diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 5c4b1ed8c10f297fdb405a1e1a0e2a95dbd1dfc0..ee69a590363c3ffd68f5a3ddd9aeb1556fe34d53 100644 --- a/analysis/current/analysis-report-hybrid.json +++ b/analysis/current/analysis-report-hybrid.json @@ -1,8 +1,8 @@ { "schema_version": "1.0", "repo": "huggingface/transformers", - "snapshot_id": "20260515T120027Z", - "generated_at": "2026-05-15T12:09:10Z", + "snapshot_id": "20260515T180026Z", + "generated_at": "2026-05-15T18:10:39Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ @@ -1148,204 +1148,6 @@ } ] }, - { - "cluster_id": "cluster-43656-4", - "summary": "Cluster of 1 issues and 3 PRs centered on issue #43824.", - "status": "open", - "confidence": 0.8, - "canonical_issue_number": 43824, - "canonical_pr_number": 43656, - "issue_numbers": [ - 43824 - ], - "pr_numbers": [ - 43656, - 43836, - 43842 - ], - "evidence_types": [ - "closing_reference", - "shared_issue_target", - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 43656, - "right_pr_number": 43836, - "code_similarity": 0.36, - "size_similarity": 0.176, - "file_overlap": 1.0, - "area_overlap": 0.051, - "patch_similarity": 0.048, - "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": [ - [ - 11, - 18 - ], - [ - 359, - 429 - ], - [ - 584, - 590 - ], - [ - 1892, - 1910 - ], - [ - 1917, - 1923 - ] - ] - } - ] - }, - { - "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 - ] - ] - } - ] - }, - { - "left_pr_number": 43836, - "right_pr_number": 43842, - "code_similarity": 0.332, - "size_similarity": 0.088, - "file_overlap": 1.0, - "area_overlap": 0.033, - "patch_similarity": 0.017, - "shared_filenames": [ - "src/transformers/cli/serve.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/cli/serve.py", - "left_ranges": [ - [ - 11, - 18 - ], - [ - 359, - 429 - ], - [ - 584, - 590 - ], - [ - 1892, - 1910 - ], - [ - 1917, - 1923 - ] - ], - "right_ranges": [ - [ - 54, - 61 - ], - [ - 587, - 593 - ] - ] - } - ] - } - ] - }, { "cluster_id": "cluster-43240-3", "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", @@ -1621,48 +1423,67 @@ ] }, { - "cluster_id": "cluster-45853-3", - "summary": "Cluster of 1 issues and 2 PRs centered on issue #45853.", + "cluster_id": "cluster-43824-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43824.", "status": "open", - "confidence": 0.75, - "canonical_issue_number": 45853, - "canonical_pr_number": 45863, + "confidence": 0.55, + "canonical_issue_number": 43824, + "canonical_pr_number": 43836, "issue_numbers": [ - 45853 + 43824 ], "pr_numbers": [ - 45863, - 45986 + 43836, + 43842 ], "evidence_types": [ - "closing_reference", - "shared_issue_target" + "closing_reference" ], "pr_comparisons": [ { - "left_pr_number": 45863, - "right_pr_number": 45986, - "code_similarity": 0.141, - "size_similarity": 0.156, - "file_overlap": 0.333, - "area_overlap": 0.0, - "patch_similarity": 0.066, + "left_pr_number": 43836, + "right_pr_number": 43842, + "code_similarity": 0.332, + "size_similarity": 0.088, + "file_overlap": 1.0, + "area_overlap": 0.033, + "patch_similarity": 0.017, "shared_filenames": [ - "src/transformers/utils/import_utils.py" + "src/transformers/cli/serve.py" ], "shared_file_areas": [ { - "filename": "src/transformers/utils/import_utils.py", + "filename": "src/transformers/cli/serve.py", "left_ranges": [ [ - 1315, - 1321 + 11, + 18 + ], + [ + 359, + 429 + ], + [ + 584, + 590 + ], + [ + 1892, + 1910 + ], + [ + 1917, + 1923 ] ], "right_ranges": [ [ - 1338, - 1345 + 54, + 61 + ], + [ + 587, + 593 ] ] } @@ -1861,281 +1682,6 @@ ] } ] - }, - { - "cluster_id": "cluster-44018-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 44068, - "issue_numbers": [], - "pr_numbers": [ - 44018, - 44068 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 44018, - "right_pr_number": 44068, - "code_similarity": 0.766, - "size_similarity": 0.939, - "file_overlap": 1.0, - "area_overlap": 0.425, - "patch_similarity": 0.866, - "shared_filenames": [ - "src/transformers/models/gpt_neo/modeling_gpt_neo.py", - "tests/models/gpt_neo/test_modeling_gpt_neo.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/models/gpt_neo/modeling_gpt_neo.py", - "left_ranges": [ - [ - 26, - 46 - ], - [ - 138, - 143 - ], - [ - 182, - 187 - ], - [ - 283, - 295 - ], - [ - 327, - 341 - ], - [ - 348, - 354 - ], - [ - 360, - 369 - ], - [ - 399, - 406 - ], - [ - 411, - 419 - ], - [ - 428, - 433 - ], - [ - 472, - 492 - ], - [ - 507, - 513 - ], - [ - 519, - 528 - ], - [ - 541, - 559 - ], - [ - 562, - 567 - ], - [ - 595, - 601 - ], - [ - 607, - 614 - ], - [ - 627, - 643 - ], - [ - 685, - 690 - ], - [ - 708, - 714 - ], - [ - 720, - 727 - ], - [ - 740, - 757 - ], - [ - 761, - 766 - ], - [ - 780, - 786 - ], - [ - 791, - 798 - ], - [ - 807, - 822 - ], - [ - 840, - 845 - ] - ], - "right_ranges": [ - [ - 26, - 31 - ], - [ - 34, - 44 - ], - [ - 136, - 141 - ], - [ - 180, - 185 - ], - [ - 281, - 293 - ], - [ - 325, - 339 - ], - [ - 346, - 352 - ], - [ - 358, - 367 - ], - [ - 397, - 404 - ], - [ - 409, - 417 - ], - [ - 426, - 434 - ], - [ - 463, - 483 - ], - [ - 498, - 504 - ], - [ - 510, - 519 - ], - [ - 532, - 550 - ], - [ - 553, - 564 - ], - [ - 586, - 592 - ], - [ - 598, - 605 - ], - [ - 618, - 634 - ], - [ - 676, - 688 - ], - [ - 699, - 705 - ], - [ - 711, - 718 - ], - [ - 731, - 748 - ], - [ - 752, - 762 - ], - [ - 771, - 777 - ], - [ - 782, - 789 - ], - [ - 798, - 813 - ], - [ - 831, - 836 - ] - ] - }, - { - "filename": "tests/models/gpt_neo/test_modeling_gpt_neo.py", - "left_ranges": [ - [ - 458, - 464 - ] - ], - "right_ranges": [ - [ - 458, - 464 - ] - ] - } - ] - } - ] } ], "duplicate_issues": [], @@ -2167,16 +1713,6 @@ "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-43656-4", - "canonical_pr_number": 43656, - "duplicate_pr_numbers": [ - 43836, - 43842 - ], - "target_issue_number": 43824, - "reason": "PRs in cluster-43656-4 are treated as duplicates because they converge on issue #43824 with closing_reference, shared_issue_target, soft_similarity evidence." - }, { "cluster_id": "cluster-43698-3", "canonical_pr_number": 43779, @@ -2186,6 +1722,15 @@ "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-43824-3", + "canonical_pr_number": 43836, + "duplicate_pr_numbers": [ + 43842 + ], + "target_issue_number": 43824, + "reason": "PRs in cluster-43824-3 are treated as duplicates because they converge on issue #43824 with closing_reference evidence." + }, { "cluster_id": "cluster-43979-11", "canonical_pr_number": 44007, @@ -2203,15 +1748,6 @@ "target_issue_number": 43979, "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." }, - { - "cluster_id": "cluster-44018-2", - "canonical_pr_number": 44068, - "duplicate_pr_numbers": [ - 44018 - ], - "target_issue_number": null, - "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-45081-3", "canonical_pr_number": 45317, @@ -2238,15 +1774,6 @@ ], "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-45853-3", - "canonical_pr_number": 45863, - "duplicate_pr_numbers": [ - 45986 - ], - "target_issue_number": 45853, - "reason": "PRs in cluster-45853-3 are treated as duplicates because they converge on issue #45853 with closing_reference, shared_issue_target evidence." } ], "best_issue": { diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index a520933bbff8491c8237052fecc6554722cdc998..61a620d562187f8c9d10fa17486998da18bc06e4 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": "20260515T120027Z", - "generated_at": "2026-05-15T12:09:10Z", + "snapshot_id": "20260515T180026Z", + "generated_at": "2026-05-15T18:10:39Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -131,409 +131,355 @@ "issue:41093|issue:43638", "issue:43792|issue:45901" ], - "prepared_review_unit_hash": "07fbf3057b486d500f7af51353f393cc719d1228ad7d54daddf8cebcd2f36987", + "prepared_review_unit_hash": "184deb9bdd0407ffb4bdbdf8985a1fe3f9bfb0dc57d12b0b5b8ff8124647d47f", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 62, - "serialized_chars": 46378, - "estimated_input_tokens": 11595, - "estimated_eval_tokens": 23446 + "serialized_chars": 46368, + "estimated_input_tokens": 11592, + "estimated_eval_tokens": 23440 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are only loosely related by subsystem or error shape. I accepted just two true duplicate clusters: the BertModel/DeepSpeed ZeRO-3 index-0 OOB reports, and the AutoTokenizer v4\u2192v5 token-id regression reports. There are no PRs in this set.", - "confidence": 0.91, - "canonical_issue_reason": "issue:45920 is the best canonical issue: it is an open, broad report of the AutoTokenizer wrong-token-IDs regression across multiple model families, so it best anchors the accepted tokenizer duplicate cluster.", + "summary": "The cluster is mostly heterogeneous, with a few clear similarity pockets. The ZeRO-3/BERT items are very similar and likely the same bug family, but the evidence is not strong enough to call 43638 an exact duplicate of 43596; 43596 remains the best anchor for that subgroup. The tokenizer-ID regression pair is a strong duplicate candidate, and the mask-shape IndexError items look like the same family as well.", + "confidence": 0.63, + "canonical_issue_reason": "issue:43596 is the cleanest anchor for the ZeRO-3 empty-tensor/index-out-of-bounds subgroup: it is the most generic BERT + ZeRO-3 report, while 43638 and 41762 read as very similar neighbors rather than proven exact duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the strongest issue candidate overall because it is open, broad, and captures the underlying tokenizer regression more generally than the narrower Granite-only report.", + "best_issue_reason": "issue:43596 is the best overall issue representative for the ZeRO-3 subgroup because it captures the shared failure mode with the broadest framing and sits at the center of the strongest similarity cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different models and failure modes: missing Qwen3_5Config field vs streaming chat/completions failure." + "reason": "Different failures: Qwen3.5 config attribute error vs streaming chat/completions behavior." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Unrelated hardware/backend errors: AMD GPU model load vs MPS histogram op not implemented." + "reason": "AMD GPU load failure and MPS histogram Int backend error are unrelated backend-specific bugs." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different bug: dtype mismatch in image preprocessing/model vs AMD GPU load failure." + "reason": "Image preprocessor dtype mismatch is a different issue from the AMD GPU device-mapping failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Both are dtype-related, but they affect different models and code paths; not the same underlying bug." + "reason": "Autocast dtype mismatch in Phi-MoE is unrelated to the AMD GPU runtime failure." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different MoE/model-specific failures; not the same concrete error path." + "reason": "Different code paths: AMD GPU support failure vs GraniteMoEHybrid calling an invalid method." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Different models and symptoms; no clear shared underlying bug." + "reason": "Qwen NVFP4 runtime error is unrelated to the AMD GPU failure." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "DeepSpeed collective mismatch is unrelated to ModernBERT ONNX export." + "reason": "DeepSpeed collective mismatch and ModernBERT ONNX export are unrelated." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "Both involve model loading/export issues, but these are different models and distinct failures." + "reason": "ModernBERT export failure and Qwen3Moe FP8 failure are different bugs." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "ONNX export error is unrelated to Qwen3 auto device mapping cudaErrorAssert." + "reason": "ONNX export error and Qwen3 auto-device-map cuda assert are unrelated." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are v5 docs cleanups, but they target different removed references and are not the same change." + "reason": "Both are docs cleanup items, but they target different removed features." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ONNX export failure is unrelated to the GenerationMixin import error." + "reason": "ModernBERT export and GenerationMixin import failure are unrelated." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Generic mask-shape mismatch is not the same as Gemma4 PLE device mismatch." + "reason": "Mask/index shape mismatch and Gemma4 PLE device mismatch are different failures." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Different failures: DeepSpeed ZeRO-3 index error vs mask/tensor shape mismatch." + "reason": "ZeRO-3 empty-tensor IndexError and mask/tensor shape mismatch are different bug families." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Different symptoms and code paths; the shared 'IndexError' wording is too broad to treat as duplicate." + "reason": "Different root causes: ZeRO-3 loading failure vs mask-shape IndexError." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Mask-shape IndexError and GPT-OSS tensor-parallel issue are unrelated." + "reason": "Mask-shape IndexError and GPT-OSS TP-only failure are unrelated." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Different model/trainer path; no evidence of the same bug." + "reason": "Unpacking embeddings with TRL SFT and the mask-shape IndexError are different bugs." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice preset bug is unrelated to ModernBERT ONNX export." + "reason": "BarkProcessor voice preset and ModernBERT export are unrelated." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both mention DeepSpeed, but one is a collective mismatch and the other is a BertModel index error." + "reason": "Collective mismatch under ZeRO-3 is a different failure from the empty-tensor index-out-of-bounds bug." }, { "left": "issue:43596", "right": "issue:43638", - "accept": true, - "reason": "Same ZeRO-3/BertModel out-of-bounds index failure; only the wording differs (init/training, pretrained/non-pretrained)." + "accept": false, + "reason": "Very similar ZeRO-3/BERT IndexError reports, but not enough evidence to treat them as an exact duplicate." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger Kernel cross-entropy call issue is unrelated to GPT-OSS tensor-parallel failure." + "reason": "Liger Kernel cross_entropy call and GPT-OSS TP-only failure are unrelated." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Different problems: detokenization regression vs LoRA merge/save/reload mismatch." + "reason": "DeepSeek detokenization and LoRA merge/save mismatch are different bugs." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Unrelated model execution error vs LoRA weight merge inconsistency." + "reason": "Qwen NVFP4 runtime failure and LoRA merge/save mismatch are unrelated." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Activation-checkpointing/FSDP recomputation error is unrelated to the mask-shape IndexError." + "reason": "Activation-checkpointing/FSDP recompute mismatch is unrelated to the mask-shape IndexError." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Both are DeepSpeed-related, but they are different runtime errors on different code paths." + "reason": "DeepSpeed weight-dimension error and collective mismatch are different failures." }, { "left": "issue:45859", "right": "issue:45902", "accept": false, - "reason": "Both involve Qwen3_5MoeTextRotaryEmbedding, but one is CPU offload compatibility and the other is uninitialized-memory materialization; not the same bug." + "reason": "Same model family, but CPU offload incompatibility and meta-to-empty uninitialized-memory are different bugs." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "Different model/setup and different error shape." + "reason": "ZeRO-3 loading failure and GPT-OSS TP-only failure are unrelated." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "FSDP checkpointing tensor-size mismatch is unrelated to the CustomTrainer signature change." + "reason": "FSDP activation checkpointing mismatch and compute_loss signature change are unrelated." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "DeepSpeed/BertModel index error is unrelated to GPT-OSS tensor-parallel behavior." + "reason": "ZeRO-3 loading failure and GPT-OSS TP-only failure are unrelated." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Different model families and failures; not the same underlying bug." + "reason": "Mixed-precision dtype mismatch and GraniteMoEHybrid invalid method are different bugs." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Gemma3 vLLM config attribute error is unrelated to the Mistral tokenizer regex crash." + "reason": "Gemma3 vLLM attribute regression and Mistral regex patch crash are unrelated." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Different Gemma4 issues: device mismatch in forward vs expensive reverse embedding lookup." + "reason": "Gemma4 device-mapping mismatch and reverse embedding lookup are different issues." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Torch version incompatibility is unrelated to Qwen-3-8B-NVFP4 runtime error." + "reason": "Torch version incompatibility and Qwen NVFP4 runtime error are unrelated." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Both are MoE-related, but the actual failures and code paths differ." + "reason": "Autocast dtype mismatch and GPT-OSS TP-only failure are different code paths." }, { "left": "issue:44918", "right": "issue:45161", "accept": false, - "reason": "TRL SFT input-embedding unpacking issue is unrelated to GPT-OSS TP failure." + "reason": "TRL SFT embedding unpacking failure is unrelated to the GPT-OSS TP issue." }, { "left": "issue:43716", "right": "issue:45685", "accept": false, - "reason": "Different model/backend errors; no shared concrete bug." - }, - { - "left": "issue:43541", - "right": "issue:43716", - "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm tracing failure is unrelated to Mistral-3 dtype mismatch." + "reason": "Mixed-precision dtype mismatch and MPS histogram Int backend error are unrelated." }, { "left": "issue:43828", "right": "issue:44292", "accept": false, - "reason": "Both are dtype/model execution issues, but they are different models and distinct failures." - }, - { - "left": "issue:45507", - "right": "issue:45685", - "accept": false, - "reason": "Invalid method call in GraniteMoEHybrid is unrelated to the MPS histogram op issue." - }, - { - "left": "issue:43716", - "right": "issue:44292", - "accept": false, - "reason": "Different model-specific runtime failures." - }, - { - "left": "issue:44292", - "right": "issue:44918", - "accept": false, - "reason": "Different model families and failure modes." - }, - { - "left": "issue:43039", - "right": "issue:44918", - "accept": false, - "reason": "Liger Kernel cross-entropy call vs TRL SFT embedding unpacking are unrelated." - }, - { - "left": "issue:41762", - "right": "issue:43638", - "accept": false, - "reason": "Both mention ZeRO-3 and index errors, but Gemma3 loading and BertModel training are different concrete bugs." - }, - { - "left": "issue:43827", - "right": "issue:44512", - "accept": false, - "reason": "Both are docs regressions, but they refer to different removed v5 artifacts and different pages." - }, - { - "left": "issue:43716", - "right": "issue:44560", - "accept": false, - "reason": "Different model-specific errors: image-preprocessor dtype mismatch vs video StopIteration." - }, - { - "left": "issue:43828", - "right": "issue:45685", - "accept": false, - "reason": "Different code paths and backends; no duplicate evidence." + "reason": "Autocast dtype mismatch and Qwen NVFP4 runtime failure are different bugs." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Video embedding StopIteration and MPS histogram implementation error are unrelated." + "reason": "Video StopIteration in Qwen3-vl-embedding is unrelated to the MPS histogram Int backend bug." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Both are model/runtime issues, but not the same bug or change." + "reason": "Liger Kernel cross_entropy call and autocast dtype mismatch are different issues." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed collective mismatch is unrelated to the trainer signature TypeError." + "reason": "DeepSpeed collective mismatch and CustomTrainer compute_loss signature mismatch are unrelated." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "Activation checkpointing/FSDP error is unrelated to GenerationMixin import failure." + "reason": "Activation-checkpointing/FSDP bug and GenerationMixin import failure are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Liger Kernel cross-entropy invocation is unrelated to mask/tensor shape IndexError." + "reason": "Liger Kernel cross_entropy call and mask-shape IndexError are unrelated." }, { "left": "issue:41762", - "right": "issue:43596", + "right": "issue:43638", "accept": false, - "reason": "Both are ZeRO-3 loading/index issues, but they affect different models and likely different root causes." + "reason": "Strongly similar ZeRO-3 empty-tensor IndexError reports, but the available evidence does not prove they are the same exact bug." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "TextClassificationPipeline docs issue is unrelated to the removed transformers run command docs issue." + "reason": "Docs around return_all_scores and docs around the removed run command are separate cleanup items." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Different models and errors: Gemma3 config attribute missing vs Qwen3 device-mapping assert." + "reason": "Gemma3 vLLM attribute regression and Qwen3 auto-device-map cuda assert are unrelated." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "GPT-OSS AMD GPU failure is unrelated to table-question-answering crash." + "reason": "AMD GPU support failure and table-question-answering crash are different bugs." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Processor failure for Voxtral is unrelated to Gemma4 device mismatch." + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are unrelated." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger Kernel cross-entropy issue is unrelated to Qwen3-vl video StopIteration." + "reason": "Liger Kernel cross_entropy call and Qwen3-vl video StopIteration are unrelated." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Same AutoTokenizer v4\u2192v5 wrong-token-ID regression affecting multiple model families; 45920 is the broader report." + "reason": "Same silent v4\u2192v5 AutoTokenizer wrong-token-ID regression, reported across different model families." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Corrupted Ovis2 checkpoint and NemotronH checkpoint-loading failure are different issues." + "reason": "Ovis2 checkpoint corruption and NemotronH checkpoint loading failure are unrelated." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "Different models and failure types; no shared root cause." + "reason": "Unit-test load failure and checkpoint corruption are not the same underlying bug." }, { "left": "issue:41093", - "right": "issue:43638", - "accept": false, - "reason": "Both are IndexErrors, but mask/tensor mismatch is not the same as the ZeRO-3 BertModel zero-size index error." + "right": "issue:44805", + "accept": true, + "reason": "Same mask-shape IndexError template; likely the same masking/indexing bug family." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper run failure and table-question-answering crash are unrelated." + "reason": "Whisper load failure and table-question-answering crash are unrelated." } ] }, "evaluator_result": { - "accept": true, - "feedback": "Grounded and conservative overall. The two accepted duplicate pairs are well supported by the packet: issue:43596/issue:43638 share the same ZeRO-3/BertModel index-out-of-bounds pattern, and issue:45812/issue:45920 clearly match the same AutoTokenizer v4\u2192v5 regression with 45920 being the broader canonical issue. The remaining pairs are mostly appropriately rejected on differing model, backend, or error-path evidence. No PR pairs are present." + "accept": false, + "feedback": "The cluster summary is mostly reasonable, but it overstates the evidence for the ZeRO-3 subgroup. In the packet, 43596 and 43638 are strongly similar, but there is no explicit proof that 43638 is an exact duplicate of 43596; likewise, calling them the 'same concrete code path' is stronger than the available evidence supports. Please tone these claims down to 'very similar' or 'likely same bug family' rather than exact duplicate." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -605,8 +551,8 @@ "issue:43425|issue:45070", "issue:43596|issue:45800", "issue:43975|issue:45901", - "issue:45362|issue:45464", "issue:43792|issue:45070", + "issue:45362|issue:45464", "issue:44292|issue:45901", "issue:43404|issue:43866", "issue:39290|issue:45823", @@ -626,8 +572,8 @@ "issue:43366|issue:45464", "issue:45020|issue:45800", "issue:43531|issue:45070", - "issue:41093|issue:43596", "issue:43541|issue:44589", + "issue:41093|issue:43596", "issue:43054|issue:43994", "issue:45542|issue:45800", "issue:45759|issue:45902", @@ -640,269 +586,272 @@ "issue:44530|issue:45005", "issue:42915|issue:45399" ], - "prepared_review_unit_hash": "3d8e8814e89a6c3e00f5b875435285b92d3c5b0534bcfba2067a918879de8a30", + "prepared_review_unit_hash": "affc09d5b1fde44ca16ec08628b0d3f70d3f30e30e357457030661e3bd868fcd", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 38, - "serialized_chars": 38764, - "estimated_input_tokens": 9691, - "estimated_eval_tokens": 19638 + "serialized_chars": 38767, + "estimated_input_tokens": 9692, + "estimated_eval_tokens": 19640 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a collection of distinct Transformers regression reports spanning different models, backends, and failure modes. The soft-similarity edges are largely false positives; at most they share broad themes like version compatibility, dtype issues, or loading problems, but not the same concrete bug.", - "confidence": 0.39, - "canonical_issue_reason": "issue:45020 is the broadest umbrella-style report in the set, centered on recent-version breakage for remote-code/custom models, so it is the best anchor for this heterogeneous cluster.", + "summary": "This cluster is mostly a grab bag of distinct model- and runtime-specific issues rather than true duplicates. The only clear soft duplicate here is the remote_code/custom-module loading pair; the rest share broad themes but not the same concrete bug path.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45020 is the broadest open report and the best umbrella for the cluster\u2019s strongest shared theme: recent-version regressions in remote_code model loading. The other issues are narrower or point to different failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the most general and reusable representative among these issues; it can subsume nearby custom-loading/version-regression reports better than the model-specific crashes.", + "best_issue_reason": "issue:45020 is the most suitable representative issue overall because it is broad, open, and framed around a user-visible regression that can subsume related custom-code loading problems.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding window, but they hit different models and different failure modes; not the same bug." + "reason": "Both mention sliding-window behavior, but they involve different model families and different failure modes; not the same bug path." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "One is a torch-version compatibility complaint, the other is a PretrainedConfig/pydantic regression; unrelated." + "reason": "One is a torch-version compatibility report, the other is a pydantic/config regression; too different to merge." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "DeepSpeed ZeRO-3 loading error and a torch/transformers version incompatibility are distinct problems." + "reason": "Both mention incompatibility-ish symptoms, but one is a ZeRO-3 index error and the other is torch/transformers version incompatibility." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Detokenization on DeepSeek Coder and a table-question-answering crash are unrelated code paths." + "reason": "Detokenization bugs and table-question-answering crashes are unrelated code paths and symptoms." }, { - "left": "issue:45362", - "right": "issue:45464", + "left": "issue:43792", + "right": "issue:45070", "accept": false, - "reason": "Both involve Qwen3.5 chat usage, but one is a model crash and the other is a streaming API failure; not clearly the same defect." + "reason": "Whisper runtime failure is not the same underlying issue as a PretrainedConfig/pydantic regression." }, { - "left": "issue:43792", - "right": "issue:45070", + "left": "issue:45362", + "right": "issue:45464", "accept": false, - "reason": "Whisper inference failure and a pydantic PretrainedConfig breakage do not match." + "reason": "Both touch Qwen3.5 chat usage, but one is a general chat crash and the other is a streaming chat/completions failure; different paths." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "Qwen NVFP4 loading/runtime issues are unrelated to table-question-answering crashes." + "reason": "Qwen-3 NVFP4 loading and table QA crashes are unrelated." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Missing tied weights in Mistral3 and a corrupted Ovis2 checkpoint are different root causes." + "reason": "A tied-weight bug in Mistral3 is not the same as a corrupted Ovis2 checkpoint report." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Gemma3 sliding_window_pattern and Gemma4 device_map mismatch are separate model bugs." + "reason": "Gemma3 sliding-window config and Gemma4 device-map mismatch are different model issues." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Qwen3.5 streaming API failure and an MPS int histogram error are unrelated." + "reason": "Streaming inference on Qwen3.5 and an MPS histogram error on MoE models are unrelated." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Mixtral torch-dynamo grouped_mm failure and an MPS backend error are different code paths." + "reason": "Grouped_mm tracing failure and an MPS Int histogram limitation are different runtime problems." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both mention numeric/dtype problems, but they affect different models and operations; not the same bug." + "reason": "Both mention dtype-ish symptoms, but one is torch dynamo tracing and the other is autocast mismatch; not the same bug." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "PaddleOCR-VL processor failure and a template-node compilation error are unrelated." + "reason": "Processor failure for Voxtral and a template-node compilation error do not share a concrete code path." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "Importing GenerationMixin and Gemma3 sliding_window_pattern are separate issues." + "reason": "Importing GenerationMixin and Gemma3 sliding-window config errors are unrelated." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are IndexError-shaped failures, but one is a mask/tensor size mismatch and the other is zero-sized loading with ZeRO-3; not the same bug." + "reason": "Both are shape/index errors, but on different tensors, models, and loading scenarios." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "A torch/transformers version incompatibility is not the same as a Gemma4 device placement mismatch." + "reason": "Torch version incompatibility and Gemma4 device mismatch are different failures." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading failure and GGUF gpt-oss support are different feature areas." + "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are not the same bug." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is a specific config regression, not the pydantic PretrainedConfig field issue." + "reason": "Missing pad_token_idx in StableLmConfig is a different config regression from the pydantic PretrainedConfig breakage." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Int4 memory growth/OOM and tied-weights handling are different problems." + "reason": "Int4 memory/OOM regression and tied-weights translation-model issues are distinct." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Grouped_mm tracing failure in Mixtral and GPT-OSS tensor-parallel issues do not share the same code path." + "reason": "Torch tracing failure and only-TP-not-working on GPT-OSS MoE are different implementation issues." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Phi dtype mismatch under autocast and Qwen3.5 streaming API failures are unrelated." + "reason": "Autocast dtype mismatch for Phi-tiny-MoE is not the same as Qwen3.5 streaming API failure." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "init_empty_weights argument handling and a pydantic PretrainedConfig regression are distinct." + "reason": "init_empty_weights argument handling and pydantic field validation are different breakages." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF architecture support and streaming chat failures on Qwen3.5 are different issues." + "reason": "GGUF architecture support and chat/completions streaming errors are unrelated." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code breakage and torch version incompatibility are only loosely related compatibility reports, not the same bug." + "reason": "remote_code loading regressions and a torch/transformers version incompatibility are too broad and different to treat as one bug." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and a pydantic config-field regression do not match." + "reason": "Qwen3-MoE sliding_window and a pydantic config regression do not share a concrete failure path." }, { - "left": "issue:41093", - "right": "issue:43596", + "left": "issue:43541", + "right": "issue:44589", "accept": false, - "reason": "Both are IndexErrors, but the model context and failure mechanics differ too much to be the same defect." + "reason": "torch dynamo grouped_mm errors and missing Float8 storage are separate float/dtype issues." }, { - "left": "issue:43541", - "right": "issue:44589", + "left": "issue:41093", + "right": "issue:43596", "accept": false, - "reason": "Mixtral grouped_mm tracing and missing Float8 storage are different backend/runtime problems." + "reason": "Both are index errors, but one is a mask-length mismatch and the other is a zero-sized dimension during ZeRO-3 init." }, { "left": "issue:43054", "right": "issue:43994", "accept": false, - "reason": "Both concern SigLIP2 quality, but one is embedding quality and the other is nonsensical AutoModel/pipeline outputs; not enough to treat as the same bug." + "reason": "Both concern SigLIP2 output quality, but one is a relative embedding-quality complaint and the other is a broader bad-results report; too imprecise to confirm same bug." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "TensorBoard-without-TensorFlow backend error and torch/transformers incompatibility are unrelated." + "reason": "TensorBoard-without-TF backend detection and torch/transformers incompatibility are unrelated." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite config unwrapping and uninitialized rotary frequency after to_empty(cuda) are different bugs." + "reason": "Both involve Qwen3.5/3.6 multimodal internals, but config unwrapping and uninitialized-memory materialization are different code paths." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "High perplexity on gpt-oss and a GLM5 issue are not the same underlying problem." + "reason": "High perplexity on gpt-oss and a GLM5 issue are unrelated model-specific reports." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Mistral-3 image-preprocessor dtype mismatch and Qwen3.5 embedding unpacking in TRL are separate failures." + "reason": "Image-preprocessor dtype mismatch and Qwen3.5 input-embedding unpacking in TRL are different failures." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP weight-shape error and Qwen3.5 embedding unpacking are different code paths." + "reason": "FSDP weight dimensionality errors and TRL embedding unpacking are not the same bug." }, { "left": "issue:45632", "right": "issue:45698", - "accept": false, - "reason": "Both involve custom code loading, but one is a cache-path collision and the other is wrong module selection after save/load; not the same concrete bug." + "accept": true, + "reason": "Both describe the same remote_code/custom-module resolution bug: cache/path collision leads to the wrong custom module being loaded after save/load." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Both mention tied weights, but they concern different model families and likely different tie-weight regressions." + "reason": "Mistral3 tied-lm_head weights and translation-model tied-weight regressions are related only at a very broad level, not the same concrete issue." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Custom-model notebook initialization and flash-attn fallback gating are unrelated." + "reason": "Jupyter custom-model initialization and flash-attn fallback checks are unrelated." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache linear_attention support and translation-model tied weights are distinct issues." + "reason": "PagedAttentionCache linear_attention failures and tied-weights translation issues are different subsystems." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "FineGrainedFP8Config failure on Qwen3Moe and flash-attn fallback blocking are different configuration problems." + "reason": "FineGrainedFP8Config failures and flash-attn fallback gating are unrelated runtime paths." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The single accepted soft pair (issue:45632 / issue:45698) is plausibly the same remote_code/custom-module loading bug, and the other rejections are appropriately cautious rather than overbroad. The representative choice of issue:45020 as the open umbrella issue is also defensible from the titles and descriptions provided." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -1002,238 +951,238 @@ "issue:41762|issue:42915", "issue:43782|issue:44960" ], - "prepared_review_unit_hash": "2a861e141f00c9594184fd25c446b41d681a31a55bf395f76f53980e0ccb6dae", + "prepared_review_unit_hash": "99deff697ff1a1ab3d20c37c5f1316f24a69253a2a4b37287d26cbfa9ecfce4b", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 33, - "serialized_chars": 36585, - "estimated_input_tokens": 9147, - "estimated_eval_tokens": 18550 + "serialized_chars": 36584, + "estimated_input_tokens": 9146, + "estimated_eval_tokens": 18548 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a loose mix of unrelated Transformers issues. The only clear duplicate family is the non-persistent-buffer regression; the rest are separate model-, loading-, or environment-specific bugs.", - "confidence": 0.87, - "canonical_issue_reason": "issue:43950 is the clearest umbrella report for the non-persistent-buffer regression, and the nearby reports (43644, 44534, 45902) read like narrower manifestations of the same bug.", + "summary": "Only one soft duplicate looks convincing: the Transformers v5 non-persistent buffer corruption regression. The other title-similar pairs appear to be different model/load-path issues and shouldn\u2019t be merged.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43950 is the broadest, most concrete report of the v5 non-persistent-buffer corruption regression; issue:44534 is the same symptom restated more generally.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 has the strongest, most general title and explicitly frames the problem as a Transformers 5.x regression, making it the best representative issue in this cluster.", + "best_issue_reason": "issue:43950 best represents the underlying bug because it names the regression directly and covers the fix surface more clearly than the narrower duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "Completely different topics: DSA indexer behavior vs docs mentioning a removed command." + "reason": "Different problems: DSA indexer ReLU discussion vs a v5 docs/run-command note." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization/offloading behavior and tied-weights translation-model issues are unrelated bugs." + "reason": "Both mention quantization/tied weights, but the concrete failures are different and not the same bug." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "Model checkpoint loading failure does not match the tied-weights translation-model regression." + "reason": "NemotronH checkpoint loading failure is unrelated to tied-weights behavior in translation models." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A cache-strategy typo and a removed-command docs issue are not the same bug." + "reason": "A typo in cache strategies is not the same issue as stale v5 docs about a removed command." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different model/code paths: TRL SFT embedding unpacking vs a GraniteMoEHybrid method call error." + "reason": "Different model/code paths: Qwen3.5 embedding unpacking vs GraniteMoEHybrid invalid method call." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Detokenization regression and Qwen NVFP4 runtime failure are distinct issues." + "reason": "Different concrete failures: detokenization bug vs NVFP4 runtime error." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Both report the same non-persistent-buffer corruption/junk regression in Transformers 5.x." + "reason": "Same underlying v5 non-persistent buffer corruption regression; 44534 is a reworded symptom of 43950." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "Unrelated model failures with different symptoms and likely different root causes." + "reason": "No shared underlying bug: Qwen-3 NVFP4 execution error vs GraniteMoEHybrid method invocation problem." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "Both mention SigLIP2, but one is a docs/example error and the other is a model quality complaint." + "reason": "SigLIP2 docs-example errors are not the same as a broad embedding-quality regression." }, { "left": "issue:43644", "right": "issue:43950", - "accept": true, - "reason": "Same underlying regression: non-persistent buffers being filled or returned with junk in Transformers 5.x." + "accept": false, + "reason": "Remote-code config field loss is a different failure mode than non-persistent buffer corruption." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Both involve precision/memory symptoms, but they are different models and different failure mechanisms." + "reason": "Autocast dtype mismatch and Qwen3.5 3D position_ids/SDPA issues are different code-path bugs." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Trainer API signature mismatch and DeepSpeed loading IndexError are unrelated." + "reason": "Trainer signature mismatch vs ZeRO-3 load failure; no concrete shared root cause." }, { "left": "issue:41628", "right": "issue:45800", "accept": false, - "reason": "Import/export issue for AutoImageProcessor is unrelated to torch-version compatibility loading problems." + "reason": "Import error for AutoImageProcessor is not the same as a torch/transformers version incompatibility." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "Checkpoint corruption and loading-log noise are not the same underlying bug." + "reason": "Corrupted checkpoint vs generic load noise; not the same underlying bug." }, { "left": "issue:43950", "right": "issue:45902", - "accept": true, - "reason": "45902 appears to be a specific manifestation of the same non-persistent-buffer materialization corruption described in 43950." + "accept": false, + "reason": "Both touch buffer initialization, but 45902 is a specific Qwen3_5Moe rotary-embedding materialization bug, not the same regression as 43950." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Different incompatibilities: bitsandbytes parameter handling vs missing Float8 storage support." + "reason": "Different quantization/backing-storage failures with different symptoms and code paths." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality and tokenization changes are different problems." + "reason": "SigLIP2 embedding quality issue is not the same as tokenizer behavior changing across versions." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error handling for Voxtral is unrelated to a Molmo attribute error." + "reason": "Bad AutoTokenizer error for Voxtral is unrelated to Molmo tied-weights metadata." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Both are SigLIP2-related, but zero probabilities and worse embeddings are not clearly the same defect." + "reason": "Both concern SigLIP2, but zero probabilities and worse embeddings are not clearly the same concrete bug." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "FlashAttention crash from 3D position_ids is unrelated to CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding." + "reason": "FlashAttention illegal access from 3D position_ids is different from CPU-offload incompatibility in rotary embedding." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "A per-layer shared_kv_states KeyError is a different code path from uninitialized rotary-embedding memory." + "reason": "FSDP2 shared_kv_states KeyError is unrelated to uninitialized rotary-embedding memory after materialization." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Different models and errors: Voxtral tokenizer loading vs GLM4V loading failure." + "reason": "AutoTokenizer loading error for Voxtral and glm-4v-9b loading failure are distinct issues." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing with LoRA and int4 quantization OOM are separate memory-related issues." + "reason": "Gradient checkpointing/LoRA ineffectiveness is not the same as int4 quantization causing higher reserved memory." }, { "left": "issue:43054", "right": "issue:43493", "accept": false, - "reason": "SigLIP2 embedding quality complaint and HF-vs-JAX discrepancy are related in theme but not clearly the same bug." + "reason": "Poor SigLIP2 embeddings and HF-vs-JAX discrepancy are related themes but not the same concrete report." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Docs/noise when loading a transformer does not match a specific model loading failure." + "reason": "Generic loading noise is not the same as an inability to load a specific model." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "A specific trust_remote_code AutoConfig field-loss bug is broader than, and not identical to, general remote_code breakage." + "reason": "Remote-code config field loss may be part of a broader class, but 45020 is too umbrella-like to treat as the same bug." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Detokenization behavior and a Python 3.13 flash_attn import KeyError are unrelated." + "reason": "DeepSeek detokenization and flash_attn import errors are unrelated." }, { "left": "issue:43975", "right": "issue:45237", "accept": false, - "reason": "DeepSeek detokenization and AMD GPU support are unrelated issues." + "reason": "Detokenization bug and AMD GPU execution failure are different problems." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "A generic loading-noise complaint is not the same as a model-specific loading failure." + "reason": "Generic load noise is not the same as a specific model failing to load." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "Both mention memory, but LoRA checkpointing inefficiency and Qwen3 MoE training memory usage are different problems." + "reason": "Both mention memory, but one is checkpointing/LoRA behavior and the other is MoE training memory usage." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Different models and failures: GLM4V loading failure vs Molmo missing all_tied_weights_keys." + "reason": "Model loading failure and missing all_tied_weights_keys are different issues." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Gemma3 ZeRO-3 loading IndexError and Qwen3 MoE FP8 failure are unrelated." + "reason": "Gemma3 ZeRO-3 loading error and Qwen3Moe FP8 failure are unrelated." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL weight_only loading error and GLM5 are unrelated." + "reason": "Qwen3VL weight_only load error is not the same as a GLM5 issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analyst output is grounded in the packet and stays conservative on the soft-edge pairs. The non-persistent-buffer regression family is the only clearly supported duplicate cluster, and the other pairings are mostly reasonable rejections based on distinct titles and bug surfaces." + "feedback": "Grounded and conservative overall. The one accepted soft duplicate pair (issue:43950 / issue:44534) is well supported by the titles and clearly points to the same Transformers v5 non-persistent-buffer corruption regression. The remaining pairs are appropriately rejected based on differing bug surfaces, and the summary does not overstate the evidence." }, "evaluator_used": true, "retried": false, @@ -1317,17 +1266,17 @@ "issue:43404|issue:43950", "issue:43742|issue:45399", "issue:42915|issue:45020", - "issue:43257|issue:43866", "issue:43638|issue:45823", + "issue:43257|issue:43866", "issue:43792|issue:44220", "issue:43872|issue:44291", - "issue:43940|issue:45313", "issue:42915|issue:43872", + "issue:43940|issue:45313", "issue:41553|issue:44991", "issue:43854|issue:44863", "issue:36331|issue:41093", - "issue:42947|issue:45923", "issue:43257|issue:45305", + "issue:42947|issue:45923", "issue:44928|issue:45588", "issue:41093|issue:44805", "issue:43872|issue:45538", @@ -1338,238 +1287,238 @@ "issue:45325|issue:45902", "issue:43366|issue:43828" ], - "prepared_review_unit_hash": "2f0ff435e57b6ca7fe6d0425da9e5fbf7d1e469a01a9d47d5510d4183333fd4c", + "prepared_review_unit_hash": "335d8be8f0c5c2f691e3e1e26decf1fc959508aa0dd29081abbce0ae0be141f0", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 38065, - "estimated_input_tokens": 9517, - "estimated_eval_tokens": 19290 + "serialized_chars": 38069, + "estimated_input_tokens": 9518, + "estimated_eval_tokens": 19292 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are unrelated model-specific regressions; only a few soft edges look like true duplicates, mainly exact error matches or the same docs cleanup.", - "confidence": 0.76, - "canonical_issue_reason": "issue:45020 is the broadest umbrella-style report in the set and best matches the recurring theme of recent-version compatibility regressions across model loading paths.", + "summary": "This set is mostly a collection of unrelated model-specific regressions and docs issues. Only a few soft pairs look like true duplicates: the removed-pipeline docs references, the `_is_hf_initialized` TypeError, and the repeated mask-shape IndexError.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 45020 is the broadest and most representative umbrella bug in the set: recent transformers versions breaking `remote_code` models across multiple concrete failures. It is a better anchor than the narrower model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the most representative issue overall: it is high-signal, broadly scoped, and can serve as the anchor for adjacent model-loading breakages.", + "best_issue_reason": "Issue 45020 is the best global issue candidate because it captures the widest recurring regression pattern and is the most suitable umbrella triage target among mostly narrow one-off bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "Different failures: MPS int-histogram support vs Gemma4 device placement mismatch." + "reason": "Different failures: an MPS histogram dtype limitation in MoE code vs a Gemma4 PLE forward device-mismatch bug." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Related Qwen3-MoE naming, but one is a LoRA compatibility break and the other is a config warning about tied embeddings." + "reason": "Both mention Qwen3.5/Qwen3 MoE, but one is a loadability regression across hf4.x/hf5.x and the other is only a warning about `tie_word_embeddings`." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Different bugs: missing Qwen2.5-VL import vs bitsandbytes/_is_hf_initialized incompatibility." + "reason": "Import error for a missing Qwen2.5-VL class is unrelated to the bitsandbytes `_is_hf_initialized` constructor incompatibility." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Broad remote_code regression vs a specific Gemma4 forward device mismatch; not the same bug." + "reason": "`remote_code` breakage is a broad loader regression, while Gemma4 PLE device placement is a model-specific forward-path issue." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Tokenizer-mapping edge case vs remote_code loading regressions; different code paths." + "reason": "Tokenizer-mapping edge case for `add-new-model-like` is a different code path from remote-code model loading failures." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading failure vs flash-attn fallback gating; unrelated issues." + "reason": "Tokenizer loading for `est-roberta` is unrelated to flash-attn fallback gating checks." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Different symptoms and subsystems: autocast dtype mismatch vs device-map forward mismatch." + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is not the same bug as Gemma4 PLE device mismatch." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Tokenizer regex patch crash vs flash-attn fallback logic; not the same underlying bug." + "reason": "Mistral regex patch crashing in tokenizer setup is separate from flash-attn fallback being blocked by other checks." }, { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "Same docs regression: both report obsolete pipeline-task references after v5 removals." + "reason": "Both report the same doc cleanup needed after v5 pipeline removals; same outdated pipeline-task references." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Generic remote_code compatibility break vs a specific Kimi-K2.5 tokenizer regression." + "reason": "General remote_code regressions are not the same as Kimi-K2.5 tokenizer codec-handling and warning behavior." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Weight tying bug in Mistral3 vs from_pretrained buffer corruption; different mechanisms." + "reason": "Untied `lm_head` weights in Mistral3 are unrelated to non-persistent buffer corruption in `from_pretrained()`." