074", + "pull_request:47051|pull_request:47074", + "pull_request:47051|pull_request:47060", + "pull_request:47060|pull_request:47074", + "pull_request:47060|pull_request:47076", + "pull_request:47074|pull_request:47076" + ], + "prepared_review_unit_hash": "c5325b77d1c18a3b3875bc7f183df6205abb44f78404af2a6a262e7c852b8095", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 9, + "serialized_chars": 15866, + "estimated_input_tokens": 3967, + "estimated_eval_tokens": 8190 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 9, + "serialized_chars": 15866, + "estimated_input_tokens": 3967, + "estimated_eval_tokens": 8190 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are six separate multimodal pipeline/processor fixes with overlapping themes but different concrete bugs and code paths. I do not see a true duplicate pair among them.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:47032", + "right": "pull_request:47051", + "accept": false, + "reason": "Both touch image_text_to_text pipeline code, but 47032 changes stop_sequence -> stop_strings behavior while 47051 changes direct generation-kwarg forwarding. Different bug classes and not mergeable as one fix." + }, + { + "left": "pull_request:47032", + "right": "pull_request:47045", + "accept": false, + "reason": "Different components and failures: image-text stop-sequence handling versus Voxtral audio-format validation/encoding." + }, + { + "left": "pull_request:47045", + "right": "pull_request:47076", + "accept": false, + "reason": "Both validate processor inputs, but one is Voxtral audio array formatting and the other is Gemma4 audio placeholder/layout validation. Distinct models and code paths." + }, + { + "left": "pull_request:47045", + "right": "pull_request:47074", + "accept": false, + "reason": "Unrelated fixes: Voxtral transcription audio handling versus AnyToAnyPipeline multimodal key normalization." + }, + { + "left": "pull_request:47051", + "right": "pull_request:47074", + "accept": false, + "reason": "Both are in AnyToAnyPipeline, but 47051 forwards generation kwargs and 47074 fixes input-key naming for videos. Separate behaviors and not the same bug." + }, + { + "left": "pull_request:47051", + "right": "pull_request:47060", + "accept": false, + "reason": "Different targets and problems: multimodal pipeline kwarg forwarding versus Qwen Omni batch post-processing decode shape handling." + }, + { + "left": "pull_request:47060", + "right": "pull_request:47074", + "accept": false, + "reason": "Qwen Omni post-processing batch decoding is unrelated to AnyToAnyPipeline input key normalization." + }, + { + "left": "pull_request:47060", + "right": "pull_request:47076", + "accept": false, + "reason": "Different models and code paths: Qwen Omni output decoding versus Gemma4 audio placeholder validation." + }, + { + "left": "pull_request:47074", + "right": "pull_request:47076", + "accept": false, + "reason": "Both are multimodal pipeline/processor changes, but one fixes AnyToAnyPipeline input keys and the other fixes Gemma4 placeholder validation. Not the same underlying bug or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 33, + "component_count": 33, + "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-47104-2", + "nodes": [ + "pull_request:47104", + "pull_request:47115" + ], + "soft_pairs": [ + "pull_request:47104|pull_request:47115" + ], + "prepared_review_unit_hash": "2b8d757150617b14aa9fefd87d239d8bf6cc877b3462ad5447837e389a80584d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6950, + "estimated_input_tokens": 1738, + "estimated_eval_tokens": 3732 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6950, + "estimated_input_tokens": 1738, + "estimated_eval_tokens": 3732 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same processor-test memory reduction work: switching processor tests to tiny Hub repos to lower CI memory usage. They overlap heavily in affected test files and appear to be alternative implementations of the same underlying change.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #47104 is the more comprehensive implementation of the same fix, bundling the processor-test tiny-repo migration with the supporting CI/memory-tracking changes and workflow updates.", + "best_issue_reason": null, + "best_pr_reason": "PR #47104 is the strongest representative of the cluster because it covers the core memory-reduction change plus the supporting infrastructure needed to validate and run it.", + "soft_edge_verdicts": [ + { + "left": "pull_request:47104", + "right": "pull_request:47115", + "accept": true, + "reason": "Same underlying change: both convert processor tests to tiny Hub repos to reduce CI memory, with heavily overlapping test files and matching intent." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/manifest.json b/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..0676c8207d711cd834c6cff283f88d6b697ce6b1 --- /dev/null +++ b/snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260707t060715z", + "artifacts": { + "hybrid": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-07-07T06:19:24Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260707T060715Z", + "variant": "hybrid" +} diff --git a/snapshots/20260707T060715Z/manifest.json b/snapshots/20260707T060715Z/manifest.json index 41843be7795e77bac432fe17f14fec5f5bf2a7c3..9665dac576d9e235bafdf16dfe404c198fdd1b0b 100644 --- a/snapshots/20260707T060715Z/manifest.json +++ b/snapshots/20260707T060715Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 972 }, "extracted_at": "2026-07-07T06:07:15Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260707t060715z", + "runs": { + "hybrid-model-20260707t060715z": { + "analysis_id": "hybrid-model-20260707t060715z", + "artifacts": { + "hybrid": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260707T060715Z/analysis-runs/hybrid-model-20260707t060715z/manifest.json", + "model": null, + "published_at": "2026-07-07T06:19:24Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260707T060715Z", "watermark": {