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "Model load/key error vs flash-attn fallback gating; not the same issue." + "reason": "A model-specific key error while loading MobileLLM-125M is different from flash-attn fallback selection logic." }, { "left": "issue:42915", "right": "issue:45020", "accept": false, - "reason": "LoRA model compatibility across versions vs broad remote_code breakage; different problems." + "reason": "LoRA compatibility across hf4.x/hf5.x is a specific Qwen3 MoE issue, not the broader remote_code regression." }, { - "left": "issue:43257", - "right": "issue:43866", + "left": "issue:43638", + "right": "issue:45823", "accept": false, - "reason": "DeepSpeed weight conversion failure vs corrupted Ovis2 checkpoint; different failure modes." + "reason": "DeepSpeed ZeRO-3 indexing failure on a non-pretrained Bert model is unrelated to Gemma4 PLE device placement." }, { - "left": "issue:43638", - "right": "issue:45823", + "left": "issue:43257", + "right": "issue:43866", "accept": false, - "reason": "Zero3 index-out-of-bounds on Bert vs Gemma4 device mismatch; unrelated." + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed and a corrupted Ovis2 checkpoint are distinct load failures." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Whisper runtime failure is too broad to equate to the specific fbank-feature helper issue." + "reason": "Whisper not running is too broad to merge with a specific `_torch_extract_fbank_features()` bug without clearer same-path evidence." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same `_is_hf_initialized` / init_empty_weights incompatibility, just described from different entry points." + "reason": "Both center on the same `_is_hf_initialized` TypeError during model materialization/loading, just from different entry points." }, { - "left": "issue:43940", - "right": "issue:45313", + "left": "issue:42915", + "right": "issue:43872", "accept": false, - "reason": "Both are ZeRO-3 weight-loading failures, but for different Qwen model variants and reported at different submodules." + "reason": "Qwen3 MoE LoRA compatibility and the `_is_hf_initialized` constructor error are different regressions." }, { - "left": "issue:42915", - "right": "issue:43872", + "left": "issue:43940", + "right": "issue:45313", "accept": false, - "reason": "LoRA/version incompatibility vs bitsandbytes init argument error; different code paths." + "reason": "Both are ZeRO-3 load problems on Qwen variants, but they appear to be model-specific missing-weight failures rather than one concrete duplicate bug." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Tokenizer error messaging for Voxtral vs tokenizer loading failure for EMBEDDIA/est-roberta; unrelated." + "reason": "A bad AutoTokenizer error message for Voxtral is not the same as failing to load EMBEDDIA/est-roberta." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Different model-load failures with different causes; no shared concrete bug is evident." + "reason": "GLM-4.7-Flash unit-test loading failure and NemotronH checkpoint loading failure are separate model-specific loader bugs." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Different TypeErrors/IndexErrors with unrelated triggers." + "reason": "Unexpected `num_items_in_batch` in custom trainer loss is unrelated to the mask/tensor shape mismatch IndexError." }, { - "left": "issue:42947", - "right": "issue:45923", + "left": "issue:43257", + "right": "issue:45305", "accept": false, - "reason": "Gradient checkpointing symptom vs missing supports_gradient_checkpointing flag; not the same bug." + "reason": "Qwen3 MoE weight conversion with deepspeed is not the same as gradient averaging missing under GAS/model_accepts_loss_kwargs." }, { - "left": "issue:43257", - "right": "issue:45305", + "left": "issue:42947", + "right": "issue:45923", "accept": false, - "reason": "DeepSpeed weight conversion bug vs gradient averaging bug; different training failures." + "reason": "LoRA gradient checkpointing being ineffective is not the same issue as a missing `supports_gradient_checkpointing` flag on a trust_remote_code model." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "RLHF NaN/SDPA fallback problem vs flash-attn auxiliary-state AttributeError; different failure modes." + "reason": "RLHF NaN explosions from Qwen3.5 SDPA/BF16 behavior are unrelated to flash_attention.py crashing on `s_aux=None`." }, { "left": "issue:41093", "right": "issue:44805", "accept": true, - "reason": "Same mask-vs-tensor shape mismatch IndexError; titles differ only in the concrete dimensions shown." + "reason": "These are the same mask-shape mismatch IndexError, with only the reported sizes/context differing." }, { "left": "issue:43872", "right": "issue:45538", "accept": false, - "reason": "bitsandbytes init-arg incompatibility vs CLIPTokenizer max-length bug; unrelated." + "reason": "The `_is_hf_initialized` loading bug is unrelated to CLIPTokenizer's `model_max_length` sentinel value." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Both are tokenizer/model loading complaints, but for different models and likely different root causes." + "reason": "Both are tokenizer-loading complaints, but the Voxtral error message issue and the sleng-bert load failure are not the same bug." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "PagedAttentionCache linear_attention crash vs NemotronH checkpoint loading failure; different bugs." + "reason": "Qwen3.5 PagedAttentionCache group-type handling and NemotronH checkpoint loading are different subsystems and failures." }, { "left": "issue:44479", "right": "issue:44560", - "accept": true, - "reason": "Both describe the same v5.3.0 Qwen video-input regression; one is the broader family report, the other a specific model manifestation." + "accept": false, + "reason": "Both are video-related, but one is a broad v5.3.0 regression across several Qwen models and the other is a specific StopIteration on Qwen3-vl-embedding." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed training weight-shape runtime error vs image processor void-segmentation-map failure; unrelated." + "reason": "DeepSpeed training weight-dimension error and void segmentation-map processing are unrelated." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both touch Qwen rotary/position-id internals, but one is a video rope-scaling regression and the other is uninitialized-memory after to_empty()." + "reason": "Qwen2.5-VL rope-index scaling and Qwen3.5 MoE rotary-embedding uninitialized-memory issues are distinct model internals." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "GGUF gpt-oss architecture support vs Phi-tiny-MoE autocast dtype mismatch; different issues." + "reason": "GGUF gpt-oss architecture support and Phi-tiny-MoE autocast dtype mismatch are separate problems." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analysis is grounded in the packet: the canonical issue choice is defensible, the summary stays conservative, and the accepted soft edges are supported by the issue titles (docs cleanup, identical `_is_hf_initialized` incompatibility, matching mask-shape IndexErrors, and the Qwen video regression family). Rejected pairs also read as appropriately cautious rather than overstated." + "feedback": "The summary is well grounded in the packet, and the soft-edge decisions are conservative. The accepted pairs are plausible same-bug duplicates/docs-cleanup matches, while the rejected pairs avoid overclaiming broader similarity. The choice of issue 45020 as the umbrella issue is also supported by its broad \u2018remote_code\u2019 regression scope." }, "evaluator_used": true, "retried": false, @@ -1673,14 +1622,14 @@ "issue:43708|issue:44075", "issue:43854|issue:43931" ], - "prepared_review_unit_hash": "7b3da719f3300b3ad40bbef27e9ce16301be60a882346540ef0e0dfe6c5e26d4", + "prepared_review_unit_hash": "ef2ac56870a0ff8745e39c714ac80192bfb5c59702f7e9a3cc3030108184900d", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, @@ -1695,195 +1644,207 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a Transformers v5 regression cluster, with a few strong same-bug pairs around buffer corruption and some closely related model-specific failures; I kept the duplicate language conservative where the packet only shows similarity, not proof of identity.", - "confidence": 0.79, - "canonical_issue_reason": "issue:43644 is the clearest representative of the non-persistent-buffer regression cluster: its title is concise and specific, and issue:44534 repeats essentially the same symptom wording, so 43644 is a good anchor without claiming an exact duplicate.", + "summary": "Mostly a mixed cluster of unrelated Transformers regressions, with a few clear duplicate pairs around the same bug family. The strongest duplicate subgroup is the non-persistent buffer corruption in Transformers 5.x.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 43950 is the clearest, most technical statement of the broad 5.x regression and covers the same non-persistent-buffer corruption described in nearby duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43644 is the best overall issue anchor because it is the most concise and generalizable report of the Transformers v5 non-persistent-buffer bug; nearby reports read as related variants or follow-ups rather than a separate class.", + "best_issue_reason": "Issue 43950 is the best canonical issue because it names the exact failing behavior, has broad applicability, and matches the duplicate subgroup more precisely than the shorter title variants.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Same model family only; one is an FSDP2 KeyError, the other is a device-map mismatch. Different code paths." + "reason": "Both are Gemma4 issues, but one is an FSDP2 shared_kv_states bug and the other is a device_map=\"auto\" PLE device-mismatch bug; different code paths." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Different models and failures: AudioFlamingo3 track leakage vs Whisper empty-transcription generation." + "reason": "Different models and failures: AudioFlamingo3 track leakage vs Whisper empty-transcription handling." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "Both are Qwen/LLM-adjacent, but one is GGUF architecture support and the other is a sliding-window bug." + "reason": "Unrelated problems: GGUF gpt-oss support versus Qwen3-MoE sliding_window behavior." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Both are loading problems, but one is non-persistent buffers and the other is long OCR index handling." + "reason": "Different subsystems entirely: non-persistent buffer corruption vs Kosmos2.5 long-OCR index error." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Different models and distinct failure modes: torch.compile/SDPA vs flash-attention illegal memory access." + "reason": "Both touch attention/compile paths, but the model, trigger, and failure mode are different; not the same concrete bug." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "ImportError for a specific class vs meta-device loading failures; no clear shared bug." + "reason": "Import/export issue for Qwen2.5-VL class vs meta-device loading regression; not the same underlying defect." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Very likely the same custom-model initialization regression, with one report adding Jupyter as the environment." + "reason": "Same custom-model initialization regression in Transformers 5.0.0, with nearly identical symptoms and scope." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "Unrelated issues: GGUF architecture support vs missing Float8 storage during load." + "reason": "GGUF gpt-oss support and Float8 storage lookup errors are unrelated." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Different regressions: Qwen3Moe FP8 failure vs generic meta-device model loading breakage." + "reason": "Qwen3Moe FP8 failure and meta-device loading are different bugs." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Both mention Qwen3.x and embeddings, but the concrete failures are different (video StopIteration vs TRL SFT unpacking)." + "reason": "Both involve Qwen3.5-family features, but one is video/StopIteration and the other is TRL embedding unpacking; not the same bug." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "3d_parallel runtime failure and GGUF gpt-oss support are unrelated." + "reason": "3d_parallel.py runtime failure is unrelated to GGUF architecture support." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "Import path breakage for GenerationMixin vs Qwen3Moe FP8 runtime failure." + "reason": "Importing GenerationMixin from transformers.generation is unrelated to Qwen3Moe FP8 behavior." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer-mapping edge case and Gemma4 device mismatch are different bugs." + "reason": "Tokenizer-mapping/model-registration failure and Gemma4 device mismatch are different issues." }, { "left": "issue:38175", "right": "issue:43994", "accept": true, - "reason": "Same SigLIP2 base model and overlapping incorrect-output symptom; this looks like a closely related same-model regression, though the packet does not prove an exact duplicate." + "reason": "Same SigLIP2 model producing bad outputs on the same checkpoint, with closely aligned symptoms." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "General _is_hf_initialized regression vs Kosmos2.5 long-input index error; no shared code path is evident." + "reason": "_is_hf_initialized breakage is unrelated to Kosmos2.5 OCR index errors." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Both are tensor-shape failures, but the affected models and mechanisms differ." + "reason": "Different failure modes and model contexts; no shared concrete code path." }, { "left": "issue:44877", "right": "issue:45030", - "accept": true, - "reason": "Both point to overly strict config validation blocking model loading; likely the same root cause." + "accept": false, + "reason": "Both are config-validation complaints, but they affect different model families and appear to be separate regressions." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "Different failures: DeepSpeed ZeRO-3 on non-pretrained BertModel vs add-new-model-like/tokenizer mapping behavior." + "reason": "DeepSpeed ZeRO-3 Bert initialization and add-new-model-like tokenizer mapping are unrelated." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "IndexError during ZeRO-3 init is unrelated to TOKENIZER_MAPPING_NAMES handling." + "reason": "Zero3/BertModel index error is a different path from the tokenizer-mapping registration bug." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer-mapping registration bug vs mask/tensor shape mismatch; no duplicate evidence." + "reason": "Tokenizer-mapping registration and mask-shape IndexError are not the same bug." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "Both involve loading/offload, but one is dequantized model offload and the other is mmap OOM on Strix Halo." + "reason": "Both relate to loading/memory behavior, but one is gpu+cpu offload dequantized loading and the other is Strix Halo mmap OOM; too different to merge." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Very close same-symptom Transformers v5 non-persistent-buffer report; the titles are nearly identical, but the metadata alone doesn't prove an exact duplicate." + "reason": "These are the same Transformers v5 non-persistent-buffer corruption regression, just phrased differently." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Same model family, but one is bad outputs and the other is a documentation example with multiple separate errors." + "reason": "Same model family, but the reported problems differ: wrong probabilities vs documentation example/model-processor mismatch and quantization failures." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Different GLM/Qwen3.5 generation issues with no evidence of a shared bug." + "reason": "GLM5 and Qwen3.5 flash-attention failures are unrelated." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "LoRA/Qwen3_moe incompatibility and BitNet packed-weight loading are unrelated." + "reason": "Qwen3-MoE LoRA compatibility on hf5.x and BitNet packed-weight loading are different bugs." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "Same model family, but one is ZeRO-3 weight loading and the other is save_pretrained visual-encoder key regression." + "reason": "Both are Qwen3.5 issues, but one is ZeRO-3 weight loading and the other is save_pretrained key serialization; distinct code paths." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Qwen3 auto device-map CUDA assert and Qwen3Moe FP8 failure are different problems." + "reason": "Auto device mapping cudaErrorAssert and FineGrainedFP8Config failure are different Qwen3-family bugs." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading failures, but the specific models differ and the bug mechanism is not shown to be the same." + "reason": "Both are model-load failures for different BERT checkpoints, but the available evidence only shows a broad symptom match, not a concrete shared root cause." }, { - "left": "issue:43854", - "right": "issue:43931", + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "LoRA hf5.x compatibility and Gemma4 resize_token_embeddings behavior are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Strix Halo mmap OOM are separate issues." + }, + { + "left": "issue:43708", + "right": "issue:44075", "accept": false, - "reason": "Distinct model-loading errors: unit-test failure for GLM-4.7-Flash vs Qwen3-VL weight-shape mismatch." + "reason": "Trainer resume/max_steps calculation and SGD argument handling are unrelated." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster summary is broadly grounded, but a few claims are too strong for the packet. In particular, calling issue:44534 an \"exact duplicate\" of issue:43644 is not supported by the metadata shown, and \"reproducible\" is also not evidenced in the packet. The accepted soft pair issue:38175/issue:43994 is similarly a bit strong for the available evidence; it should be softened to something like \"closely related same-model regression\" rather than \"same underlying regression.\" Please keep the canonical/best-issue wording more conservative and based only on the observed similarity and titles." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative about the mixed cluster. The canonical issue choice is well supported by the non-persistent-buffer regression subgroup, and the accepted soft-edge pairs all look like plausible same-bug duplicates/mergeable fix targets. No obvious overstatement in the negative verdicts." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1964,16 +1925,16 @@ "issue:42491|issue:45292", "issue:45276|issue:45335", "issue:44164|issue:44291", - "issue:43404|issue:43454", "issue:36010|issue:39404", + "issue:43404|issue:43454", "issue:41720|issue:44155", "issue:44402|issue:44488", "issue:39692|issue:44960", "issue:43425|issue:43957", "issue:42947|issue:45704", "issue:45663|issue:45759", - "issue:43873|issue:44387", "issue:43335|issue:45663", + "issue:43873|issue:44387", "issue:45663|issue:45859", "issue:44291|issue:44756", "issue:44155|issue:45823", @@ -1985,239 +1946,242 @@ "issue:45276|issue:45292", "issue:45684|issue:45698" ], - "prepared_review_unit_hash": "ed0cb8e771db94991ca45cc6934afc841c9f5f17403d8587a07ac9f60a2542e6", + "prepared_review_unit_hash": "91a222db6499e3e33eaded5c536a325a04a9e077cc97554153984314deec5032", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37767, - "estimated_input_tokens": 9442, - "estimated_eval_tokens": 19140 + "serialized_chars": 37771, + "estimated_input_tokens": 9443, + "estimated_eval_tokens": 19142 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: tokenizer regressions, quantization/offload issues, save/load bugs, and several model-specific generation/loading failures. A few pairs are superficially similar, especially the resize_token_embeddings reports, but none are safe duplicate merges on the evidence provided.", - "confidence": 0.84, - "canonical_issue_reason": "No single canonical issue is safe here; the cluster spans multiple unrelated code paths and model families, so there isn\u2019t a trustworthy umbrella duplicate target.", - "canonical_pr_reason": "No PRs in the cluster.", - "best_issue_reason": "Issue 45276 is the strongest local representative of the only weak mini-cluster (resize_token_embeddings propagation), because it is specific and actionable, but it still does not safely subsume the broader cluster.", - "best_pr_reason": "No PRs in the cluster.", + "summary": "Mostly a loose collection of unrelated Transformers bug reports spanning model loading, tokenization, quantization, and model-specific regressions. Only one clear duplicate pair stands out: the resize_token_embeddings/output_embeddings bug.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45292 is the broader duplicate target: it describes the generic resize_token_embeddings/output_embeddings failure, while issue:45276 is a Gemma4-specific manifestation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the best representative because it is the most general statement of the duplicate resize_token_embeddings bug and subsumes the narrower Gemma4 report.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and failure modes; not the same bug." + "reason": "Both are tokenizer regressions, but one is DeepSeek/LlamaTokenizer pre-tokenizer behavior and the other is Granite token IDs; different failures and likely different fixes." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2, but one is a docs/example mismatch and the other is a model/pipeline correctness issue." + "reason": "Both involve SigLIP2, but one is a docs/example mismatch with quantization errors and the other is incorrect inference results in AutoModel/pipeline." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Different models and different loading paths; no sign of a shared underlying defect." + "reason": "Different models and different loading problems: Qwen3 MoE weight conversion under accelerate/deepspeed vs GLM-4.7-Flash loading in unit tests." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer-loading problems, but they hit different checkpoints and likely different model-specific fixes." + "reason": "Both are tokenizer loading failures, but they concern different models and no shared concrete bug is evident." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "Both involve Qwen3, but one is LoRA/hf4.x compatibility and the other is a Qwen3-VL weight-shape mismatch." + "reason": "Unrelated issues: Qwen3 MoE LoRA compatibility across branches vs Qwen3-VL weight shape mismatch." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is quantization/offloading behavior." + "reason": "Both mention memory, but one is Qwen3 MoE training inefficiency and the other is quantization/offloading behavior." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "Both are Transformers 5.0 regressions, but they describe unrelated bugs in different subsystems." + "reason": "Both are Transformers 5.0 regressions, but one is junk-filled non-persistent buffers and the other is custom-model initialization in Jupyter." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different models and different failures: chat crashing for Qwen3.5 versus a Gemma4 processor attribute error." + "reason": "Different model families and different runtime failures: Qwen3.5 chat crash vs Gemma4Processor missing _tokenizer in serve." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both touch memory usage, but one is training inefficiency and the other is int4 quantization causing OOM." + "reason": "Both are memory-related, but the causes differ: MoE training inefficiency vs int4 quantization reserved-memory OOM." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Both involve flash attention, but the concrete bug and trigger are different." + "reason": "Different flash-attention bugs: Qwen3.5 position_ids misinterpretation vs s_aux=None AttributeError on sink-less models." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "LoRA/branch compatibility and resize_token_embeddings propagation are unrelated bugs." + "reason": "LoRA branch compatibility is unrelated to the resize_token_embeddings/output_embeddings bug." }, { "left": "issue:45276", "right": "issue:45335", "accept": false, - "reason": "These are similar at a high level, but they affect different model families and different embedding members, so not a safe duplicate." + "reason": "Both are resize_token_embeddings issues, but they hit different model implementations and different target modules, so the underlying bug is not clearly the same." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "One is extra_state handling in save/from_pretrained, the other is init_empty_weights passing an unexpected argument." + "reason": "Different save/load paths: extra_state handling vs init_empty_weights argument mismatch." }, { - "left": "issue:43404", - "right": "issue:43454", + "left": "issue:36010", + "right": "issue:39404", "accept": false, - "reason": "Both are weight-tying complaints, but they concern different models and likely distinct implementation paths." + "reason": "Unrelated import error vs Whisper pipeline return_language regression." }, { - "left": "issue:36010", - "right": "issue:39404", + "left": "issue:43404", + "right": "issue:43454", "accept": false, - "reason": "An import error for GenerationMixin is unrelated to Whisper return_language pipeline behavior." + "reason": "Both are weight-tying bugs, but they are reported for different multimodal model classes and are not clearly the same concrete code-path." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Different models and unrelated runtime failures; no common code-path bug." + "reason": "Different subsystems and symptoms: Qwen3 device-map CUDA assert vs AudioFlamingo3 batched inference token leak." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Tokenizer vocab-size mismatch versus a model loading failure; not the same issue." + "reason": "One is a tokenizer vocab-size mismatch, the other is a tokenizer load failure; related theme but not the same bug." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Both mention model examples, but the affected models and failures are entirely different." + "reason": "Both touch model families with similar naming, but the bugs are about docs/example errors vs a GLM5-specific issue." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "One is a Torch version compatibility complaint, the other is a meta-device loading regression." + "reason": "Different issues: Torch 2.10 compatibility vs meta-device model loading regressions." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing/LoRA and an apex RMSNorm memory leak are unrelated." + "reason": "Gradient checkpointing with LoRA is unrelated to the apex.FusedRMSNorm memory leak in T5." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Both involve composite models, but one is an FSDP2 KeyError and the other is AutoModel config unwrapping." + "reason": "Different bugs: Gemma4 FSDP2 shared_kv_states failure vs Qwen 3.5/3.6 multimodal config unwrapping." }, { - "left": "issue:43873", - "right": "issue:44387", + "left": "issue:43335", + "right": "issue:45663", "accept": false, - "reason": "Both are memory/quantization themed, but the symptoms and code paths differ." + "reason": "SwitchTransformers sparse-layer configuration is unrelated to Gemma4 FSDP2 attention state handling." }, { - "left": "issue:43335", - "right": "issue:45663", + "left": "issue:43873", + "right": "issue:44387", "accept": false, - "reason": "SwitchTransformers sparse-layer construction is unrelated to Gemma4 FSDP2 shared_kv_states rebuilding." + "reason": "Different memory regressions with different triggers: quantization offloading vs int4 CUDA reserved-memory growth." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Both are model-internal bugs, but one is Gemma4 FSDP2 and the other is Qwen3 MoE CPU offload." + "reason": "Gemma4 FSDP2 attention state bug is unrelated to Qwen3.5 MoE CPU offload incompatibility." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "Meta-initialization argument handling and Strix Halo mmap OOM are unrelated." + "reason": "TypeError from init_empty_weights is unrelated to disabling mmap for Strix Halo OOM." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "Different model families and different runtime errors; no duplicate evidence." + "reason": "AudioFlamingo3 track leakage and Gemma4 PLE device mismatch are different code paths." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Both are quantization load issues, but they involve different models and quantization formats." + "reason": "Different quantization problems: NVFP4 loading error vs MXFP4 fallback to bf16." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Int4 quantization OOM and mmap-related OOM are different underlying problems." + "reason": "Both are OOM-related, but one is int4 quantization memory growth and the other is mmap-specific Strix Halo OOM." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "A model loading/test failure and a save_pretrained regression are different bugs." + "reason": "Different regressions: GLM-4.7-Flash loading in tests vs Qwen3.5 save_pretrained checkpoint corruption." }, { "left": "issue:44743", "right": "issue:44960", "accept": false, - "reason": "Recurrent-state caching in modular_qwen3_5 is unrelated to GLM5." + "reason": "Qwen3 modular recurrent-state caching bug is unrelated to the generic GLM5 issue." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading regressions are unrelated to disabling mmap for Strix Halo." + "reason": "Meta-device loading failures and Strix Halo mmap OOM are unrelated." }, { "left": "issue:45276", "right": "issue:45292", - "accept": false, - "reason": "Both are resize_token_embeddings regressions, but the affected outputs/components differ and there is no evidence of one shared code-path bug." + "accept": true, + "reason": "Same underlying resize_token_embeddings bug: both report output embeddings not being updated, with 45276 being the Gemma4-specific reproduction and 45292 the broader generic report." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Both concern save_pretrained/from_pretrained with custom modules, but one is permissions propagation and the other is wrong module loading." + "reason": "Both concern save/load of custom models, but one is read-only file permissions from save_pretrained and the other is loading the wrong custom module after save_pretrained." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft-edge pair, issue:45276 vs issue:45292, is well supported by the nearly identical resize_token_embeddings/output_embeddings wording. The other pairwise rejections are cautious and the narrative does not overclaim broader duplication." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -2292,8 +2256,8 @@ "issue:45588|issue:45663", "issue:45292|issue:45335", "issue:43645|issue:44164", - "issue:43257|issue:44863", "issue:44402|issue:44451", + "issue:43257|issue:44863", "issue:43828|issue:43957", "issue:43425|issue:43828", "issue:44589|issue:44756", @@ -2305,8 +2269,8 @@ "issue:43611|issue:43645", "issue:43425|issue:43716", "issue:44530|issue:45084", - "issue:43901|issue:44509", "issue:39404|issue:44912", + "issue:43901|issue:44509", "issue:44265|issue:44898", "issue:44479|issue:44805", "issue:43994|issue:44079", @@ -2319,20 +2283,20 @@ "issue:43596|issue:44683", "issue:44671|issue:44991" ], - "prepared_review_unit_hash": "9a838c87bdddb78bb342934dea77559ef8e1548a93a3b8429eed72f4262300ea", + "prepared_review_unit_hash": "d0f3842138cdd2f3be79f405fd829a1a71bf305138a745cd48aa01b7914effe0", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37971, + "serialized_chars": 37972, "estimated_input_tokens": 9493, "estimated_eval_tokens": 19242 }, @@ -2341,222 +2305,222 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issues with one likely duplicate pair around resize_token_embeddings behavior. No PRs were provided. Best representative issue is the broader embedding-resize report.", - "confidence": 0.78, - "canonical_issue_reason": "issue:45292 is the broadest and clearest representative of the resize_token_embeddings propagation bug, and issue:45335 looks like the same defect in the t5gemma path.", + "summary": "This cluster is dominated by broad but mostly unrelated Transformers regressions (model loading, dtype/device mismatches, docs, training edge cases). Only one pair looks like a true duplicate: the resize_token_embeddings propagation bug.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45292 is the broader, model-agnostic report of the resize_token_embeddings bug, while issue:45335 is the T5Gemma-specific manifestation.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 has the most general title and would best anchor a fix that also covers the narrower t5gemma-specific report.", + "best_issue_reason": "issue:45292 is the clearest and most general representative of the one confirmed duplicate pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Both mention training/gradient topics, but one is eval-mode checkpointing and the other is a Qwen3.5 RLHF NaN issue with a different root cause." + "reason": "Different bugs: gradient checkpointing in eval mode vs Qwen3.5 RLHF NaN/SDPA fallback instability." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "save/from_pretrained extra_state handling and Strix Halo mmap OOM are unrelated bugs." + "reason": "Unrelated failures: extra_state save/load handling vs mmap/OOM behavior on Strix Halo." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Both involve gradient checkpointing, but LoRA ineffectiveness and eval-mode incompatibility are different problems." + "reason": "Both mention gradient checkpointing, but one is LoRA inefficiency and the other is eval-mode incompatibility." }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Video input regression in Qwen VL models and strict config loading for granite_speech are unrelated." + "reason": "Different model/config regressions: Qwen VL video-input path vs strict config blocking granite_speech loading." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization in notebooks is unrelated to Trainer max_steps recalculation on checkpoint resume." + "reason": "Custom model init in notebooks is unrelated to Trainer resume/max_steps calculation." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "One is a flash_attention None-handling crash; the other is a Gemma4 FSDP2 shared_kv_states KeyError." + "reason": "Different code paths and symptoms: flash_attention s_aux handling vs Gemma4 FSDP2 shared_kv_states rebuild." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Both report resize_token_embeddings failing to update the output-side embeddings; 45335 is a model-specific manifestation of the same bug." + "reason": "Same underlying bug: resize_token_embeddings does not propagate to the active embedding/output embedding; 45335 is the T5Gemma-specific case." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Jupyter custom-model initialization and extra_state save/load failure do not describe the same code path." + "reason": "Notebook custom-model initialization is unrelated to extra_state save/from_pretrained handling." }, { - "left": "issue:43257", - "right": "issue:44863", + "left": "issue:44402", + "right": "issue:44451", "accept": false, - "reason": "Qwen3 MOE conversion under accelerate/deepspeed and NemotronH checkpoint loading are distinct model-loading issues." + "reason": "Tokenizer vocab-size issue for one model is not the same as a general load failure for another model." }, { - "left": "issue:44402", - "right": "issue:44451", + "left": "issue:43257", + "right": "issue:44863", "accept": false, - "reason": "Both are tokenizer-loading complaints, but they concern different models and likely different tokenizer defects." + "reason": "Both are loading problems, but they affect different architectures and failure mechanisms (Qwen3 MoE conversion vs NemotronH checkpoint loading)." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Autocast dtype mismatch on Phi-tiny-MoE and meta-device loading failures are different loading/runtime issues." + "reason": "Autocast dtype mismatch is a different problem from meta-device loading failures." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch 2.10 incompatibility and an autocast dtype mismatch are not the same defect." + "reason": "Torch version incompatibility is not the same as a Phi-tiny-MoE autocast dtype mismatch." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup failure and Strix Halo mmap OOM are unrelated." + "reason": "Float8 storage lookup failure is unrelated to Strix Halo mmap/OOM behavior." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 device-map mismatch are separate issues." + "reason": "Different subsystems: compiled flex_attention on torch>=2.9 vs Gemma4 PLE device placement." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper return_language pipeline regression and tokenization changes across versions are different problems." + "reason": "Whisper return_language pipeline regression is not the same as general tokenization changes." }, { "left": "issue:43931", "right": "issue:44863", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and NemotronH checkpoint loading are unrelated model-loading bugs." + "reason": "Different model families and symptoms: Qwen3-VL shape mismatch vs NemotronH checkpoint loading." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "An isolated reproduction of another issue and a PagedAttentionCache linear_attention crash do not match." + "reason": "Isolated reproduction of an earlier issue is not the same as the Qwen3.5 PagedAttentionCache crash." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "Torch version incompatibility and Gemma4 forward device mismatch are different defects." + "reason": "Torch compatibility issue does not match Gemma4 forward-device mismatch." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "base_model_prefix loading breakage and custom model initialization in notebooks are not the same root cause." + "reason": "Base_model_prefix loading regression is different from notebook-defined custom model initialization." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "General Torch incompatibility and a specific Mistral-3 dtype mismatch are unrelated." + "reason": "Torch version incompatibility is unrelated to Mistral-3 image preprocessor/model dtype mismatch." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and non-template-node compile TypeError are different code paths." + "reason": "PagedAttentionCache group-type crash is unrelated to template-node compilation errors." }, { - "left": "issue:43901", - "right": "issue:44509", + "left": "issue:39404", + "right": "issue:44912", "accept": false, - "reason": "Both are docs-related, but they cover different deprecated pipeline/task references." + "reason": "Whisper pipeline language return regression is unrelated to MXFP4 quantization fallback." }, { - "left": "issue:39404", - "right": "issue:44912", + "left": "issue:43901", + "right": "issue:44509", "accept": false, - "reason": "Whisper pipeline language return regression and MXFP4 quantization fallback are unrelated." + "reason": "Both are docs issues, but they cover different obsolete pipeline/task mentions." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export/torch_compilable_check failure and Perceiver interpolation failure are different model/export issues." + "reason": "torch.export failure with torch_compilable_check is unrelated to Perceiver non-default-resolution inference." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Qwen VL video-input regression and a generic mask-shape IndexError are not the same bug." + "reason": "Different regressions: Qwen VL video-input path vs generic mask/index shape mismatch." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "SigLIP2 nonsensical outputs and ModelOutput key assignment are unrelated." + "reason": "Nonsensical model outputs are not the same bug as incorrect ModelOutput key assignment." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Both touch Qwen3-VL loading, but one is a version-specific MoE loading regression and the other is a specific weight-shape mismatch report; not enough evidence they are the same bug." + "reason": "Both concern Qwen3-VL loading, but one is a MoE/version regression and the other is a specific weight-shape mismatch." }, { "left": "issue:43425", "right": "issue:44918", "accept": false, - "reason": "Torch 2.10 incompatibility and Qwen3.5 embedding unpacking in TRL are different problems." + "reason": "Torch compatibility is unrelated to Qwen3.5 embedding unpacking with trl SFT trainer." }, { "left": "issue:39404", "right": "issue:44671", "accept": false, - "reason": "Whisper pipeline language handling and CamemBERT masked LM prediction errors are unrelated." + "reason": "Whisper pipeline behavior and CamemBERT masked-LM predictions are different code paths." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage lookup and non-template-node compile errors do not match." + "reason": "Float8 storage deserialization failure is unrelated to non-template-node compilation." }, { "left": "issue:44164", "right": "issue:45759", "accept": false, - "reason": "extra_state save/load failure and AutoModelForCausalLM.from_config text_config unwrapping are different loading behaviors." + "reason": "extra_state handling and AutoModelForCausalLM.from_config text_config unwrapping are different loading bugs." }, { "left": "issue:36032", "right": "issue:43452", "accept": false, - "reason": "T5Tokenizer special-tokens conflict and gguf_file breaking from_pretrained are unrelated tokenizer/model-loading issues." + "reason": "Tokenizer special-token conflict is unrelated to gguf_file loading behavior." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "Zero3 init IndexError with BertModel and compiled flex_attention torch>=2.9 are unrelated." + "reason": "Deepspeed ZeRO3/BertModel index error is unrelated to compiled flex_attention on newer torch." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "CamemBERT MLM prediction regression and tokenizer loading failure for EMBEDDIA/est-roberta are different bugs." + "reason": "CamemBERT masked-LM predictions and tokenizer load failure for EMBEDDIA/est-roberta are different issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the one accepted soft-edge pair (45292/45335) is a conservative duplicate call based on the shared resize_token_embeddings propagation failure. The other verdicts stay appropriately negative given the titles alone, and no PR mergeability issue applies because no PRs were provided." + "feedback": "The judgments are well grounded in the packet and stay conservative. The only accepted pair, issue:45292 / issue:45335, is supported by the titles as the same resize_token_embeddings propagation bug with a model-specific manifestation. The rest of the soft-edge rejections are cautious and not overstated, and the cluster summary matches the mostly heterogeneous issue titles." }, "evaluator_used": true, "retried": false, @@ -2660,58 +2624,58 @@ "issue:43089|issue:45987", "issue:45084|issue:45507" ], - "prepared_review_unit_hash": "935733371e0282169a8f50808adc185ebaf51fbd4f63d5ea2c1dd9c8ba5dcae9", + "prepared_review_unit_hash": "8664bfbb92e3c2987d4422374eafded9829395d2986d47fe25538558739bc4ab", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37262, - "estimated_input_tokens": 9316, - "estimated_eval_tokens": 18888 + "serialized_chars": 37267, + "estimated_input_tokens": 9317, + "estimated_eval_tokens": 18890 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous set of Transformers issue reports with overlapping buzzwords (v5 regressions, loading, tokenizers, device/offload), but no pair looks like the same concrete bug/change. No PRs were present.", - "confidence": 0.71, - "canonical_issue_reason": "issue:43644 is the broadest, most central regression report in the set (v5.0.0 non-persistent buffer corruption) and is the best umbrella anchor among these mostly unrelated issues.", + "summary": "The cluster is heterogeneous and the soft evidence does not justify any duplicate merges. In particular, the two most tempting pairs, 44206/44220 and 43452/43482, are only thematically similar; the packet marks both as non-duplicates, so they should not be treated as confirmed dupes.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43611 is the broadest representative of the v5 loading/init regression subgroup, though it does not capture the whole mixed cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43644 is the best global issue candidate because it describes a core, actionable v5 regression with potential downstream impact on model initialization/loading paths.", + "best_issue_reason": "issue:43611 is the best single representative because it sits in the largest coherent subgroup of v5-related loading regressions.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:32090", "right": "issue:39404", "accept": false, - "reason": "Different failures: Trainer GPU broadcast NoneType error vs Whisper pipeline return_language regression." + "reason": "Different subsystems: Trainer GPU broadcast TypeError vs Whisper pipeline return_language regression." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is Strix Halo mmap OOM; not the same bug." + "reason": "Unrelated problems: Qwen3 MoE training memory usage vs Strix Halo mmap OOM avoidance." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Both concern tokenizer loading, but one is an est-roberta tokenizer failure and the other is a Mistral regex patch crash; different code paths." + "reason": "Both are tokenizer-loading failures, but the models and failure modes differ." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both involve device placement, but XLNet CPU arange overhead and Qwen3_5Moe CPU offload incompatibility are distinct issues." + "reason": "Both involve device/offload themes, but the actual code paths are different." }, { "left": "issue:44164", @@ -2723,169 +2687,172 @@ "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Apple Silicon caching_allocator_warmup TypeError and the _is_hf_initialized regression are separate failures." + "reason": "Apple Silicon caching allocator warmup TypeError is unrelated to the _is_hf_initialized regression." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Different model-loading bugs; shared tokenization context is too broad to treat as one defect." + "reason": "Both are tokenizer-load bugs, but for different models and different root causes." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Tokenizer mapping/add-new-model-like failure is unrelated to compiled flex_attention on torch>=2.9." + "reason": "Model-mapping logic and compiled flex_attention are unrelated." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity regression on gpt-oss-20b is not the same as runtime post-processor updates for special tokens." + "reason": "Perplexity on gpt-oss-20b is unrelated to runtime post-processor updates in TokenizersBackend." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Both are model-loading complaints, but they target different models and failure modes." + "reason": "Both are model-load failures, but there is no concrete shared bug or code path." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Both are v5 regressions, but base_model_prefix loading and non-persistent buffer junk are different code-path bugs." + "reason": "Both are v5 regressions, but base_model_prefix loading and non-persistent buffer junk are distinct bugs." }, { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "FeatureExtractor center-arg regression and _torch_extract_fbank_features problems are related audio code, but not the same concrete issue." + "reason": "Same audio feature-extraction area, but the packet does not support a single shared underlying bug." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Mistral image dtype mismatch is unrelated to the template-node compilation TypeError." + "reason": "Mistral image-preprocessor dtype mismatch is unrelated to the template compilation error." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Missing all_tied_weights_keys in Molmo is unrelated to loading sleng-bert." + "reason": "Missing all_tied_weights_keys in Molmo is unrelated to a BERT tokenizer load failure." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "AutoConfig trust_remote_code field loss and Qwen3.5 PagedAttentionCache invalid group type are unrelated." + "reason": "AutoConfig trust_remote_code field loss is unrelated to Qwen3.5 PagedAttentionCache linear_attention failures." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "T5 tokenizer special-token conflict and Qwen3Moe FP8 config failure are different bugs." + "reason": "T5 tokenizer special-token conflict is unrelated to Qwen3 MoE FineGrainedFP8Config failures." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both are Whisper/special-token themed, but batch_decode skip_special_tokens is distinct from empty-transcription generation failure." + "reason": "Both are Whisper-related, but they describe different bugs and code paths." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "TrainerState token counter bug and Qwen2.5-VL iterable dataset multi-image finetuning failure are unrelated." + "reason": "TrainerState token counting and Qwen2.5-VL multi-image finetuning are unrelated." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is not the same as StaticCache.get_seq_length returning a tensor." + "reason": "Chunked generation inconsistency and StaticCache.get_seq_length return-type mismatch are not the same issue." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Related v5 initialization complaints, but one is buffer corruption and the other is custom model initialization; not the same defect." + "reason": "Both are v5 initialization regressions, but the concrete failures differ." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Different model-loading failures for different repositories; no evidence of one shared underlying bug." + "reason": "Different model-loading regressions for different checkpoints; no shared underlying code path is evident." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "Both involve GGUF loading, but one is a general gguf_file API break and the other is a Qwen2.5-GGUF-specific failure." + "reason": "Both are GGUF/loading regressions in v5, but the titles point to different failure modes and models." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Mixtral grouped_mm tracing error and template-node compile TypeError are distinct compiler/runtime problems." + "reason": "Grouped_mm tracing failure and non-template node compilation error are unrelated." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "Different Qwen-family issues: FP8 training failure vs PagedAttentionCache linear_attention crash." + "reason": "Different Qwen runtime issues: FP8 training config vs PagedAttentionCache linear_attention crash." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "device_map auto offload loading failure and Qwen2.5-GGUF v5 loading failure are not the same code path." + "reason": "Dequantized model offload failure is unrelated to GGUF loading regression." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Qwen3.5 output_hidden_states bug and Qwen2.5-VL rope indexing bug are unrelated model-specific regressions." + "reason": "Qwen3.5 hidden_states behavior and Qwen2.5-VL rope-index scaling are different model-specific bugs." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "T5 tokenizer special-token conflict is unrelated to custom model initialization in notebooks." + "reason": "Tokenizer special-token conflict is unrelated to Jupyter custom-model initialization." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Different model/tokenizer loading failures for different checkpoints; no shared concrete bug." + "reason": "Both fail on model loading, but they concern different models and different failure modes." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Molmo tied-weights attribute error is unrelated to tokenizer loading for EMBEDDIA/est-roberta." + "reason": "Molmo tied-weights attribute error is unrelated to loading EMBEDDIA/est-roberta." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Mistral3 lm_head tying bug and lm_head serialization inconsistency are related to weights, but not the same failure mode." + "reason": "Mistral3 lm_head tying and device-dependent serialization are related to weights, but not the same concrete bug." }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation GPU sync/dispatch overhead is unrelated to StaticCache.get_seq_length return type." + "reason": "Generation overhead from sync/dispatch and StaticCache seq-length type mismatch are different issues." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Template compilation error and GraniteMoEHybrid invalid method calls are different bugs in different subsystems." + "reason": "Template compilation TypeError and GraniteMoEHybrid invalid method call are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The summary overstates the evidence. In the packet, every soft-pair verdict is deterministic_accept=false, including the two pairs the summary calls true duplicates (`44206/44220` and `43452/43482`). That makes the claim that these are duplicates unsupported. The rest of the summary is broadly conservative, but this specific conclusion is too strong for the provided evidence." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -2984,14 +2951,14 @@ "issue:43632|issue:44534", "issue:42175|issue:45542" ], - "prepared_review_unit_hash": "a989dbb9af60ea95d02438feb2c78afe783af51168a5c35f029f8d3d36996c42", + "prepared_review_unit_hash": "e5590337e6b1dce4771794e143091280b210af4af60bf9b4cfa7ded89fab51cc", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, @@ -3006,168 +2973,168 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a collection of unrelated bug reports with only broad topical overlap (especially around Transformers v5 and Qwen variants). None of the soft pairs look like the same concrete underlying bug or change, so I would not merge any of them as duplicates.", - "confidence": 0.91, - "canonical_issue_reason": "issue:43646 is the broadest, most umbrella-like Transformers v5 regression in the set, so it is the best canonical issue anchor if one must be chosen.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers regressions, not one duplicate family. I only see two strong duplicate-like pairs: the Qwen3.5 save_pretrained regression and the Qwen video/temporal-position regression; the rest are too semantically distant to merge.", + "confidence": 0.71, + "canonical_issue_reason": "If one issue must anchor the cluster, issue 44479 is the broadest concrete regression report and best representative of the Qwen video/position-id bug subset. It is still not a good umbrella for the many unrelated loading/init/backend issues in the set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43646 is the most generally framed issue and the most plausible duplicate target among the items, even though the cluster is not truly a duplicate set.", + "best_issue_reason": "Issue 44479 is the best overall issue candidate because it describes a specific regression affecting multiple Qwen variants and is a plausible duplicate target for the narrower Qwen2.5-VL position-id report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Different subsystems and failure modes: audio chat template sample-rate handling vs MIMI encoder padding-mask batching." + "reason": "Both are audio-related, but one is a chat-template sampling-rate default bug and the other is a padding-mask/ batching correctness bug in MIMI; different code paths." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Both are vision/model-quality reports, but one is nonsensical outputs from AutoModel/pipeline and the other is a processor size mismatch for OmDet-Turbo." + "reason": "Different models and failures: SigLIP2 pipeline output quality vs OmDet processor image-size mismatch." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Both touch Transformers v5 init/load behavior, but one is custom-model initialization and the other is non-persistent buffer corruption; not the same bug." + "reason": "Custom model initialization failure is not the same underlying bug as non-persistent buffers being filled with junk." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Llava Next loading segfault and Qwen3OmniMoe video ValueError are unrelated code paths." + "reason": "Different multimodal model families and different symptoms: Llava Next load segfault vs Qwen3OmniMoe video ValueError." }, { "left": "issue:43645", "right": "issue:44534", "accept": false, - "reason": "Notebook/custom model initialization breakage is a different issue from junk-filled non-persistent buffers." + "reason": "Notebook/custom-model init regression is unrelated to the non-persistent buffer initialization bug." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Generic loading noise and a Mistral tokenizer regex patch crash are not the same defect." + "reason": "Generic loading noise vs a Mistral tokenizer regex crash; not the same bug." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "MIMI batching/padding-mask mismatch is unrelated to DAC latent decoding without STE." + "reason": "MIMI batching/padding-mask bug and DAC from_latents STE mismatch are unrelated." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "Qwen3Moe FP8 config failure and template-compilation errors have different root causes and code paths." + "reason": "Qwen3MoE FP8 config failure and compile-time template-node error are different failures." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export with torch_compilable_check and OmDet processor resolution mismatch are distinct issues." + "reason": "torch.export with torch_compilable_check and OmDet processor image sizing are unrelated." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Model loading failure for ScandiBERT and tokenizer backend attribute crash are unrelated." + "reason": "ScandiBERT loading failure and Mistral regex patch crash are different model-specific bugs." }, { "left": "issue:45216", "right": "issue:45357", - "accept": false, - "reason": "Both concern Qwen3.5 save_pretrained regressions, but one is incorrect checkpointing generally and the other is specific visual encoder key naming; not clearly the same bug." + "accept": true, + "reason": "Same Qwen3.5 save_pretrained regression: 45357 is a later report of incorrect visual encoder keys from the same underlying bug." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Physical-intelligence/fast loading regression and Mistral tokenizer regex patch crash are different problems." + "reason": "Physical-intelligence/fast loading failure is not the same as the Mistral tokenizer backend_tokenizer crash." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Audio feature extraction bug and compiled flex_attention failure on newer torch are unrelated." + "reason": "Audio feature extraction issue and compiled flex_attention failure are unrelated." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Untied lm_head weights in Mistral3 and LoRA merge collapse after vocab extension are different model-state issues." + "reason": "Tied lm_head weights in Mistral3ForConditionalGeneration is not the same as LoRA merge collapse with extended vocab." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "FineGrainedFP8Config failure and unexpected _is_hf_initialized argument during init_empty_weights are distinct Transformers v5 regressions." + "reason": "Different root causes: Qwen3MoE FP8 config vs init_empty_weights/_is_hf_initialized argument handling." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "_is_hf_initialized flag breakage and mmap/OOM on Strix Halo are unrelated." + "reason": "_is_hf_initialized flag regression and mmap/OOM on Strix Halo are unrelated." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "Tokenizers AddedToken special-flag conflict does not match the physical-intelligence/fast loading regression." + "reason": "AddedToken special-argument duplication and loading physical-intelligence/fast are different bugs." }, { "left": "issue:44479", "right": "issue:45325", - "accept": false, - "reason": "Video-input regression across Qwen models and still-image temporal position scaling are separate Qwen2.5-VL/Qwen3-VL bugs." + "accept": true, + "reason": "45325 is a narrower manifestation of the same Qwen video/temporal-position regression described broadly in 44479." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "T5Tokenizer special-token method conflict and MobileLLM key error are different loading failures." + "reason": "T5Tokenizer add_special_tokens conflict and MobileLLM-125M key error are unrelated loading problems." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "Qwen3Moe FP8 config failure and missing Float8 storage lookup are not the same error path." + "reason": "Qwen3MoE FineGrainedFP8Config failure is unrelated to missing Float8 storage support." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "PaddleOCR-VL loading failure and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "PaddleOCR-VL load failure and Qwen3-MoE sliding_window issue are different model-specific regressions." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "IterableDataset multi-image finetuning failure and Qwen2.5-VL still-image rope scaling are different multimodal bugs." + "reason": "IterableDataset multi-image fine-tuning failure is not the same as the Qwen2.5-VL temporal-position bug." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Qwen3-MoE sliding_window bug and granite_speech strict-config loading failure do not share a concrete root cause." + "reason": "Qwen3-MoE sliding_window behavior and granite_speech strict config loading are unrelated." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "base_model_prefix loading regression is not the same as non-persistent buffer corruption." + "reason": "base_model_prefix loading breakage is a different issue from non-persistent buffer corruption." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "Tokenizer attribute conflict and Qwen3 auto device mapping cudaErrorAssert are unrelated." + "reason": "Tokenizer add_special_tokens conflict and Qwen3 device-mapping cudaErrorAssert are unrelated." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch version incompatibility and generic loading noise are not duplicates." + "reason": "Torch 2.10 incompatibility is not the same as generic loading noise." }, { "left": "issue:41553", @@ -3179,32 +3146,35 @@ "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "ModelOutput key assignment bug and Gemma4 FSDP2 shared_kv_states rebuild failure are different issues." + "reason": "ModelOutput key assignment and Gemma4 FSDP2 shared_kv_states KeyError are different code paths." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "High perplexity on gpt-oss-20b and Whisper batch_decode skip_special_tokens behavior are unrelated." + "reason": "High perplexity on gpt-oss-20b and Whisper batch_decode skip_special_tokens are unrelated." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "The _is_hf_initialized flag regression and non-persistent buffer junk issue are separate Transformers v5 breakages." + "reason": "These are both Transformers v5 regressions, but they affect different mechanisms and are not the same underlying bug." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "TensorFlow missing from a torch-only install and undefined tf backend from tensorboard-only install are related packaging themes but not the same defect." + "reason": "Both mention TensorFlow packaging/backend issues, but one is missing tf in the torch extra and the other is a tensorboard-only runtime backend error; too different to treat as duplicates." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The chosen anchor issue 44479 is a reasonable broad representative for the Qwen video/position-id subset, and the only accepted duplicate-like pair (45216/45357) is plausibly the same save_pretrained regression. The rejections mostly rest on clearly different model families or bug mechanisms, so I don\u2019t see an overstatement." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -3289,9 +3259,9 @@ "issue:43377|issue:45491", "issue:43749|issue:44079", "issue:40444|issue:45381", - "issue:43644|issue:44079", - "issue:41720|issue:45081", "issue:43531|issue:44291", + "issue:41720|issue:45081", + "issue:43644|issue:44079", "issue:44265|issue:45072", "issue:45325|issue:45381", "issue:43632|issue:45072", @@ -3301,14 +3271,14 @@ "issue:43454|issue:44466", "issue:45325|issue:45834" ], - "prepared_review_unit_hash": "d008e177a7ef4168cbc1f22b10d2fcbd333f817f4f08d893bd171aece99f1e4b", + "prepared_review_unit_hash": "349dad4f140efb7bc3ff5a08bc0a839e2ee5e9ef8db2f482929c436ab23ee9fd", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, @@ -3323,192 +3293,195 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated issue reports, many around Transformers v5 regressions. None of the proposed soft pairs look like true duplicates of the same concrete bug. The broadest umbrella issue in the set is the remote_code regression report.", - "confidence": 0.86, - "canonical_issue_reason": "issue:45020 is the broadest, most umbrella-style report and has active discussion/references; it best represents the cluster if one issue must be chosen.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers v5 regressions, with the strongest umbrella being the open remote_code loading breakage. The Qwen2.5-VL position-id pair should stay rejected: the titles point to related model family bugs, but one is still-image temporal position_id scaling and the other is video vision_position_ids, with no explicit shared code path in the packet.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45020 is the best umbrella issue: it is broad, open, and explicitly covers recent-version regressions breaking model loading with remote_code, which matches the cluster's overall theme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the most suitable representative issue because it spans multiple affected models and a common regression class rather than a narrow model-specific failure.", + "best_issue_reason": "issue:45020 is the strongest representative for triage because it is the broadest open report and the most plausible place for cross-cutting follow-up links or fixes.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer-related, but one is about updating the post-processor after runtime special-token changes while the other is about add_special_tokens not adding BOS/EOS; different bugs." + "reason": "Related tokenizer/special-token area, but one is runtime post-processor refresh and the other is add_special_tokens behavior; different bugs." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Unrelated failure modes: Whisper model loading versus missing Float8 storage handling." + "reason": "Different failure modes and code paths: Whisper loading versus a Float8 storage lookup error." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Different models and defects: Video-LLaVA tower architecture issue versus Qwen3.5 save_pretrained key serialization." + "reason": "Different models and bugs: Video-LLaVA tower/weight sharing versus Qwen3.5 visual-encoder save_pretrained keys." }, { "left": "issue:44683", "right": "issue:45020", "accept": false, - "reason": "Both involve recent version regressions, but one is a torch.compile/flex_attention breakage and the other is a remote_code model-loading regression." + "reason": "Torch flex_attention compatibility issue, not the remote_code loading regression." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "One is generic loading noise, the other is a tokenizer-loading regression for a specific model; not the same bug." + "reason": "Generic loading noise versus a specific tokenizer loading regression for EMBEDDIA/est-roberta." }, { "left": "issue:44462", "right": "issue:45812", "accept": false, - "reason": "Both concern AutoTokenizer behavior, but one ignores tokenizer.json while the other returns wrong token IDs for Granite models." + "reason": "Both are AutoTokenizer regressions, but they affect different model families and concrete failure modes." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Different scenarios: multi-GPU training for Qwen2-VL versus IterableDataset failures with multiple images per prompt." + "reason": "Both are Qwen-VL training issues, but multi-GPU training and multiple-images-per-prompt are distinct bugs." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Both affect classification training, but the triggers differ: flash_attention_3 degeneration versus num_labels=1 zero-loss behavior." + "reason": "Both affect classification training, but flash_attention_3 behavior and num_labels=1 loss handling are different paths." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "Different problems: device_map/offload loading failure versus inefficient memory usage during Qwen3 MoE training." + "reason": "Device_map/offload loading failure versus MoE training memory efficiency; not the same bug." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "Different tokenizer/model-loading issues: GGUF loading failure versus CLIPTokenizer max_length behavior." + "reason": "Different areas: GGUF loading with v5 versus CLIPTokenizer model_max_length behavior." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "One is eval-mode gradient checkpointing usage, the other is a missing supports_gradient_checkpointing flag on a remote-code variant." + "reason": "Related to gradient checkpointing, but eval-mode usage and a missing capability flag on a remote-code model are different issues." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Different generation/training contexts and different root causes; not the same bug." + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL multi-image finetuning." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "Different internal behaviors: ModelOutput key assignment versus non-persistent buffer serialization." + "reason": "ModelOutput key assignment and non-persistent buffer serialization are separate internal regressions." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "Both are v5 compatibility issues, but one is init_empty_weights argument handling and the other is Float8 storage resolution." + "reason": "Unexpected _is_hf_initialized init arg and Float8 storage lookup are different v5 breakages." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Both are loading issues, but one is GGUF/v5 compatibility and the other is missing pad_token_id on Llama4Config." + "reason": "GGUF loading failure and missing pad_token_id on Llama4Config are unrelated." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Both mention batched/padding-related failures, but they affect different models and code paths." + "reason": "Padding-mask correctness in MIMI encoder versus Gemma3 NaN embeddings from sliding-window attention." }, { "left": "issue:43749", "right": "issue:44079", "accept": false, - "reason": "FSDP RAM-efficient loading and ModelOutput key assignment are unrelated." + "reason": "FSDP CPU RAM efficient loading and ModelOutput key assignment are distinct regressions." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Different Qwen2.5-VL bugs: multi-image IterableDataset finetuning versus video vision_position_ids." + "reason": "Related model family, but the packet only shows different paths: multi-image training versus video vision_position_ids." }, { - "left": "issue:43644", - "right": "issue:44079", + "left": "issue:43531", + "right": "issue:44291", "accept": false, - "reason": "Non-persistent buffer junk filling is unrelated to ModelOutput key bookkeeping." + "reason": "Sliding_window behavior in Qwen3-MoE is unrelated to init_empty_weights/_is_hf_initialized." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Different failures in different models: cudaErrorAssert on Qwen3 auto device mapping versus Mistral tokenizer regex patch crash." + "reason": "Auto device-mapping cudaErrorAssert and a Mistral tokenizer regex crash are unrelated." }, { - "left": "issue:43531", - "right": "issue:44291", + "left": "issue:43644", + "right": "issue:44079", "accept": false, - "reason": "Sliding-window handling in Qwen3-MoE is not the same as unexpected _is_hf_initialized argument in init_empty_weights." + "reason": "Non-persistent buffers filled with junk and ModelOutput key assignment are different internal bugs." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "torch.export with torch_compilable_check and bfloat16 dtype mismatches are different bugs." + "reason": "torch.export with torch_compilable_check is unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel." }, { "left": "issue:45325", "right": "issue:45381", "accept": false, - "reason": "Same model family, but one is still-image temporal position scaling and the other is video vision_position_ids; not the same concrete defect." + "reason": "Related Qwen2.5-VL bugs, but the packet suggests different code paths: still-image temporal position_ids scaling versus video vision_position_ids." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "Different v5 regressions: _is_hf_initialized flag handling versus bfloat16 dtype mismatch." + "reason": "_is_hf_initialized breaks model init/loading, while this is a dtype mismatch in inference." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Different vision-model failures: siglip2 nonsensical outputs versus Perceiver resolution/interpolation issue." + "reason": "Both are vision-model correctness issues, but they involve different models and failure mechanisms." }, { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Both are v5 model-init regressions, but one is base_model_prefix loading and the other is custom initialization behavior." + "reason": "Both are v5 loading/init regressions, but one is about base_model_prefix loading and the other custom model initialization." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Package backend inclusion and PaddleOCR-VL load failure are unrelated." + "reason": "Package/backend composition issue versus a specific PaddleOCR-VL loading failure." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "One is a specific AyaVision weight-tying bug; the other is inconsistent lm_head serialization across devices." + "reason": "Both involve lm_head weights, but one is a missing tie in a model config and the other is device-dependent serialization inconsistency." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Both involve Qwen2.5-VL OCR/position behavior, but one is temporal position scaling and the other is long OCR input index errors." + "reason": "Different model families and symptoms: Qwen2.5-VL position IDs versus Kosmos2.5 OCR long-input index errors." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but one soft-edge verdict is too strong: the acceptance of issue:45325 \u2194 issue:45381 goes beyond the packet evidence. The titles suggest related Qwen2.5-VL position-id bugs, but they refer to different paths (still-image temporal position_ids scaling vs. video vision_position_ids), and the packet provides no explicit linkage or shared codepath. Please keep that pair rejected or qualify it more conservatively." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -3584,8 +3557,8 @@ "issue:43881|issue:44488", "issue:43606|issue:44265", "issue:44291|issue:44877", - "issue:44568|issue:45920", "issue:43262|issue:43746", + "issue:44568|issue:45920", "issue:40990|issue:45584", "issue:34634|issue:35707", "issue:43873|issue:45701", @@ -3599,207 +3572,207 @@ "issue:39692|issue:40444", "issue:43572|issue:45759", "issue:43454|issue:45072", - "issue:43257|issue:45923", "issue:43377|issue:44898", + "issue:43257|issue:45923", "issue:43526|issue:43883", "issue:40990|issue:42947" ], - "prepared_review_unit_hash": "06c12f4633bba68c0ccf7ddfe73859d54b9c8f1fc2dd090180f41cb05c3944ba", + "prepared_review_unit_hash": "75297cbafcc2eb0ea32309ab413ad86c06c538f209d3d94d2606202133b39171", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35605, - "estimated_input_tokens": 8902, - "estimated_eval_tokens": 18060 + "serialized_chars": 35603, + "estimated_input_tokens": 8901, + "estimated_eval_tokens": 18058 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a heterogeneous mix of unrelated bugs/regressions across model loading, tokenization, offloading, multimodal processors, and training utilities. None of the soft pairs look like the same concrete underlying issue, so there is no single canonical duplicate to keep for this cluster.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "This cluster is very heterogeneous: the items span audio, vision, tokenization, training, checkpointing, and multiple unrelated loading/offload regressions. I don\u2019t see any soft pair that clearly shares the same concrete bug path, so all candidate duplicate links should be rejected.", + "confidence": 0.34, + "canonical_issue_reason": "issue:43873 is the broadest and most active open report here, and it best represents the loose loading/offloading/quantization subtheme among the items.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:43873 is the strongest single issue to keep as the representative bug report: it is open, discussion-heavy, and describes a broadly scoped regression rather than a one-off model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Both involve loading/runtime failures, but one is Whisper-specific and the other is a broad meta-device loading issue affecting different models; not the same bug." + "reason": "Whisper runtime failure and meta-device loading breakage are different code paths and different failure modes." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "Qwen3 MoE FP8 config failure and Qwen2.5 GGUF load failure under v5 are different code paths and symptoms." + "reason": "Different models, different formats, and different root causes: FP8 config failure vs GGUF loading failure." }, { "left": "issue:43296", "right": "issue:45800", "accept": false, - "reason": "One is a PaddleOCR-VL load problem in vLLM; the other is a general torch/transformers version incompatibility. Too broad to be the same fix." + "reason": "Both mention incompatibility/loading, but one is a PaddleOCR-VL/vLLM model-load issue and the other is a torch/transformers version mismatch; not the same bug." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Both mention OOM, but one is an int4 quantization reserved-memory regression and the other is a video model memory issue." + "reason": "Both can end in OOM, but one is an int4 quantization memory regression and the other is a large video model running out of memory." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL rope index scaling for still-image temporal positions." + "reason": "Checkpoint resume key mismatch and Qwen2.5-VL rope-index scaling are unrelated training vs multimodal-positioning bugs." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Video-input regression in Qwen VL models and Gemma4 PLE device mismatch are separate forward-path bugs." + "reason": "Both involve multimodal models, but video-input regression and Gemma4 device mismatch are distinct failures." }, { "left": "issue:43881", "right": "issue:44488", "accept": false, - "reason": "Different model families and failure modes: glm-4v-9b load failure versus sleng-bert load failure." + "reason": "Two separate loading failures on different models with no shared concrete code path." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU offload device mismatch for Bark is unrelated to torch.export failures caused by torch_compilable_check." + "reason": "CPU-offload device mismatch and torch.export failing on torch_compilable_check are unrelated subsystems." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Unexpected init_empty_weights argument and strict config loading for granite_speech are distinct loading/config issues." + "reason": "init_empty_weights argument handling and strict config loading are different initialization/config bugs." }, { - "left": "issue:44568", - "right": "issue:45920", + "left": "issue:43262", + "right": "issue:43746", "accept": false, - "reason": "Specific special-token insertion regression for one tokenizer is not the same as wrong token IDs across several models." + "reason": "Audio chat-template sample-rate handling and PEFT checkpoint loading are unrelated." }, { - "left": "issue:43262", - "right": "issue:43746", + "left": "issue:44568", + "right": "issue:45920", "accept": false, - "reason": "Audio chat-template sampling-rate default is unrelated to PEFT checkpoint loading for GraniteSpeech." + "reason": "A specific special-tokens regression for one tokenizer is not the same concrete bug as a broad AutoTokenizer ID regression across multiple models." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "High perplexity on gpt-oss-20b and Whisper empty-transcription failure are unrelated generation issues." + "reason": "Perplexity evaluation issue and Whisper empty-transcription generation failure are unrelated model behaviors." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "BarkProcessor voice_preset malfunction and progressive generation with inputs_embeds/past_key_values are different APIs and bugs." + "reason": "BarkProcessor voice_preset and progressive generation with inputs_embeds/past_key_values do not share the same bug." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Quantization/offloading behavior and tokenizer-version changes do not share the same concrete failing code path." + "reason": "Quantization/offloading regression and tokenizer-version behavior change are unrelated." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer crash on NER/padding is unrelated to torch.export with torch_compilable_check." + "reason": "Tokenizer crash on NER/padding is not the same as torch.export failing on compilability checks." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Progressive generation with inputs_embeds is unrelated to Qwen3 auto device mapping cudaErrorAssert." + "reason": "Progressive generation and Qwen3 auto device mapping/cudaErrorAssert are different code paths." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "Both are model-runtime bugs, but CPU offload device mismatch and DAC.from_latents/STE mismatch are separate implementations." + "reason": "CPU offload device mismatch and DAC latent/STE mismatch are distinct audio-model bugs." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Auto device mapping assert on Qwen3 and Gemma4 PLE device mismatch are different model-specific failures." + "reason": "Both mention device mismatch, but they affect different models and different execution paths." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "Weight tying/generation corruption in AyaVision is unrelated to Bamba-9B-v2 torch.compile SDPA failure." + "reason": "lm_head tying and torch.compile SDPA are unrelated model setup vs compilation issues." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "Missing padding-mask support in MIMI encoder and OmDet-Turbo processor image-size mismatch are different bugs." + "reason": "Missing padding-mask support in MIMI and OmDet-Turbo processor size mismatch are different preprocessing/model-shape bugs." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Loading dequantized models with device_map=auto and general offloading-not-working reports are related in theme but not the same concrete issue." + "reason": "Both are about offloading/quantization, but one is a dequantized-model load failure and the other is a broader offloading behavior regression; not enough to call the same bug." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "SigLIP2 doc-example/model-processor mismatch and Qwen2.5-VL iterable multi-image finetuning failure are unrelated." + "reason": "Documentation/example errors and finetuning with multiple images per prompt are unrelated." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig and text_config unwrapping for composite Qwen multimodal configs are distinct config issues." + "reason": "Missing pad_token_idx and text_config unwrapping are separate config-compatibility bugs." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "AyaVision lm_head tying bug and bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel are different failures." + "reason": "Weight tying and bfloat16 dtype mismatch are different issues." }, { - "left": "issue:43257", - "right": "issue:45923", + "left": "issue:43377", + "right": "issue:44898", "accept": false, - "reason": "Qwen3 MoE weight conversion with accelerate+deepspeed is unrelated to a missing supports_gradient_checkpointing flag." + "reason": "Batched-padding mask behavior in MIMI and Perceiver interpolate_pos_encoding failure are unrelated." }, { - "left": "issue:43377", - "right": "issue:44898", + "left": "issue:43257", + "right": "issue:45923", "accept": false, - "reason": "MIMI padding-mask divergence and Perceiver non-default-resolution failure are separate model/processor problems." + "reason": "MoE weight conversion under accelerate/deepspeed and a missing supports_gradient_checkpointing flag are different fixes." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels bug and Molmo missing all_tied_weights_keys are unrelated." + "reason": "Label reduction in BEiT image processing and missing all_tied_weights_keys in Molmo are unrelated." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity evaluation issue and ineffective gradient checkpointing with PEFT LoRA are unrelated training/inference bugs." + "reason": "High perplexity on one model and ineffective gradient checkpointing with LoRA are not the same underlying bug." } ] }, @@ -3881,8 +3854,8 @@ "issue:43296|issue:44877", "issue:41720|issue:44977", "issue:44008|issue:45823", - "issue:43404|issue:45357", "issue:42617|issue:42915", + "issue:43404|issue:45357", "issue:40990|issue:45701", "issue:44186|issue:45072", "issue:45632|issue:45684", @@ -3892,8 +3865,8 @@ "issue:43873|issue:44756", "issue:36331|issue:42617", "issue:44877|issue:44991", - "issue:36331|issue:42915", "issue:43550|issue:43720", + "issue:36331|issue:42915", "issue:43450|issue:45701", "issue:44610|issue:44898", "issue:44186|issue:44610", @@ -3907,238 +3880,220 @@ "issue:44671|issue:44743", "issue:41720|issue:45161" ], - "prepared_review_unit_hash": "75511809c79ebbb90733b24087c08a93c04f6557362370d47acfe4cc46c3f1d2", + "prepared_review_unit_hash": "7dde2dea9dab4da0c95b0fc3caf13477864ac3fb3a357f5e112b74f8d1d59e90", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37665, - "estimated_input_tokens": 9417, - "estimated_eval_tokens": 19090 + "serialized_chars": 37656, + "estimated_input_tokens": 9414, + "estimated_eval_tokens": 19084 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is dominated by superficially similar Transformers issues, but the paired items mostly describe distinct bugs, models, or code paths. I do not see any soft pair that is a safe duplicate merge candidate.", - "confidence": 0.91, - "canonical_issue_reason": "Best representative is issue 43873: it is open, has the most discussion/inbound references, and covers a broad, active offloading+quantization problem.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bugs across loading, tokenization, multimodal processing, quantization, and trainer behavior. No soft pair looks like the same underlying defect, so all candidate duplicate links should be rejected.", + "confidence": 0.87, + "canonical_issue_reason": "Issue 43749 is the strongest anchor: it has the highest discussion/inbound reference signal and describes a concrete, user-facing loading/offloading regression that could plausibly attract related reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the strongest cluster anchor because it is current, broadly scoped, and appears to be the most central active regression in the set.", + "best_issue_reason": "Issue 43749 is the best issue to use as the cluster representative because it is well-scoped, high-traffic, and clearly about a core broken code path rather than a narrow model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both mention save/pretrained and tied weights, but they target different models and different failures (serialization inconsistency vs incorrect visual encoder keys)." + "reason": "Different bugs: one is lm_head weight tying/serialization, the other is incorrect visual encoder keys on save_pretrained." }, { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "Unrelated problems: one is a model loading failure for glm-4v-9b, the other is tokenizer loading for est-roberta." + "reason": "Unrelated failures: model loading for glm-4v-9b versus tokenizer loading for EMBEDDIA/est-roberta." }, { "left": "issue:43749", "right": "issue:44534", "accept": false, - "reason": "Different bugs: FSDP CPU-efficient loading regression versus non-persistent buffers being filled with junk." + "reason": "One is FSDP CPU RAM efficient loading, the other is non-persistent buffers being filled with junk; different code paths." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "Different code paths and symptoms: weight tying in AyaVision versus unpacking packed BitNet weights during accelerate loading." + "reason": "Different model and failure mode: AyaVision lm_head tying vs BitNet packed-weight unpacking during accelerate loading." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Both are model-load failures, but they are different models and failure modes (cuda assert on A800 vs segmentation fault on Llava Next)." + "reason": "Both are crashes, but on different models and causes: Qwen3 auto device mapping cuda assert vs Llava Next segmentation fault." }, { "left": "issue:43296", "right": "issue:44877", "accept": false, - "reason": "Different issues: vLLM/PaddleOCR-VL load failure versus strict config blocking granite_speech config loading." + "reason": "A model load failure and a strict-config compatibility issue are not the same underlying bug." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Same general Qwen area, but the concrete bugs differ: auto device mapping assert versus flash-attention generation regression." + "reason": "Device-mapping cuda assert on Qwen3 is distinct from a flash-attention generation regression in Qwen3.5." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Different Gemma models and different failures: variable-name collision in Gemma3n versus device mismatch in Gemma4 PLE." + "reason": "Different model families and failures: Gemma3n variable collision vs Gemma4 PLE device mismatch." }, { - "left": "issue:43404", - "right": "issue:45357", + "left": "issue:42617", + "right": "issue:42915", "accept": false, - "reason": "Both involve save_pretrained/tied weights, but one is about lm_head tying in Mistral3ForConditionalGeneration and the other about visual encoder keys in Qwen3.5." + "reason": "Completely different problems: a script execution failure versus Qwen3Moe failing with FineGrainedFP8Config." }, { - "left": "issue:42617", - "right": "issue:42915", + "left": "issue:43404", + "right": "issue:45357", "accept": false, - "reason": "Different failures: running 3d_parallel.py versus FineGrainedFP8Config on Qwen3Moe." + "reason": "One is tied lm_head weights in Mistral3, the other is incorrect visual encoder key serialization in Qwen3.5." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "Tokenization may affect perplexity, but these are not the same underlying bug; one is a perplexity report, the other is a tokenizer-version behavior change." + "reason": "High perplexity on one model is not the same as a tokenizer-behavior change across versions." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "Different subsystems entirely: tokenizer padding/NER crashes versus dtype mismatches in bfloat16 inference." + "reason": "Tokenizer crash in NER/padding is unrelated to dtype mismatches during bfloat16 inference." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Both touch save/load workflows, but one is a cache-path collision and the other is permission propagation from custom-model source files." + "reason": "Cache-path collisions for trust_remote_code are unrelated to permission propagation in save_pretrained." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both are serve-time multimodal crashes, but the actual root causes differ (string `.to()` misuse vs missing `_tokenizer` on Gemma4Processor)." + "reason": "Different multimodal serve crashes: continuous batching with Qwen3.5 versus Gemma4Processor missing _tokenizer." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "Unrelated: BLIP2 dtype loading bug versus resume_from_checkpoint max_steps calculation." + "reason": "One is dtype loading for BLIP2, the other is resuming training and recomputing max_steps." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Both are memory-related, but one is int4 quantization OOM behavior and the other is an apex RMSNorm leak in T5." + "reason": "Int4 quantization OOM and a T5 Apex RMSNorm memory leak are separate issues." }, { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Different memory/offload issues: quantization offloading failure versus disabling mmap on Strix Halo to avoid OOM." + "reason": "Quantized offloading behavior and a Strix Halo mmap OOM workaround are not the same bug." }, { "left": "issue:36331", "right": "issue:42617", "accept": false, - "reason": "Completely different failures: CustomTrainer API break versus 3d_parallel.py runtime issue." + "reason": "CustomTrainer compute_loss signature mismatch is unrelated to the 3d_parallel.py execution failure." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Different config/loading bugs: strict Granite config handling versus tokenizer loading for est-roberta." + "reason": "Strict config loading for granite_speech and tokenizer loading for est-roberta are distinct." }, { - "left": "issue:36331", - "right": "issue:42915", + "left": "issue:43550", + "right": "issue:43720", "accept": false, - "reason": "No shared underlying bug; trainer keyword-argument regression is unrelated to Qwen3Moe FP8 loading." + "reason": "Bamba torch.compile with SDPA and BitNet packed-weight unpacking are different failure modes." }, { - "left": "issue:43550", - "right": "issue:43720", + "left": "issue:36331", + "right": "issue:42915", "accept": false, - "reason": "Different problems in different execution paths: torch.compile+SDPA on Bamba versus accelerate loading of packed BitNet weights." + "reason": "A trainer API signature regression is unrelated to Qwen3Moe FP8 loading." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Video processor batch-shape bug is unrelated to tokenization-version differences." + "reason": "Video processor batch-shape bug does not match a version-dependent tokenization change." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Both involve image size/processor expectations, but they affect different models and are not the same defect." + "reason": "Both are image-size related, but one is a processor/model size mismatch and the other is a non-default resolution failure in Perceiver." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Tokenizer NER/padding crash and OmDet-Turbo image-size mismatch are different issues." + "reason": "Tokenizer crashes and OmDet-Turbo image sizing issues affect different subsystems." }, { "left": "issue:44479", "right": "issue:44918", "accept": false, - "reason": "Different Qwen3.5-related bugs: video-input regression versus unpacking embeddings with trl SFT trainer." + "reason": "Qwen video-input regression and Qwen3.5 embedding unpacking with TRL are separate problems." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Different device/execution problems: torch.compile+SDPA versus CPU-offload device mismatch." + "reason": "torch.compile/SDPA failure in Bamba is unrelated to Bark-small CPU offload device mismatch." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "Different models and failure modes; no shared code-path bug is evident." + "reason": "Qwen3 auto device mapping and 3d_parallel.py startup failure are not the same underlying issue." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Both concern config behavior, but one is an API discrepancy and the other is a rotary_pct reload regression." + "reason": "Gemma3TextConfig API discrepancy and GPTNeoXConfig rotary_pct reload behavior are different config bugs." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Different issues: Qwen3Moe FineGrainedFP8Config failure versus dequantized model loading with gpu+cpu offload." + "reason": "Qwen3Moe FineGrainedFP8Config failure is unrelated to dequantized model loading with device_map=auto and CPU offload." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Processor/tokenizer regression in v4.57.5 is not the same as version-dependent tokenization changes." + "reason": "Processor/tokenizer regression is not the same as a general tokenization change across versions." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "Different models and different defects: BeitImageProcessorFast label reduction versus glm-4v-9b loading failure." - }, - { - "left": "issue:44671", - "right": "issue:44743", - "accept": false, - "reason": "Masked LM prediction bug in CamemBERT is unrelated to Qwen3.5 recurrent-state reset behavior." - }, - { - "left": "issue:43873", - "right": "issue:44756", - "accept": false, - "reason": "The memory symptoms differ enough that they are not the same underlying bug; one is quantization offload, the other is mmap-induced OOM." - }, - { - "left": "issue:41720", - "right": "issue:45161", - "accept": false, - "reason": "Both mention Qwen/GPT-OSS style loading/runtime issues, but the concrete bug here is auto device mapping on Qwen3, not TP on GPT-OSS MoE." + "reason": "BeitImageProcessorFast label reduction bug and glm-4v-9b loading failure are unrelated." } ] }, @@ -4222,11 +4177,11 @@ "issue:43388|issue:45059", "issue:45406|issue:45464", "issue:33357|issue:35707", - "issue:44492|issue:44509", "issue:43421|issue:45701", "issue:43606|issue:43720", - "issue:30064|issue:34634", "issue:41720|issue:43746", + "issue:30064|issue:34634", + "issue:44492|issue:44509", "issue:43550|issue:43653", "issue:43873|issue:45704", "issue:44514|issue:44545", @@ -4242,20 +4197,20 @@ "issue:44898|issue:45072", "issue:44451|issue:44991" ], - "prepared_review_unit_hash": "a41525236d4f2eb9e8c505fc5d0514aa1567eda3841a9b76a40850b1232467e7", + "prepared_review_unit_hash": "f30c2c8bf4f5de9e851e3c32572edf06262e4d9386afbf94fe5ac5ccaa14a54b", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36655, + "serialized_chars": 36656, "estimated_input_tokens": 9164, "estimated_eval_tokens": 18584 }, @@ -4264,195 +4219,195 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is still mostly a set of unrelated Transformers issues that only share broad thematic similarity. The only clear duplicate pair is issue:44514/issue:44545. For the tokenization-regression subset, issue:45812 is a strong representative, but it should be described as one of the more referenced issues rather than the most referenced one.", - "confidence": 0.93, - "canonical_issue_reason": "issue:45812 is a strong canonical issue because it is open, specific, and describes a concrete silent AutoTokenizer regression in Granite models; it is one of the more referenced issues in the packet.", - "canonical_pr_reason": "No PRs are present in this cluster.", - "best_issue_reason": "issue:45812 is the best issue representative because it is the clearest concrete regression in the tokenization subset and can anchor that theme without overgeneralizing.", - "best_pr_reason": "No PRs are present in this cluster.", + "summary": "This cluster is mostly heterogeneous. Only the two Qwen2_5_VLProcessor apply_chat_template reports appear to be the same underlying bug; the rest are broad subsystem/model similarities, not true duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "issue:44514 and issue:44545 describe the same batched apply_chat_template crash with padding=False; 44514 is the earlier, more detailed report.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44514 is the best representative because it is the earliest duplicate report and contains the fuller reproduction context.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45081", "right": "issue:45812", "accept": false, - "reason": "Different tokenizer bugs: Mistral regex/backend-tokenizer crash vs Granite token-ID regression." + "reason": "Both are tokenizer-related, but they concern different models and different failure modes (Mistral regex patch crash vs Granite token IDs regression)." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Different config/model bugs: Qwen3-MoE sliding_window vs StableLmConfig pad_token_idx." + "reason": "Unrelated bugs: Qwen3-MoE sliding_window behavior vs StableLmConfig pad_token_idx regression." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Different Qwen failures and code paths: video-input processing vs weight_only loading." + "reason": "Different Qwen multimodal paths and different errors; not the same bug." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to the 3d_parallel.py runtime failure." + "reason": "GenerationMixin import error and a 3d_parallel.py runtime failure are unrelated." }, { "left": "issue:43421", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer-related, but runtime post-processor updates and Granite wrong token IDs are different bugs." + "reason": "Both touch tokenization, but one is runtime post-processor updates and the other is wrong token IDs for Granite; different underlying problems." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer crashes affecting different models and inputs." + "reason": "Both are tokenizer crashes, but on different tokenizer classes and different code paths." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "FP8Config behavior in Qwen3Moe is unrelated to the Llava Next load-time segfault." + "reason": "Qwen3Moe FP8 config failure and Llava Next segmentation fault are unrelated." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Trainer metric label handling and SAM3 text/bbox behavior are unrelated." + "reason": "Different subsystems: metric gathering for tuple labels vs SAM3 text/bbox behavior." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Different processors and failures: Gemma4Processor missing _tokenizer vs Qwen3.5 streaming API crash." + "reason": "Gemma4Processor missing _tokenizer and Qwen3.5 streaming inference failure are not the same bug." }, { "left": "issue:33357", "right": "issue:35707", "accept": false, - "reason": "MacOS bus error with CLIP is unrelated to progressive generation with inputs_embeds/past_key_values." - }, - { - "left": "issue:44492", - "right": "issue:44509", - "accept": false, - "reason": "A typo in cache strategy text is not the same as stale docs for removed pipeline tasks." + "reason": "MacOS CLIP bus error and progressive generation with inputs_embeds/past_key_values are unrelated." }, { "left": "issue:43421", "right": "issue:45701", "accept": false, - "reason": "Generic tokenization changes and runtime special-token post-processing are too broad to be the same bug." + "reason": "Too broad a match; one is a specific special-token post-processor bug, the other is a general tokenization version-change report." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "Different accelerate/offload bugs affecting different models and loading paths." + "reason": "Different loading/offload issues in different model paths; not mergeable as one fix." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "CUDA assert during auto device mapping and GraniteSpeech PEFT local checkpoint loading are unrelated." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Image segmentation-map processing and BarkProcessor voice_preset are unrelated." + "reason": "ImageProcessor void segmentation maps and BarkProcessor voice_preset are unrelated." }, { - "left": "issue:41720", - "right": "issue:43746", + "left": "issue:44492", + "right": "issue:44509", "accept": false, - "reason": "CUDA assert during auto device mapping is unrelated to PEFT checkpoint loading for GraniteSpeech." + "reason": "Both are documentation/cleanup flavored, but they are different edits in different areas and not the same bug." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "Torch.compile/SDPA failure and tokenizer special-token registration are unrelated." + "reason": "Different issues: torch.compile/SDPA failure vs tokenizer special-token registration." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantization offloading behavior is unrelated to the T5 apex RMSNorm memory leak." + "reason": "Quantization offloading behavior and T5 apex RMSNorm memory leak are unrelated." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same concrete crash in Qwen2_5_VLProcessor.apply_chat_template with batched input and padding=False." + "reason": "Same crash report: Qwen2_5_VLProcessor.apply_chat_template fails on batched input when padding=False." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Different classification bugs: num_labels propagation vs degenerate zero loss for num_labels=1." + "reason": "Both concern classification config, but one is missing num_labels propagation in Qwen3.5 and the other is a broader zero-loss issue across multiple models." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "CPU offload device mismatch is unrelated to BigBirdTokenizer mask token registration." + "reason": "CPU-offload device mismatch and BigBirdTokenizer decode output are different problems." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Sliding_window behavior in Qwen3-MoE is unrelated to from_config not unwrapping text_config." + "reason": "Different Qwen config/model bugs: sliding_window behavior vs text_config unwrapping for AutoModelForCausalLM.from_config." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Trailing-slash import warning is unrelated to the undefined-name video branch bug." + "reason": "Local-path trailing slash warning and undefined variables in multimodal token counting are unrelated." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "BarkProcessor voice_preset bug is unrelated to Qwen3 auto device mapping CUDA asserts." + "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping are unrelated." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token counting and progressive generation are different subsystems and failures." + "reason": "TrainerState token counter not updating and progressive generation with past_key_values are unrelated." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Runtime special-token post-processing and save/load losing custom token IDs are different bugs." + "reason": "Runtime special-token post-processor updates vs save/load losing custom token IDs are different issues." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "A tie_word_embeddings warning during LoRA fine-tuning is unrelated to incorrect visual encoder keys in save_pretrained." + "reason": "Tie-word-embeddings warning and incorrect visual encoder keys on save_pretrained are unrelated." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Different Qwen VL bugs: weight_only loading error vs video vision_position_ids mismatch." + "reason": "Different Qwen multimodal bugs: weight_only loading vs video vision_position_ids." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver resolution handling and dtype mismatch in inference are unrelated." + "reason": "Perceiver resolution/interpolation failure and bfloat16 dtype mismatch are unrelated." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Different tokenizer-loading regressions for different models; not the same underlying bug." + "reason": "Both are model/tokenizer loading regressions, but for different models and with no evidence of the same underlying failure." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster summary is mostly conservative, but the canonical/best-issue rationale overstates the evidence: issue:45812 does not have the most inbound references in the packet (issue:45701 has 6 vs 45812\u2019s 5). Please revise the representative selection or soften that claim. The duplicate call on issue:44514/44545 looks grounded." + "accept": true, + "feedback": "The cluster summary is well grounded in the packet: only issue:44514 and issue:44545 have clearly matching titles and dates support the choice of 44514 as the representative. The remaining pair verdicts are conservative and avoid overclaiming duplicates when the evidence is only broad subsystem/model similarity." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -4475,6 +4430,7 @@ "issue:39692", "issue:40444", "issue:42886", + "issue:42915", "issue:43381", "issue:43421", "issue:43525", @@ -4509,7 +4465,6 @@ "issue:45081", "issue:45183", "issue:45356", - "issue:45381", "issue:45479", "issue:45588", "issue:45663", @@ -4527,8 +4482,8 @@ "issue:44062|issue:44451", "issue:45704|issue:45923", "issue:39692|issue:44610", - "issue:44918|issue:44936", "issue:43421|issue:43653", + "issue:44918|issue:44936", "issue:45663|issue:45923", "issue:45588|issue:45759", "issue:42886|issue:44843", @@ -4543,239 +4498,236 @@ "issue:43746|issue:43873", "issue:43596|issue:44220", "issue:40444|issue:43746", - "issue:43746|issue:45381", + "issue:42915|issue:43746", "issue:43720|issue:43819", "issue:43540|issue:44008", "issue:44220|issue:44661", "issue:45183|issue:45479", - "issue:43618|issue:43697", "issue:30064|issue:36010", + "issue:43618|issue:43697", "issue:43653|issue:43720" ], - "prepared_review_unit_hash": "aa8273ba0c49a27ab7db44218cd286c8526dc51224cc6f35f92dc353e37fadd3", + "prepared_review_unit_hash": "fce444d8daf69ff3b5d28c9d2bf79c5e895cc735e6b610540967f22c19e39f48", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37377, - "estimated_input_tokens": 9345, - "estimated_eval_tokens": 18946 + "serialized_chars": 37340, + "estimated_input_tokens": 9335, + "estimated_eval_tokens": 18926 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most candidates are unrelated model/tokenizer bugs. The only strong duplicate-looking pair is the offline tokenizer regression in 42886 and 44843; the rest are distinct issues in different code paths.", - "confidence": 0.79, - "canonical_issue_reason": "44843 most directly identifies the root cause: `_patch_mistral_regex` unconditionally calling `model_info()`, which matches the offline/cache failure described in 42886.", + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, vision, multimodal, training, and loading/offline bugs. None of the soft pairs look like the same concrete fix target, so I rejected all merges.", + "confidence": 0.93, + "canonical_issue_reason": "issue:43421 is the closest thing to a hub here because it sits in the densest tokenizer-special-token subtheme, but the cluster is too mixed to treat it as a true duplicate canonical.", "canonical_pr_reason": null, - "best_issue_reason": "44843 is the clearest representative of the one real overlap in this cluster and gives the most actionable root-cause description.", + "best_issue_reason": "issue:43421 is the best representative of the most coherent subcluster (runtime tokenizer/post-processor behavior), but it should not be used as a global stand-in for the whole set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "Different models and failures: patchtsmixer `post_init` policy vs Qwen3.5 `num_labels` propagation." + "reason": "Different problems: optional post_init on patchtsmixer vs missing num_labels propagation in Qwen3.5 config." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Both are loading/processor issues, but they describe different bugs and code paths; not the same underlying change." + "reason": "Unrelated failures: loading physical-intelligence/fast vs PIL image processors wrongly requiring torchvision." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Different Qwen 3.5/3.6 problems: config unwrapping vs CPU-offload incompatibility in rotary embedding." + "reason": "Both involve Qwen 3.5, but one is composite config unwrapping and the other is CPU-offload rotary embedding behavior; not the same bug." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Unrelated failures: Qwen3-MoE sliding window logic vs Float8 storage deserialization." + "reason": "Qwen3-MoE sliding-window behavior is unrelated to the Float8 storage load error." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "A tokenizer constructor argument conflict is not the same bug as a model-specific load failure for ScandiBERT." + "reason": "AddedToken keyword-collision constructor bug is not the same as ScandiBERT loading regression." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "T5 RMSNorm memory leak and missing gradient-checkpointing flag are unrelated." + "reason": "T5 RMSNorm memory leak and missing supports_gradient_checkpointing are different code paths." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both involve vision processors, but the specific model mismatch and failure modes differ." + "reason": "SigLIP2 doc/example errors and OmDet-Turbo processor size mismatch are separate vision issues." }, { - "left": "issue:44918", - "right": "issue:44936", + "left": "issue:43421", + "right": "issue:43653", "accept": false, - "reason": "Different trainer failures: embedding unpacking with SFT trainer vs evaluate-after-train lifecycle issue." + "reason": "Both are tokenizer/special-token related, but one is post-processor refresh on runtime edits and the other is BigBird mask-token registration/empty decode; not the same concrete bug." }, { - "left": "issue:43421", - "right": "issue:43653", + "left": "issue:44918", + "right": "issue:44936", "accept": false, - "reason": "Both mention special tokens, but one is runtime post-processor updates and the other is BigBird mask-token registration." + "reason": "Qwen3.5 embedding unpacking in TRL SFT is unrelated to trainer.evaluate() failing after train()." }, { "left": "issue:45663", "right": "issue:45923", "accept": false, - "reason": "FSDP2 shared-kv state rebuilding and a trust_remote_code checkpointing flag are not the same bug." + "reason": "Gemma-4 FSDP2 shared_kv_states bug is unrelated to the missing gradient_checkpointing flag." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Flash-attention sink handling and multimodal config unwrapping are different code paths." + "reason": "flash_attention s_aux=None crash is not the same as Qwen 3.5 multimodal config unwrapping." }, { "left": "issue:42886", "right": "issue:44843", - "accept": true, - "reason": "Same offline tokenizer regression: both point to tokenizer loading failing because `_patch_mistral_regex`/`model_info()` is invoked in offline mode." + "accept": false, + "reason": "Both touch offline/tokenizer loading, but one is cache loading with HF_HUB_OFFLINE and the other is an unconditional model_info() call in _patch_mistral_regex." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Eval-mode checkpointing restriction is unrelated to ZeRO-3 gradient inflation during training." + "reason": "Eval-mode gradient checkpointing restriction is unrelated to exploding gradients under HfDeepSpeedConfig + ZeRO-3." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Both are tokenizer bugs, but the concrete failures differ and do not look like one fix." + "reason": "Special-token post-processor update bug is not the same as LayoutLMv2Tokenizer crashing on NER/padding/truncation." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different models and behaviors: SigLIP2 bad outputs vs CamemBERT MLM regression." + "reason": "Different models and symptoms: SigLIP2 nonsensical outputs vs CamemBERT masked-LM prediction regression." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Different subsystems: missing `pad_token_id` on Llama4Config vs Qwen3-MoE sliding-window handling." + "reason": "Missing pad_token_id on Llama4Config is unrelated to Qwen3-MoE sliding_window handling." }, { "left": "issue:45704", "right": "issue:45859", "accept": false, - "reason": "RMSNorm memory leak and CPU-offload rotary embedding are unrelated." + "reason": "Apex RMSNorm memory leak and Qwen3_5MoeTextRotaryEmbedding CPU offload compatibility are different defects." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both touch `_patch_mistral_regex`, but one is a crash on loading Mistral tokenizer and the other is a Kimi-K2.5 regression/warning issue; not clearly the same bug." + "reason": "Both mention fix_mistral_regex, but one is a tokenizer crash and the other is a Kimi-K2.5 codec/regression/warning issue; not the same fix target." }, { "left": "issue:45356", "right": "issue:45907", "accept": false, - "reason": "Tokenizer-regex regression vs `list_repo_files` revision forwarding are unrelated." + "reason": "Kimi tokenizer regression is unrelated to list_repo_files not forwarding revision." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "SigLIP2 example issues and quantization offloading are different problems." + "reason": "SigLIP2 example/doc issues are unrelated to quantization offloading behavior." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "Local checkpoint loading with PEFT adapters is unrelated to quantized offloading behavior." + "reason": "Local checkpoint PEFT loading and quantized offloading are different loading problems." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "DeepSpeed/BertModel initialization error vs audio feature extraction bug." + "reason": "DeepSpeed ZeRO-3/BertModel index error is unrelated to _torch_extract_fbank_features()." }, { "left": "issue:40444", "right": "issue:43746", "accept": false, - "reason": "Multimodal iterable dataset training failure is unrelated to GraniteSpeech checkpoint loading." + "reason": "IterableDataset with multiple images per prompt is unrelated to GraniteSpeech PEFT local checkpoint loading." }, { - "left": "issue:43746", - "right": "issue:45381", + "left": "issue:42915", + "right": "issue:43746", "accept": false, - "reason": "PEFT local checkpoint loading and Qwen2.5-VL video position IDs are different issues." + "reason": "Qwen3Moe FineGrainedFP8Config failure is unrelated to GraniteSpeech local-checkpoint loading." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "BitNet packed-weight loading and DAC `from_latents`/STE mismatch are unrelated." + "reason": "BitNet packed-weight loading and DAC.from_latents/forward mismatch are unrelated model bugs." }, { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Qwen3OmniMoe video-processing error and Gemma 3n variable collision are different bugs." + "reason": "Both are multimodal, but one is Qwen3OmniMoe video-input processing and the other is Gemma 3n audio tensor name collision; not the same bug." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Audio feature extraction and `add-new-model-like` tokenizer mapping logic are unrelated." + "reason": "Audio fbank extraction failure is unrelated to add-new-model-like failing inside TOKENIZER_MAPPING_NAMES." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Input validation messaging and sequence-classification zero-loss behavior are separate issues." + "reason": "Input validation XOR messaging and single-label-classification zero-loss behavior are different issues." }, { - "left": "issue:43618", - "right": "issue:43697", + "left": "issue:30064", + "right": "issue:36010", "accept": false, - "reason": "CLIP attention output regression and RTDetrV2 output drift are different model behaviors." + "reason": "Image segmentation-map processing and GenerationMixin import error are unrelated." }, { - "left": "issue:30064", - "right": "issue:36010", + "left": "issue:43618", + "right": "issue:43697", "accept": false, - "reason": "Void segmentation-map processing and a `GenerationMixin` import error are unrelated." + "reason": "CLIPOutput attentions assignment and RTDetrV2 output differences in v5 are not the same code-path bug." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "BigBird special-token decode bug and BitNet packed-weight loading are unrelated." + "reason": "Tokenizer special-token registration and BitNet accelerate loading are unrelated." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The analysis is conservative and grounded in the packet. The single accepted overlap, 42886/44843, is directly supported by the titles: both concern `_patch_mistral_regex`/`model_info()` breaking offline tokenizer loading. The remaining soft-edge rejections are appropriately cautious and do not overstate evidence for duplicate bugs or mergeable PRs." - }, - "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 }, @@ -4842,8 +4794,8 @@ ], "soft_pairs": [ "issue:43335|issue:43756", - "issue:43716|issue:44008", "issue:43295|issue:45834", + "issue:43716|issue:44008", "issue:43550|issue:45072", "issue:43550|issue:43819", "issue:44062|issue:44488", @@ -4854,8 +4806,8 @@ "issue:43531|issue:43643", "issue:43653|issue:43819", "issue:43643|issue:44877", - "issue:43577|issue:43818", "issue:43421|issue:44811", + "issue:43577|issue:43818", "issue:43540|issue:45381", "issue:43782|issue:44977", "issue:44448|issue:44671", @@ -4865,23 +4817,23 @@ "issue:39692|issue:45834", "issue:44821|issue:45084", "issue:30064|issue:36331", - "issue:41720|issue:44991", - "issue:44485|issue:44509" + "issue:44485|issue:44509", + "issue:41720|issue:44991" ], - "prepared_review_unit_hash": "2c6a970724dfc40de87dc9b6a17b08caab0a1433266bd724db94ac812edf2fc3", + "prepared_review_unit_hash": "5c958e05d6399e8e4e1240da7182bbb34aa49c7bfd0185bff7b9fb787f45d314", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 35169, + "serialized_chars": 35170, "estimated_input_tokens": 8793, "estimated_eval_tokens": 17842 }, @@ -4890,168 +4842,168 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issue reports across model-specific regressions, tokenizer/config edge cases, and docs/errors. No soft pair looks like the same concrete bug, though several are loosely in the remote_code / model-loading regression family.", - "confidence": 0.94, - "canonical_issue_reason": null, + "summary": "The cluster is highly heterogeneous: most items are distinct regressions across different models, tokenizers, and APIs. None of the soft edges look like the same underlying bug closely enough to merge.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45020 is the broadest umbrella report ('Recent transformers versions break models using `remote_code`'), so it is the most plausible single anchor among otherwise distinct issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the broadest representative of the recurring remote_code/model-loading regression family; it best anchors the cluster even though the overall set is heterogeneous.", + "best_issue_reason": "issue:45020 is the best representative by scope and potential to subsume related remote_code/config-loading regressions, though the cluster is not truly duplicate-like.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "Different model bugs: sparse-layer construction vs RoPE-layer dropping." + "reason": "Different bugs: sparse-layer creation in SwitchTransformers vs RoPE layer-dropping in Smollm3." }, { - "left": "issue:43716", - "right": "issue:44008", + "left": "issue:43295", + "right": "issue:45834", "accept": false, - "reason": "Different code paths and models: image preprocessor dtype mismatch vs Gemma3n variable-name collision." + "reason": "Both are multimodal, but one is a processor/tokenizer regression and the other is a long OCR input index error in Kosmos2.5." }, { - "left": "issue:43295", - "right": "issue:45834", + "left": "issue:43716", + "right": "issue:44008", "accept": false, - "reason": "Both are processor-related, but one is a regression in processor/tokenizer access while the other is a long-OCR index error in Kosmos2.5." + "reason": "Unrelated failures: image preprocessor/model dtype mismatch vs Gemma3n audio variable-name collision." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Different models and failures: Bamba torch.compile+SDPA vs SwitchTransformers/TimmWrapper bfloat16 mismatches." + "reason": "Different code paths: torch.compile+SDPA failure in Bamba-9B-v2 vs bfloat16 dtype mismatch in SwitchTransformers/TimmWrapperModel." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Unrelated bugs in different subsystems: Bamba compilation failure vs DAC latent reconstruction mismatch." + "reason": "Different underlying problems: Bamba-9B-v2 compile/SDPA issue vs DAC.from_latents not matching forward pass." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "Both touch tokenizers, but one is an AddedToken argument error and the other is a tokenizer load failure for a specific model." + "reason": "Tokenizer-related but not the same bug: duplicate AddedToken kwargs vs loading a specific tokenizer model." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Different backend/runtime issues: Llava Next load segfault vs MPS histogram not implemented for int." + "reason": "Completely different subsystems and failures: Llava Next load segfault vs MPS histogram int support in MoE." }, { "left": "issue:43950", "right": "issue:45987", "accept": false, - "reason": "Both are API regressions, but one corrupts non-persistent buffers and the other returns the wrong type from StaticCache.get_seq_length()." + "reason": "One is from_pretrained buffer corruption; the other is a StaticCache return-type contract bug." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Different models and symptoms: LayoutLMv2 tokenizer crashes vs Perceiver image-classification failures." + "reason": "Different areas: LayoutLMv2 tokenization/padding crash vs Perceiver image classification failing at non-default resolution." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Same helper area (_patch_mistral_regex), but these are different failures: offline model_info access vs missing backend_tokenizer attribute." + "reason": "Both involve Mistral regex patching, but one is offline Hub access and the other is an AttributeError on a different tokenizer object shape." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Unrelated: Qwen3-MoE sliding-window behavior vs trust_remote_code config field loss." + "reason": "Distinct issues: Qwen3-MoE sliding_window behavior vs AutoConfig trust_remote_code missing fields." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Different bugs: BigBird special-token registration vs DAC forward/from_latents mismatch." + "reason": "BigBird special-token registration bug is unrelated to DAC.from_latents forward-pass mismatch." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "Both involve config loading, but one is missing fields under trust_remote_code and the other is strict config rejection for granite_speech." + "reason": "Both are config-loading problems, but one is missing fields from trust_remote_code and the other is strict config rejection for granite_speech." }, { - "left": "issue:43577", - "right": "issue:43818", + "left": "issue:43421", + "right": "issue:44811", "accept": false, - "reason": "Different model-specific issues: BLIP2 dtype propagation vs Video-LLaVA temporal-attention/weight-sharing problem." + "reason": "One is runtime post-processor updates for special tokens; the other is Whisper batch_decode ignoring skip_special_tokens." }, { - "left": "issue:43421", - "right": "issue:44811", + "left": "issue:43577", + "right": "issue:43818", "accept": false, - "reason": "Both involve token handling, but one is runtime post-processor updates and the other is Whisper batch_decode skip_special_tokens behavior." + "reason": "Different model bugs: BLIP-2 dtype propagation vs Video-LLaVA missing temporal attention/shared weights." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both are video-input bugs in Qwen models, but they appear to be distinct failures with different versions and code paths." + "reason": "Both are video-input issues, but they affect different Qwen model families and different internal failures." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Different Qwen models and failures: weight_only loading error vs flash-attention generation issue." + "reason": "Qwen3VL loading with weight_only=True is not the same as Qwen3.5 flash-attention generation failure." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Both are output regressions, but they affect different models/tasks and do not share a concrete code path." + "reason": "Both are v5 output regressions, but for different models and different code paths." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Unrelated: Whisper empty-transcription generation vs Kosmos2.5 long-input index error." + "reason": "Whisper empty-transcription generation bug is unrelated to Kosmos2.5 long-OCR index errors." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "Different issues in the pipeline stack: a bad error message vs incorrect/nonsensical model outputs." + "reason": "One is an error-message/docs mismatch in pipeline(); the other is an inference correctness issue for SigLIP2." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Related theme, but not the same concrete bug: one is a specific is_torch_fx_available regression, the other is a broader remote_code breakage report." + "reason": "Related theme, but not the same concrete bug: a specific missing helper breaks trust_remote_code models, while the other is a broad umbrella report." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "Different problems: SigLIP2 docs/example errors vs Kosmos2.5 OCR long-input failure." + "reason": "Different models and failures: SigLIP2 docs/example errors vs Kosmos2.5 OCR index error." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "Different failures: AutoImageProcessor URL loading vs template-node compilation error." + "reason": "AutoImageProcessor URL loading and template-node compilation are unrelated failure modes." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Completely unrelated: segmentation-map processing vs CustomTrainer compute_loss signature mismatch." + "reason": "Unrelated bugs: void segmentation map processing vs CustomTrainer.compute_loss signature regression." }, { - "left": "issue:41720", - "right": "issue:44991", + "left": "issue:44485", + "right": "issue:44509", "accept": false, - "reason": "Different models and subsystems: Qwen3 auto device mapping crash vs tokenizer loading regression for EMBEDDIA/est-roberta." + "reason": "GLM-5 RoPE implementation discussion is unrelated to v5 pipeline task documentation leftovers." }, { - "left": "issue:44485", - "right": "issue:44509", + "left": "issue:41720", + "right": "issue:44991", "accept": false, - "reason": "GLM-5 RoPE discussion is unrelated to docs still mentioning removed pipeline tasks." + "reason": "Different failures in different components: Qwen3 device-mapping CUDA assert vs tokenizer loading for est-roberta." } ] }, @@ -5149,207 +5101,207 @@ "issue:45459|issue:45803", "issue:34634|issue:36331", "issue:45084|issue:45538", - "issue:43653|issue:44912", "issue:43742|issue:45020", + "issue:43653|issue:44912", "issue:43329|issue:45600", "issue:43638|issue:44936" ], - "prepared_review_unit_hash": "6de2f9fdad2a73d5134527e1afc5b2ee8da69ab339b3b9caccfed4d27b8fc7c1", + "prepared_review_unit_hash": "3a95c784ed6dc7a6c761780fde3572c79ce68b114f166ca271787979b60ee35a", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36473, - "estimated_input_tokens": 9119, - "estimated_eval_tokens": 18494 + "serialized_chars": 36471, + "estimated_input_tokens": 9118, + "estimated_eval_tokens": 18492 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers issues. The soft links mostly share broad themes like loading, tokenization, multimodal inputs, or dtype regressions, but not the same concrete bug or code path.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue stands out: the cluster spans many unrelated subsystems and failure modes, so any representative would be arbitrary.", + "summary": "This cluster is mostly a loose set of unrelated bug reports that only share broad themes like loading, tokenizers, multimodal models, or v5 regressions. None of the soft pairs look like the same underlying bug/change, so all soft edges should be rejected. No PRs are present; the most representative issue is the broad v5/regression report around remote_code loading.", + "confidence": 0.59, + "canonical_issue_reason": "issue:45020 is the broadest, most central issue in the set: it describes recent transformers versions breaking remote_code models, which matches the general regression-focused nature of the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "No strong global best issue for deduping; the items are too heterogeneous to anchor a single duplicate cluster.", + "best_issue_reason": "issue:45020 is open, broad, and impacts multiple models/paths, making it the best representative issue for the cluster even though the rest are mostly unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Both involve loading/memory behavior, but one is junk non-persistent buffers and the other is broken FSDP CPU RAM-efficient loading; different bugs." + "reason": "Both are loading/runtime regressions, but one is about non-persistent buffers and the other about FSDP CPU RAM efficient loading; different failure modes and code paths." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "Tokenizer-related failures, but different tokenizer classes and different crash conditions; not the same code path." + "reason": "Both are tokenizer issues, but they affect different tokenizers and different bugs (NER/batched padding vs AutoTokenizer load failure)." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Both are multimodal input bugs, but they affect different model families and different preprocessing/indexing paths." + "reason": "Both are multimodal input issues, but one is a v5.3.0 video regression in Qwen VL models and the other is a long OCR index error in Kosmos2.5." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "Completely different problems: BLIP2 dtype handling vs Apple Silicon allocator warmup TypeError." + "reason": "Blip2 dtype handling and Apple Silicon allocator warmup are unrelated bugs." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "Both mention inference mismatches, but they are model-specific and caused by different mechanisms." + "reason": "Both mention inference/loading problems, but one is a Bark CPU offload device mismatch and the other is a dtype mismatch in different models." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Related to v5 pipeline removals, but one is a runtime error message and the other is docs cleanup; not one bug." + "reason": "The first is an incorrect pipeline error message; the second is outdated docs. Related topic, but not the same underlying bug." }, { "left": "issue:44805", "right": "issue:44936", "accept": false, - "reason": "Generic runtime errors in training/eval, but no shared concrete defect or code path." + "reason": "Different failures: a mask/tensor shape IndexError versus trainer.evaluate breaking after train()." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "One is a CamemBERT masked-LM regression; the other is a broad tokenization change report." + "reason": "One is incorrect masked LM predictions for CamemBERT; the other is a generic tokenization-change report." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "Unrelated: Bark voice_preset issue vs GenerationMixin import error." + "reason": "Bark voice_preset and GenerationMixin import failure are unrelated." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Different models and different failures; input-size mismatch is unrelated to dtype mismatch." + "reason": "OmDet-Turbo image size mismatch and dtype mismatches in other models are different issues." }, { "left": "issue:43782", "right": "issue:45325", "accept": false, - "reason": "Both are Qwen VL issues, but they concern different functions and distinct bugs." + "reason": "Both are Qwen VL-related, but one is a weight_only load error and the other is a rope-index temporal scaling bug." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "Packed-weight loading failure is unrelated to bfloat16 dtype mismatch." + "reason": "BitNet packed-weight loading failure and bfloat16 dtype mismatches are unrelated." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Backend packaging/install issue vs tokenizer regex patch crash; unrelated." + "reason": "An installation/backend packaging issue is not the same as a tokenizer regex patch crash." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Both are attribute errors during loading, but they hit different objects and different fixes." + "reason": "Both are AttributeErrors during loading, but they occur in different components and for different missing attributes." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "Both are model-loading problems, but the root causes are different initialization/loading APIs." + "reason": "FSDP CPU RAM loading and init_empty_weights/_is_hf_initialized are separate loading bugs." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "The first is a nonspecific title; the second is a reproduction of another issue, not the same bug." + "reason": "A placeholder 'ADD THE DATA' issue is unrelated to an isolated reproduction report." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Both mention tied-weights warnings, but the specific warning and root cause differ." + "reason": "Both mention tied-weights warnings, but one is a concrete wrong tie in Qwen3-Next and the other is a finetuning warning about config cleanup." }, { "left": "issue:43582", "right": "issue:44857", "accept": false, - "reason": "Allocator warmup TypeError on Apple Silicon is unrelated to AMP/CUDA loss crash." + "reason": "Apple Silicon allocator warmup and CUDA AMP loss crashes are unrelated." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Both are audio-related, but one is chat-template sampling rate and the other is DAC latent reconstruction." + "reason": "Audio processor chat-template sampling-rate handling and DAC latent reconstruction are different audio bugs." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "Vision-model regressions, but different components and different symptoms." + "reason": "BEiT label reduction and RTDetrV2 output drift are different vision bugs." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "No meaningful overlap beyond being correctness issues." + "reason": "Image label reduction and an MPS attention correctness issue do not describe the same problem." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve exception handling that hides errors, but they are different functions and different failure modes." + "reason": "Both are error-handling bugs, but they affect different code paths and different swallowed errors." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "Completely unrelated bug reports." + "reason": "BarkProcessor voice preset and CustomTrainer.compute_loss signature mismatch are unrelated." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template compilation error vs tokenizer model_max_length issue; unrelated." + "reason": "A template compilation TypeError is unrelated to CLIPTokenizer model_max_length behavior." }, { - "left": "issue:43653", - "right": "issue:44912", + "left": "issue:43742", + "right": "issue:45020", "accept": false, - "reason": "Tokenizer special-token bug vs quantization fallback bug; no shared underlying defect." + "reason": "A key error loading one model is not the same underlying issue as broad remote_code breakage." }, { - "left": "issue:43742", - "right": "issue:45020", + "left": "issue:43653", + "right": "issue:44912", "accept": false, - "reason": "Specific model loading key error vs broad remote_code regression claim; too different to be duplicates." + "reason": "BigBirdTokenizer special-token registration and MXFP4 quantization fallback are unrelated." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Both are CI/regression flavored, but they affect different modules and unrelated code paths." + "reason": "A multimodal token counting bug and stale auto_mappings references are different regressions." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Different training/evaluation failures with no shared root cause." + "reason": "DeepSpeed zero3 indexing failure and trainer.evaluate-after-train failure are different training bugs." } ] }, @@ -5451,217 +5403,235 @@ "issue:43606|issue:45859", "issue:44849|issue:45701", "issue:43697|issue:45357", - "issue:44987|issue:45538", - "issue:44291|issue:45042" + "issue:44291|issue:45042", + "issue:44987|issue:45538" ], - "prepared_review_unit_hash": "b0878c291f3d3af8a807a818a106e91faf5316d082284c0215ebf9d527d2ec82", + "prepared_review_unit_hash": "716667f9d1bcdc8f0b26c981c067718a21d3c1308281a83bc139da5e3d9daa70", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36973, - "estimated_input_tokens": 9244, - "estimated_eval_tokens": 18744 + "serialized_chars": 36979, + "estimated_input_tokens": 9245, + "estimated_eval_tokens": 18746 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft pair looks like the same underlying bug or change; most are only loosely related by subsystem (tokenization, video, loading, offload, etc.). If a single issue must anchor the cluster, the broad open tokenization regression report is the best representative.", - "confidence": 0.83, - "canonical_issue_reason": "issue:45701 is the broadest, most umbrella-like report in the set, with the most inbound references and a title that can absorb nearby tokenization regressions better than the more narrowly scoped bugs.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers regressions that only share broad vocabulary like tokenization, loading, video, or offload; the soft pairs do not look like true duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45701 is the broadest, most referenced representative of the loose tokenization-regression subset; most other issues here are narrower or unrelated.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the best overall representative because it is open, broad in scope, and appears most central to the tokenization-related slice of this cluster.", + "best_issue_reason": "Issue 45701 is the best umbrella representative because it is open, has the most inbound references, and its title matches the cluster\u2019s broad tokenization theme better than the more model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "Different failure modes: gguf_file handling for from_pretrained vs a tokenizer-loading regression for a specific model." + "reason": "Both involve loading tokenizers/models, but the failure modes and models differ." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "Unrelated bugs: PEFT/local checkpoint loading vs Whisper batch_decode skip_special_tokens behavior." + "reason": "PEFT local-checkpoint loading and Whisper batch_decode are unrelated bugs." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Different subsystems and symptoms: float8 storage lookup vs PIL image processors requiring torchvision." + "reason": "Float8 storage lookup and PIL image processor torchvision requirements are different issues." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Both are vision-related, but one is zero probabilities in SigLIP2 and the other is an OmDet processor/model image-size mismatch." + "reason": "Both are vision-related, but one is output/probability behavior and the other is processor image size mismatch." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "Different models and issues: GLM-4.7-Flash loading test failure vs Qwen3.5 flash-attention generation regression." + "reason": "Different models and different failure modes; not the same code-path bug." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "DAC latent reconstruction mismatch is unrelated to DiaConfig losing custom token IDs after save/load." + "reason": "DAC forward mismatch and DiaConfig save/load token IDs are unrelated." }, { "left": "issue:39692", "right": "issue:44912", "accept": false, - "reason": "Both mention quantization, but one is a SigLIP2 docs example with model/processor mismatch and the other is a git-oss-20b MXFP4 loading fallback." + "reason": "Docs/example errors and MXFP4 quantization fallback are separate problems." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Distinct code paths: Qwen3 MoE accelerate/deepspeed weight conversion vs multimodal token counting in a video branch." + "reason": "Qwen3 MoE loading/conversion and multimodal video token counting are different bugs." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "ImportError for GenerationMixin is not the same as notebook-specific custom model initialization breakage." + "reason": "Both are version-regression style reports, but they hit different APIs and causes." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Different model/runtime problems: torch.compile with SDPA vs MXFP4 quantization fallback on a different model." + "reason": "torch.compile SDPA failure and quantization fallback are unrelated." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "Packed-weight unpacking during accelerate loading is unrelated to an image processor output-size mismatch." + "reason": "BitNet accelerate loading and OmDet-Turbo processor sizing do not share a concrete bug." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "MoE weight conversion/loading bug is not the same as a CUDA OOM report for GPT-oss-20b." + "reason": "MoE weight conversion and OOM are distinct issues." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "AyaVision lm_head tying issue is unrelated to Blip2 dtype propagation for the main model and qformer." + "reason": "lm_head tying failure and dtype staying float32 are different model-loading bugs." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "Related video area, but not the same bug: undefined helpers in token counting vs a broader Qwen video-input regression." + "reason": "Both are video-related, but one is undefined helper usage and the other is a broader regression." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE has no connection to PIL backend torchvision requirements." + "reason": "Qwen3-MoE sliding_window and PIL/torchvision image processing are unrelated." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting is unrelated to Whisper batch_decode token-skipping." + "reason": "Audio chat-template sampling-rate handling and Whisper batch_decode special-token handling differ." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Local-path trailing-slash import warning is not the same as Qwen3 MoE accelerate/deepspeed weight conversion." + "reason": "Trailing-slash path warning and Qwen3 MoE weight conversion are unrelated." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "Importing GenerationMixin and compute_loss signature changes are separate trainer/library issues." + "reason": "GenerationMixin import failure and custom_trainer loss signature breakage are separate API regressions." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Different tokenizer regressions affecting different models and behaviors: BOS/EOS insertion vs codec/warning handling." + "reason": "Tokenizer special-token insertion and Kimi codec handling are different tokenizer bugs." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "FSDP CPU RAM-efficient loading breakage is not the same as disabling mmap to avoid OOM on Strix Halo." + "reason": "FSDP RAM-efficient loading and mmap OOM on Strix Halo are different loading/resource issues." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "Missing fields from trust_remote_code config loading is distinct from unwrapping text_config for composite multimodal configs." + "reason": "Missing config fields with trust_remote_code and text_config unwrapping are not the same defect." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both are tokenization regressions, but they affect different tokenizer implementations and different concrete behaviors." + "reason": "BOS/EOS insertion failure and pre-tokenizer override are different tokenization behaviors." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "TrainerState token counters and GPT-oss-20b perplexity are unrelated problems." + "reason": "Trainer token-count tracking and GPT-oss perplexity are unrelated." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Undefined helpers in video token counting are different from batched video processor shape errors." + "reason": "Video token counting undefined helpers and batched video processor shape are different bugs." }, { "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "MoE weight conversion during accelerate/deepspeed loading is separate from Qwen3-VL weight-shape mismatch on load." + "reason": "Qwen3 MoE loading conversion and Qwen3-VL weight-shape mismatch are separate model-loading issues." }, { "left": "issue:43606", "right": "issue:45859", "accept": false, - "reason": "Both mention CPU offload, but they involve different models and different failure mechanisms." + "reason": "Both mention CPU offload, but the model paths and concrete failures differ." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Qwen3.5 hidden-state output bug is not the same as a broader tokenization change regression." + "reason": "Hidden-states output bug and tokenization version change are unrelated." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RTDetrV2 output differences and Qwen3.5 save_pretrained visual-key regression are unrelated model bugs." + "reason": "RTDetrV2 output differences and Qwen3.5 save_pretrained key issues do not match." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized and PIL/torchvision are different regressions." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "A model-loading failure for physical-intelligence/fast is not the same as CLIPTokenizer's model_max_length bug." + "reason": "Model loading failure and CLIPTokenizer max_length bug are unrelated." }, { - "left": "issue:44291", - "right": "issue:45042", + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Tokenizer special-token insertion and ByteLevel-to-Metaspace override are different defects." + }, + { + "left": "issue:44991", + "right": "issue:45538", + "accept": false, + "reason": "Different tokenizers and different symptoms; not the same bug." + }, + { + "left": "issue:45488", + "right": "issue:45701", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized TypeError is unrelated to PIL backend processors depending on torchvision." + "reason": "Broad tokenization regression vs specific LlamaTokenizer pre-tokenizer override; too imprecise to merge." } ] }, @@ -5734,14 +5704,14 @@ "issue:45987" ], "soft_pairs": [ - "issue:43257|issue:43756", - "issue:44479|issue:45381", "issue:42915|issue:43742", + "issue:44479|issue:45381", + "issue:43257|issue:43756", "issue:41720|issue:43540", "issue:43452|issue:43582", - "issue:44987|issue:45907", - "issue:33357|issue:34567", "issue:42915|issue:44821", + "issue:33357|issue:34567", + "issue:44987|issue:45907", "issue:42915|issue:45538", "issue:43927|issue:44912", "issue:43317|issue:45538", @@ -5760,26 +5730,26 @@ "issue:43742|issue:44291", "issue:42915|issue:43749", "issue:44737|issue:45588", - "issue:43526|issue:45479", "issue:43540|issue:45823", - "issue:43756|issue:45325", + "issue:43526|issue:45479", "issue:43582|issue:43782", + "issue:43756|issue:45325", "issue:28282|issue:29127" ], - "prepared_review_unit_hash": "314aaaeef15f08bed36cc02b76409f702aea634ba7c99538d92ca4ee87491c90", + "prepared_review_unit_hash": "417844dc8116336f94f910c4d7ed4ee7d432c5b9e9013adb5f16c39e72c79e1e", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36533, + "serialized_chars": 36534, "estimated_input_tokens": 9134, "estimated_eval_tokens": 18524 }, @@ -5788,207 +5758,204 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is still mostly a grab bag of distinct Transformer bugs. Most soft edges are appropriately rejected; the only correction is to treat 44479 \u2194 45381 as merely related Qwen VL video-input regressions, not confirmed duplicates.", - "confidence": 0.69, - "canonical_issue_reason": "Issue 44479 remains the broadest representative of the Qwen video-input regression cluster, but the evidence is not strong enough to collapse 45381 into it as the same underlying bug.", + "summary": "This cluster is heterogeneous, with a loose concentration around Transformers model-loading and Qwen/VL video-input regressions, but no pair is convincingly the same underlying bug/change.", + "confidence": 0.31, + "canonical_issue_reason": "issue:44479 is the broadest representative of the recurring Qwen/VL video-input regression theme, with a clear concrete failure surface and the widest scope among the related reports.", "canonical_pr_reason": null, - "best_issue_reason": "44479 is still the best cluster representative because it captures the broad v5.3.0 Qwen VL video regression framing more generally than 45381, even though the two should only be considered possibly related.", + "best_issue_reason": "issue:44479 is the most representative issue for this cluster\u2019s strongest recurring theme (video-input regressions in Qwen/VL models).", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43257", - "right": "issue:43756", + "left": "issue:42915", + "right": "issue:43742", "accept": false, - "reason": "Different bugs: MoE weight conversion during load vs Smollm3 RoPE layer selection." + "reason": "Different models and failures: Qwen3MoE FP8 config vs MobileLLM key error; same loading area only." }, { "left": "issue:44479", "right": "issue:45381", "accept": false, - "reason": "The titles point to related Qwen2.5-VL video-input regressions, but the packet does not show enough shared traceback, code path, or concrete overlap to call them the same bug." + "reason": "Both involve Qwen/VL video input, but one is a broad multi-model regression and the other is a specific vision_position_ids bug; not clearly the same root cause." }, { - "left": "issue:42915", - "right": "issue:43742", + "left": "issue:43257", + "right": "issue:43756", "accept": false, - "reason": "Unrelated failures: Qwen3Moe FineGrainedFP8Config vs MobileLLM key error on load." + "reason": "Different subsystems: MoE weight conversion with accelerate/deepspeed vs Smollm3 RoPE layer handling." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Both Qwen-related, but different code paths and failures: auto device mapping CUDA assert vs video-input ValueError." + "reason": "Different Qwen3 failures: auto device mapping cudaErrorAssert vs video-input ValueError in Qwen3OmniMoe." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "Tokenizer/model gguf loading issue vs Apple Silicon allocator warmup TypeError; not the same bug." + "reason": "Unrelated issues: gguf tokenizer/model loading vs Apple Silicon allocator warmup TypeError." }, { - "left": "issue:44987", - "right": "issue:45907", + "left": "issue:42915", + "right": "issue:44821", "accept": false, - "reason": "Different concerns: model loading failure vs repo file listing not forwarding revision." + "reason": "Different problem areas: Qwen3MoE FP8 loading vs AutoImageProcessor URL loading." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "Mac CLIP bus error and TrainerState token counting are unrelated." + "reason": "Different bugs: MacOS CLIP bus error vs TrainerState token counter not updating." }, { - "left": "issue:42915", - "right": "issue:44821", + "left": "issue:44987", + "right": "issue:45907", "accept": false, - "reason": "Qwen3Moe FP8 load failure vs AutoImageProcessor URL loading problem." + "reason": "Different failures: model loading regression vs list_repo_files needing revision forwarding." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "FP8 config failure vs CLIPTokenizer max_length behavior; different artifacts and code paths." + "reason": "Different areas: Qwen3MoE FP8 loading vs CLIPTokenizer max-length constant." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "Config save/load token ID loss is not the same as MXFP4 quantization fallback." + "reason": "Different concrete bugs: DiaConfig save/load token IDs vs MXFP4 quantization fallback behavior." }, { "left": "issue:43317", "right": "issue:45538", "accept": false, - "reason": "device_map/offload loading failure vs tokenizer max_length constant; unrelated." + "reason": "Different failures: device_map/offload load path vs tokenizer max_length constant." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Packaging/backend availability issue vs missing Qwen2_5_VL class import." + "reason": "Different issues: missing TensorFlow backend in torch install vs missing Qwen2_5_VL import." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both video-related, but different models and errors; not enough to treat as the same concrete bug." + "reason": "Both touch video input, but they target different model families and distinct failure modes." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Different video-input failures in different Qwen variants; no clear shared code-path bug from the titles alone." + "reason": "Different Qwen video bugs: Qwen3OmniMoe video ValueError vs Qwen3-vl-embedding StopIteration." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "AddedToken special-arg error vs tokenizer loading regression for a specific model." + "reason": "Different tokenizer problems: AddedToken keyword collision vs tokenizer loading failure for EMBIDDIA/est-roberta." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "BLIP2 dtype persistence bug and CI dtype mismatches in other models are not the same issue." + "reason": "Both are dtype-related, but one is BLIP2 loading dtype propagation and the other is inference dtype mismatch in different models." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "Backend-install packaging problem vs Python 3.14 import failure." + "reason": "Different compatibility bugs: install extras/backend packaging vs Python 3.14 import/load failure." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "Unexpected zero probabilities and Perceiver interpolation failure are distinct model behaviors." + "reason": "Different vision-model bugs: SigLIP2 zero probabilities vs Perceiver non-default resolution handling." }, { "left": "issue:43404", "right": "issue:45987", "accept": false, - "reason": "lm_head tying bug vs StaticCache seq-length return type mismatch." + "reason": "Different code paths: tied lm_head weights in Mistral3 vs StaticCache return type contract." }, { "left": "issue:42175", "right": "issue:44488", "accept": false, - "reason": "Package backend composition issue vs inability to load a specific model." + "reason": "Different failures: backend packaging omission vs model loading failure for cjvt/sleng-bert." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 deque error is unrelated to the PEFT version pinning problem." + "reason": "Different bugs: DeepSpeed ZeRO3 deque error vs PEFT version pin mismatch." }, { "left": "issue:43906", "right": "issue:45381", "accept": false, - "reason": "One is a repro of an older issue; the other is a specific Qwen2.5-VL video regression." + "reason": "No evidence they share the same root cause; isolated reproduction of another issue vs Qwen2.5-VL video position_ids bug." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "Missing PyTorch ImportError vs Llama 3.2 Vision loading regression are different causes." + "reason": "Different import/load problems: missing PyTorch dependency vs a specific Llama 3.2 Vision loading regression." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "MobileLLM key error and unexpected _is_hf_initialized argument are separate load-time bugs." + "reason": "Both involve model loading, but one is a KeyError on a specific model and the other is a generic init_empty_weights argument mismatch." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Qwen3Moe FP8 issue and FSDP CPU RAM efficient loading are unrelated." + "reason": "Different failures: Qwen3MoE FP8 config vs FSDP CPU RAM efficient loading." }, { "left": "issue:44737", "right": "issue:45588", "accept": false, - "reason": "XLNet CPU arange issue vs flash_attention sink-less crash are distinct implementation bugs." - }, - { - "left": "issue:43526", - "right": "issue:45479", - "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to the sequence-classification zero-loss issue." + "reason": "Different bugs in different areas: XLNet CPU tensor creation vs flash_attention sink handling." }, { "left": "issue:43540", "right": "issue:45823", "accept": false, - "reason": "Qwen3OmniMoe video processing error and Gemma4 device mismatch are different failures." + "reason": "Different model families and symptoms: Qwen3OmniMoe video processing vs Gemma4 device mismatch." }, { - "left": "issue:43756", - "right": "issue:45325", + "left": "issue:43526", + "right": "issue:45479", "accept": false, - "reason": "Smollm3 RoPE-layer behavior and Qwen2.5-VL temporal position scaling are unrelated." + "reason": "Different semantics: BeitImageProcessorFast label reduction vs sequence-classification loss degeneracy." }, { "left": "issue:43582", "right": "issue:43782", "accept": false, - "reason": "Apple Silicon caching allocator TypeError vs Qwen3VL weight_only load error are different bugs." + "reason": "Unrelated: Apple Silicon warmup TypeError vs Qwen3VL from_pretrained weight_only error." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Different RoPE-related issues in different models; no evidence of a shared underlying bug." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "AutoModel missing PyTorch and LayoutLMv3 box-information message clarity are unrelated." + "reason": "Different issues: missing PyTorch ImportError vs LayoutLMv3 box-data error message clarity." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The cluster-level summary is mostly grounded, but one soft-edge verdict is too strong for the evidence in the packet. The 44479 \u2194 45381 link is plausible from the titles, yet the packet does not provide enough direct overlap to assert they are the same underlying bug; this should be framed more conservatively (e.g. \"likely related\" / \"possible duplicate\") or left unaccepted. The rest of the verdicts look appropriately conservative." - }, - "evaluator_used": true, - "retried": true, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -6071,218 +6038,218 @@ "issue:43697|issue:45216", "issue:43688|issue:43749", "issue:44514|issue:45381", - "issue:44484|issue:44704", "issue:43720|issue:44898", + "issue:44484|issue:44704", "issue:41762|issue:44936", "issue:24643|issue:29127", "issue:44625|issue:44877", "issue:44514|issue:45290", - "issue:43452|issue:44843", "issue:43452|issue:43577", + "issue:43452|issue:44843", "issue:44466|issue:45005", "issue:43295|issue:45356", "issue:43819|issue:44811", "issue:44936|issue:45823" ], - "prepared_review_unit_hash": "632b64d2460915bba5e36985386642433964426f1eb9a33b8aa2c70da634243d", + "prepared_review_unit_hash": "0607cf2ab6e3a17269883f36ec5062013a5ccf58df0a03db3cb60eee18423ff9", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36184, - "estimated_input_tokens": 9046, - "estimated_eval_tokens": 18348 + "serialized_chars": 36186, + "estimated_input_tokens": 9047, + "estimated_eval_tokens": 18350 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of Transformers regressions around loading, tokenization, configs, and save_pretrained behavior. A few pairs look like true duplicates, but most are only broadly similar.", - "confidence": 0.71, - "canonical_issue_reason": "issue:45920 is the broadest current regression report in the set: it spans multiple model families and the tokenizer/output corruption theme that recurs elsewhere in the cluster.", + "summary": "This cluster is mostly a mix of unrelated issues. Based on titles alone, only one pair looks tentatively like the same underlying SigLIP2 correctness problem; the rest are better treated as separate bugs or too broad to merge safely.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 43493 is the best representative of the only plausible duplicate family here: it describes a broader SigLIP2 discrepancy, while 43994 appears to be a more specific symptom involving a particular checkpoint/model path.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best representative issue for this cluster because it is the most general, user-facing tokenizer regression and could absorb related follow-ups.", + "best_issue_reason": "Issue 43493 is the strongest issue anchor for triage because it is the broadest SigLIP2 report in the set and most plausibly covers the related model-specific complaint in 43994.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43493", "right": "issue:43994", "accept": true, - "reason": "Same SigLIP2 implementation bug: both report incorrect outputs from the HF SigLIP2 path, just from different entry points." + "reason": "These both point to SigLIP2 producing incorrect results; one is a general implementation-vs-JAX discrepancy and the other is a model-specific bad-output report, so they tentatively look like the same underlying bug." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "Trainer bookkeeping bug vs custom compute_loss signature mismatch; different code paths and symptoms." + "reason": "Different Trainer problems: token counting not updating versus a custom compute_loss signature mismatch." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "AddedToken constructor conflict vs Mistral regex patch crash; unrelated tokenizer failures." + "reason": "Both involve tokenizer loading, but the failure modes are different and do not look like the same bug." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized loading error vs AutoImageProcessor URL loading; different subsystems." + "reason": "Different loading paths and different error shapes; no strong sign they are the same underlying issue." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 save_pretrained visual-encoder key serialization vs tokenizer ID regression across models; not the same bug." + "reason": "Both are Qwen-related, but the titles indicate different regression surfaces: save_pretrained serialization versus tokenizer ID generation." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Generic trust_remote_code AutoConfig field loss vs a specific Qwen3.5 vision config field omission; different defects." + "reason": "Both are config-related, but they describe different config-handling failures and do not appear to be duplicates." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "GGUF loading regression vs a model-specific load failure; different mechanisms." + "reason": "Different from_pretrained failures with different root symptoms; too little evidence that they are the same bug." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Janus generate_images test failure vs Phi-4 multimodal model load failure; different models and code paths." + "reason": "Both are multimodal/model-loading related, but the titles suggest different model families and different failures." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image processor torchvision dependency regression vs tokenizer regex crash; unrelated." + "reason": "They share a tokenizer-adjacent area, but the specific issues are different enough that merging them would be unsafe." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "MobileLLM key error vs Float8 storage deserialization issue; distinct loading bugs." + "reason": "Different low-level errors: a key lookup problem versus a missing storage object type." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding_window bug vs image-processor URL loading; unrelated." + "reason": "Unrelated issue surfaces: Qwen3-MoE sliding window behavior versus AutoImageProcessor URL loading." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "PEFT version pin problem vs removed config references in auto_mappings; different maintenance issues." + "reason": "Maintenance/config change versus stale mapping references; not the same bug." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "RTDetrV2 output drift vs Qwen3.5 save_pretrained regression; different models and code paths." + "reason": "Different model families and different regressions; no concrete overlap beyond both being correctness issues." }, { "left": "issue:43688", "right": "issue:43749", "accept": false, - "reason": "Aux-loss normalization bug vs FSDP CPU RAM efficient loading; different training/load problems." + "reason": "Separate bugs affecting different training/loading paths." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is a chat-template batching crash and the other is wrong vision_position_ids; not the same concrete defect." + "reason": "Same broader Qwen2.5-VL area, but the titles indicate different failure modes." }, { - "left": "issue:44484", - "right": "issue:44704", + "left": "issue:43720", + "right": "issue:44898", "accept": false, - "reason": "save_pretrained shard-size question vs AutoProcessor kwarg forwarding bug; not duplicates." + "reason": "Different model-specific problems with no clear shared code-path from the titles alone." }, { - "left": "issue:43720", - "right": "issue:44898", + "left": "issue:44484", + "right": "issue:44704", "accept": false, - "reason": "BitNet packed-weight loading vs Perceiver interpolation/classification issue; unrelated." + "reason": "A save_pretrained size question is unrelated to AutoProcessor passing kwargs to cached_file." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "DeepSpeed ZeRO-3 loading error vs trainer.evaluate() failing after train; different issues." + "reason": "Different Trainer/DeepSpeed behaviors with distinct symptoms." }, { "left": "issue:24643", "right": "issue:29127", "accept": false, - "reason": "DeepSpeed training runtime error vs LayoutLMv3 error-message clarity; unrelated." + "reason": "DeepSpeed training runtime error versus LayoutLMv3 box-related error messaging; unrelated." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation bug vs strict config rejecting granite_speech; different config paths." + "reason": "Both involve config handling, but the underlying failures are different." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both hit apply_chat_template, but the triggers differ: batched padding=False crash vs tool-call/no-content crash." + "reason": "Same API surface, but the batched padding crash and the assistant tool-call/content crash look distinct." }, { "left": "issue:43452", - "right": "issue:44843", + "right": "issue:43577", "accept": false, - "reason": "GGUF loading regression vs offline model_info call in _patch_mistral_regex; different loading bugs." + "reason": "Different loading issues: gguf_file handling versus dtype propagation in BLIP2." }, { "left": "issue:43452", - "right": "issue:43577", + "right": "issue:44843", "accept": false, - "reason": "GGUF/model loading regression vs BLIP2 dtype propagation; unrelated." + "reason": "Tokenizer/loading-related, but the specific failure modes are different." }, { "left": "issue:44466", "right": "issue:45005", - "accept": true, - "reason": "Both describe the same tied-weights/save_pretrained regression in v5, with device-dependent lm_head serialization and translation-model tied weights." + "accept": false, + "reason": "Related to tied weights, but the titles suggest different manifestations and too broad a scope to treat as one bug." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Processor.tokenizer regression vs Kimi-K2.5 codec/fix_mistral_regex warning issue; different bugs." + "reason": "Different regression surfaces: processor/tokenizer access versus tokenizer codec/regex handling." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "DAC latent round-trip bug vs Whisper batch_decode skip_special_tokens handling; unrelated." + "reason": "Different bugs in different model areas; no sign of a shared underlying defect." }, { "left": "issue:44936", "right": "issue:45823", "accept": false, - "reason": "trainer.evaluate() after train vs Gemma4 PLE device mismatch; different runtime problems." + "reason": "Trainer lifecycle failure versus a device mismatch in Gemma4 forward pass." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is grounded in the packet and stays appropriately cautious. The two accepted soft pairs are plausible duplicates, and the rejected pairs are consistently conservative rather than overstated. The canonical issue choice (45920) is well supported as the broadest tokenizer regression in the set." + "accept": false, + "feedback": "The triage is mostly conservative, but it still overstates what is grounded in the packet. In particular, \"clearly reproducible\" and \"captures the underlying SigLIP2 mismatch\" are not supported by the packet itself (we only have titles/body lengths here), so they read as stronger than the evidence allows. Also, the summary says \"a couple of pairs\" look like duplicates even though only one pair is actually marked accept=true. Please soften those claims and keep the duplicate judgment explicitly tentative." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -6307,13 +6274,13 @@ "issue:33666", "issue:34567", "issue:35141", + "issue:36010", "issue:38175", "issue:41720", "issue:43122", "issue:43278", "issue:43295", "issue:43452", - "issue:43493", "issue:43504", "issue:43531", "issue:43550", @@ -6326,7 +6293,6 @@ "issue:43720", "issue:43742", "issue:43749", - "issue:43825", "issue:43906", "issue:43927", "issue:44361", @@ -6350,19 +6316,19 @@ "issue:45704" ], "soft_pairs": [ - "issue:43653|issue:44568", "issue:38175|issue:43295", + "issue:43653|issue:44568", "issue:43653|issue:44361", - "issue:43643|issue:44561", "issue:44361|issue:44442", "issue:30064|issue:33666", + "issue:43643|issue:44561", "issue:44464|issue:44743", "issue:41720|issue:43906", "issue:43688|issue:45440", "issue:43688|issue:45704", "issue:28282|issue:32090", - "issue:43577|issue:43697", "issue:28282|issue:33357", + "issue:43577|issue:43697", "issue:28282|issue:35141", "issue:44843|issue:44991", "issue:43646|issue:43749", @@ -6371,8 +6337,8 @@ "issue:43531|issue:43742", "issue:43550|issue:43927", "issue:44464|issue:44977", - "issue:44479|issue:44625", "issue:43550|issue:44361", + "issue:44479|issue:44625", "issue:43122|issue:43906", "issue:43577|issue:44554", "issue:43577|issue:45588", @@ -6385,252 +6351,246 @@ "issue:43452|issue:44451", "issue:43742|issue:44821", "issue:43278|issue:43688", - "issue:43493|issue:43825" + "issue:33666|issue:36010" ], - "prepared_review_unit_hash": "a47a6e14523fc9450088f5c09f8c07974418c6a1b1ddd5eda9880393c4a5b236", + "prepared_review_unit_hash": "3fd1233f82948697ab3814f1d84c4efa37218912a27db3db22e0b14b02110161", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 36, - "serialized_chars": 38477, - "estimated_input_tokens": 9620, - "estimated_eval_tokens": 19496 + "serialized_chars": 37895, + "estimated_input_tokens": 9474, + "estimated_eval_tokens": 19204 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All soft edges look like false positives: they share broad subsystem themes, but not the same concrete bug or fix path. The set is heterogeneous across tokenizer loading, model init, training, generation, and device/runtime issues.", - "confidence": 0.96, - "canonical_issue_reason": "No single issue is a clear canonical representative; the items are too heterogeneous to form one duplicate cluster.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports, with a few loose tokenizer/model-loading regressions but no true duplicate pair among the soft edges. I would keep one concrete tokenizer regression as the representative issue, and there are no PRs to consider.", + "confidence": 0.84, + "canonical_issue_reason": "issue:43122 is a concrete, user-visible tokenizer regression with version-specific repro and inbound references, making it the cleanest representative issue in this set.", "canonical_pr_reason": null, - "best_issue_reason": "If forced to pick an anchor, issue 45701 is the broadest tokenization-regression report and has the most inbound references, but it still does not represent the rest of the mixed set well.", + "best_issue_reason": "issue:43122 is the strongest issue candidate because it is specific, already closed, and describes a reproducible version-to-version behavior change rather than a vague umbrella symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43653", - "right": "issue:44568", - "accept": false, - "reason": "Both concern tokenizers and special tokens, but they are different tokenizers with different failures and code paths." - }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Unrelated problems: SigLIP2 probability behavior vs processor/tokenizer API regression." + "reason": "Different failures: zero probabilities in SigLIP2 vs processor/tokenizer access regression in custom model code." }, { "left": "issue:43653", - "right": "issue:44361", + "right": "issue:44568", "accept": false, - "reason": "Different tokenizer classes and symptoms; special-token decode bug is not the same as an AttributeError on tasks." + "reason": "Both are tokenizer-special-token bugs, but they affect different models and different symptoms/code paths." }, { - "left": "issue:43643", - "right": "issue:44561", + "left": "issue:43653", + "right": "issue:44361", "accept": false, - "reason": "Both involve trust_remote_code, but one is missing config fields and the other is a removed helper breaking imports." + "reason": "BigBird special-token decode issue vs MLukeTokenizer AttributeError; not the same bug." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "Separate tokenizer-loading failures for different models and failure modes." + "reason": "Tokenizer AttributeError on tasks vs AutoTokenizer failing to load a specific tokenizer class; different root causes." }, { "left": "issue:30064", "right": "issue:33666", "accept": false, - "reason": "Image-processor segmentation-map bug vs multi-GPU training issue; no shared underlying bug." + "reason": "Image processor segmentation-map handling and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both touch trust_remote_code/v5 behavior, but one is missing config fields and the other is a removed utility breaking models." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Both are generation/cache related, but one is compiled-forward inconsistency and the other is recurrent-state reset logic." + "reason": "Chunked generation with compiled forward is a different code path from recurrent-state reset in modular Qwen3.5." }, { "left": "issue:41720", "right": "issue:43906", "accept": false, - "reason": "An A800 cuda assert with auto device mapping is not the same as an isolated reproduction of a different issue." + "reason": "Different CUDA/Qwen3 device-mapping crash versus an isolated reproduction of another issue; no concrete same defect." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Different MoE model bugs with different implementations and likely different fixes." + "reason": "Both involve MoE models, but one is auxiliary-loss normalization and the other is a remote/native implementation divergence." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Aux-loss normalization bug is unrelated to a T5 apex RMSNorm memory leak." + "reason": "Auxiliary-loss normalization and an apex RMSNorm memory leak are unrelated bugs." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "PyTorch-missing ImportError vs Trainer GPU broadcast NoneType TypeError; unrelated." + "reason": "Missing PyTorch import vs NoneType passed to gpu broadcast are distinct runtime errors." }, { - "left": "issue:43577", - "right": "issue:43697", + "left": "issue:28282", + "right": "issue:33357", "accept": false, - "reason": "Dtype loading mismatch for BLIP2 is not the same as RT-DETR v5 output divergence." + "reason": "PyTorch-not-found import error and MacOS bus error on CLIP are different failures." }, { - "left": "issue:28282", - "right": "issue:33357", + "left": "issue:43577", + "right": "issue:43697", "accept": false, - "reason": "ImportError about missing PyTorch and a Mac bus error are unrelated." + "reason": "BLIP-2 dtype propagation and RTDetrV2 output drift are different model-specific issues." }, { "left": "issue:28282", "right": "issue:35141", "accept": false, - "reason": "No common code path beyond generic loading/model usage." + "reason": "ImportError from missing torch and embedding reinitialization after resize are unrelated." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer loading regressions, but they stem from different mechanisms and models." + "reason": "Offline hub lookup bug and tokenizer load failure for a specific model are different problems." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "Custom model init breakage and FSDP RAM-efficient loading are distinct failures." + "reason": "Custom model initialization breakage is not the same as FSDP RAM-efficient loading failure." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Aux-loss normalization in specific MoE models is not the same as a broad tokenization-version change." + "reason": "Auxiliary-loss normalization and broad tokenization changes are not the same underlying bug." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "Tokenizer special-token decode bug and DiaConfig custom token ID persistence are different problems." + "reason": "BigBird special-token registration and DiaConfig custom token-ID loss are both tokenizer/config issues, but they are separate code paths." }, { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Qwen3-MoE sliding-window logic and a MobileLLM key error are unrelated." + "reason": "Qwen3-MoE sliding-window behavior and MobileLLM loading key error are unrelated." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "torch.compile SDPA failure and config token-ID persistence do not share the same fix path." + "reason": "torch.compile/SDPA failure in Bamba is unrelated to DiaConfig token-ID persistence." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Both mention generation, but one is compiled-forward inconsistency and the other is flash-attention behavior in Qwen3.5." + "reason": "Compiled-forward chunked generation and flash-attention generation regression are different execution paths." }, { - "left": "issue:44479", - "right": "issue:44625", + "left": "issue:43550", + "right": "issue:44361", "accept": false, - "reason": "Video input regression and num_labels propagation are separate issues." + "reason": "Model compile/SDPA failure and tokenizer AttributeError are unrelated." }, { - "left": "issue:43550", - "right": "issue:44361", + "left": "issue:44479", + "right": "issue:44625", "accept": false, - "reason": "Different models and failure modes; no common underlying bug." + "reason": "Video-input regression in Qwen VL models and num_labels propagation are different bugs." }, { "left": "issue:43122", "right": "issue:43906", "accept": false, - "reason": "Generic tokenization change report vs an isolated reproduction of another issue." + "reason": "Tokenizer behavior change across versions is not the same as an isolated reproduction of another issue." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "BLIP2 dtype loading bug and MPS attention correctness issue are unrelated." + "reason": "BLIP-2 dtype mismatch and MPS attention correctness are different issues." }, { "left": "issue:43577", "right": "issue:45588", "accept": false, - "reason": "Different subsystems: dtype propagation vs flash_attention crash on sink-less models." + "reason": "BLIP-2 dtype propagation and flash_attention crashing on s_aux=None are unrelated." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Legacy-field model loading bug and packed-weight unpacking bug are different code paths." - }, - { - "left": "issue:43504", - "right": "issue:43550", - "accept": false, - "reason": "Different model families and distinct loading/compile issues." + "reason": "BEiT legacy-field loading and BitNet packed-weight loading are different loaders and bugs." }, { "left": "issue:32090", "right": "issue:34567", "accept": false, - "reason": "Trainer broadcast NoneType error vs token-count state not updating; not the same defect." + "reason": "gpu_broadcast NoneType failure and TrainerState token counter not updating are unrelated trainer issues." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Auto device mapping cuda assert and chunked generation inconsistency are unrelated." + "reason": "Qwen3 auto device-mapping CUDA assert and compiled-forward generation inconsistency are different failures." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "Pretrained preset loading failure and processor size mismatch are separate bugs." + "reason": "BEiT preset loading bug and OmDet processor size mismatch are unrelated." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer GPU broadcast error and Qwen2-VL multi-GPU training question are not the same issue." + "reason": "NoneType broadcast crash and Qwen2-VL multi-GPU training are not the same bug." }, { "left": "issue:43452", "right": "issue:44451", "accept": false, - "reason": "GGUF loading breakage and a model-specific tokenizer load failure are different root causes." + "reason": "gguf_file loading breakage and ScandiBERT tokenizer loading failure are different model-loading problems." }, { "left": "issue:43742", "right": "issue:44821", "accept": false, - "reason": "Model config loading key error vs AutoImageProcessor-from-URL failure." + "reason": "MobileLLM key error and AutoImageProcessor URL loading failure are unrelated." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "Training/eval dtype mismatch is unrelated to MoE auxiliary-loss normalization." + "reason": "Embedding dtype drift during evaluate and auxiliary-loss normalization in MoE models are different bugs." }, { - "left": "issue:43493", - "right": "issue:43825", + "left": "issue:33666", + "right": "issue:36010", "accept": false, - "reason": "SigLIP2 implementation discrepancy and an error-message wording bug are not the same bug." + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure are unrelated." } ] }, @@ -6657,7 +6617,6 @@ "issue:28218", "issue:33357", "issue:33666", - "issue:36010", "issue:36331", "issue:39692", "issue:43257", @@ -6676,6 +6635,7 @@ "issue:43720", "issue:43756", "issue:43824", + "issue:43825", "issue:43844", "issue:43927", "issue:44291", @@ -6704,11 +6664,11 @@ "issue:45987" ], "soft_pairs": [ - "issue:33666|issue:36010", + "issue:43493|issue:43825", "issue:44849|issue:45357", "issue:43606|issue:43927", - "issue:44568|issue:45812", "issue:43720|issue:43927", + "issue:44568|issue:45812", "issue:43525|issue:43572", "issue:43262|issue:43295", "issue:43756|issue:44913", @@ -6719,15 +6679,15 @@ "issue:43540|issue:44977", "issue:43493|issue:43844", "issue:44964|issue:45405", - "issue:45803|issue:45987", "issue:43334|issue:43572", + "issue:45803|issue:45987", "issue:43278|issue:43844", "issue:44625|issue:45584", - "issue:43257|issue:43299", "issue:43756|issue:45083", - "issue:44488|issue:44843", - "issue:45200|issue:45229", + "issue:43257|issue:43299", "issue:43334|issue:45759", + "issue:45200|issue:45229", + "issue:44488|issue:44843", "issue:43824|issue:44589", "issue:39692|issue:43295", "issue:43646|issue:44466", @@ -6738,14 +6698,14 @@ "issue:33357|issue:33666", "issue:43927|issue:44361" ], - "prepared_review_unit_hash": "996e68aed3375b661a22c6be681166785e69cdfe3ec414640df964eab05a2ce8", + "prepared_review_unit_hash": "70615c1120c9439e055843c6214f1a47823326b4fc5660c43e8199c3ffe3980e", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, @@ -6760,210 +6720,210 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The soft pairs are broadly heterogeneous; none look like the same underlying bug or change, so no duplicates should be merged.", - "confidence": 0.95, - "canonical_issue_reason": null, + "summary": "Mostly heterogeneous issue set: many items share tokenizer/config/model-loading language, but the underlying bugs differ by model, subsystem, and failure mode. I found no convincing duplicate clusters, so the best anchor is a broad, user-visible tokenizer regression rather than a narrow model-specific report.", + "confidence": 0.22, + "canonical_issue_reason": "Issue 45812 is the broadest and most cross-cutting regression in this set: it affects all Granite models, is user-visible, and has multiple inbound references, making it the best single anchor if one issue must represent the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45812 is the best canonical-style issue: it is open, broad in impact, and clearly describes a specific regression with multiple references.", + "best_issue_reason": "45812 is the strongest representative issue because it is broadly scoped, current, and tied to a concrete tokenizer behavior regression rather than a niche model-specific edge case.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:33666", - "right": "issue:36010", + "left": "issue:43493", + "right": "issue:43825", "accept": false, - "reason": "Different problems: Qwen2-VL training vs a GenerationMixin import failure." + "reason": "Different bugs: SigLIP2 implementation mismatch vs a pipeline error message about translation tasks." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen3.5-related, but one is output_hidden_states behavior and the other is save_pretrained key serialization." + "reason": "Different Qwen3.5 problems: hidden-states behavior vs incorrect save_pretrained visual encoder keys." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch is unrelated to DiaConfig token ID loss on save/load." + "reason": "Device-mismatch/offload loading bug vs config losing custom token IDs; unrelated code paths." }, { - "left": "issue:44568", - "right": "issue:45812", + "left": "issue:43720", + "right": "issue:43927", "accept": false, - "reason": "Tokenizer regressions, but different models and different failure modes." + "reason": "Packed-weight conversion during accelerate loading is unrelated to DiaConfig token ID persistence." }, { - "left": "issue:43720", - "right": "issue:43927", + "left": "issue:44568", + "right": "issue:45812", "accept": false, - "reason": "BitNet accelerate loading bug is unrelated to DiaConfig token serialization." + "reason": "Both are tokenizer bugs, but they affect different models and different behaviors (special tokens vs Granite token IDs)." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Both involve missing pad token fields, but they affect different configs and code paths." + "reason": "Similar symptom name, but different model configs and different missing fields (`pad_token_id` vs `pad_token_idx`)." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Audio chat template sampling-rate default is unrelated to processor.tokenizer/image handling." + "reason": "Audio chat-template sampling-rate default bug vs processor.tokenizer/image-handling regression." }, { "left": "issue:43756", "right": "issue:44913", "accept": false, - "reason": "One is a RoPE layer-count bug, the other is config reload losing rotary_pct." + "reason": "Different config/architecture issues: RoPE layer dropping vs rotary_pct not persisting on reload." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "Packed-weight loading and tokenizer AttributeError are different bugs." + "reason": "Different tokenizer/loading failures: packed-weight unpacking vs MLukeTokenizer AttributeError on tasks." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer spacing bug and MacOS bus error are unrelated." + "reason": "Tokenizer spacing bug vs Mac bus error on a CLIP model; no shared underlying defect." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Multi-GPU training support is unrelated to compute_loss signature mismatch." + "reason": "Multi-GPU training support issue vs CustomTrainer compute_loss signature mismatch." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Same model family, but doc example mistakes are not the same as implementation discrepancies." + "reason": "Docs/example errors are not the same bug as the SigLIP2 HF-vs-JAX implementation discrepancy." }, { "left": "issue:43540", "right": "issue:44977", "accept": false, - "reason": "Video input processing in Qwen3OmniMoe is unrelated to flash-attention generation issues." + "reason": "Video-input processing error in Qwen3OmniMoe vs flash-attention generation bug in Qwen3.5." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation mismatch and ZeRO-3 gradient growth are unrelated." + "reason": "SigLIP2 implementation discrepancy is unrelated to the DeepSpeed gradient growth report." }, { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "Model loading failure and an unreleased PEFT version bump are unrelated." + "reason": "Model-loading failure for Phi-4 multimodal is unrelated to the PEFT version bump issue." }, { - "left": "issue:45803", - "right": "issue:45987", + "left": "issue:43334", + "right": "issue:43572", "accept": false, - "reason": "Bare except in tensor conversion is unrelated to StaticCache return-type mismatch." + "reason": "Both mention missing pad token fields, but in different model families and different config attributes." }, { - "left": "issue:43334", - "right": "issue:43572", + "left": "issue:45803", + "right": "issue:45987", "accept": false, - "reason": "Both mention missing token IDs, but they are different model/config regressions." + "reason": "Bare except swallowing interrupts is unrelated to StaticCache returning a tensor instead of an int." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "Embedding dtype drift and gradient instability under ZeRO-3 are different issues." + "reason": "Embedding dtype drift in eval is not the same as gradient explosion under ZeRO-3." }, { "left": "issue:44625", "right": "issue:45584", "accept": false, - "reason": "Qwen3.5 num_labels propagation is unrelated to Whisper empty-transcription behavior." + "reason": "Qwen3.5 num_labels propagation bug vs Whisper empty-transcription generation failure; different paths." }, { - "left": "issue:43257", - "right": "issue:43299", + "left": "issue:43756", + "right": "issue:45083", "accept": false, - "reason": "Both involve Qwen MoE loading, but the concrete failure modes and model variants differ." + "reason": "RoPE-layer omission in Smollm3 is unrelated to the qwen3_omni_moe feature-length helper behavior." }, { - "left": "issue:43756", - "right": "issue:45083", + "left": "issue:43257", + "right": "issue:43299", "accept": false, - "reason": "RoPE-layer truncation and feature-length helper behavior are unrelated." + "reason": "Both are Qwen/MoE loading regressions, but they hit different model variants and different failure causes." }, { - "left": "issue:44488", - "right": "issue:44843", + "left": "issue:43334", + "right": "issue:45759", "accept": false, - "reason": "Model load failure and offline model_info regression are different code paths." + "reason": "Pad-token-id missing in Qwen3-VL config is unrelated to from_config not unwrapping text_config." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Text-only mm_token_type_ids defaulting and multi-GPU CUDA OOM are unrelated." + "reason": "Text-only mm_token_type_ids defaulting bug vs multi-GPU inference OOM; distinct issues." }, { - "left": "issue:43334", - "right": "issue:45759", + "left": "issue:44488", + "right": "issue:44843", "accept": false, - "reason": "Missing pad_token_id and text_config unwrapping are different multimodal config bugs." + "reason": "Loading a specific model failing is not the same as AutoTokenizer calling model_info() offline." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "Import error for a model class is unrelated to Float8 storage lookup failure." + "reason": "ImportError for Qwen2.5-VL symbol vs Float8 storage lookup TypeError; unrelated failures." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "Documentation example errors are unrelated to processor.tokenizer regression." + "reason": "Documentation example mistakes are unrelated to the processor.tokenizer regression." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model init breakage and lm_head serialization inconsistency are different regressions." + "reason": "Custom model initialization breakage is different from inconsistent lm_head.weight serialization across devices." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "OOM during inference and a device mismatch bug are not the same failure." + "reason": "CUDA OOM in Gemma4 inference vs device mismatch in Gemma4 PLE forward; different concrete bugs." }, { "left": "issue:43504", "right": "issue:43606", "accept": false, - "reason": "Legacy-field load failure in BEiT is unrelated to bark-small CPU offload issues." + "reason": "Legacy-field load failure in Beit semantic segmentation is unrelated to CPU offload device mismatch." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing pad_token_idx and unexpected _is_hf_initialized are separate loading regressions." + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to unexpected _is_hf_initialized argument in init_empty_weights." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Float8 storage lookup and flash_attn KeyError are unrelated." + "reason": "Float8 storage import error vs flash_attn KeyError on Python 3.13; different dependency issues." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "MacOS bus error on CLIP and Qwen2-VL multi-GPU training are unrelated." + "reason": "Mac bus error with a CLIP model vs Qwen2-VL multi-GPU training; no overlap in root cause." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "Config token persistence bug and MLukeTokenizer task AttributeError are different issues." + "reason": "DiaConfig token ID loss and MLukeTokenizer task AttributeError are unrelated bugs." } ] }, @@ -7063,190 +7023,190 @@ "issue:43295|issue:44610", "issue:44554|issue:45356" ], - "prepared_review_unit_hash": "e3f355efbedb0a8e6c7c661c0628ffec379763b934922cc42472b212446efbc7", + "prepared_review_unit_hash": "d06800b383e02227a032c9c1c493ff61edc5f94c3d92becc84103a06ba8228c0", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 26, - "serialized_chars": 34969, - "estimated_input_tokens": 8743, - "estimated_eval_tokens": 17742 + "serialized_chars": 34964, + "estimated_input_tokens": 8741, + "estimated_eval_tokens": 17738 }, "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 bug reports across tokenization, generation, multimodal processing, device/dtype issues, and CI failures. None of the soft pairs look like true duplicates of the same underlying bug; a few are superficially similar but differ in concrete failure mode or code path.", - "confidence": 0.62, - "canonical_issue_reason": "issue:44521 is the most central active report in the only loosely related multimodal/chat-template subgroup and has inbound references, making it the best representative issue for this mixed cluster.", + "summary": "The set is heterogeneous: the soft pairs mostly share only broad subsystem vocabulary (generation, loading, tokenizers, CI, multimodal), but not the same concrete bug or change. I would not merge any of the proposed soft edges, and there is no single canonical issue/PR for the whole set.", + "confidence": 0.91, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the strongest issue anchor here: open, referenced by others, and thematically closest to several multimodal/tokenization reports without being too specific.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Both are serve/runtime generation crashes, but one is a multimodal `.to()` attribute error and the other is KV-cache tensor indexing; different bugs." + "reason": "Both are serving-related crashes, but one is a multimodal `'str'.to` error and the other is KV-cache continuation indexing; different failure modes and code paths." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "BEiT legacy-field loading and BigBirdTokenizer special-token decoding are unrelated subsystems and failure modes." + "reason": "Both mention loading/token handling, but BEiT legacy-field loading and BigBirdTokenizer mask-token registration are unrelated bugs." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is not the same as empty-transcription failure after `align_special_tokens`." + "reason": "Both involve generation, but chunked generation with compiled forward and empty-transcription whisper failure after token alignment are different issues." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "Both mention device mismatch, but one is a real CPU-offload bug and the other is a stale CI test for GraniteSpeech." + "reason": "Both are CI failures, but CPU offload device mismatch in bark-small and a stale GraniteSpeech device-override test are not the same bug." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 output discrepancies and unexpected position-id keys are different model behaviors with no shared code-path evidence." + "reason": "SigLIP2 nonsensical outputs and generic 'unexpected key with position id' regressions are too different in symptom and likely root cause." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "Tensor-parallel OOM for a model load and an unmet PEFT version requirement are unrelated." + "reason": "Qwen2 TP OOM and a PEFT version bump/release issue are unrelated; one is memory pressure, the other dependency metadata." }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "BF16/FP32 dtype drift in evaluation is not the same as resume-from-checkpoint key mismatch." + "reason": "One is embedding dtype changing between train/eval, the other is a resume_from_checkpoint key mismatch; no shared underlying defect." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "Janus image-generation test failure and PEFT version pinning are unrelated." + "reason": "A Janus image-generation test failure and a missing released PEFT version are unrelated." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both touch `apply_chat_template`, but one is all-zero assistant masks for multimodal inputs while the other is a crash on tool-call assistant messages; different defects." + "reason": "Both concern chat/template processing, but all-zero assistant masks for multimodal inputs and tool-call assistant messages with no content are distinct edge cases." }, { "left": "issue:29942", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 test failures and MLflow job-status reporting are different CI/training issues." + "reason": "Flash Attention 2 test failures and MLflow not being reported after job failure are different CI/training issues with different surfaces." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Import warning for trailing-slash local paths is unrelated to CUDA OOM on GPT-oss-20b." + "reason": "Local-path import warnings and GPT-oss CUDA OOM are unrelated; one is path handling, the other memory exhaustion." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient explosion and Phi-4 multimodal load failure are different problems." + "reason": "Gradient growth with HfDeepSpeedConfig/ZeRO-3 and inability to load Phi-4 multimodal are different model/runtime problems." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Tokenizer extra-space behavior and TrainerState token-count updates do not describe the same bug." + "reason": "Tokenizer added-space behavior and TrainerState token counting are both token-related but not the same bug." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness and a PyTorch version check in flex_attention are unrelated." + "reason": "MPS attention correctness when head dims differ and a PyTorch version check for AuxRequest import are unrelated." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Non-template-node compilation error and missing `flash_attn` import key on Python 3.13 are different import/runtime issues." + "reason": "Non-template node compilation error and flash_attn import key error on Python 3.13 are different failure modes." }, { "left": "issue:43582", "right": "issue:44977", "accept": false, - "reason": "Apple Silicon allocator TypeError and Qwen3.5 flash-attention generation bug are separate platform/model issues." + "reason": "Apple Silicon warmup TypeError and Qwen3.5 flash-attention generation bugs are separate platform/model-specific issues." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Video-input processing error in Qwen3OmniMoe and Gemma4 token-type defaults are not the same underlying bug." + "reason": "Qwen3OmniMoe video-processing ValueError and Gemma 4 mm_token_type_ids defaulting are different multimodal input problems." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero probabilities are different vision-model issues." + "reason": "TorchScript export GeneratorExp limitation and SigLIP2 zero probabilities are unrelated model/runtime bugs." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwargs update after `sync_gpus` and Qwen3.5 hidden-states output are distinct generation-path bugs." + "reason": "Generation kwargs update after sync_gpus and Qwen3.5 hidden_states output bug are distinct generation path issues." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "bfloat16 dtype mismatches in inference and Wav2Vec2 save/tokenization failures are unrelated." + "reason": "bfloat16 dtype mismatches in inference and Wav2Vec2 save/tokenization failures do not look like one underlying bug." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Missing flash-attention helper import and meta-device loading breakage are different compatibility issues." + "reason": "Missing flash_supports_window_size import and meta-device loading regressions are both loading-related but clearly different problems." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Gemma4 text-only token-type defaults and PLE device mismatch with `device_map=\"auto\"` are separate forward-pass problems." + "reason": "Gemma 4 token type ids defaulting and Gemma4 PLE device mismatch are separate issues in the same model family, not the same code-path defect." }, { "left": "issue:44315", "right": "issue:44849", "accept": false, - "reason": "Liger Kernel not applied in `model_init` is unrelated to Qwen3.5 hidden-state output behavior." + "reason": "Liger Kernel not applied with model_init and Qwen3.5 hidden_states output bug are unrelated training/model behavior issues." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 HF-vs-JAX discrepancy and checkpoint resume key mismatch are not the same bug." + "reason": "SigLIP2 implementation discrepancy and resume checkpoint key mismatch are different classes of bug with no clear shared root cause." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Processor/tokenizer regression for multimodal custom code and OmDet-Turbo image-size mismatch are different processor bugs." + "reason": "processor.tokenizer regression with images and OmDet-Turbo input-size mismatch are both multimodal, but not the same defect." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS attention correctness issue and Kimi tokenizer codec regression are unrelated." + "reason": "MPS attention correctness and Kimi-K2.5 tokenizer regression are unrelated subsystems." } ] }, @@ -7324,9 +7284,9 @@ "issue:44466|issue:45127", "issue:44792|issue:45563", "issue:43572|issue:44987", - "issue:45200|issue:45325", - "issue:43572|issue:43957", "issue:37428|issue:43824", + "issue:43572|issue:43957", + "issue:45200|issue:45325", "issue:43844|issue:44936", "issue:43295|issue:43450", "issue:43334|issue:43525", @@ -7335,8 +7295,8 @@ "issue:43388|issue:43526", "issue:38175|issue:45412", "issue:42915|issue:43824", - "issue:44514|issue:44521", "issue:38175|issue:43493", + "issue:44514|issue:44521", "issue:44442|issue:45198", "issue:28218|issue:35141", "issue:43278|issue:43856", @@ -7351,214 +7311,214 @@ "issue:43262|issue:43450", "issue:44610|issue:44857" ], - "prepared_review_unit_hash": "03f6cf3f0514350557fb16e8e0ae9058a824da5400dc878b8a0d08eeb84e0790", + "prepared_review_unit_hash": "ebb89661c4e59e1121fbb5348d46376b9147c84bfbc0bc01d4fa6d0fd137f7e3", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36949, - "estimated_input_tokens": 9238, - "estimated_eval_tokens": 18732 + "serialized_chars": 36948, + "estimated_input_tokens": 9237, + "estimated_eval_tokens": 18730 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items do not form a true duplicate cluster; they span unrelated tokenizer, import, processor, training, and model-specific bugs. I reject all soft edges and pick the most concrete, well-documented issue as the best standalone representative.", - "confidence": 0.9, - "canonical_issue_reason": "issue:43824 is the strongest standalone report here: it has the highest engagement, a precise import failure, and a clear reproduction path.", + "summary": "These items are largely unrelated issue reports across different models/subsystems. I don\u2019t see any soft-edge pair that clearly represents the same underlying bug or change, so none should be merged as duplicates.", + "confidence": 0.28, + "canonical_issue_reason": "No single underlying bug clearly ties the set together; if forced to pick a representative, issue 44521 is the most central because it is open, recently active, and has the strongest reference activity in the processor/chat-template area.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43824 is the best issue candidate because it is specific, broadly actionable, and better evidenced than the others.", + "best_issue_reason": "Issue 44521 is the best overall issue candidate by activity and recency, but this cluster is too heterogeneous to treat it as a true canonical duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Different bugs: Perceiver resolution handling vs RT-DETR memory cleanup." + "reason": "Different bugs: Perceiver image-classification resolution failure vs RT-DETR memory not being released." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Both involve weights/serialization, but one is device-dependent lm_head serialization and the other is LoRA merge collapse." + "reason": "Both touch tied weights, but one is serialization/device-dependent while the other is LoRA merge collapse with extended vocab; not the same bug." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "Janus image-generation test failure is unrelated to a stale generate() warning." + "reason": "Janus image-generation test failure and a stale generate() warning are unrelated symptoms." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "StableLm pad_token_idx regression is unrelated to loading physical-intelligence/fast." + "reason": "Missing pad_token_idx in StableLmConfig is not the same as loading a specific model failing in v5.1.0." }, { - "left": "issue:45200", - "right": "issue:45325", + "left": "issue:37428", + "right": "issue:43824", "accept": false, - "reason": "Gemma 4 token-type defaults and Qwen2.5-VL temporal rope scaling are different code paths." + "reason": "Both are ImportErrors, but they concern different missing symbols in different code paths." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is not the same as meta-device model loading failures." + "reason": "Config attribute regression vs meta-device model-loading failure; different problems." }, { - "left": "issue:37428", - "right": "issue:43824", + "left": "issue:45200", + "right": "issue:45325", "accept": false, - "reason": "Different missing imports in different modules and model families." + "reason": "Gemma 4 token-type defaults and Qwen2.5-VL rope-index scaling are distinct multimodal bugs." }, { "left": "issue:43844", "right": "issue:44936", "accept": false, - "reason": "DeepSpeed gradient growth and trainer.evaluate() failure are unrelated training issues." + "reason": "Gradient explosion under ZeRO-3 is unrelated to trainer.evaluate() failing after trainer.train()." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Processor.tokenizer regression with images is not the same as batched video shape output." + "reason": "Processor.tokenizer regression is different from incorrect batched video output shapes." }, { "left": "issue:43334", "right": "issue:43525", "accept": false, - "reason": "Both mention pad_token_id, but they are separate model configs and not clearly one bug." + "reason": "Same missing-attribute class of failure, but for different model configs and likely separate fixes." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "Import error for Qwen2.5-VL is unrelated to DeepSeek detokenization." + "reason": "An import error for a VL class is unrelated to a detokenization regression in deepseek-coder." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Processor API regression is not the same as quantization/offloading behavior." + "reason": "Custom processor API regression and quantization/offloading behavior are not the same code-path problem." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Both are label-handling bugs, but in different utilities with different failure modes." + "reason": "Both involve labels, but one drops tuple elements in gather_for_metrics while the other truncates Beit labels." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 numerical issue and RT-DETR memory retention are unrelated." + "reason": "SigLIP2 zero-probability outputs and RT-DETR memory leaks are unrelated." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Qwen3Moe FP8 config failure is unrelated to a Qwen2.5-VL import error." + "reason": "FineGrainedFP8Config failure for Qwen3Moe is unrelated to a missing Qwen2.5-VL import." }, { - "left": "issue:44514", - "right": "issue:44521", + "left": "issue:38175", + "right": "issue:43493", "accept": false, - "reason": "Batched padding crash and zero assistant masks are different apply_chat_template failures." + "reason": "Both mention SigLIP2, but one is zero probabilities and the other is HF-vs-JAX implementation discrepancy; not clearly the same bug." }, { - "left": "issue:38175", - "right": "issue:43493", + "left": "issue:44514", + "right": "issue:44521", "accept": false, - "reason": "SigLIP2 implementation discrepancy is not the same as a separate vision-model processor bug." + "reason": "Both are about apply_chat_template, but one is a batching/padding crash and the other is all-zero assistant masks; too different to treat as one fix." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "Different tokenizer failures for different model families and different operations." + "reason": "Different speech-model failures: tokenizer loading vs save_pretrained/tokenization." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer spacing bug is unrelated to embedding reinitialization after resize." + "reason": "Added-token spacing and post_init reinitializing output embeddings are separate tokenizer/model-weight bugs." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Embedding dtype drift in eval is unrelated to Qwen3 MoE memory usage." + "reason": "Training/eval dtype drift is unrelated to Qwen3 MoE memory usage." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "Local-path import warning is unrelated to SigLIP2 implementation discrepancy." + "reason": "Local-path import warning and SigLIP2 implementation discrepancy are unrelated." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "BigBirdTokenizer special-token registration and GraniteSpeech CI failure are different issues." + "reason": "Tokenizer special-token registration bug is not the same as a stale CI test failure." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs-related, but they concern different pipeline APIs and different behavior changes." + "reason": "Both are docs issues around pipelines, but they cover different deprecations and behaviors." }, { "left": "issue:43450", "right": "issue:43575", "accept": false, - "reason": "Batched video output shape and Qwen2-57B TP OOM are unrelated." + "reason": "Batched video shape bug and tensor-parallel OOM are unrelated." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Feature-extractor center-arg regression and Whisper empty-transcription failure are different code paths." + "reason": "Unsupported feature-extractor arg crash and Whisper empty-transcription generation failure are different issues." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "LoRA merge/tied-weights collapse is unrelated to category-count limits." + "reason": "LoRA/tied-weights collapse is unrelated to a hard category-count limit." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "DAC latent/STE mismatch is unrelated to Qwen3 Omni MoE feature-length helper behavior." + "reason": "DAC latent reconstruction mismatch and qwen3_omni_moe feature-length behavior are different code paths." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "Different subsystems: DAC quantization logic vs GraniteSpeech device override test." + "reason": "No shared bug: DAC latent mismatch vs GraniteSpeech CI test instability." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate default and batched video processor shape are unrelated." + "reason": "Audio chat-template sample-rate defaulting and video batched-shape handling are unrelated." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Processor image-size mismatch and AMP/CUDA loss crash are different bugs." + "reason": "Processor output resolution mismatch and AMP/CUDA loss crash are different bugs." } ] }, @@ -7633,16 +7593,16 @@ ], "soft_pairs": [ "issue:37428|issue:43425", - "issue:45290|issue:45381", "issue:43335|issue:43550", "issue:43606|issue:45083", + "issue:45290|issue:45381", "issue:44448|issue:44743", "issue:44857|issue:45593", "issue:45200|issue:45588", "issue:44360|issue:44561", "issue:43278|issue:43825", - "issue:29942|issue:32090", "issue:37428|issue:44351", + "issue:29942|issue:32090", "issue:43611|issue:43931", "issue:44206|issue:45823", "issue:39692|issue:43526", @@ -7650,8 +7610,8 @@ "issue:43819|issue:44493", "issue:43550|issue:44112", "issue:43335|issue:43504", - "issue:45310|issue:45478", "issue:44855|issue:45083", + "issue:45310|issue:45478", "issue:43335|issue:45083", "issue:29127|issue:32090", "issue:45588|issue:45859", @@ -7661,253 +7621,253 @@ "issue:38175|issue:44493", "issue:43335|issue:45072", "issue:44462|issue:44561", - "issue:44561|issue:44779", "issue:43720|issue:44112", - "issue:43335|issue:45593", + "issue:44561|issue:44779", + "issue:43335|issue:44112", "issue:44977|issue:45229", "issue:42490|issue:43854" ], - "prepared_review_unit_hash": "7653befb2093dfb26871099e882f47404bdf00e442b3a4a8eb1530bdb3c08198", + "prepared_review_unit_hash": "f82bb94aa980ddac020333cb79f8fa90141d7603f2ea2d472a11c38b12ea8eb4", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37851, - "estimated_input_tokens": 9463, - "estimated_eval_tokens": 19182 + "serialized_chars": 37857, + "estimated_input_tokens": 9465, + "estimated_eval_tokens": 19186 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous and does not look like a single duplicate set. The only clear duplicate soft edge is the Qwen3.5 MoE from_pretrained regression pair; the rest are distinct bugs/regressions across unrelated code paths.", - "confidence": 0.61, - "canonical_issue_reason": "issue:43611 is the broadest, clearest user-facing regression in this set: model loading breaks generically for models using `base_model_prefix`, which makes it a strong cluster anchor.", + "summary": "This cluster is still mostly heterogeneous. The only clear duplicate pair is the Qwen3.5 MoE `from_pretrained` report: issue:45310 and issue:45478 share the same title and underlying failure mode, with 45310 being the earlier, more detailed report and 45478 a later repost.", + "confidence": 0.94, + "canonical_issue_reason": "issue:45310 is the better canonical issue because it is earlier and has the longer, more detailed report for the same Qwen3.5 MoE `from_pretrained` error; issue:45478 appears to be a later repost with the same title.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43611 is the most representative and broadly applicable issue here, with a concrete loading-path failure that could plausibly attract related duplicates.", + "best_issue_reason": "issue:45310 is the strongest issue candidate overall for the duplicate cluster because it most clearly captures the underlying bug and predates issue:45478, which looks like a shorter follow-up repost.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "Flash-attention import failure vs a general Torch 2.10 compatibility issue; different root causes." - }, - { - "left": "issue:45290", - "right": "issue:45381", - "accept": false, - "reason": "Chat template crash on tool-call messages vs Qwen2.5-VL video position-id bug; unrelated code paths." + "reason": "Both are runtime/compatibility failures, but they point to different missing-symbol vs. torch-version problems." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug vs Bamba torch.compile/SDPA failure; not the same bug." + "reason": "SwitchTransformers sparse-layer initialization and Bamba torch.compile/SDPA failures are different code paths." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU offload device mismatch in bark-small vs qwen3_omni_moe output-length helper behavior; different failures." + "reason": "Bark CPU-offload device mismatch is unrelated to the qwen3_omni_moe feature-length helper bug." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat-template tool-call handling and Qwen2.5-VL video position IDs are distinct issues." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Pegasus v4/v5 output regression vs recurrent-state reset in modular_qwen3_5; separate model issues." + "reason": "Pegasus output drift and Qwen3.5 recurrent-state reset are separate model regressions." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "AMP/CUDA loss crash vs missing auxiliary losses when denoising is off; unrelated functionality." + "reason": "LwDetr AMP/CUDA crash is not the same as D-FINE's missing auxiliary losses when denoising is off." }, { "left": "issue:45200", "right": "issue:45588", "accept": false, - "reason": "Gemma4 token-type defaulting issue vs flash_attention s_aux=None crash; different affected components." + "reason": "Gemma 4 token-type defaults and flash_attention `s_aux=None` crashes are different failures." }, { "left": "issue:44360", "right": "issue:44561", "accept": false, - "reason": "DSA indexer ReLU comment vs removal of `is_torch_fx_available`; no shared underlying defect." + "reason": "DSA indexer activation behavior and removal of `is_torch_fx_available` breaking remote-code models are unrelated." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "BF16-to-FP32 eval dtype drift vs pipeline translation error messaging; not the same problem." + "reason": "Embedding dtype regression in evaluate is not the same as a pipeline error-message issue." }, { - "left": "issue:29942", - "right": "issue:32090", + "left": "issue:37428", + "right": "issue:44351", "accept": false, - "reason": "Flash Attention 2 test failures vs `_gpu_broadcast_one` NoneType error in Trainer; distinct issues." + "reason": "Both are import errors, but they involve different missing symbols and different compatibility breaks." }, { - "left": "issue:37428", - "right": "issue:44351", + "left": "issue:29942", + "right": "issue:32090", "accept": false, - "reason": "Missing `_flash_supports_window_size` import vs missing `HybridCache` import; separate import regressions." + "reason": "Flash Attention 2 test failures and `_gpu_broadcast_one` NoneType handling are separate problems." }, { "left": "issue:43611", "right": "issue:43931", "accept": false, - "reason": "Generic model-loading regression vs Qwen3-VL weight-shape mismatch; different failure modes." + "reason": "Base-model-prefix loading breakage and Qwen3-VL weight-shape mismatch are not the same bug." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "LaserFeatureExtractor `center` arg crash vs Gemma4 PLE device mismatch; unrelated." + "reason": "Unsupported `center` argument in a feature extractor is unrelated to Gemma4 device mismatch during forward." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "SigLIP2 docs example problems vs BeitImageProcessorFast `reduce_labels` bug; different artifacts." + "reason": "SigLIP2 docs example errors and BeitImageProcessorFast `reduce_labels` behavior are different artifacts." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "SwitchTransformers sparse-layer creation bug vs BigBirdTokenizer special-token decode issue; not the same underlying bug." + "reason": "SwitchTransformers sparse-layer creation and BigBirdTokenizer special-token registration are unrelated." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC `from_latents` mismatch vs unexpected position-id key behavior; unrelated model internals." + "reason": "DAC latent-forward mismatch and unexpected position-id keys are different regressions." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "Bamba SDPA/compile bug vs GraniteSpeech stale device-override CI failure; different scopes." + "reason": "Bamba SDPA/torch.compile behavior and GraniteSpeech CI stale-device testing are not the same bug." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "SwitchTransformers sparse-layer config issue vs Beit semantic-segmentation legacy-field loading issue; separate bugs." - }, - { - "left": "issue:45310", - "right": "issue:45478", - "accept": true, - "reason": "Same title and same Qwen3.5 MoE from_pretrained failure; these appear to be duplicates of the same regression." + "reason": "SwitchTransformers sparse-layer initialization and Beit legacy-field loading failure are unrelated." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 IndentationError in DebertaV2Model import vs qwen3_omni_moe feature-length helper bug; unrelated." + "reason": "Python 3.13 `IndentationError` from scripted code is unrelated to the qwen3_omni_moe helper issue." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same title, same model family, and same `from_pretrained` error; 45478 looks like a later repost of 45310." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "SwitchTransformers sparse-layer bug vs qwen3_omni_moe helper behavior; different code paths." + "reason": "SwitchTransformers sparse-layer creation and qwen3_omni_moe helper behavior are different issues." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 error-message clarity issue vs Trainer `_gpu_broadcast_one` NoneType crash; unrelated." + "reason": "LayoutLMv3 message clarity and Trainer `_gpu_broadcast_one` NoneType failure are unrelated." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "flash_attention sink-less `s_aux=None` crash vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility; separate failures." + "reason": "Both touch Qwen3.5/MoE/offload, but they describe different failure modes and code paths." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 docs example errors vs unexpected position-id key warnings; not the same bug." + "reason": "Docs example failures are not the same as runtime unexpected key/position-id behavior." }, { "left": "issue:44292", "right": "issue:45520", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime issue vs Python 3.13 `flash_attn` import KeyError; distinct causes." + "reason": "Qwen-3-8B-NVFP4 runtime failure and Python 3.13 flash-attn import KeyError are unrelated." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Pegasus output drift across versions vs Deepseek tokenizer regression in v5; different models and behaviors." + "reason": "Pegasus output differences and Deepseek tokenizer regressions are different bugs." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 zero-probability output vs position-id key warning; unrelated symptoms." + "reason": "Unexpected zero probabilities in SigLIP2 and unexpected position-id keys are different symptoms and subsystems." }, { "left": "issue:43335", "right": "issue:45072", "accept": false, - "reason": "Sparse-layer creation bug vs bfloat16 dtype mismatch in inference; not the same code-path problem." + "reason": "SwitchTransformers sparse-layer creation and bfloat16 dtype mismatches are unrelated." }, { "left": "issue:44462", "right": "issue:44561", "accept": false, - "reason": "AutoTokenizer ignoring `tokenizer.json` vs removal of `is_torch_fx_available`; unrelated loading regressions." + "reason": "Tokenizer repository-file selection and removal of a helper function are different problems." }, { - "left": "issue:44561", - "right": "issue:44779", + "left": "issue:43720", + "right": "issue:44112", "accept": false, - "reason": "`is_torch_fx_available` removal breakage vs Deepseek tokenizer regression; no shared root cause." + "reason": "BitNet packed-weight loading and a GraniteSpeech CI device-override test do not describe the same bug." }, { - "left": "issue:43720", - "right": "issue:44112", + "left": "issue:44561", + "right": "issue:44779", "accept": false, - "reason": "BitNet packed-weight unpacking bug vs GraniteSpeech stale device override CI failure; separate issues." + "reason": "A removed helper breaking remote-code models and a Deepseek tokenizer regression are unrelated." }, { "left": "issue:43335", - "right": "issue:45593", + "right": "issue:44112", "accept": false, - "reason": "SwitchTransformers sparse-layer bug vs D-FINE auxiliary-loss behavior; unrelated model logic." + "reason": "SwitchTransformers sparse-layer creation and GraniteSpeech CI device-override failure are different issues." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Qwen3.5 flash-attention generation bug vs Gemma4 multi-GPU CUDA OOM; different failure modes." + "reason": "Qwen3.5 flash-attention generation bug and Gemma4 multi-GPU OOM are not the same underlying problem." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Trailing-slash local-path import warning vs GLM-4.7-Flash unit-test load failure; not the same bug." + "reason": "Trailing-slash import warning and GLM-4.7-Flash unit-test load failure are different loading issues." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The analysis is grounded in the packet and stays conservative on the soft edges. The single accepted duplicate pair (issue:45310 / issue:45478) is well-supported by the identical title and same Qwen3.5 MoE from_pretrained regression, and the remaining rejections do not overclaim shared root cause." + "accept": false, + "feedback": "The duplicate call on issue:45310/45478 is plausible, but the justification overstates the evidence: 45478 actually has more discussion activity than 45310, so it is not grounded to say 45310 has 'slightly more discussion/context.' A safer statement would be that 45310 is earlier and slightly longer, while 45478 appears to be a later repost with the same title. The rest of the soft-edge rejections look conservative." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -7928,6 +7888,7 @@ "issue:10000", "issue:15354", "issue:28218", + "issue:29127", "issue:29942", "issue:32090", "issue:33357", @@ -7937,7 +7898,6 @@ "issue:40444", "issue:41720", "issue:42490", - "issue:43065", "issue:43278", "issue:43329", "issue:43412", @@ -7950,7 +7910,6 @@ "issue:43873", "issue:43927", "issue:43976", - "issue:43994", "issue:44448", "issue:44493", "issue:44554", @@ -7977,20 +7936,20 @@ "soft_pairs": [ "issue:40444|issue:43526", "issue:34689|issue:35141", - "issue:43606|issue:44857", "issue:44448|issue:44977", + "issue:43606|issue:44857", "issue:43723|issue:44554", "issue:44857|issue:45072", "issue:44977|issue:45563", - "issue:44568|issue:45491", "issue:28218|issue:32090", "issue:43927|issue:45698", + "issue:44568|issue:45491", "issue:43701|issue:45083", "issue:43550|issue:45083", "issue:45198|issue:45803", "issue:37428|issue:44987", - "issue:45372|issue:45588", "issue:29942|issue:35141", + "issue:45372|issue:45588", "issue:45459|issue:45712", "issue:43873|issue:44493", "issue:32090|issue:33357", @@ -7999,214 +7958,202 @@ "issue:43412|issue:45588", "issue:41720|issue:45229", "issue:43278|issue:43329", - "issue:43976|issue:45440", "issue:10000|issue:15354", + "issue:43976|issue:45440", "issue:45372|issue:45406", "issue:42490|issue:43575", - "issue:43065|issue:43994" + "issue:28218|issue:29127" ], - "prepared_review_unit_hash": "05b3d6e92b6291034d8febae792ffb359f3897317620cba0cfa05fc5e85ffbe9", + "prepared_review_unit_hash": "520a581f8117e9e5129cd400c489b7d0c44e16462206dc8a13ed6c861c941775", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36594, - "estimated_input_tokens": 9149, - "estimated_eval_tokens": 18554 + "serialized_chars": 36076, + "estimated_input_tokens": 9019, + "estimated_eval_tokens": 18294 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the soft pairs mostly share broad subsystem vocabulary, but they do not look like the same underlying bug or change. I found no strong duplicate PR/issue center, so the best representative is only a loose cluster hub rather than a true canonical duplicate.", - "confidence": 0.67, - "canonical_issue_reason": "issue:43873 is the broadest open infra bug in the set ('offloading not working as expected with quantization') and has the most inbound references, so it is the closest thing to a hub; the other items are mostly model-specific and not true duplicates.", + "summary": "The cluster is mostly a set of unrelated Transformer bugs that only share broad subsystem similarity (tokenizers, multimodal, loading, attention, offload). I did not find any soft pair that looks like the same underlying defect closely enough to merge.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43873 has the highest engagement and inbound references, making it the most defensible representative issue if a single canonical item is needed.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the safest global representative because it is broad, still active, and most likely to attract related reports, even though the cluster overall is too mixed to form a real duplicate group.", + "best_issue_reason": "issue:43873 is the strongest standalone issue in the set: high discussion activity, multiple inbound references, and a broad runtime/infrastructure impact.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "Different problems: multimodal finetuning with multiple images vs a BeitImageProcessorFast label-reduction bug." + "reason": "Both are multimodal/processor-related, but one is an IterableDataset multi-image failure and the other is a BeitImageProcessorFast label-reduction bug; different code paths and symptoms." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "One is a model-loading regression for Llama 3.2 Vision, the other is embedding reinitialization during resize/eval." + "reason": "Model-loading failure for Llama 3.2 vision is not the same defect as token embedding reinitialization after resize when tie_word_embeddings is false." }, { - "left": "issue:43606", - "right": "issue:44857", + "left": "issue:44448", + "right": "issue:44977", "accept": false, - "reason": "CPU offload device mismatch and float16 AMP loss crashes are unrelated failure modes." + "reason": "Different models and failure modes: Pegasus output drift across versions versus Qwen3.5 flash-attention generation problems." }, { - "left": "issue:44448", - "right": "issue:44977", + "left": "issue:43606", + "right": "issue:44857", "accept": false, - "reason": "Different concrete bugs: Pegasus output drift across versions vs Qwen3.5 flash-attention generation failure." + "reason": "CPU offload device mismatch for bark-small is unrelated to the float16 AMP crash in LwDetrImageLoss." }, { "left": "issue:43723", "right": "issue:44554", "accept": false, - "reason": "Tokenizer loading in v5 and an MPS attention correctness issue are not the same code-path." + "reason": "Tokenizer loading in v5 and an MPS attention correctness issue are separate code paths and bugs." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "These are different dtype-related issues in different paths: AMP loss crash vs inference dtype mismatch." + "reason": "Both mention dtype/precision, but one is a loss crash under AMP and the other is a bfloat16 inference mismatch in different models." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Generation correctness with flash-attention is not the same as a stale warning about num_return_sequences." - }, - { - "left": "issue:44568", - "right": "issue:45491", - "accept": false, - "reason": "Tokenizer special-token omission and Gemma3 NaN embeddings are unrelated bugs." + "reason": "Flash-attention generation failure and a stale warning about num_return_sequences are unrelated." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Added-token spacing behavior and Trainer broadcast NoneType errors are unrelated." + "reason": "Tokenizer spacing behavior and Trainer GPU broadcast NoneType errors are not the same underlying issue." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Both touch save/load, but one loses custom token IDs in config while the other restores the wrong custom module; different serialization paths." + "reason": "Both involve save/load behavior, but one loses custom token IDs in config while the other loads the wrong custom module after save_pretrained." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Missing BOS/EOS insertion in a tokenizer is unrelated to Gemma3 NaN embeddings from sliding-window attention." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint resume key mismatch is unrelated to qwen3_omni_moe feature-length helper behavior." + "reason": "Checkpoint key mismatch and a qwen3_omni_moe helper-function bug are different defects." }, { "left": "issue:43550", "right": "issue:45083", "accept": false, - "reason": "torch.compile/SDPA failure and feature-length calculation bugs are different issues." + "reason": "torch.compile/SDPA failure in Bamba-9B-v2 does not match the qwen3_omni_moe length helper issue." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure and a bare-except bug in Fuyu tensor conversion are not the same defect." + "reason": "Wav2Vec2 save/tokenization failure and FuyuBatchFeature's bare except swallowing errors are related to tokenization but not the same concrete bug." }, { "left": "issue:37428", "right": "issue:44987", "accept": false, - "reason": "The first is a missing flash-attn import symbol; the second is a generic model-loading failure with no clear same root cause." + "reason": "Both involve loading/import issues, but there is no evidence they share the same failing code path or root cause." }, { - "left": "issue:45372", - "right": "issue:45588", + "left": "issue:29942", + "right": "issue:35141", "accept": false, - "reason": "Gemma4 processor import dependency failure and flash_attention.py s_aux handling are distinct bugs." + "reason": "Failing Flash Attention 2 tests are not the same defect as embedding reinitialization after resize." }, { - "left": "issue:29942", - "right": "issue:35141", + "left": "issue:45372", + "right": "issue:45588", "accept": false, - "reason": "Flash Attention 2 test failures and embedding reinitialization on resize are unrelated." + "reason": "Gemma4 processor loading breaks because of a missing mistral_common symbol; flash_attention.py crashes on s_aux=None in sink-less models. Different dependency and runtime bugs." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Tokenizer error masking from protobuf handling and dummy-class cleanup/dir leakage are different concerns." + "reason": "Tokenizer exception masking from protobuf handling is unrelated to leftover dummy class cleanup and dir(transformers) leakage." }, { "left": "issue:43873", "right": "issue:44493", "accept": false, - "reason": "Quantization/offloading behavior and unexpected position-id key warnings do not share a concrete root cause." + "reason": "Quantization/offloading behavior and unexpected position-id key warnings are different problems." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Trainer broadcast NoneType and a MacOS CLIP bus error are not the same failure." + "reason": "Trainer _gpu_broadcast_one NoneType errors and MacOS bus errors on a CLIP model are separate runtime failures." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Both involve large-model resource issues, but offloading/quantization misbehavior is not the same as multi-GPU OOM." - }, - { - "left": "issue:45245", - "right": "issue:45563", - "accept": false, - "reason": "A category-count runtime limit and a stale generation warning are unrelated." - }, - { - "left": "issue:43412", - "right": "issue:45588", - "accept": false, - "reason": "Executorch export problems and a sink-less flash-attention AttributeError are different code paths." + "reason": "Both touch memory/offload concerns, but one is a quantization offloading bug and the other is a Gemma4 multi-GPU OOM; not the same concrete issue." }, { "left": "issue:41720", "right": "issue:45229", "accept": false, - "reason": "Auto device-mapping cudaErrorAssert and multi-GPU OOM are different resource failures." + "reason": "CUDA assert with auto device mapping on Qwen3 is a different failure mode from multi-GPU OOM on Gemma4 31B." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "BF16-to-FP32 dtype drift and an undefined-variable multimodal token helper bug are unrelated." + "reason": "Embedding dtype regression in eval is unrelated to an undefined-variable bug in the video multimodal token counting path." }, { - "left": "issue:43976", - "right": "issue:45440", + "left": "issue:10000", + "right": "issue:15354", "accept": false, - "reason": "Python version incompatibility and DeepSeekV3MoE divergence are not the same bug." + "reason": "German DistilBertModel issue and GeneratorExp unsupported by torch.jit.script are unrelated legacy bugs." }, { - "left": "issue:10000", - "right": "issue:15354", + "left": "issue:43976", + "right": "issue:45440", "accept": false, - "reason": "German DistilBertModel error and torch.jit GeneratorExp export support are unrelated." + "reason": "Python version compatibility and DeepseekV3MoE divergence are not the same issue." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Both mention Gemma4, but one is an import dependency issue and the other is a missing _tokenizer attribute in serve." + "reason": "Both concern Gemma4 processor loading, but one fails on an import from mistral_common and the other crashes in serve due to a missing _tokenizer attribute; insufficient evidence of a shared root cause." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "A trailing-slash import warning and tensor-parallel OOM are unrelated." + "reason": "Local-path import warning and Qwen2-57B-A14B-Instruct tensor-parallel OOM are unrelated." }, { - "left": "issue:43065", - "right": "issue:43994", + "left": "issue:28218", + "right": "issue:29127", "accept": false, - "reason": "Dummy Conv2d cleanup in Sam3PixelDecoder and SigLIP2 nonsensical outputs are different issues." + "reason": "Tokenizer spacing behavior and LayoutLMv3 box validation error messages are different bugs." } ] }, @@ -8231,7 +8178,6 @@ "cluster_id": "cluster-10000-402", "nodes": [ "issue:25251", - "issue:28218", "issue:29127", "issue:29942", "issue:30064", @@ -8251,11 +8197,13 @@ "issue:43421", "issue:43653", "issue:43756", + "issue:43824", "issue:43873", "issue:43994", "issue:44186", "issue:44206", "issue:44265", + "issue:44279", "issue:44361", "issue:44448", "issue:44451", @@ -8279,7 +8227,7 @@ "issue:45823" ], "soft_pairs": [ - "issue:28218|issue:29127", + "issue:43065|issue:43994", "issue:43065|issue:45412", "issue:45372|issue:45459", "issue:44857|issue:45588", @@ -8287,223 +8235,230 @@ "issue:43653|issue:44743", "issue:45588|issue:45823", "issue:43278|issue:43381", - "issue:43066|issue:44462", "issue:44206|issue:44479", - "issue:43994|issue:45412", "issue:25251|issue:32090", + "issue:43994|issue:45412", "issue:44186|issue:45198", + "issue:43066|issue:44462", "issue:44451|issue:45520", "issue:44361|issue:45198", "issue:42915|issue:42994", "issue:44871|issue:44977", "issue:44265|issue:45198", "issue:43756|issue:45440", - "issue:43421|issue:45491", "issue:29942|issue:30064", + "issue:43421|issue:45491", "issue:43257|issue:43278", "issue:33357|issue:39692", "issue:29127|issue:34567", "issue:44610|issue:45412", "issue:32090|issue:33290", "issue:42490|issue:43873", - "issue:45440|issue:45593" + "issue:45440|issue:45593", + "issue:43824|issue:44279" ], - "prepared_review_unit_hash": "3af4d7545024d9f5120b1043c2dce1be20a68aa6d9d4a7fed36bc6f661835c34", + "prepared_review_unit_hash": "b6020c409e02a6276ce1ebe25d854f04d4b09696ef67431b779dbff2ed0d9124", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 28, - "serialized_chars": 35486, - "estimated_input_tokens": 8872, - "estimated_eval_tokens": 18000 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36352, + "estimated_input_tokens": 9088, + "estimated_eval_tokens": 18432 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster looks heterogeneous rather than like one duplicate family; all soft-pair candidates are distinct bugs/features and should be kept separate.", - "confidence": 0.86, - "canonical_issue_reason": "issue:44462 is the most representative standalone bug here: concrete, reproducible, user-facing, and with inbound references; the rest of the items are a mix of unrelated tokenizer, training, vision, quantization, and runtime issues.", + "summary": "The cluster is highly heterogeneous: the soft pairs mostly share only broad subsystem labels (tokenizers, quantization, trainer, vision, etc.) but not the same concrete failure mode or fix. I would not merge any of the proposed pairs as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "If one issue must anchor the cluster, issue #43873 has the strongest engagement (highest inbound refs/discussion) and is open, but it is not a true duplicate representative of the rest.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44462 is the best anchor for triage because it is narrowly scoped, clearly described, and has external references that suggest real impact.", + "best_issue_reason": "Issue #43873 is the best single anchor by activity and references, though the cluster does not contain a clear canonical duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:28218", - "right": "issue:29127", + "left": "issue:43065", + "right": "issue:43994", "accept": false, - "reason": "Different subsystems and failure modes: tokenizer spacing vs. LayoutLMv3 error-message clarity." + "reason": "Different models and failures: Sam3PixelDecoder dummy Conv2d vs SigLIP2 incorrect outputs." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "No shared code-path: Sam3PixelDecoder dummy layer is unrelated to RT-DETR memory cleanup." + "reason": "Dummy layer implementation issue vs RT-DETR memory not released; unrelated bug classes." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma processor import breakage is unrelated to tokenizer error masking when protobuf is missing." + "reason": "Gemma 4 processor import breakage due to missing ReasoningEffort vs tokenizer error masking when protobuf is absent." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "Different bugs in different paths: LwDetr loss under AMP vs flash-attention crash on sink-less models." + "reason": "AMP/CUDA loss crash vs flash_attention AttributeError on sink-less models; different code paths." }, { "left": "issue:44448", "right": "issue:44466", "accept": false, - "reason": "Both are version/regression complaints, but one is generation output drift and the other is serialization of tied weights." + "reason": "Both are v5 regressions, but one is generation-output difference and the other is serialization of tied weights." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "Tokenizer special-token registration bug is unrelated to Qwen3.5 recurrent-state reset logic." + "reason": "Tokenizer special-token registration bug vs Qwen3.5 recurrent-state reset bug; unrelated." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "One is a flash-attention attribute error; the other is a device-map forward mismatch in Gemma4 PLE." + "reason": "Flash-attention aux-state crash vs Gemma4 device_map forward mismatch; not the same defect." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "Embedding dtype mismatch in eval is not the same issue as gradient checkpointing being disallowed in eval mode." + "reason": "Eval-mode dtype drift vs gradient checkpointing forbidden in eval; different symptoms and fixes." }, { - "left": "issue:43066", - "right": "issue:44462", + "left": "issue:44206", + "right": "issue:44479", "accept": false, - "reason": "Tokenizer decoder-type regression is a different loading bug than ignoring tokenizer.json from a repo." + "reason": "Unsupported feature argument in a feature extractor vs video-input regression for Qwen VL models." }, { - "left": "issue:44206", - "right": "issue:44479", + "left": "issue:25251", + "right": "issue:32090", "accept": false, - "reason": "Feature extractor center-arg crash and Qwen video-input regression are unrelated code paths." + "reason": "Pipeline top_k nesting behavior vs trainer broadcast NoneType error; unrelated." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 model/pipeline output corruption is unrelated to RT-DETR memory release behavior." + "reason": "SigLIP2 incorrect inference outputs vs RT-DETR memory leak; no shared underlying bug." }, { - "left": "issue:25251", - "right": "issue:32090", + "left": "issue:44186", + "right": "issue:45198", "accept": false, - "reason": "Pipeline top_k nesting bug is unrelated to Trainer broadcast failing on NoneType." + "reason": "LayoutLMv2Tokenizer NER/padding crash vs Wav2Vec2 save_pretrained/tokenization failure." }, { - "left": "issue:44186", - "right": "issue:45198", + "left": "issue:43066", + "right": "issue:44462", "accept": false, - "reason": "LayoutLMv2 tokenizer NER/padding crash is unrelated to Wav2Vec2 save_pretrained/tokenization failures." + "reason": "Tokenizer decoder type in v5 vs AutoTokenizer ignoring tokenizer.json; both tokenizer-related but different failures." }, { "left": "issue:44451", "right": "issue:45520", "accept": false, - "reason": "Model-loading failure for a specific checkpoint is unrelated to a flash_attn import KeyError on Python 3.13." + "reason": "Model loading failure for ScandiBERT vs Python 3.13 flash_attn import mapping issue." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "MLukeTokenizer task AttributeError and Wav2Vec2 save/tokenization failure are distinct tokenizer/model issues." + "reason": "MLukeTokenizer task-time AttributeError vs Wav2Vec2 save/tokenization failure." }, { "left": "issue:42915", "right": "issue:42994", "accept": false, - "reason": "One concerns Qwen3 MoE loading with FP8 config; the other is quantized model saving." + "reason": "Quantization-related, but one is FP8 loading/runtime failure and the other is quantized model saving." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "EOS-token config inconsistency is unrelated to Qwen3.5 flash-attention generation problems." + "reason": "Gemma3 eos token config mismatch vs Qwen3.5 flash-attention generation regression." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export failure from torch_compilable_check is a different export path than Wav2Vec2 save/tokenization." + "reason": "torch.export failure with torch_compilable_check vs Wav2Vec2 tokenization/save_pretrained issues." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Different models and different defects: Smollm3 RoPE-layer mismatch vs DeepseekV3MoE implementation divergence." + "reason": "RoPE layer count mismatch in Smollm3 vs DeepseekV3MoE implementation divergence; different models and defects." }, { - "left": "issue:43421", - "right": "issue:45491", + "left": "issue:29942", + "right": "issue:30064", "accept": false, - "reason": "Runtime special-token post-processor update bug is unrelated to Gemma3 NaN embeddings in mixed-length batches." + "reason": "FlashAttention test failures vs void segmentation map preprocessing failure; unrelated." }, { - "left": "issue:29942", - "right": "issue:30064", + "left": "issue:43421", + "right": "issue:45491", "accept": false, - "reason": "Flash Attention 2 CI failures and void segmentation-map processing are unrelated regressions." + "reason": "Tokenizer post-processor not refreshed after special token changes vs NaN embeddings with sliding-window attention." }, { "left": "issue:43257", "right": "issue:43278", "accept": false, - "reason": "Qwen3 MoE weight-conversion failure is unrelated to embedding dtype changing between train and eval." + "reason": "Qwen3 MoE weight conversion problem vs embedding dtype mismatch during eval; different subsystems." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "MacOS CLIP bus error and SigLIP2 docs-example errors are not the same bug or code path." + "reason": "MacOS bus error on community CLIP model vs SigLIP2 documentation/example problems." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "Clarity of a layoutlmv3 error message is unrelated to TrainerState token-count tracking not updating." + "reason": "LayoutLMv3 error messaging issue vs TrainerState token counter not updating; unrelated." }, { "left": "issue:44610", "right": "issue:45412", "accept": false, - "reason": "OmDet-Turbo processor resolution mismatch is unrelated to RT-DETR object lifetime/memory release." + "reason": "Processor image-size mismatch vs model memory leak; no common bug." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer NoneType broadcast error and DeepSpeed Adafactor OOM are different training failures." + "reason": "Trainer broadcast NoneType error vs DeepSpeed Adafactor OOM; both training-related but not the same issue." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Local-path import warning is unrelated to quantization offloading behavior." + "reason": "Trailing-slash local model import warning vs quantization offloading not working; different code paths." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "DeepseekV3MoE implementation drift and D-FINE auxiliary-loss omission are different model-specific issues." + "reason": "DeepseekV3MoE implementation mismatch vs D-FINE auxiliary-loss bug; unrelated model families." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Specific missing Qwen2_5_VL import vs generic dependency issue; too broad to be the same bug." } ] }, @@ -8548,14 +8503,13 @@ "issue:43701", "issue:43723", "issue:43761", - "issue:43824", "issue:43931", "issue:44062", "issue:44077", "issue:44079", "issue:44186", + "issue:44188", "issue:44206", - "issue:44279", "issue:44387", "issue:44442", "issue:44464", @@ -8572,236 +8526,244 @@ "issue:45479", "issue:45491", "issue:45588", + "issue:45803", "issue:45834", "issue:45859", "issue:45907" ], "soft_pairs": [ - "issue:43824|issue:44279", "issue:43653|issue:44442", "issue:43761|issue:45859", "issue:43526|issue:45834", - "issue:45103|issue:45372", "issue:43381|issue:43701", - "issue:43526|issue:44186", + "issue:45103|issue:45372", "issue:29942|issue:36010", + "issue:43526|issue:44186", "issue:29127|issue:33357", "issue:43334|issue:43531", "issue:44977|issue:45245", "issue:29127|issue:35141", "issue:44464|issue:45491", - "issue:42915|issue:43931", "issue:44442|issue:45072", + "issue:42915|issue:43931", "issue:43262|issue:44206", "issue:43526|issue:44857", "issue:42490|issue:43381", "issue:44062|issue:45907", "issue:43441|issue:45588", - "issue:43232|issue:43388", "issue:44743|issue:44811", - "issue:43646|issue:44077", + "issue:43232|issue:43388", "issue:43575|issue:43701", + "issue:43646|issue:44077", "issue:43761|issue:44079", "issue:43575|issue:44387", "issue:45003|issue:45440", - "issue:43723|issue:45479" + "issue:43723|issue:45479", + "issue:44442|issue:45803", + "issue:43761|issue:44188" ], - "prepared_review_unit_hash": "72f91e550386cdbb74aee4678444bebbd63dcfe75dee13476ebd6039d6d6da79", + "prepared_review_unit_hash": "c045f41cb130e5f3765fff2ec841f8f08dfa8aaef1c59f226f3200b926418f7e", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36208, - "estimated_input_tokens": 9052, - "estimated_eval_tokens": 18360 + "soft_pair_count": 29, + "serialized_chars": 36586, + "estimated_input_tokens": 9147, + "estimated_eval_tokens": 18550 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the items span unrelated tokenizer, generation, model-loading, training, flash-attention, and processor bugs. None of the soft pairs look like the same underlying defect, so no duplicate merges are warranted.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "The set is heterogeneous: these are not duplicate reports, but unrelated regressions and edge cases across tokenizers, generation, model loading, attention kernels, and multimodal processors. No soft pair looks safe to merge as the same underlying bug.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 43761 is the best anchor if one is needed: it is a clear, self-contained regression with a precise API, expected behavior, and reproducible symptom.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45003 is the best representative issue because it is a concrete, self-contained core-library bug in modeling_utils with a clear code path and broad impact.", + "best_issue_reason": "43761 is the most actionable and well-scoped issue in the set; the bug is concrete and isolated enough to serve as a representative anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43824", - "right": "issue:44279", - "accept": false, - "reason": "Different failure modes: Qwen2.5-VL import error vs a vague dependency issue; no shared code path." - }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "Both involve tokenizers, but one is a BigBird special-token decoding bug and the other is an AutoTokenizer load failure for a different model." + "reason": "Both involve tokenizers, but one is a special-token decode bug and the other is an AutoTokenizer loading failure; different code paths." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIPVision hidden_states regression and Qwen3_5Moe CPU offload incompatibility are unrelated model-specific bugs." + "reason": "A CLIPVision hidden-state regression and a Qwen3 MoE rotary/cpu-offload issue are unrelated model-specific bugs." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "Different processors and symptoms: Beit reduce_labels bug vs Kosmos2.5 OCR index error." - }, - { - "left": "issue:45103", - "right": "issue:45372", - "accept": false, - "reason": "auto_docstring annotation parsing crash is unrelated to the Gemma 4 processor import failure." + "reason": "One is a BEiT Fast image label reduction bug; the other is a Kosmos2.5 long-OCR index error. No same underlying failure." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Training-loop issues in different places: gradient checkpointing eval-mode restriction vs checkpoint key mismatch." + "reason": "Gradient checkpointing in eval mode and checkpoint resume key mismatch are distinct training-state issues." }, { - "left": "issue:43526", - "right": "issue:44186", + "left": "issue:45103", + "right": "issue:45372", "accept": false, - "reason": "Both are vision/text preprocessing issues, but they hit different classes and code paths with different root causes." + "reason": "Auto-docstring annotation parsing and Gemma processor loading/import dependency issues are different subsystems." }, { "left": "issue:29942", "right": "issue:36010", "accept": false, - "reason": "Flash Attention test failures and a GenerationMixin import error are not the same bug." + "reason": "Flash Attention 2 test failures and GenerationMixin import errors are unrelated regression classes." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Both touch layout/image tokenization, but the BEiT label reduction bug and LayoutLMv2 NER/padding crash are not the same defect." }, { "left": "issue:29127", "right": "issue:33357", "accept": false, - "reason": "A poor error message in LayoutLMv3 is unrelated to a MacOS bus error loading a CLIP model." + "reason": "LayoutLMv3 box-content validation and a macOS bus error loading CLIP are completely different failures." }, { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Both mention Qwen3, but one is missing pad_token_id and the other is a sliding_window regression in MoE models." + "reason": "Missing pad_token_id in Qwen3-VL text config and sliding_window behavior in Qwen3-MoE are separate model bugs." }, { "left": "issue:44977", "right": "issue:45245", "accept": false, - "reason": "Qwen3.5 flash-attention generation bug and category-count limit error are completely different issues." + "reason": "Flash-attention generation corruption and category-count overflow are unrelated; no shared code path." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "LayoutLMv3 box-validation messaging and token-embedding reinit after resize are unrelated." + "reason": "One is an error-message clarity issue for LayoutLMv3; the other is embedding reinitialization during post_init. Different bugs." }, { "left": "issue:44464", "right": "issue:45491", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is not the same as Gemma3 NaN embeddings from sliding-window attention." + "reason": "Compiled chunked generation inconsistency and Gemma3 mixed-length NaNs stem from different attention/generation paths." }, { - "left": "issue:42915", - "right": "issue:43931", + "left": "issue:44442", + "right": "issue:45072", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8Config failure and Qwen3-VL weight-shape mismatch are distinct loading/configuration problems." + "reason": "Tokenizer loading failure and bfloat16 dtype mismatch in inference are different classes of problems." }, { - "left": "issue:44442", - "right": "issue:45072", + "left": "issue:42915", + "right": "issue:43931", "accept": false, - "reason": "Tokenizer loading failure and bfloat16 dtype mismatch in model wrappers do not share the same underlying defect." + "reason": "Fine-grained FP8 config failure for Qwen3Moe and Qwen3-VL weight-shape mismatch are separate model-loading bugs." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Audio chat-template sampling-rate default bug and LasrFeatureExtractor center-arg crash are unrelated processor issues." + "reason": "Audio chat-template sampling-rate handling and feature-extractor unsupported-arg crash are unrelated processor issues." }, { "left": "issue:43526", "right": "issue:44857", "accept": false, - "reason": "Beit reduce_labels truncation and LwDetrImageLoss AMP/CUDA crash are different components and failure modes." + "reason": "A label-reduction bug in BEiT fast processing does not match an AMP/CUDA image-loss crash." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Import warning on local-path loading is unrelated to gradient checkpointing in eval mode." + "reason": "Local-path trailing-slash import warning and gradient-checkpointing eval-mode restriction are unrelated." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are different API bugs." + "reason": "AddedToken constructor argument duplication and list_repo_files revision forwarding are different API bugs." }, { "left": "issue:43441", "right": "issue:45588", "accept": false, - "reason": "Different flash-attention problems: Ministral v5 RC failure vs s_aux=None crash in integrations/flash_attention.py." - }, - { - "left": "issue:43232", - "right": "issue:43388", - "accept": false, - "reason": "Generation kwargs sync after sync_gpus and metric-gathering label truncation are separate training/eval bugs." + "reason": "Ministral FlashAttention failure and sink-less flash_attention.py s_aux=None crash are distinct attention implementations/fixes." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Qwen3_5 recurrent-state reset and Whisper batch_decode skip_special_tokens are unrelated behavioral bugs." + "reason": "Qwen3.5 recurrent-state reset and Whisper batch_decode skip_special_tokens handling are not the same underlying issue." }, { - "left": "issue:43646", - "right": "issue:44077", + "left": "issue:43232", + "right": "issue:43388", "accept": false, - "reason": "Custom model init breakage in Transformers 5.0.0 and patchtsmixer's post_init constraint are not the same issue." + "reason": "Generation kwargs synchronization after sync_gpus and metric-gather label truncation are different training/generation paths." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "Tensor-parallel OOM on Qwen2-57B and resume_from_checkpoint key mismatch do not overlap." + "reason": "Tensor-parallel OOM on model load and resume_from_checkpoint key mismatch are unrelated." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Transformers 5 custom initialization breakage and patchtsmixer post_init allowance are not the same code-path problem." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "A CLIPVision forward regression is not the same as a generic ModelOutput key-assignment bug." + "reason": "Hidden_states missing from CLIPVision forward and ModelOutput key assignment when a value was previously None are separate bugs." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Both mention OOM, but one is TP load OOM for a specific model and the other is increased reserved memory under int4 quantization." + "reason": "Qwen2-57B tp OOM on load and increased reserved memory under int4 quantization are different memory-pressure issues." }, { "left": "issue:45003", "right": "issue:45440", "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils and DeepseekV3MoE divergence from remote implementation are different code paths." + "reason": "Unsafe sys.modules access in modeling_utils and DeepseekV3MoE divergence from remote implementation are unrelated." }, { "left": "issue:43723", "right": "issue:45479", "accept": false, - "reason": "AutoTokenizer loading failure and zero-loss classification behavior are unrelated problems." + "reason": "Tokenizer loading in v5 and degenerate zero-loss sequence classification are different APIs and symptoms." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "AutoTokenizer loading failure and FuyuBatchFeature exception swallowing are unrelated tokenizer/feature-processing bugs." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVision hidden_states regression and attention-kernel divergence under torch.compile are separate model execution issues." } ] }, @@ -8825,6 +8787,8 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-402", "nodes": [ + "issue:25251", + "issue:28282", "issue:33290", "issue:34567", "issue:34689", @@ -8836,6 +8800,7 @@ "issue:42175", "issue:42490", "issue:42907", + "issue:43065", "issue:43316", "issue:43381", "issue:43404", @@ -8847,17 +8812,16 @@ "issue:43643", "issue:43720", "issue:43749", - "issue:43761", "issue:43827", "issue:43873", "issue:43931", "issue:44077", - "issue:44188", "issue:44190", "issue:44442", "issue:44479", "issue:44496", "issue:44514", + "issue:44617", "issue:44743", "issue:44855", "issue:44871", @@ -8870,24 +8834,21 @@ "issue:45372", "issue:45491", "issue:45698", - "issue:45759", - "issue:45803" + "issue:45759" ], "soft_pairs": [ - "issue:44442|issue:45803", "issue:44479|issue:45290", - "issue:43761|issue:44188", "issue:43316|issue:43525", - "issue:42907|issue:45698", "issue:44442|issue:44898", + "issue:42907|issue:45698", "issue:42490|issue:44743", "issue:33290|issue:34567", "issue:34689|issue:43611", - "issue:39692|issue:45491", "issue:33290|issue:35141", "issue:33290|issue:34689", - "issue:43931|issue:45216", + "issue:39692|issue:45491", "issue:43381|issue:43873", + "issue:43931|issue:45216", "issue:43493|issue:43827", "issue:44077|issue:44479", "issue:43316|issue:45759", @@ -8903,216 +8864,218 @@ "issue:43540|issue:44514", "issue:43316|issue:44877", "issue:42175|issue:43316", - "issue:43404|issue:43441" + "issue:43065|issue:44617", + "issue:43404|issue:43441", + "issue:25251|issue:28282" ], - "prepared_review_unit_hash": "03b9cab88a02228737fbda707b821ebb276a1152d6ba851ef14d84c2e35fec37", + "prepared_review_unit_hash": "3fc865d5a30c3dee3013d22b6a262ec7c7df624f6f4817c4603faf4b29296f50", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36437, - "estimated_input_tokens": 9110, - "estimated_eval_tokens": 18476 + "serialized_chars": 36864, + "estimated_input_tokens": 9216, + "estimated_eval_tokens": 18688 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the soft pairs cover unrelated bugs across tokenizers, vision, config reloads, multimodal chat, trainer APIs, and loading/saving regressions. None of the proposed soft edges look like true duplicates of the same underlying defect, so I reject them all.", + "summary": "These are not duplicates as a cluster: the items cover many unrelated issues across loading, config, training, docs, quantization, and multimodal/video paths. All soft-similarity pairs are superficial and should be rejected.", "confidence": 0.95, - "canonical_issue_reason": "No clear canonical duplicate exists in this set. If one issue must serve as an anchor, issue #43749 has the strongest triage signal (discussion plus inbound references) and is a broad runtime/regression report.", - "canonical_pr_reason": null, - "best_issue_reason": "Issue #43749 is the best single representative by external signal and breadth of impact, but it still does not match the rest closely enough to act as a true duplicate hub.", - "best_pr_reason": null, + "canonical_issue_reason": "No canonical issue: the set is heterogeneous and does not represent one underlying bug or change.", + "canonical_pr_reason": "No PRs are present in this cluster, so no canonical PR can be selected.", + "best_issue_reason": "No single best issue for deduping; none of the issues is clearly central enough to serve as a representative for the others.", + "best_pr_reason": "No PRs are present in this cluster, so no best PR exists.", "soft_edge_verdicts": [ - { - "left": "issue:44442", - "right": "issue:45803", - "accept": false, - "reason": "Different bugs: FastSpeech2Conformer tokenizer loading vs a Fuyu bare-except masking errors/KeyboardInterrupt." - }, { "left": "issue:44479", "right": "issue:45290", "accept": false, - "reason": "Both are multimodal/chat regressions, but one is a video-input regression and the other is an assistant tool-call chat_template crash." - }, - { - "left": "issue:43761", - "right": "issue:44188", - "accept": false, - "reason": "Unrelated code paths: CLIPVision hidden_states handling vs attention-kernel divergence under torch.compile." + "reason": "Both involve video/chat-related failures, but one is a video-input regression and the other is a chat-template crash on tool-call messages; different code paths and symptoms." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Both concern config APIs, but they are different model-specific attribute mismatches with different symptoms and fixes." + "reason": "Both are config/API discrepancies, but they concern different models and different missing fields; not the same bug." }, { - "left": "issue:42907", - "right": "issue:45698", + "left": "issue:44442", + "right": "issue:44898", "accept": false, - "reason": "Save/load are both involved, but one is dequantized model saving and the other is loading the wrong custom module after save_pretrained." + "reason": "Tokenizer loading failure vs image-classification interpolation bug; no shared underlying code path." }, { - "left": "issue:44442", - "right": "issue:44898", + "left": "issue:42907", + "right": "issue:45698", "accept": false, - "reason": "Different model families and failures: tokenizer import/loading vs Perceiver image-classification resize/interpolation behavior." + "reason": "Both touch save/load behavior, but one is about saving dequantized weights and the other about loading the wrong custom module after save_pretrained; distinct failures." }, { "left": "issue:42490", "right": "issue:44743", "accept": false, - "reason": "Completely different areas: local-path import warning vs recurrent state reset with cache in modular Qwen3.5 code." + "reason": "Local-path import warning and recurrent-state reset in Qwen3.5 are unrelated." }, { "left": "issue:33290", "right": "issue:34567", "accept": false, - "reason": "AdaFactor/DeepSpeed OOM is unrelated to TrainerState.num_input_tokens_seen not updating." + "reason": "DeepSpeed/Adafactor OOM and TrainerState token-count tracking are unrelated training issues." }, { "left": "issue:34689", "right": "issue:43611", "accept": false, - "reason": "Both are loading regressions, but one is a Llama vision model breakage and the other is a base_model_prefix-related loading issue." - }, - { - "left": "issue:39692", - "right": "issue:45491", - "accept": false, - "reason": "Docs-example errors are unrelated to Gemma3 NaN embeddings in mixed-length GPU batching." + "reason": "Both are model-loading regressions, but they involve different loading mechanisms and different root causes." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "Different subsystems: optimizer memory blow-up vs embedding resize/post_init reinitialization." + "reason": "OOM in optimizer setup vs embedding reinitialization during post_init; not the same defect." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "No shared bug: DeepSpeed optimizer OOM is unrelated to model-loading failure." + "reason": "Different failures: optimizer memory usage versus specific model-loading breakage." }, { - "left": "issue:43931", - "right": "issue:45216", + "left": "issue:39692", + "right": "issue:45491", "accept": false, - "reason": "Both mention Qwen models, but one is a shape-mismatch load failure and the other is an incorrect save_pretrained regression." + "reason": "Docs example errors for SigLIP2 versus Gemma3 NaN embeddings; unrelated surfaces and code paths." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Gradient checkpointing in eval mode is unrelated to quantization/offloading behavior." + "reason": "Gradient checkpointing in eval mode and quantization/offloading behavior are different bugs." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Weight-shape mismatch on load versus incorrect checkpoint saving; related to persistence broadly but not the same concrete issue." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "A model-implementation discrepancy is not the same as stale docs still referencing removed pipeline() APIs." + "reason": "SigLIP2 implementation discrepancy and pipeline-doc removal are unrelated." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "PatchTSTMixer post_init policy and Qwen video-input regression are different bugs." + "reason": "Optional post_init enforcement for patchtsmixer and a video-input regression are unrelated." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Both involve config shape, but Gemma3TextConfig API inconsistency is not the same as Qwen multimodal config unwrapping in from_config." + "reason": "Both mention config handling, but one is an API discrepancy in Gemma3TextConfig and the other is unwrapping text_config for composite Qwen configs; different fixes." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight loading and tokenizer import failure are unrelated." + "reason": "Accelerate loading of packed BitNet weights versus tokenizer loading failure; no overlap beyond generic model loading." }, { "left": "issue:41628", "right": "issue:45372", "accept": false, - "reason": "Both are import errors, but they come from different symbols, dependencies, and processor-loading paths." + "reason": "Different import errors for different components and dependencies; not the same underlying bug." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Different config persistence bugs: trust_remote_code missing fields vs GPTNeoX rotary_pct resetting on reload." + "reason": "Remote-code config missing fields on load versus a specific GPTNeoX config value not persisting; distinct serialization issues." }, { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "FSDP CPU RAM-efficient loading and Qwen3-VL weight-shape mismatch are different loading failures." + "reason": "FSDP CPU RAM-efficient loading breakage and a Qwen3-VL weight mismatch are different loading problems." }, { "left": "issue:34689", "right": "issue:36010", "accept": false, - "reason": "Model-loading breakage and GenerationMixin import failure are not the same underlying issue." + "reason": "Both are import/load regressions, but they affect different symbols and code paths." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to Gemma-3 eos_token_id config inconsistency." + "reason": "Quantization/offloading behavior and Gemma-3 eos_token_id inconsistency are unrelated." }, { "left": "issue:33290", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed OOM is unrelated to CustomTrainer.compute_loss receiving an unexpected keyword argument." + "reason": "OOM with Adafactor/DeepSpeed is unrelated to the CustomTrainer compute_loss signature break." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in an image-classification script is unrelated to an unrecognized model/config_type error." + "reason": "Dataset-loading issue in an example script versus unrecognized model type in config.json; different layers entirely." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 torch.jit parsing/IndentationError is unrelated to DeepSpeed ZeRO3 deque underflow." + "reason": "Python 3.13 import parsing error and DeepSpeed deque underflow are unrelated." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Both are Qwen multimodal issues, but one is video-input processing and the other is chat-template handling for tool calls." + "reason": "Both are multimodal/chat-related, but one is video-input validation in Qwen3OmniMoe and the other is apply_chat_template crashing on tool-call messages; not the same bug." }, { "left": "issue:43316", "right": "issue:44877", "accept": false, - "reason": "Different config issues for different models; not the same API discrepancy." + "reason": "Both involve config handling, but the specific models and failure modes differ completely." }, { "left": "issue:42175", "right": "issue:43316", "accept": false, - "reason": "TensorFlow backend packaging is unrelated to Gemma3TextConfig behavior." + "reason": "Packaging/backend dependency omission and Gemma3 config API mismatch are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Same model family, but dummy Conv2d in Sam3PixelDecoder and Sam3Video CUDA OOM are different issues." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "Mistral3 lm_head tying and Ministral FlashAttention failure are different bugs with different code paths." + "reason": "Related to Mistral/Ministral, but one is a missing lm_head tie and the other is a FlashAttention failure; not the same code-path problem." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k nesting bug and AutoModel PyTorch import error are entirely unrelated." } ] }, @@ -9138,8 +9101,6 @@ "nodes": [ "issue:16998", "issue:22355", - "issue:25251", - "issue:28282", "issue:32090", "issue:35141", "issue:41950", @@ -9168,6 +9129,7 @@ "issue:44464", "issue:44479", "issue:44496", + "issue:44521", "issue:44534", "issue:44610", "issue:44617", @@ -9179,24 +9141,23 @@ "issue:44977", "issue:45127", "issue:45200", + "issue:45381", "issue:45406", "issue:45563", "issue:45584", "issue:45698" ], "soft_pairs": [ - "issue:43065|issue:44617", "issue:43673|issue:44743", "issue:43295|issue:43825", - "issue:25251|issue:28282", "issue:44162|issue:44792", "issue:43388|issue:43688", "issue:41950|issue:43475", "issue:44077|issue:44534", "issue:16998|issue:22355", "issue:43550|issue:43701", - "issue:44871|issue:45584", "issue:43606|issue:43701", + "issue:44871|issue:45584", "issue:43761|issue:44479", "issue:45127|issue:45200", "issue:43295|issue:44871", @@ -9209,221 +9170,216 @@ "issue:44617|issue:45584", "issue:43931|issue:44060", "issue:43761|issue:44242", - "issue:45563|issue:45584", "issue:32090|issue:35141", + "issue:45563|issue:45584", "issue:44464|issue:44871", "issue:43065|issue:44610", "issue:44792|issue:44977", - "issue:44479|issue:44734" + "issue:44479|issue:44734", + "issue:44521|issue:45381" ], - "prepared_review_unit_hash": "5aaf6fc09974e85d2fda2229f6e2f9723f80cc45baa7f826133142af95d25183", + "prepared_review_unit_hash": "868263f649e9596f951ca7a658eb90481e95275a28881f017c3c9a6c851ce861", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36084, - "estimated_input_tokens": 9021, - "estimated_eval_tokens": 18298 + "soft_pair_count": 29, + "serialized_chars": 35779, + "estimated_input_tokens": 8945, + "estimated_eval_tokens": 18146 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is heterogeneous: the issues span unrelated Transformers bugs across loading, generation, processors, training, and model-specific regressions. None of the soft-similarity pairs look like the same underlying bug or change.", - "confidence": 0.94, - "canonical_issue_reason": "issue:45698 is the broadest and most active report in the set, but the overall collection is too heterogeneous to treat as a true duplicate cluster.", + "summary": "This cluster is highly heterogeneous: the soft pairs mostly share only broad terminology (generation, loading, multimodal, training) but not the same concrete bug or code path. I would not merge any of the proposed soft edges as duplicates. If a representative issue is needed, issue:44521 is the strongest activity-wise, but it is not a true canonical for the whole set.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue fits this cluster; the items span unrelated bugs across loading, generation, multimodal processing, serving, and training.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45698 has the strongest discussion/inbound-reference signal and is the best single anchor if one must be chosen, though it does not represent the unrelated reports well.", + "best_issue_reason": "issue:44521 is the most active/open item and has the strongest discussion footprint, but it is only representative of one narrow multimodal bug, not the cluster as a whole.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43065", - "right": "issue:44617", - "accept": false, - "reason": "Both are SAM3/video-related, but one is a dummy layer issue and the other is an OOM report; different code paths and failure modes." - }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both involve cache behavior, but one is missing GenerationMixin cache in chunked prefill and the other resets recurrent states in modular_qwen3_5; not the same bug." + "reason": "Both involve cache-related generation behavior, but one is about chunked_prefill cache availability and the other about recurrent states resetting with seq_len>1; different symptoms and code paths." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "Both are v5 regressions, but one is a processor/tokenizer API break and the other is a pipeline error-message problem; unrelated fixes." - }, - { - "left": "issue:25251", - "right": "issue:28282", - "accept": false, - "reason": "Top-k output nesting and a missing PyTorch import are unrelated issues." + "reason": "One is a regression in processor/tokenizer access and image passing; the other is an error-message mismatch about supported translation tasks. Different bug types." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 breakage and a Janus image-generation test failure are model-specific and unrelated." + "reason": "ESM2 workflow breakage and a Janus image-generation test failure are unrelated models and failures." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "One is a last-batch label-dropping bug in gather_for_metrics; the other is auxiliary-loss normalization in MoE models." + "reason": "Dropping label elements in gather_for_metrics is unrelated to auxiliary-loss normalization in expert models." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "A pipeline processor-selection bug and a missing Sam3VisionEncoderOutput attribute are different failures." + "reason": "A pipeline selecting the wrong processor type is not the same as a Sam3VisionEncoderOutput missing fpn_position_embeddings." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "Optional post_init handling and non-persistent buffer junk are separate initialization/state bugs." + "reason": "Optional post_init handling for patchtsmixer and junk-filled non-persistent buffers are different initialization bugs." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "Model_max_length on DeBERTa-V3 and a missing transformers.onnx module are unrelated." + "reason": "Model max length behavior for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "A torch.compile/SDPA OOM in Bamba and a resume_from_checkpoint key mismatch are different problems." + "reason": "torch.compile/SDPA failure for Bamba and a resume_from_checkpoint key mismatch are different problems." }, { - "left": "issue:44871", - "right": "issue:45584", + "left": "issue:43606", + "right": "issue:43701", "accept": false, - "reason": "Both concern generation/token IDs, but Gemma eos_token_id inconsistency and Whisper empty-transcription failure are distinct bugs." + "reason": "CPU offload device mismatch in bark-small is not the same as the checkpoint key mismatch bug." }, { - "left": "issue:43606", - "right": "issue:43701", + "left": "issue:44871", + "right": "issue:45584", "accept": false, - "reason": "CPU offload device mismatch and checkpoint key mismatch are unrelated." + "reason": "EOS token config inconsistency in Gemma-3 and Whisper empty-transcription failure after align_special_tokens are different model-specific issues." }, { "left": "issue:43761", "right": "issue:44479", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and a Qwen video-input regression affect different models/code paths." + "reason": "CLIPVisionModel hidden_states regression and Qwen VL video-input regression are separate model families and failure modes." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "LoRA merge collapse with tied embeddings and Gemma 4 mm_token_type_ids defaults are different bugs." + "reason": "LoRA merging with extended vocabulary and Gemma 4 mm_token_type_ids defaults are unrelated training/configuration bugs." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Processor/tokenizer regression and eos_token_id configuration mismatch are not the same underlying issue." + "reason": "Both mention processor/tokenizer behavior, but one is a custom model regression and the other is an EOS-token config mismatch; not the same bug." }, { "left": "issue:42617", "right": "issue:43575", "accept": false, - "reason": "Running 3d_parallel.py and loading Qwen2-57B with tensor parallelism are different distributed/runtime failures." + "reason": "Running 3d_parallel.py and OOM when loading Qwen2-57B with tensor parallelism are different issues." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "gguf_file loading breaks specific from_pretrained paths, while the other is an unrecognized model_type config issue." + "reason": "A gguf_file path breaking tokenizer/model loading is not the same as failing because a config.json lacks model_type." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "Qwen2-57B TP OOM and MobileLLM key-error loading are unrelated model-loading failures." + "reason": "OOM on a very large Qwen model with tp and a key error loading MobileLLM-125M are unrelated load-time failures." }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "KV-cache continuation indexing in serve and a missing _tokenizer attribute on Gemma4Processor are different code paths." + "reason": "KV-cache indexing in transformers serve and Gemma4Processor missing _tokenizer are different serving bugs." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Custom model initialization breakage in v5 and GPTNeoXConfig rotary_pct reload drift are unrelated." + "reason": "Breaking custom model initialization in v5 and GPTNeoXConfig rotary_pct not persisting are distinct configuration/init issues." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "Legacy preset loading for BEiT segmentation and wrong custom-module loading after save/load are both loading-related but not the same bug." + "reason": "Legacy-field loading failure in BeitForSemanticSegmentation and wrong custom module selection after save_pretrained are different pretrained-loading bugs." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "Sam3Video OOM and Whisper empty-transcription handling are unrelated." + "reason": "Sam3Video CUDA OOM and Whisper empty transcription failure are unrelated." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and Qwen3-Next tied-weights warning are different model/config issues." + "reason": "A weight-shape mismatch when loading Qwen3-VL is not the same as an incorrect tied-weights warning in Qwen3-Next." }, { "left": "issue:43761", "right": "issue:44242", "accept": false, - "reason": "A CLIP vision hidden-states regression and a missing load-balancing loss are unrelated." + "reason": "Hidden_states=None in CLIPVisionModel and missing load-balancing loss when output_router_logits=False are unrelated." }, { - "left": "issue:45563", - "right": "issue:45584", + "left": "issue:32090", + "right": "issue:35141", "accept": false, - "reason": "A stale num_return_sequences warning and Whisper align_special_tokens failure are different generation bugs." + "reason": "GPU broadcast TypeError in Trainer and embedding reinitialization after resize are separate bugs." }, { - "left": "issue:32090", - "right": "issue:35141", + "left": "issue:45563", + "right": "issue:45584", "accept": false, - "reason": "GPU broadcast NoneType handling and token-embedding resize reinitialization are unrelated." + "reason": "A stale warning in paged generate and Whisper generation failing on empty transcription are not the same underlying issue." }, { "left": "issue:44464", "right": "issue:44871", "accept": false, - "reason": "Chunked-generation inconsistency with compiled forward and Gemma eos_token_id mismatch are not the same bug." + "reason": "Compiled chunked generation inconsistency and Gemma-3 EOS configuration mismatch are unrelated." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "SAM3 dummy Conv2d and OmDet-Turbo processor/model size mismatch are different vision-model issues." + "reason": "Dummy Conv2d in Sam3PixelDecoder and OmDet-Turbo processor/model size mismatch are both vision-related but clearly different bugs." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "A Janus image-generation test failure and a Qwen3.5 flash-attention generation bug are unrelated." + "reason": "Janus image-generation test failure and Qwen3.5 flash-attention generation failure are different models and code paths." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Qwen video-input regression and serve KV-cache continuation crash are distinct regressions." + "reason": "A multimodal video-input regression and a serve KV-cache continuation indexing crash are unrelated." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both are multimodal, but one is assistant mask construction in apply_chat_template and the other is vision_position_ids for Qwen2.5-VL video input; not the same bug." } ] }, @@ -9450,6 +9406,7 @@ "issue:35141", "issue:39692", "issue:41720", + "issue:41950", "issue:42490", "issue:43335", "issue:43388", @@ -9462,6 +9419,7 @@ "issue:43906", "issue:43976", "issue:43994", + "issue:44016", "issue:44077", "issue:44079", "issue:44112", @@ -9470,7 +9428,6 @@ "issue:44462", "issue:44464", "issue:44496", - "issue:44521", "issue:44534", "issue:44734", "issue:44737", @@ -9486,7 +9443,6 @@ "issue:45290", "issue:45357", "issue:45362", - "issue:45381", "issue:45542", "issue:45563", "issue:45584", @@ -9496,7 +9452,6 @@ "issue:45907" ], "soft_pairs": [ - "issue:44521|issue:45381", "issue:43906|issue:45290", "issue:43827|issue:43994", "issue:44464|issue:45563", @@ -9512,210 +9467,217 @@ "issue:35141|issue:42490", "issue:44462|issue:45907", "issue:44077|issue:44079", - "issue:44938|issue:45542", "issue:43335|issue:44855", + "issue:44938|issue:45542", "issue:43550|issue:44190", "issue:43976|issue:44855", "issue:43644|issue:44913", "issue:43540|issue:44734", "issue:44112|issue:45600", - "issue:43650|issue:44220", "issue:44496|issue:45803", + "issue:43650|issue:44220", "issue:43335|issue:44190", "issue:45276|issue:45584", "issue:39692|issue:44190", - "issue:45245|issue:45584" + "issue:45245|issue:45584", + "issue:41950|issue:44016" ], - "prepared_review_unit_hash": "7638ebc92705cea9ef77496ffc819edcd8892c59cfc3d743e30af37b43eaa4b4", + "prepared_review_unit_hash": "76fb2c28234ac7c80fb3cba196541da178f6ebf7ae23bee59945582b892e8af5", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 35918, - "estimated_input_tokens": 8980, - "estimated_eval_tokens": 18216 + "serialized_chars": 35902, + "estimated_input_tokens": 8976, + "estimated_eval_tokens": 18208 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly unrelated bug reports; none of the suggested soft pairs are true duplicates. The only loosely related subgroup is around token-embedding resize/output-embedding behavior, but even there the evidence is not strong enough to merge the other pairs.", - "confidence": 0.94, - "canonical_issue_reason": "issue:35141 is the best anchor for the only coherent subgroup here: it states the underlying resize_token_embeddings/post_init bug in general terms, while the other embedding-related report is a model-specific variant.", + "summary": "This set is mostly heterogeneous and does not look like a tight duplicate cluster. The only clear thematic overlap is around token-embedding resizing, but the listed soft pairs are otherwise separate bugs or unrelated regressions.", + "confidence": 0.92, + "canonical_issue_reason": "issue:35141 is the clearest, most general concrete bug report in the set: it names the exact code path (`post_init` after `resize_token_embeddings`) and the failure mode (`output embedding` reinitialized when `tie_word_embedding=False`).", "canonical_pr_reason": null, - "best_issue_reason": "issue:35141 is the strongest issue to keep as the representative bug report because it describes a concrete code-path regression with a clear root cause and broader applicability than the more model-specific variants.", + "best_issue_reason": "issue:35141 is the best anchor for the only plausible duplicate family here because it is specific, actionable, and directly describes the underlying embedding-resize bug more cleanly than the other items.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44521", - "right": "issue:45381", - "accept": false, - "reason": "Both are multimodal-related, but one is about all-zero assistant masks in apply_chat_template while the other is about wrong vision_position_ids for Qwen2.5-VL video input; different failures and likely different fixes." - }, { "left": "issue:43906", "right": "issue:45290", "accept": false, - "reason": "The first is an isolated reproduction of an older issue, while the second is a crash with tool-call assistant messages; they do not describe the same bug." + "reason": "Both mention chat/template-style behavior, but one is just an isolated reproduction of another issue while the other is a concrete crash on assistant tool-call messages; not enough evidence of the same bug." }, { "left": "issue:43827", "right": "issue:43994", "accept": false, - "reason": "Docs still referencing pipeline() and SigLIP2 producing bad outputs are unrelated problems." + "reason": "Documentation still referencing `pipeline()` after v5 removals is unrelated to SigLIP2 producing bad outputs with AutoModel/pipeline." }, { "left": "issue:44464", "right": "issue:45563", "accept": false, - "reason": "Chunked generation inconsistencies with compiled forward and a stale paged-generate warning are different generation issues." + "reason": "Chunked generation inconsistency under compiled forward and a stale `num_return_sequences` warning in paged generate are different problems." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk and GPTNeoX rotary_pct reload behavior affect different code paths." + "reason": "One is about non-persistent buffers being filled with junk; the other is about a config value reverting on reload." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Dataset-loading failure in an example script and a model config/load failure for Phi-4 multimodal instruct are not the same bug." + "reason": "Local dataset loading in an example script and model loading for Phi-4 multimodal are separate failures." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "A missing import from image_utils is unrelated to Qwen3.5 chat crashes." + "reason": "A bad import from `image_utils` is unrelated to Qwen3.5 chat crashes." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Auto device mapping on Qwen3 and a categories-limit RuntimeError are different failures." + "reason": "CUDA assert in Qwen3 auto device mapping and a category-limit runtime error are different failure modes." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "A Janus generation test failure and stale auto_mappings references to removed configs do not share the same underlying defect." + "reason": "A janus image-generation test failure and broken `auto_mappings.py` references are not the same underlying bug." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Python 3.13 import parsing failure in DebertaV2 and a flash_attention crash on s_aux=None are unrelated." + "reason": "Python 3.13 import parsing/`torch.jit.script` issue is unrelated to flash-attention crashing on `s_aux=None`." }, { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "CPU placement in XLNet relative positional encoding and a bare except swallowing errors in FuyuBatchFeature are different bugs." + "reason": "Missing `device=` in `torch.arange` causing CPU work is unrelated to the bare `except:` swallowing errors." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "LoRA merge collapse with extended vocabulary and incorrect visual encoder keys in save_pretrained are not the same code-path issue." + "reason": "LoRA/extended-vocab collapse with tied embeddings and incorrect visual-encoder keys in save_pretrained are different bugs." }, { "left": "issue:43388", "right": "issue:43582", "accept": false, - "reason": "A last-batch label handling bug and an Apple Silicon TypeError in caching_allocator_warmup are unrelated." + "reason": "A metrics label-truncation bug and an Apple Silicon `caching_allocator_warmup` TypeError do not match." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "resize_token_embeddings/post_init behavior and a trailing-slash import warning are unrelated." + "reason": "Embedding reinitialization after resize and a trailing-slash local-path import warning are unrelated." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json and list_repo_files not forwarding revision target different repository/file-resolution paths." + "reason": "Ignoring `tokenizer.json` and failing to forward `revision` in `list_repo_files` are different repository-loading issues." }, { "left": "issue:44077", "right": "issue:44079", "accept": false, - "reason": "patchtsmixer post_init policy and ModelOutput key assignment with None are different issues." + "reason": "These are separate core-library bugs: `post_init` policy for patchtsmixer versus `ModelOutput` key assignment." }, { - "left": "issue:44938", - "right": "issue:45542", + "left": "issue:43335", + "right": "issue:44855", "accept": false, - "reason": "Python 3.14 load failure and a tf-backend warning when only tensorboard is installed are separate compatibility problems." + "reason": "SwitchTransformers sparse-layer construction and a Python 3.13 import parsing error are unrelated." }, { - "left": "issue:43335", - "right": "issue:44855", + "left": "issue:44938", + "right": "issue:45542", "accept": false, - "reason": "SwitchTransformers sparse-layer construction and DebertaV2 import parsing are unrelated." + "reason": "Python 3.14 import/load failure and the TensorBoard-without-TensorFlow backend error are different compatibility problems." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "torch.compile/SDPA failure in Bamba and a local dataset-loading example failure are not the same bug." + "reason": "Bamba torch.compile/SDPA failure is unrelated to local dataset loading in an example script." }, { "left": "issue:43976", "right": "issue:44855", "accept": false, - "reason": "Python version support mismatch and DebertaV2 import parsing under Python 3.13 are different compatibility failures." + "reason": "General Python-version compatibility and a specific DebertaV2 import indentation/parsing bug are not the same issue." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer initialization junk and GPTNeoX rotary_pct reload are separate state-loading issues." + "reason": "Non-persistent buffer initialization junk and `rotary_pct` not persisting on reload are separate state-handling bugs." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video processing ValueError and KV-cache continuation tensor indexing are different runtime failures." + "reason": "Qwen3OmniMoe video input processing and serve-side KV-cache tensor indexing are different code paths." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "A stale CI test in GraniteSpeech and broken auto_mappings entries for removed configs are unrelated maintenance issues." + "reason": "A stale CI test and broken auto-mapping references are both CI-related but not the same underlying change." + }, + { + "left": "issue:44496", + "right": "issue:45803", + "accept": false, + "reason": "Model-type resolution failure when loading a repo and a bare exception swallowing errors are unrelated." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "The placeholder issue and the _torch_extract_fbank_features bug are not duplicates." + "reason": "`ADD THE DATA` is too vague to tie to the audio feature extraction bug." }, { - "left": "issue:44496", - "right": "issue:45803", + "left": "issue:45276", + "right": "issue:45584", "accept": false, - "reason": "Unrecognized model loading via missing model_type and a bare-except bug in FuyuBatchFeature are unrelated." + "reason": "Gemma4 resize-token-embedding behavior and Whisper empty-transcription failure are unrelated." }, { - "left": "issue:43335", + "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "SwitchTransformers sparse-layer behavior and local dataset loading in an example script are unrelated." + "reason": "SigLIP2 docs/model-processor mismatch and local dataset loading are different documentation/example issues." }, { - "left": "issue:45276", + "left": "issue:45245", "right": "issue:45584", "accept": false, - "reason": "Gemma4 resize_token_embeddings/output_embeddings behavior and Whisper empty-transcription generation after align_special_tokens are different bugs." + "reason": "Category-count runtime error and Whisper empty-transcription crash are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "Video-classification pipeline processor selection and a notebook syntax error do not share the same bug." } ] }, @@ -9743,7 +9705,6 @@ "issue:38175", "issue:38617", "issue:41628", - "issue:41950", "issue:42175", "issue:42490", "issue:42907", @@ -9789,15 +9750,14 @@ "issue:45632" ], "soft_pairs": [ - "issue:41950|issue:44016", "issue:43479|issue:45375", "issue:43295|issue:45276", "issue:36246|issue:41628", "issue:36246|issue:38617", "issue:44186|issue:44734", "issue:38175|issue:43065", - "issue:44683|issue:45446", "issue:42175|issue:44279", + "issue:44683|issue:45446", "issue:44016|issue:44220", "issue:44466|issue:44913", "issue:43404|issue:45216", @@ -9808,8 +9768,8 @@ "issue:43643|issue:44496", "issue:44964|issue:45600", "issue:44112|issue:44964", - "issue:42907|issue:43441", "issue:43854|issue:44964", + "issue:42907|issue:43441", "issue:44016|issue:44062", "issue:43976|issue:44987", "issue:43295|issue:45335", @@ -9820,214 +9780,208 @@ "issue:43526|issue:43901", "issue:43299|issue:45216" ], - "prepared_review_unit_hash": "06324f74e335087e67daa6dc63a556e7384fc527dde3e2c5b1b69571dff62c4e", + "prepared_review_unit_hash": "21a6295b2542464e886e796dc5b61e86b9f5b8dd4cfa068152c89ad033495881", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36738, - "estimated_input_tokens": 9185, - "estimated_eval_tokens": 18626 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35884, + "estimated_input_tokens": 8971, + "estimated_eval_tokens": 18198 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These issues are largely heterogeneous; none of the soft pairs look like the same underlying bug or change, so I would not collapse any of them as duplicates. If a representative issue must be chosen, issue 43824 is the strongest anchor because it is a concrete import regression with the most external signal.", - "confidence": 0.84, - "canonical_issue_reason": "Issue 43824 is the best canonical anchor: it has the clearest, concrete failure mode (missing top-level import) and the strongest signal (multiple inbound refs and active discussion), making it more representative than the others.", + "summary": "This set is not a true duplicate cluster: it mixes unrelated issues across imports, model loading, configs, attention, tokenizers, docs, and CI. All soft-edge pairs should be rejected.", + "confidence": 0.89, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 43824 is the most actionable and well-evidenced issue in the set; the rest are scattered across unrelated models, config fields, pipelines, and regressions.", + "best_issue_reason": "issue:44964 is the broadest user-facing regression report here (a concrete model-loading failure with recent transformers) and has relatively strong discussion activity, but it still only loosely represents this heterogeneous set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:41950", - "right": "issue:44016", - "accept": false, - "reason": "Video-classification pipeline lookup bug vs notebook syntax error; unrelated symptoms and code paths." - }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Both involve multimodal config handling, but they target different models and different bugs (default config init vs strict field dropping)." + "reason": "Both are multimodal-config bugs, but they affect different models and different fields/code paths; not the same underlying defect." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Different regressions: processor/tokenizer image handling vs resize_token_embeddings not updating Gemma4 embeddings." + "reason": "Both mention resize/token embedding behavior, but the regressions are on different models and symptoms; not mergeable as one fix." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but for different exported symbols and different breakages." + "reason": "Different missing imports for different symbols and modules; same broad theme, not the same bug." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Unrelated import failures: Qwen2.5 VL image processor vs layer_type_validation." + "reason": "Separate import errors in unrelated modules/symbols, so they are not the same underlying issue." }, { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "Different failures in different subsystems: tokenizer padding/NER vs serving KV-cache continuation indexing." + "reason": "Tokenizer NER/padding crash vs server KV-cache continuation indexing bug; different code paths and failures." }, { "left": "issue:38175", "right": "issue:43065", "accept": false, - "reason": "Siglip2 zero-probabilities issue and Sam3PixelDecoder dummy Conv2d are unrelated model bugs." + "reason": "Unrelated model behavior bug vs dummy Conv2d implementation issue; no shared defect." }, { - "left": "issue:44683", - "right": "issue:45446", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "Both touch flex_attention/PyTorch gating, but one is a compile/runtime failure on torch>=2.9 and the other is an incorrect version check for an import path." + "reason": "Both are dependency-related, but one is about optional backend installation and the other is a vague dependency problem; too different to deduplicate." }, { - "left": "issue:42175", - "right": "issue:44279", + "left": "issue:44683", + "right": "issue:45446", "accept": false, - "reason": "Dependency installation issue vs vague dependency issue; no shared concrete code-path or fix." + "reason": "Both touch flex_attention, but they are distinct torch/version-check problems in different code paths." }, { "left": "issue:44016", "right": "issue:44220", "accept": false, - "reason": "Notebook syntax error vs _torch_extract_fbank_features failure; unrelated." + "reason": "Notebook syntax error and fbank feature extraction bug are unrelated." }, { "left": "issue:44466", "right": "issue:44913", "accept": false, - "reason": "Both concern serialization/reload, but one is tied lm_head weights and the other is a GPTNeoXConfig field reset." + "reason": "Tied-weight serialization/device behavior vs rotary_pct persistence in GPTNeoXConfig; different components and fixes." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Different model families and different bugs: Mistral3 weight tying vs Qwen3.5 checkpoint save regression." + "reason": "Both involve saving/tied weights, but they concern different models and different save-pretrained regressions." }, { "left": "issue:44561", "right": "issue:45632", "accept": false, - "reason": "Both involve trust_remote_code, but one is a removed API breakage and the other is a local-model cache-path collision." + "reason": "Removal of is_torch_fx_available breaking remote-code models is unrelated to local cache-path collisions." }, { "left": "issue:44734", "right": "issue:45632", "accept": false, - "reason": "Serving KV-cache continuation crash vs trust_remote_code cache collision; different mechanisms and fixes." + "reason": "Server request crash and trust_remote_code cache collision are separate bugs with no shared code path." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import performance issue vs unsafe sys.modules access; not the same bug." + "reason": "Import-time slowdown and unsafe sys.modules access are related only loosely; they are not the same bug." }, { "left": "issue:42490", "right": "issue:45632", "accept": false, - "reason": "Both are local path handling problems, but one is a trailing-slash warning and the other is a cache collision between leaf-name matches." + "reason": "Import-warning behavior for paths ending in slash is unrelated to cache collisions between local model paths." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "Both are config/loading failures, but one is trust_remote_code missing fields and the other is an unrecognized model/type key problem." + "reason": "Missing fields from trust_remote_code config loading vs unrecognized model/config.json issue; different failure modes." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "Different model-loading/CI failures with no shared code-path." + "reason": "Model-loading failure for one checkpoint vs CI break from removed Sam3LiteText configs; different underlying problems." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "CI test flake in GraniteSpeech vs Phi-4 multimodal loading regression; unrelated." + "reason": "A stale CI test failure is not the same as a real model-loading regression." }, { - "left": "issue:42907", - "right": "issue:43441", + "left": "issue:43854", + "right": "issue:44964", "accept": false, - "reason": "Both mention Ministral/Devstral, but save-dequantized failure and FlashAttention failure are distinct bugs." + "reason": "Both are load-related, but one is a specific test/model loading issue and the other is a different checkpoint load failure." }, { - "left": "issue:43854", - "right": "issue:44964", + "left": "issue:42907", + "right": "issue:43441", "accept": false, - "reason": "Different model load failures affecting different model families and failure points." + "reason": "Saving dequantized models and FlashAttention failures are unrelated." }, { "left": "issue:44016", "right": "issue:44062", "accept": false, - "reason": "Notebook syntax error vs AddedToken keyword-argument TypeError; unrelated." + "reason": "Notebook syntax error vs AddedToken keyword conflict; unrelated bugs." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "Python-version compatibility issue vs a specific model-loading failure; not the same underlying defect." + "reason": "Python-version compatibility and a specific model-loading failure are not the same defect." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Different embedding-update regressions in different models; not mergeable as one fix." + "reason": "Both mention resize_token_embeddings, but on different model families and different affected embeddings; not a single code-path bug." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Generation cache bug vs ANSI formatting in loading_report; unrelated." + "reason": "Generation cache absence during chunked prefill vs ANSI logging behavior; unrelated." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior vs stale auto_mappings references; different areas and fixes." + "reason": "Qwen3-MoE sliding_window behavior and removed auto_mapping configs are separate regressions." }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Both are import-related, but one is a missing model class export and the other is a missing image_utils export; different symbols and root causes." + "reason": "Different import errors from different modules/symbols; same class of problem, not the same bug." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "Unrecognized model/config issue vs PIL image processors incorrectly requiring torchvision; unrelated." + "reason": "Unrecognized-model loading error vs unnecessary torchvision dependency in PIL image processors; different issues." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug vs documentation drift for return_all_scores; not the same issue." + "reason": "Image-processor label reduction bug vs documentation mismatch in TextClassificationPipeline; unrelated." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Both are version regressions for model loading/saving, but they affect different Qwen variants and different code paths." + "reason": "Qwen3-MoE loading regression and Qwen3.5 save_pretrained serialization bug are different model-specific failures." } ] }, @@ -10057,6 +10011,7 @@ "issue:41950", "issue:43295", "issue:43299", + "issue:43335", "issue:43381", "issue:43441", "issue:43479", @@ -10100,6 +10055,7 @@ "issue:45907" ], "soft_pairs": [ + "issue:43335|issue:45593", "issue:44075|issue:44368", "issue:45335|issue:45584", "issue:43976|issue:44279", @@ -10115,223 +10071,229 @@ "issue:43479|issue:43525", "issue:44779|issue:45812", "issue:41950|issue:45823", + "issue:38175|issue:44617", "issue:43577|issue:45216", "issue:44683|issue:45542", - "issue:38175|issue:44617", "issue:39692|issue:43901", "issue:43299|issue:43854", "issue:44279|issue:45405", - "issue:41950|issue:43650", "issue:43295|issue:44779", + "issue:41950|issue:43650", "issue:44462|issue:45081", "issue:44297|issue:44462", "issue:45081|issue:45907", "issue:44016|issue:44291", - "issue:43441|issue:45834", - "issue:33290|issue:45656" + "issue:33290|issue:45656", + "issue:43441|issue:45834" ], - "prepared_review_unit_hash": "0a634433dd61ed2343f10287f8d37e82bee7123eafe43d635d84b633ab24be8f", + "prepared_review_unit_hash": "221957cd85ef245666f23dcc872c9b372578d769bcb0bfdf572d004c13eed1dc", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 35828, - "estimated_input_tokens": 8957, - "estimated_eval_tokens": 18170 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36740, + "estimated_input_tokens": 9185, + "estimated_eval_tokens": 18626 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous rather than duplicate-heavy: it mixes unrelated Transformers issues spanning tokenizers, multimodal configs, pipelines, docs, DeepSpeed/optimizer behavior, and model-loading regressions. None of the soft edges looks like the same underlying bug or change.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue: the cluster spans unrelated bugs/regressions across tokenizers, multimodal configs, pipelines, docs, and DeepSpeed/optimizer behavior.", + "summary": "This cluster is highly heterogeneous: it mixes unrelated tokenizer, multimodal, config, optimizer, docs, and runtime regression reports. The soft edges are mostly superficial keyword matches, not the same underlying bug, so none should be merged.", + "confidence": 0.95, + "canonical_issue_reason": "No safe canonical issue: the items span many distinct bugs and regressions, so collapsing them into one would be misleading.", "canonical_pr_reason": null, - "best_issue_reason": "No issue cleanly serves as an umbrella; the broadest report (#45020) is still too general to represent the rest of the items.", + "best_issue_reason": "issue:45020 is the broadest umbrella regression report here ('Recent transformers versions break models using `remote_code`') and is the closest thing to a representative issue, though it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers config construction bug vs D-FINE auxiliary-loss bug; different model families and code paths." + }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Optimizer-arg handling is unrelated to the tie_word_embeddings warning; different symptoms and code paths." + "reason": "Optimizer SGD args ignored vs LoRA tie_word_embeddings warning; unrelated training/config behaviors." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "resize_token_embeddings on t5gemma decoder is unrelated to Whisper empty-transcription failure after align_special_tokens." + "reason": "t5gemma embedding resize bug vs Whisper empty-transcription failure; different models and failure modes." }, { "left": "issue:43976", "right": "issue:44279", "accept": false, - "reason": "Python-version compatibility and a generic dependency issue are not the same defect." + "reason": "Python version support issue vs generic dependency issue; not the same concrete bug." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient checkpointing in eval mode is unrelated to D-FINE auxiliary-loss behavior when denoising is off." + "reason": "Gradient checkpointing eval-mode restriction vs D-FINE auxiliary-loss omission; unrelated runtime semantics." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "Model loading regression and save_pretrained key regression affect different code paths." + "reason": "Qwen3VL model-loading regression vs Qwen3.5 save_pretrained key serialization bug; different code paths." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to an unrecognized-model/config loading error." + "reason": "Docs mismatch for return_all_scores vs model loading/config recognition failure; unrelated concerns." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "One is config/model recognition failure; the other is from_config failing to unwrap text_config in multimodal configs." + "reason": "Unrecognized model config vs AutoModelForCausalLM text_config unwrapping; different loader failures." }, { "left": "issue:44521", "right": "issue:45325", "accept": false, - "reason": "All-zero assistant masks and rope-index scaling are distinct multimodal internals." + "reason": "assistant_masks all-zero in chat template vs Qwen2.5-VL rope index scaling bug; both multimodal, but different subsystems." }, { "left": "issue:41950", "right": "issue:45020", "accept": false, - "reason": "A pipeline processor lookup bug is not the same as a broad remote_code compatibility regression." + "reason": "Video-classification pipeline processor lookup vs broad remote_code breakage; not the same bug." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "Missing HybridCache import and broad import errors after upgrade are not a single concrete bug." + "reason": "Missing HybridCache import vs broad import errors after upgrade; too vague and not concrete enough to duplicate." }, { "left": "issue:43976", "right": "issue:45405", "accept": false, - "reason": "Python support breakage is unrelated to the PEFT minimum-version bump." + "reason": "Python support mismatch vs peft version floor bump; unrelated packaging/release issues." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Missing load-balancing loss and default mm_token_type_ids are separate modeling issues." + "reason": "Load-balancing loss omission vs Gemma 4 mm_token_type_ids defaulting; different model training paths." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Default multimodal config initialization is unrelated to Llama4Config missing pad_token_id." + "reason": "Phi4Multimodal default config initialization vs Llama4Config missing pad_token_id; distinct config bugs." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but for different model families and failure modes; no evidence of one shared bug." + "reason": "Deepseek tokenizer regression vs Granite tokenizer regression; both tokenizer-related but different models and symptoms." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "Video pipeline processor lookup and Gemma4 PLE device mismatch are unrelated." + "reason": "Video pipeline processor lookup vs Gemma4 PLE device mismatch; unrelated failures." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "SigLIP2 zero-probabilities inference issue vs Sam3Video CUDA OOM; completely different symptoms and code paths." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "Blip2 dtype propagation and Qwen3.5 save_pretrained corruption are different bugs." + "reason": "BLIP2 dtype propagation bug vs Qwen3.5 save_pretrained regression; unrelated model-loading behaviors." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "torch>=2.9 flex_attention failure is unrelated to the tensorboard/TensorFlow backend error." - }, - { - "left": "issue:38175", - "right": "issue:44617", - "accept": false, - "reason": "SigLIP2 zero-probability outputs and Sam3Video OOM are clearly different model-specific problems." + "reason": "Compiled flex_attention on torch>=2.9 vs TensorBoard/tf backend install issue; different dependency/runtime areas." }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "Both involve docs, but they concern different APIs and unrelated fixes." + "reason": "SigLIP2 docs example errors vs TextClassificationPipeline docs mismatch; both documentation issues but not the same bug." }, { "left": "issue:43299", "right": "issue:43854", "accept": false, - "reason": "Qwen3VL loading breakage and GLM-4.7-Flash unit-test loading failures are different models and code paths." + "reason": "Qwen3VL loading regression vs GLM-4.7-Flash unit-test load failure; different models and loaders." }, { "left": "issue:44279", "right": "issue:45405", "accept": false, - "reason": "A dependency issue and a PEFT version requirement bump are not the same defect." + "reason": "Generic dependency issue vs PEFT minimum version bump; packaging problems, but not the same underlying issue." }, { - "left": "issue:41950", - "right": "issue:43650", + "left": "issue:43295", + "right": "issue:44779", "accept": false, - "reason": "The video-classification processor bug is unrelated to the placeholder 'ADD THE DATA' issue." + "reason": "processor.tokenizer regression in v4.57.5 vs Deepseek tokenizer incorrect results in v5; different regressions and APIs." }, { - "left": "issue:43295", - "right": "issue:44779", + "left": "issue:41950", + "right": "issue:43650", "accept": false, - "reason": "Processor/tokenizer regression in v4.57.5 is unrelated to DeepSeek tokenizer output changes in v5." + "reason": "video-classification processor lookup vs nonspecific 'ADD THE DATA' issue; no shared underlying bug." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json and Mistral regex patch crashes are different tokenizer bugs." + "reason": "AutoTokenizer ignoring tokenizer.json vs Mistral regex patch crash; different tokenizer pipeline failures." }, { "left": "issue:44297", "right": "issue:44462", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch is separate from AutoTokenizer ignoring tokenizer.json." + "reason": "tokenizer_class mismatch on save_pretrained vs AutoTokenizer ignoring repository tokenizer.json; separate save/load bugs." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "A tokenizer regex crash and list_repo_files not forwarding revision are unrelated." + "reason": "Mistral regex patch crash vs list_repo_files not forwarding revision; unrelated utility functions." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error and init_empty_weights argument handling are unrelated." - }, - { - "left": "issue:43441", - "right": "issue:45834", - "accept": false, - "reason": "FlashAttention failure in Ministral-3 and long-input OCR index errors in Kosmos2.5 are distinct issues." + "reason": "Notebook syntax error vs init_empty_weights TypeError; different failure types and code paths." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "DeepSpeed-related optimizer OOM and optimizer step being called twice are different failure modes." + "reason": "OOM with Adafactor in DeepSpeed vs optimizer.step called twice in DeepSpeed; related subsystem but different bugs and symptoms." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Ministral FlashAttention failure vs Kosmos2.5 OCR index error; unrelated model-specific issues." } ] }, @@ -10405,8 +10367,8 @@ "issue:45823" ], "soft_pairs": [ - "issue:45600|issue:45712", "issue:43450|issue:44279", + "issue:45600|issue:45712", "issue:43065|issue:44898", "issue:42994|issue:43317", "issue:44466|issue:45276", @@ -10427,216 +10389,219 @@ "issue:43901|issue:43994", "issue:44987|issue:45542", "issue:43122|issue:43650", - "issue:43824|issue:43976", "issue:44297|issue:45397", - "issue:43525|issue:43881", + "issue:43824|issue:43976", "issue:44188|issue:44336", + "issue:43525|issue:43881", "issue:43525|issue:43643", "issue:43450|issue:44617" ], - "prepared_review_unit_hash": "e9a4a570caf81a6c125f97c9b8073c3f1c4adb608c09311853339cc809d9a709", + "prepared_review_unit_hash": "395336265ca2c1e82e5d499715cca80fd8d7d76f37f7d39c192c1679a2079774", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 35940, - "estimated_input_tokens": 8985, - "estimated_eval_tokens": 18226 + "serialized_chars": 35943, + "estimated_input_tokens": 8986, + "estimated_eval_tokens": 18228 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the issues share broad Transformers/v5 regression language, but the paired items point to different code paths and failure modes, so I would not deduplicate any soft-edge pair. The strongest representative issue is the open multimodal `apply_chat_template` regression.", - "confidence": 0.84, - "canonical_issue_reason": "issue:44521 is the clearest ongoing, well-documented regression in the set: open, actively discussed, with a concrete repro and multiple references.", + "summary": "Only one soft edge looks like a true duplicate pair; the rest are distinct issues across separate subsystems or different failure modes.", + "confidence": 0.72, + "canonical_issue_reason": "43643 is the broader root-cause report: `trust_remote_code=True` returns config objects with missing fields; 43525 looks like a specific `pad_token_id` symptom of that bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the most actionable representative issue in this mixed cluster because it is open and has the richest current discussion/repro context.", + "best_issue_reason": "43643 has the clearest general reproduction of the config-field-loss bug and is the best anchor for deduping the pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45600", - "right": "issue:45712", + "left": "issue:43450", + "right": "issue:44279", "accept": false, - "reason": "Both are repo-hygiene/dummy-object problems, but one is stale auto-mapping references and the other is leftover dummy classes; different fixes." + "reason": "Different bugs: batched video processor shape vs a generic quantized-model save failure." }, { - "left": "issue:43450", - "right": "issue:44279", + "left": "issue:45600", + "right": "issue:45712", "accept": false, - "reason": "Video processor batch shape bug vs a dependency issue; unrelated failure modes." + "reason": "Both are cleanup/CI-related, but they target different stale artifacts and different breakages." }, { "left": "issue:43065", "right": "issue:44898", "accept": false, - "reason": "Both touch vision models, but one is a dummy Conv2d in Sam3PixelDecoder and the other is Perceiver resizing behavior; not the same bug." + "reason": "Unrelated model issues: dummy Conv2d in Sam3PixelDecoder vs Perceiver resolution handling." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Both involve quantization/dequantization, but one is save failure and the other is device_map offload load failure; distinct code paths." + "reason": "Save-failure for quantized models vs load/offload failure for dequantized models; different code paths." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "lm_head serialization/tied-weights issue vs Gemma4 embedding resize propagation; different underlying problems." + "reason": "Both involve Gemma4 weights, but one is serialization inconsistency and the other is resize_token_embeddings behavior." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "MLA layernorm epsilon precision issue vs torch.split return handling in GlmMoeDsaIndexer; unrelated." + "reason": "Different GLM/MoE bugs: missing rms_norm_eps precision issue vs torch.split return handling." }, { "left": "issue:43650", "right": "issue:45290", "accept": false, - "reason": "Generic placeholder issue vs a specific tool-call assistant message crash in apply_chat_template; no same bug." + "reason": "Generic low-signal issue vs a specific chat_template crash on tool-call messages; not the same bug." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Missing processor save files vs Gemma-4 ZeRO-3 from_pretrained loading; different symptoms and paths." + "reason": "processor.save_pretrained missing files is separate from gemma-4 zero3 from_pretrained loading failures." }, { "left": "issue:43441", "right": "issue:44683", "accept": false, - "reason": "Model-specific FlashAttention failure vs compiled flex_attention failure on torch>=2.9; not the same fix target." + "reason": "Different attention-backend failures: FlashAttention on Ministral-3 vs compiled flex_attention on torch>=2.9." }, { "left": "issue:44623", "right": "issue:44749", "accept": false, - "reason": "Processor save_pretrained missing files vs a severe data filtering slowdown; unrelated." + "reason": "Missing processor save files vs a tokenizer/filtering performance regression; unrelated symptoms." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "BigBirdTokenizer special-token registration bug vs D-FINE auxiliary-loss behavior; different subsystems." + "reason": "Tokenizer special-token registration vs D-FINE auxiliary-loss logic; different components." }, { "left": "issue:43262", "right": "issue:44521", "accept": false, - "reason": "Both involve apply_chat_template, but one is audio sampling-rate defaulting and the other is multimodal assistant masks; distinct bugs." + "reason": "Both touch apply_chat_template, but one is audio sampling-rate defaults and the other is multimodal assistant masks." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer loading regression vs ESM2 model breakage; no shared concrete code path." + "reason": "Tokenizer loading regression vs ESM2 model breakage; no shared underlying defect." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error vs Gemma4 device-mismatch runtime bug; unrelated." + "reason": "Notebook syntax error is unrelated to Gemma4 device-mismatch runtime behavior." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache/chunked_prefill issue vs CLIPVision hidden_states regression; different components." + "reason": "Generation cache regression and CLIP hidden_states regression are separate model-path bugs." }, { "left": "issue:42175", "right": "issue:43976", "accept": false, - "reason": "Packaging/backend dependency problem vs Python version compatibility issue; separate causes." + "reason": "TF backend packaging issue vs Python version compatibility regression; different root causes." }, { "left": "issue:44361", "right": "issue:45593", "accept": false, - "reason": "MLukeTokenizer task AttributeError vs D-FINE auxiliary-loss logic; unrelated." + "reason": "MLukeTokenizer task AttributeError is unrelated to D-FINE auxiliary-loss behavior." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "Both are v5 regressions affecting outputs, but one is masked LM predictions and the other is tokenizer correctness; not the same underlying bug." + "reason": "Different failures: CamemBERT masked-LM predictions vs Deepseek tokenizer correctness." }, { "left": "issue:43784", "right": "issue:45542", "accept": false, - "reason": "sentence-transformers import NameError vs missing TensorFlow backend due to only tensorboard installed; different root causes." + "reason": "Sentence-transformers import NameError and TF backend packaging error are separate import/dependency issues." }, { "left": "issue:43901", "right": "issue:43994", "accept": false, - "reason": "Docs mismatch for return_all_scores vs a bad model/pipeline output regression; not duplicates." + "reason": "Docs mismatch for return_all_scores vs a model/pipeline output quality issue; not duplicate bugs." }, { "left": "issue:44987", "right": "issue:45542", "accept": false, - "reason": "Specific model loading failure vs TensorFlow backend installation problem; unrelated." + "reason": "Loading a specific model failing in v5.1.0 is not the same as the TensorFlow backend dependency problem." }, { "left": "issue:43122", "right": "issue:43650", "accept": false, - "reason": "Tokenizer version regression vs a placeholder/data request issue; no overlap in bug." - }, - { - "left": "issue:43824", - "right": "issue:43976", - "accept": false, - "reason": "Missing exported class import vs Python 3.9+ compatibility problem; different issues." + "reason": "Tokenizer output changed across versions vs an unrelated low-signal issue." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer save metadata mismatch vs Gemma-4 zero3 loading failure; distinct behavior." + "reason": "Tokenizer metadata saved incorrectly vs gemma-4 zero3 loading; separate save/load issues." }, { - "left": "issue:43525", - "right": "issue:43881", + "left": "issue:43824", + "right": "issue:43976", "accept": false, - "reason": "Missing pad_token_id on Llama4Config vs glm-4v-9b loading failure; not the same concrete defect." + "reason": "ImportError for a missing model class vs Python 3.9 support regression; different failure modes." }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "torch.compile attention-branch divergence vs ANSI code emission in loading_report; unrelated." + "reason": "Attention kernel branching under torch.compile is unrelated to ANSI code emission in loading_report." }, { "left": "issue:43525", - "right": "issue:43643", + "right": "issue:43881", "accept": false, - "reason": "The latter is a general missing-fields issue with trust_remote_code, but the former is a specific Llama4Config attribute absence; too weak to merge as one bug." + "reason": "Llama4Config missing pad_token_id is not the same as GLM-4V loading failure." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": true, + "reason": "Both describe missing fields on configs returned by AutoConfig/trust_remote_code; 43525 is a concrete missing-pad_token_id instance of the broader bug." }, { "left": "issue:43450", "right": "issue:44617", "accept": false, - "reason": "Batched video shape bug vs CUDA OOM in Sam3Video; different failure modes." + "reason": "Batched video processor shape bug is unrelated to Sam3Video CUDA OOM." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative overall. The single accepted soft edge, issue:43525 with issue:43643, is a plausible duplicate based on the shared missing-config-field failure mode, and the other rejections are framed as distinct failure modes or subsystems without overstating the evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -10707,8 +10672,8 @@ "issue:41628|issue:43976", "issue:43901|issue:44246", "issue:45561|issue:45563", - "issue:42491|issue:43299", "issue:44336|issue:44373", + "issue:42491|issue:43299", "issue:44242|issue:45593", "issue:44295|issue:45397", "issue:41950|issue:44220", @@ -10716,9 +10681,9 @@ "issue:43531|issue:45362", "issue:43352|issue:45250", "issue:36246|issue:42757", + "issue:44336|issue:44485", "issue:43526|issue:45003", "issue:36246|issue:44351", - "issue:44336|issue:44485", "issue:45003|issue:45561", "issue:43673|issue:44561", "issue:44246|issue:44871", @@ -10733,24 +10698,24 @@ "issue:45292|issue:45701", "issue:43232|issue:44336", "issue:44987|issue:45459", - "issue:43065|issue:43408", "issue:44496|issue:45003", + "issue:43065|issue:43408", "issue:45375|issue:45759" ], - "prepared_review_unit_hash": "1ed288d19888d5534e09e7dc41e84d48d4a0d1010a630bdd555692f00a1a02ef", + "prepared_review_unit_hash": "2576f80e9e2645550051fc4e979259e8b8e963f483c9f9b342950b5e116c298c", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37278, + "serialized_chars": 37279, "estimated_input_tokens": 9320, "estimated_eval_tokens": 18896 }, @@ -10759,210 +10724,210 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly loosely related Transformers issues, but the suggested soft pairs are not true duplicates: they mix different models, different code paths, or unrelated docs/perf/config problems.", - "confidence": 0.9, - "canonical_issue_reason": "issue:43299 is the most concrete version-regression model-loading report in the set, so it works best as a representative anchor; the rest are mostly separate bugs rather than one duplicate thread.", + "summary": "Mostly heterogeneous regression reports around Transformers v5, model/config loading, and unrelated utility bugs; I did not find any pair that clearly describes the same underlying fix target.", + "confidence": 0.31, + "canonical_issue_reason": "issue:44561 is the broadest, clearest compatibility regression report in the set: a concrete v5 removal breaking trust_remote_code model loading, which makes it the best umbrella issue among otherwise unrelated model-specific failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43299 has a clear reproduction scope and a specific failing path (Qwen3VL MoE loading on v5.0.0.dev0), making it the strongest single issue to triage from this mixed cluster.", + "best_issue_reason": "issue:44561 is the strongest representative issue because it has a specific failure mode, clear upgrade context, and broad downstream impact compared with the more narrow or ambiguous reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Different failures: float8 storage lookup vs a Qwen3.5 chat crash." + "reason": "Different failures: a missing Float8 storage class versus a Qwen chat crash; no shared code path is evident." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Import error for AutoImageProcessor vs a Python-version compatibility problem." + "reason": "ImportError vs Python-version compatibility; same package area, but not the same bug." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Docs mismatch in a pipeline vs slow package import; unrelated code paths." + "reason": "Docstring mismatch for text classification is unrelated to intermittent import slowness." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "pytest-xdist file race vs a stale generation warning; no shared bug." + "reason": "Pytest-xdist file race and a stale generation warning are unrelated behaviors." }, { - "left": "issue:42491", - "right": "issue:43299", + "left": "issue:44336", + "right": "issue:44373", "accept": false, - "reason": "Both involve Qwen/MoE, but they are different model families and different load failures." + "reason": "ANSI output leakage and a wrong docstring are separate documentation/runtime issues." }, { - "left": "issue:44336", - "right": "issue:44373", + "left": "issue:42491", + "right": "issue:43299", "accept": false, - "reason": "ANSI output in loading_report vs a wrong docstring; not the same issue." + "reason": "Both involve Qwen3 MoE, but one is LoRA loading on hf4.x/hf5.x and the other is multimodal model loading; too different to merge." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Different loss-missing bugs in different models and settings." + "reason": "Missing load-balancing loss and missing auxiliary losses in D-FINE are different training-loss bugs." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Position_ids buffer access vs Gemma-4 Zero3 loading; unrelated." + "reason": "Buffer access on position_ids and a Gemma-4 ZeRO-3 loading problem are unrelated." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video pipeline image-processor lookup vs audio fbank feature extraction; different subsystems." + "reason": "Video-classification pipeline processor lookup and fbank feature extraction are different pipeline/audio issues." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import latency vs unrecognized model/config loading; not duplicates." + "reason": "Import performance and unrecognized model config are not the same underlying bug." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Different Qwen bugs: sliding window behavior vs a chat crash." + "reason": "Both mention Qwen models, but one is sliding-window behavior and the other is a chat crash; different code paths." }, { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "Specific Nemotron Flash Attention support bug vs a generic Flash Attention issue." + "reason": "A model-specific Flash Attention support gap is not the same as the generic Flash Attention 2.0 issue." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Different missing-symbol imports from different packages." + "reason": "Different missing symbols from different libraries; both are import errors but not one duplicate bug." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI-code output and GLM-5 RoPE implementation are unrelated." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction vs sys.modules access in modeling_utils." + "reason": "Image label reduction logic and unsafe sys.modules access are distinct bugs." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different exported symbols and different failure modes." - }, - { - "left": "issue:44336", - "right": "issue:44485", - "accept": false, - "reason": "Terminal ANSI leakage vs GLM-5 RoPE implementation; unrelated." + "reason": "Missing Qwen2.5 VL export and missing HybridCache export are separate symbol-export regressions." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Unsafe sys.modules access vs a test-file race condition; different bugs." + "reason": "sys.modules access safety and captured-info file races do not share a fix target." }, { "left": "issue:43673", "right": "issue:44561", "accept": false, - "reason": "Generation cache handling vs removal of an import helper for trust_remote_code models." + "reason": "Both are v5 regressions, but one is generation cache/chunked prefill and the other is trust_remote_code import behavior; too broad to merge." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Slow import vs inconsistent eos_token_id config; unrelated." + "reason": "Import slowdown and Gemma EOS-token config inconsistency are unrelated." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "Missing load-balancing loss vs a float16 CUDA loss crash; not the same code path." + "reason": "Load-balancing loss omission and float16 CUDA crash are different loss/runtime failures." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage vs Gemma-4 token-type defaults; different models and fixes." + "reason": "ESM2 breakage and Gemma-4 mm_token_type_ids defaulting are separate model issues." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "Buffer registration error vs a tokenizer/filtering slowdown after upgrade." + "reason": "Position_ids buffer access and post-upgrade data-filter slowness are unrelated." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "ImportError on Qwen image processor vs a zero3/BertModel IndexError." + "reason": "ImportError for a missing processor class and a zero-sized tensor IndexError are unrelated." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Strict config loading granite_speech vs missing deepstack_visual_indexes in Qwen vision config." + "reason": "Both involve strict config handling, but they are different model configs and different symptoms; too broad to treat as one duplicate." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Tokenizer output regression vs resize_token_embeddings not updating output embeddings." + "reason": "Tokenizer output regression and resize_token_embeddings/output_embeddings mismatch are separate tokenization/model-editing bugs." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Embedding resize bug vs Whisper empty-transcription generation failure." + "reason": "Embedding resize behavior and Whisper empty-transcription generation failure are unrelated." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Pipeline docs mismatch vs unsafe sys.modules access; no overlap." + "reason": "TextClassificationPipeline docs and unsafe sys.modules access are unrelated." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "Embedding resize/output mismatch vs version-dependent tokenization changes." + "reason": "Output-embedding resize bug and version-dependent tokenization changes are not the same issue." }, { "left": "issue:43232", "right": "issue:44336", "accept": false, - "reason": "Generation kwarg update bug vs terminal ANSI output in loading_report." + "reason": "Generation kwargs after sync_gpus and ANSI-code emission are unrelated." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Model loading failure on v5.1.0 vs tokenizer error masking when protobuf is missing." + "reason": "A model-loading failure on transformers>=5.1.0 and hidden tokenizer errors from protobuf handling are different bugs." }, { - "left": "issue:43065", - "right": "issue:43408", + "left": "issue:44496", + "right": "issue:45003", "accept": false, - "reason": "Dummy conv in Sam3PixelDecoder vs a model-type mismatch warning for sam3_video/sam3_tracker." + "reason": "Model config recognition and sys.modules access safety are different parts of the loading stack." }, { - "left": "issue:44496", - "right": "issue:45003", + "left": "issue:43065", + "right": "issue:43408", "accept": false, - "reason": "Config/model-type resolution failure vs unsafe sys.modules access; distinct problems." + "reason": "SAM3 pixel-decoder implementation and SAM3 tracker/model-type warning are different issues." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Missing strict field in a Qwen vision config vs AutoModelForCausalLM not unwrapping text_config; related domain but different concrete bugs." + "reason": "A strict-config field omission and AutoModelForCausalLM text_config unwrapping are separate composite-config bugs." } ] }, @@ -11037,14 +11002,14 @@ ], "soft_pairs": [ "issue:42886|issue:45229", - "issue:43257|issue:45137", "issue:43262|issue:44246", + "issue:43257|issue:45137", "issue:44995|issue:45200", - "issue:43475|issue:43540", "issue:44060|issue:44077", + "issue:43475|issue:43540", "issue:44466|issue:45335", - "issue:43653|issue:44162", "issue:45397|issue:45656", + "issue:43653|issue:44162", "issue:44995|issue:45588", "issue:44368|issue:45563", "issue:44368|issue:44977", @@ -11069,20 +11034,20 @@ "issue:44038|issue:44623", "issue:43475|issue:45381" ], - "prepared_review_unit_hash": "b99a8b016c653fb8a6af484cfad62016b4495c0fb83f4d55adb645fc70a44571", + "prepared_review_unit_hash": "a3b5e6c0b97ba765baaeade95f6f6192787dd2fe77833838d6eb66f5bfc010b7", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37146, + "serialized_chars": 37147, "estimated_input_tokens": 9287, "estimated_eval_tokens": 18830 }, @@ -11091,204 +11056,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: loading/import errors, generation-cache bugs, tokenizer/processor save issues, multimodal model-specific failures, and unrelated deepspeed/CI regressions. I don\u2019t see a single duplicate family here, so all soft-edge pairs should be rejected.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "This cluster is mostly a grab-bag of unrelated Transformers regression reports, many centered on v5-era model loading/generation/tokenizer issues. I don\u2019t see any pair that clearly describes the same concrete bug or a mergeable PR-sized fix.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43646 is the broadest representative regression report: it frames a general v5.0 breakage in custom model initialization, which fits the overall upgrade-regression theme better than the more model-specific tickets.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:43646 is the best single issue to anchor the cluster because it is general, user-visible, and closest to the shared theme of Transformers v5 compatibility breakages.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Offline tokenizer cache loading and Gemma 4 multi-GPU OOM are unrelated failure modes." + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU OOM are unrelated failures in different code paths." }, { - "left": "issue:43257", - "right": "issue:45137", + "left": "issue:43262", + "right": "issue:44246", "accept": false, - "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed is not the same bug as the ZeRO3 deque IndexError." + "reason": "Audio processor sampling-rate defaults and import slowdown are not the same bug." }, { - "left": "issue:43262", - "right": "issue:44246", + "left": "issue:43257", + "right": "issue:45137", "accept": false, - "reason": "Audio processor sample-rate defaulting and slow import are different subsystems and symptoms." + "reason": "Both involve DeepSpeed, but one is Qwen3 MOE weight conversion and the other is a ZeRO3 deque error." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "GlmMoeDsa stale indexer cache is unrelated to Gemma 4 mm_token_type_ids defaults." + "reason": "A stale second-forward cache crash is unrelated to the mm_token_type_ids defaulting issue." }, { - "left": "issue:43475", - "right": "issue:43540", + "left": "issue:44060", + "right": "issue:44077", "accept": false, - "reason": "Both are multimodal, but SAM3 encoder attribute failure and Qwen3OmniMoe video-processing ValueError are different code paths." + "reason": "Incorrect tied-weights warnings and patchtsmixer post_init allowance are different model/config behaviors." }, { - "left": "issue:44060", - "right": "issue:44077", + "left": "issue:43475", + "right": "issue:43540", "accept": false, - "reason": "Incorrect tied-weights warning and patchtsmixer post_init policy are unrelated model/config issues." + "reason": "SAM3 video encoder missing attribute and Qwen3OmniMoe video processing ValueError are different bugs." }, { "left": "issue:44466", "right": "issue:45335", "accept": false, - "reason": "Both mention tied embeddings, but one is serialization inconsistency and the other is resize_token_embeddings not updating decoder embeddings." + "reason": "Serialization of tied lm_head weights is not the same as resize_token_embeddings failing to update decoder embeddings." }, { - "left": "issue:43653", - "right": "issue:44162", + "left": "issue:45397", + "right": "issue:45656", "accept": false, - "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are unrelated." + "reason": "Gemma4 from_pretrained with ZeRO3 and duplicated optimizer steps are different DeepSpeed problems." }, { - "left": "issue:45397", - "right": "issue:45656", + "left": "issue:43653", + "right": "issue:44162", "accept": false, - "reason": "Gemma-4 Zero3 from_pretrained failure is not the same as deepspeed calling optimizer.step twice." + "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are unrelated tokenizer/model issues." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "Stale indexer cache in GlmMoeDsa and flash_attention.py crashing on s_aux=None are separate bugs." + "reason": "A model cache crash and a flash-attention sinkless AttributeError are different failure modes." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "A tied-word-embeddings warning in ms-swift fine-tuning is unrelated to paged generate() warning behavior." + "reason": "A tie_word_embeddings warning and a stale num_return_sequences warning are not the same underlying issue." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "Qwen3.5 tied-embedding warning and flash-attention generation failure do not share the same underlying defect." + "reason": "A config warning is not the same as Qwen3.5 flash-attention generation failure." }, { "left": "issue:43673", "right": "issue:44734", "accept": false, - "reason": "Both involve generation/cache, but chunked_prefill cache handling and /v1/responses continuation indexing are different code paths." + "reason": "Generation cache missing and KV-cache continuation indexing are both generation-related, but they are distinct code-path bugs." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "ZeRO3 deque underflow and auto_mappings references to removed Sam3LiteText configs are unrelated." + "reason": "A ZeRO3 deque error and stale auto_mappings references are unrelated." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and Qwen3.5 tied-embeddings warning are different model initialization problems." + "reason": "Qwen3-VL weight-shape mismatch is not the same as a tied-embeddings warning." }, { "left": "issue:44933", "right": "issue:45459", "accept": false, - "reason": "Missing image_utils import and protobuf-decode-error handling are different import/error-reporting bugs." + "reason": "A missing import from image_utils and suppressed tokenizer decode errors are different import/error-handling bugs." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper word-timestamp crash and Kimi-K2.5 tokenizer codec regression are both tokenizer-related but not the same bug." + "reason": "Whisper word-timestamp decode crash and Kimi-K2.5 codec regression are both tokenizer-adjacent, but not the same defect." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "Tokenizer save_pretrained writes the wrong tokenizer_class, while processor.save_pretrained is missing files; these are distinct save-path issues." + "reason": "Tokenizer save_pretrained config mismatch and processor save_pretrained missing files are separate serialization issues." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "sync_gpus generation-kwargs update and missing GenerationMixin cache in chunked_prefill are separate generation bugs." + "reason": "sync_gpus generation kwargs handling and chunked_prefill cache missing are different generation regressions." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer task AttributeError and a data-filtering slowdown after upgrade are unrelated." + "reason": "MLukeTokenizer task AttributeError and slowdown after upgrade are unrelated symptoms." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Transformers 5 custom model init breakage and ESM2 breakage are different model-loading failures." + "reason": "Custom model initialization breakage and ESM2 breakage are both broad regressions, but not the same concrete bug." }, { "left": "issue:43673", "right": "issue:45278", "accept": false, - "reason": "A generation cache bug and broad import errors after upgrade are not the same underlying issue." + "reason": "Missing GenerationMixin cache in chunked_prefill is not the same as many import errors after upgrade." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Processor save_pretrained missing files is unrelated to deepspeed double optimizer stepping." + "reason": "Processor serialization missing files and duplicated optimizer steps do not share a code path." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Performance regression in filtering does not match a deepspeed optimizer-step duplication bug." + "reason": "Performance slowdown after upgrade and repeated optimizer stepping are unrelated." }, { "left": "issue:44749", "right": "issue:45310", "accept": false, - "reason": "The filtering slowdown and Qwen3.5 MoE from_pretrained error are separate issues." + "reason": "Data filtering slowdown and Qwen3.5 MOE from_pretrained failure are different bugs." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are opposite directions and different formats." + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different directions of the serialization/load pipeline." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "PyTorch-missing AutoModel ImportError and Sentence Transformers loading hang are not the same bug." + "reason": "PyTorch-missing ImportError and Sentence Transformers hanging on load are unrelated issues." }, { "left": "issue:45230", "right": "issue:45397", "accept": false, - "reason": "A generic bug report cannot be merged with a specific Gemma-4 Zero3 from_pretrained failure." + "reason": "The generic bug report provides no concrete overlap with Gemma4 Zero3 from_pretrained." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 configuration guidance and InternVL2 meta-initialization incompatibility are unrelated." + "reason": "TF32 settings guidance and InternVL2 meta-initialization incompatibility are unrelated." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "SAM3 video encoder missing attribute and trust_remote_code AutoConfig field loss are different failures." + "reason": "SAM3 video encoder attribute loss and trust_remote_code AutoConfig field loss are different failures." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe transformers v5 bug and processor.save_pretrained missing files are unrelated." - }, - { - "left": "issue:43475", - "right": "issue:45381", - "accept": false, - "reason": "SAM3 Video missing fpn_position_embeddings and Qwen2.5-VL wrong vision_position_ids are separate video-model bugs." + "reason": "Qwen3-VL-Moe v5 issue and processor save_pretrained missing files are not the same bug." } ] }, @@ -11331,11 +11290,12 @@ "issue:43526", "issue:43673", "issue:43704", + "issue:43784", + "issue:43824", "issue:43873", "issue:43976", "issue:44038", "issue:44079", - "issue:44164", "issue:44188", "issue:44222", "issue:44242", @@ -11376,15 +11336,15 @@ "issue:44222|issue:45310", "issue:43673|issue:44188", "issue:36010|issue:41628", - "issue:42994|issue:44164", + "issue:42994|issue:43824", "issue:42222|issue:45200", "issue:42175|issue:45712", "issue:43232|issue:44792", "issue:44038|issue:45310", "issue:43526|issue:45561", - "issue:43504|issue:45561", - "issue:44496|issue:45561", "issue:43377|issue:44242", + "issue:44496|issue:45561", + "issue:43504|issue:45561", "issue:44964|issue:45137", "issue:42175|issue:44933", "issue:42898|issue:44568", @@ -11392,235 +11352,233 @@ "issue:43976|issue:44933", "issue:43479|issue:44869", "issue:43479|issue:44079", - "issue:44623|issue:45310" + "issue:44623|issue:45310", + "issue:43784|issue:43824" ], - "prepared_review_unit_hash": "7957d1c363d0912b679721c9d5578077f12198ad39f70297c1fdc3ac73dbfcec", + "prepared_review_unit_hash": "aae2b7f7a365c5119c8a224012c96782590731b73970ca4049d28b676ffa7711", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36576, - "estimated_input_tokens": 9144, - "estimated_eval_tokens": 18544 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37509, + "estimated_input_tokens": 9378, + "estimated_eval_tokens": 19012 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous and does not contain any true duplicate pairs. The strongest apparent overlaps are superficial topical similarities around import regressions, Qwen/MoE loading, and save/load behavior, but each points to a different code path.", - "confidence": 0.74, - "canonical_issue_reason": "issue:41628 is the clearest representative of the import/export regression mini-theme and the most broadly phrased among the similar-looking issues, even though the cluster overall is not a single duplicate set.", + "summary": "Most pairs are only loosely similar and should stay separate. The only accepted soft-edge duplicate is issue:42673 / issue:43704, which share the same title and appear to describe the same Qwen3ForCausalLM VRAM leak, with 42673 showing more discussion. No PR cluster is present.", + "confidence": 0.88, + "canonical_issue_reason": "issue:42673 is the better canonical issue because it is the earlier report and has substantially more discussion than issue:43704, while the titles indicate the same Qwen3ForCausalLM VRAM leak scenario.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41628 is the best single anchor for triage because it is concrete, library-wide, and sits closest to the other import-related reports.", + "best_issue_reason": "issue:42673 is the best issue overall because it is the clearest anchor for the duplicate pair and the later issue:43704 looks like a repeat report of the same bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "Both are import failures, but they target different missing exports/modules (`AutoImageProcessor` vs `image_utils`) and likely different fixes." + "reason": "Both are import-related, but they name different missing symbols and do not show the same concrete bug." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Different failure modes: Mistral regex patching vs auto-docstring annotation handling." + "reason": "Tokenizer regex crash vs auto-docstring AttributeError are different code paths and different failures." }, { "left": "issue:43475", "right": "issue:44661", "accept": false, - "reason": "SAM3 vision output attribute breakage vs tokenizer-mapping/model-registration logic; not the same bug." + "reason": "SAM 3 vision encoder attribute error and tokenizer-mapping model-addition failure are unrelated." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized model saving failure and `Llama4Config.pad_token_id` are separate config/persistence problems." + "reason": "Quantized model saving failure and missing pad_token_id on Llama4Config are different bugs." }, { "left": "issue:45375", "right": "issue:45478", "accept": false, - "reason": "Same model family, but one is a missing config field and the other is a `from_pretrained` loading regression; different underlying issues." + "reason": "Same model family, but the visible reports describe different failure modes and not one shared root cause." }, { "left": "issue:43232", "right": "issue:44188", "accept": false, - "reason": "One is generation kwargs handling after `sync_gpus`; the other is attention-kernel branching under `torch.compile`." + "reason": "Generation kwargs handling after sync_gpus and attention-kernel divergence under torch.compile are different problems." }, { "left": "issue:42673", "right": "issue:43704", - "accept": false, - "reason": "Same title, but the reports are separate instances and there is no evidence they are the same concrete leak." + "accept": true, + "reason": "Identical title and closely matching symptom indicate the same Qwen3ForCausalLM VRAM leak bug." }, { "left": "issue:45310", "right": "issue:45375", "accept": false, - "reason": "`from_pretrained` error for Qwen3.5 MoE vs missing `deepstack_visual_indexes` config field; not one fix." + "reason": "Both involve Qwen3.5 MoE, but one is a from_pretrained error and the other is a missing config field; not enough to treat as the same bug." }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization/offloading behavior and pipeline-parallelism output correctness are unrelated code paths." + "reason": "Quantization/offloading behavior and pipeline-parallel output correctness are different issues." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different symbols (`AutoImageProcessor` vs `HybridCache`) and likely different export points." + "reason": "Both are import errors, but they refer to different missing exports and likely different fixes." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "FP8 accuracy regression vs FP8 save/pretrained failure; one is numerical correctness, the other serialization." + "reason": "FP8 accuracy issue versus FP8 save_pretrained bug are related thematically but not the same concrete change." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Pipeline parallelism output mismatch vs Qwen3_5 image infinite loop; unrelated behaviors." + "reason": "Pipeline-parallel output mismatch and an infinite loop on a specific image are distinct bugs." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen3.5/MoE, but one is FP8 saving and the other is loading via `from_pretrained`; different bugs." + "reason": "FP8 save/pretrained issue and Qwen3.5 MoE from_pretrained error do not point to one shared code-path problem." }, { "left": "issue:43673", "right": "issue:44188", "accept": false, - "reason": "Chunked-prefill cache handling vs compile-time attention-kernel branching are distinct generation internals." + "reason": "Generation cache missing in chunked prefill and attention-kernel branching under torch.compile are different code paths." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "Both are import-related, but the missing symbol and module path differ (`GenerationMixin` vs `AutoImageProcessor`)." + "reason": "Both are import failures, but for different symbols and different breakages." }, { "left": "issue:42994", - "right": "issue:44164", + "right": "issue:43824", "accept": false, - "reason": "Quantized saving failure and `extra_state` handling in save/from_pretrained are separate persistence bugs." + "reason": "Quantized model saving and missing Qwen2_5_VL import are unrelated." }, { "left": "issue:42222", "right": "issue:45200", "accept": false, - "reason": "Different model families and different fixes: ViTPose breakage vs Gemma 4 multimodal token-type defaults." + "reason": "Broken vitpose models and Gemma 4 token-type-id defaults are separate model-specific bugs." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "Packaging/backend dependency issue vs dummy class cleanup in repo checks; unrelated." + "reason": "Packaging/backend dependency issue and leftover dummy classes leaking into dir(transformers) are not the same bug." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs update bug vs Janus image-generation test failure; different models and paths." + "reason": "Generation kwargs update after sync_gpus and a janus test failure do not share a concrete root cause." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Related Qwen/MoE family, but one is Qwen3-VL-Moe loading and the other is Qwen3.5 MoE `from_pretrained`; not the same concrete bug." + "reason": "Different Qwen model variants with different failure modes; too broad to merge." }, { "left": "issue:43526", "right": "issue:45561", "accept": false, - "reason": "Beit image processor label reduction vs pytest-xdist file race; unrelated subsystems." + "reason": "Processor label-reduction bug and pytest-xdist file-race bug are unrelated." }, { - "left": "issue:43504", - "right": "issue:45561", + "left": "issue:43377", + "right": "issue:44242", "accept": false, - "reason": "Legacy-field pretrained loading vs pytest-xdist race; no shared code path." + "reason": "Missing padding-mask support in MIMI encoder and missing load balancing loss are different bugs." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Unrecognized-model/config loading failure vs testing-utils race condition; unrelated." + "reason": "Unrecognized model load failure and test-worker file race are unrelated." }, { - "left": "issue:43377", - "right": "issue:44242", + "left": "issue:43504", + "right": "issue:45561", "accept": false, - "reason": "MIMI padding-mask discrepancy vs MoE load-balancing loss omission; different model components." + "reason": "Beit legacy-field loading issue and xdist race on captured_info.txt are unrelated." }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Phi-4 multimodal loading issue vs DeepSpeed ZeRO3 deque error; unrelated." + "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque underflow are different problems." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Tensorflow extra/backend packaging issue vs nonexistent import from `image_utils`; unrelated." + "reason": "Missing TensorFlow backend dependency and nonexistent image_utils import are different issues." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Tokenizer cleanup behavior change vs BOS/EOS special-token insertion bug; distinct tokenization behaviors." + "reason": "clean_up_tokenization_spaces behavior change and missing BOS/EOS with add_special_tokens are distinct tokenizer regressions." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Sentence Transformers loading hang vs embedding resize/reinit bug; unrelated." + "reason": "Sentence-Transformers loading hang and embedding reinitialization after resize are unrelated." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python-version compatibility problem vs image_utils import failure; unrelated." + "reason": "Python version compatibility breakage and image_utils import error are not the same bug." }, { "left": "issue:43479", "right": "issue:44869", "accept": false, - "reason": "Phi4 multimodal default-config initialization vs Whisper timestamp decode crash; different subsystems." + "reason": "Multimodal config default initialization bug and Whisper timestamp decoding crash are unrelated." }, { "left": "issue:43479", "right": "issue:44079", "accept": false, - "reason": "Default multimodal config init vs `ModelOutput` key assignment semantics; unrelated." - }, - { - "left": "issue:44623", - "right": "issue:45310", - "accept": false, - "reason": "Processor save_pretrained missing files vs Qwen3.5 MoE loading error; persistence vs loading bug." + "reason": "Config defaulting bug and ModelOutput key assignment bug are different code paths." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, + "evaluator_result": { + "accept": false, + "feedback": "The core duplicate call on issue:42673 / issue:43704 looks plausible from the identical title, but the writeup overstates the evidence in a few places. The packet does not show their full descriptions, so phrases like \u201cexact same \u2026 in title and description\u201d are too strong. Also, the justification about having \u201cmore discussion/inbound references\u201d is only partly grounded: 42673 clearly has more discussion, but both items show 0 inbound references in the packet. The soft-edge verdicts themselves are mostly conservative, but the overall summary should be tightened to stay within the visible evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -11654,8 +11612,6 @@ "issue:43479", "issue:43644", "issue:43723", - "issue:43784", - "issue:43824", "issue:43874", "issue:44060", "issue:44188", @@ -11667,6 +11623,7 @@ "issue:44297", "issue:44373", "issue:44448", + "issue:44483", "issue:44484", "issue:44568", "issue:44623", @@ -11681,20 +11638,20 @@ "issue:45310", "issue:45356", "issue:45405", + "issue:45406", "issue:45656", "issue:45712", "issue:45812", "issue:45834" ], "soft_pairs": [ - "issue:43784|issue:43824", "issue:44188|issue:44373", "issue:42898|issue:45812", "issue:43278|issue:45137", "issue:44734|issue:45103", "issue:30333|issue:33290", - "issue:43723|issue:44568", "issue:43874|issue:45092", + "issue:43723|issue:44568", "issue:42898|issue:45356", "issue:44060|issue:45125", "issue:42898|issue:44779", @@ -11712,241 +11669,242 @@ "issue:41628|issue:42757", "issue:22355|issue:28282", "issue:30333|issue:36331", - "issue:44297|issue:45310", "issue:30333|issue:34689", + "issue:44297|issue:45310", "issue:44222|issue:44484", "issue:43232|issue:45834", "issue:43723|issue:44977", "issue:28282|issue:30333", "issue:44263|issue:45092", - "issue:30333|issue:35141" + "issue:30333|issue:35141", + "issue:44483|issue:45406" ], - "prepared_review_unit_hash": "337840a5810e4c271382568acc62e5505032ce82763fc387c9fd450a30dd46c3", + "prepared_review_unit_hash": "ed00c924f7aec4a08ed9223bb721939528da7fc92f02b078a0c2bc444d7d6a83", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37572, - "estimated_input_tokens": 9393, - "estimated_eval_tokens": 19042 + "serialized_chars": 37555, + "estimated_input_tokens": 9389, + "estimated_eval_tokens": 19034 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of Transformers issues with superficial overlap around tokenizer/regression reports; none of the soft pairs look like the same underlying bug or change, so no duplicate merges are warranted.", - "confidence": 0.43, - "canonical_issue_reason": "issue:42898 is the closest representative of the recurring v5 tokenizer-regression theme and has the broadest, most explicit behavioral description among the set, though it is not a duplicate of the others.", + "summary": "These items are a heterogeneous grab bag of unrelated bugs/regressions (tokenizers, generation, deepspeed, serving, save/load, docs). None of the soft pairs look like the same underlying change or defect, so all are rejected.", + "confidence": 0.99, + "canonical_issue_reason": "No true canonical duplicate emerges from this heterogeneous set; if one issue must anchor the cluster, issue 45812 is the best-documented open regression with broad impact.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42898 is the strongest standalone issue to anchor the cluster because it describes a concrete, versioned tokenizer behavior change and is central to the small tokenizer-related subtheme here.", + "best_issue_reason": "Issue 45812 is the strongest standalone report: open, detailed, reproducible, and impacts all Granite models.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43784", - "right": "issue:43824", - "accept": false, - "reason": "Different import failures in unrelated packages/models: sentence-transformers import vs missing Qwen2_5_VL class." - }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "One is a torch.compile attention-kernel divergence; the other is a docstring mismatch for position_ids." + "reason": "Attention-kernel divergence under torch.compile is unrelated to a docstring-only position_ids report." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Both involve tokenization regressions, but one is about clean_up_tokenization_spaces behavior and the other about Granite token IDs; different bugs/code paths." + "reason": "Tokenization-space cleanup behavior change is unrelated to Granite token ID regression." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Embedding dtype drift in evaluation vs a DeepSpeed ZeRO3 deque error are unrelated failures." + "reason": "Embedding dtype drift during eval is unrelated to a DeepSpeed ZeRO3 deque IndexError." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "KV-cache indexing crash in serving is unrelated to auto_docstring annotation handling." + "reason": "KV-cache continuation crash in serve is unrelated to an auto_docstring AttributeError under future annotations." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLflow reporting on training failure and adafactor OOM in deepspeed are different training problems." + "reason": "MLflow not reporting failed jobs is unrelated to an Adafactor/DeepSpeed OOM." }, { - "left": "issue:43723", - "right": "issue:44568", + "left": "issue:43874", + "right": "issue:45092", "accept": false, - "reason": "Both are tokenizer regressions in v5, but one is a generic from_pretrained loading issue and the other is add_special_tokens/BOS-EOS behavior." + "reason": "Missing image-processor method for GLM46V is unrelated to remote-code checkpoint/meta-init incompatibility." }, { - "left": "issue:43874", - "right": "issue:45092", + "left": "issue:43723", + "right": "issue:44568", "accept": false, - "reason": "Missing image patch count in a fast image processor is not the same bug as remote-code checkpoint/meta-init incompatibility." + "reason": "Tokenizer loading failure and missing BOS/EOS insertion are different tokenizer regressions." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but the specific failures differ: whitespace cleanup behavior vs codec/fix_mistral_regex handling." + "reason": "General v5 tokenization-space behavior change is unrelated to Kimi-K2.5 codec handling regression." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Incorrect tied-weights warning for Qwen3-Next is unrelated to Qwen3_5Moe missing _tp_plan." + "reason": "Incorrect tied-weights warning is unrelated to missing tensor-parallel plan metadata." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both mention tokenizer regressions, but Deepseek tokenization output mismatch is a different concrete issue than clean_up_tokenization_spaces." + "reason": "Tokenization cleanup behavior change is unrelated to Deepseek tokenizer output corruption." }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "A dependency-version bump and leftover dummy classes failing repo checks are unrelated." + "reason": "PEFT version bump issue is unrelated to leftover dummy-class cleanup." }, { "left": "issue:42898", "right": "issue:44448", "accept": false, - "reason": "Pegasus output drift across v4/v5 is a generation/model-output regression, not the same as clean_up_tokenization_spaces." + "reason": "Tokenization cleanup behavior change is unrelated to Pegasus output regression." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "A position_ids buffer-reading error and a double optimizer-step issue with DeepSpeed are different failure modes." + "reason": "Position_ids buffer access error is unrelated to a double optimizer-step DeepSpeed bug." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Missing import from image_utils is unrelated to dummy classes leaking into dir(transformers)." + "reason": "Missing image_utils import is unrelated to dummy-class leakage into dir(transformers)." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Both involve save_pretrained, but one is FP8 MoE serialization and the other is tokenizer config metadata; not the same bug." + "reason": "FP8 save_pretrained issue is unrelated to tokenizer class metadata mismatch on save." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer config mismatch has no relation to DeepSpeed calling optimizer.step twice." + "reason": "Tokenizer save metadata mismatch is unrelated to duplicate optimizer steps." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Example-script multi-label classification output bug and local dataset loading failure are different script-level issues." + "reason": "Empty multi-label classification results are unrelated to local dataset loading failure." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Bad default multimodal config initialization is not the same as non-persistent buffers being filled with junk." + "reason": "Multimodal config default-init bug is unrelated to non-persistent buffer junk values." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both are save_pretrained-related, but they affect different objects and different serialization outputs." + "reason": "FP8 MoE save_pretrained behavior is unrelated to processor.save_pretrained missing files." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FineGrainedFP8 accuracy degradation is unrelated to processor save_pretrained missing files." + "reason": "FineGrainedFP8 accuracy issue is unrelated to processor save_pretrained file omission." }, { "left": "issue:44206", "right": "issue:45103", "accept": false, - "reason": "Feature extractor center-arg regression and auto_docstring annotation crash do not share a code path." + "reason": "Unsupported center arg in feature extractor is unrelated to auto_docstring crashing on annotations." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Different missing imports from different packages: transformers vs huggingface_hub." + "reason": "AutoImageProcessor import failure is unrelated to is_offline_mode import failure from huggingface_hub." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Missing transformers.onnx module and missing PyTorch dependency are separate environment/import problems." + "reason": "Missing transformers.onnx module is unrelated to AutoModel requiring absent PyTorch." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "MLflow reporting on failure and Trainer.compute_loss signature breakage are unrelated trainer issues." + "reason": "MLflow reporting on failed jobs is unrelated to CustomTrainer.compute_loss signature mismatch." }, { - "left": "issue:44297", - "right": "issue:45310", + "left": "issue:30333", + "right": "issue:34689", "accept": false, - "reason": "Tokenizer save metadata mismatch is not the same as Qwen3.5 MoE from_pretrained load failure." + "reason": "MLflow failure handling is unrelated to Llama 3.2 Vision model-loading regression." }, { - "left": "issue:30333", - "right": "issue:34689", + "left": "issue:44297", + "right": "issue:45310", "accept": false, - "reason": "Failed MLflow reporting and Llama 3.2 Vision model loading regression are unrelated." + "reason": "Tokenizer save metadata mismatch is unrelated to Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "FP8 MoE save_pretrained bug and max_shard_size default question are different save_pretrained concerns." + "reason": "FP8 MoE save behavior is unrelated to the default max_shard_size question." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation kwargs update after sync_gpus is unrelated to long-ocr-input index errors in Kosmos2.5." + "reason": "Generation kwargs sync_gpus bug is unrelated to Kosmos2.5 OCR long-input index error." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Generic tokenizer loading issues and Qwen3.5 flash-attention generation failures are different problems." + "reason": "Tokenizer loading issue is unrelated to Qwen3.5 flash-attention generation failure." }, { "left": "issue:28282", "right": "issue:30333", "accept": false, - "reason": "AutoModel missing PyTorch and MLflow-not-reported-on-failure are unrelated." + "reason": "Missing PyTorch import error is unrelated to failed-job MLflow reporting." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "torch.split return handling in a MoE indexer is unrelated to meta-initialization incompatibility." + "reason": "GlmMoeDsaIndexer torch.split issue is unrelated to InternVL2 meta-init incompatibility." }, { "left": "issue:30333", "right": "issue:35141", "accept": false, - "reason": "MLflow reporting on failed training and embedding reinitialization during resize are distinct bugs." + "reason": "Training-job MLflow reporting is unrelated to embedding reinitialization after resize." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat-completions request rejection is unrelated to Gemma4Processor missing _tokenizer crash." } ] }, @@ -11997,16 +11955,17 @@ "issue:44261", "issue:44263", "issue:44295", + "issue:44297", "issue:44360", "issue:44373", "issue:44442", - "issue:44483", "issue:44485", "issue:44514", "issue:44521", "issue:44568", "issue:44623", "issue:44704", + "issue:44743", "issue:44871", "issue:44913", "issue:44991", @@ -12015,12 +11974,10 @@ "issue:45278", "issue:45310", "issue:45357", - "issue:45406", "issue:45593", "issue:45753" ], "soft_pairs": [ - "issue:44483|issue:45406", "issue:44263|issue:44360", "issue:44373|issue:44485", "issue:44295|issue:44623", @@ -12048,210 +12005,218 @@ "issue:42371|issue:44263", "issue:22355|issue:28218", "issue:44117|issue:44704", - "issue:30990|issue:33357" + "issue:30990|issue:33357", + "issue:44295|issue:44297", + "issue:43232|issue:44743" ], - "prepared_review_unit_hash": "26b759fa9f72b604900b7dbc09793524144ce3820d4e67ca39562d3753991247", + "prepared_review_unit_hash": "17ae752da0a7d4660a9a1a654bea36c8a277f1e0b1d11f2941c8f61ca2994646", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 35994, - "estimated_input_tokens": 8999, - "estimated_eval_tokens": 18254 + "soft_pair_count": 30, + "serialized_chars": 36309, + "estimated_input_tokens": 9078, + "estimated_eval_tokens": 18412 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly a grab-bag of unrelated Transformers regression reports; the only strong common thread is tokenizer/loading-related breakage, but the pairs are still too specific to merge together confidently.", - "confidence": 0.68, - "canonical_issue_reason": "issue:43723 is the broadest and most representative user-facing regression here, centered on tokenizer loading via AutoTokenizer.from_pretrained in v5.", + "summary": "The cluster is a mixed bag of tokenizer, import, config, and model-specific regression reports rather than a single duplicate thread. None of the soft pairs look like the same underlying bug or change; the broad tokenizer-loading regression issue is the best umbrella node.", + "confidence": 0.88, + "canonical_issue_reason": "issue:43723 is the broadest, most umbrella-like tokenizer regression here and best represents the tokenizer-loading side of this heterogeneous cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43723 is the best anchor for this cluster because it is generic, high-signal, and closest to several tokenizer-related reports.", + "best_issue_reason": "issue:43723 is the most general issue in the set and can anchor the cluster better than the highly specific model/config tickets.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44483", - "right": "issue:45406", - "accept": false, - "reason": "Different failures and code paths: chat/completions request validation vs Gemma4Processor missing _tokenizer in transformers serve." - }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "Both mention DSA, but one is about torch.split return values and the other about a missing ReLU in the indexer; not the same bug." + "reason": "Both mention DSA/indexer internals, but one is about torch.split return handling and the other about a missing ReLU; different bugs." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Docstring/position_ids documentation issue vs GLM-5 RoPE implementation bug; unrelated symptoms." + "reason": "A docstring mismatch is not the same as a RoPE implementation discussion; no shared concrete failure." }, { "left": "issue:44295", "right": "issue:44623", "accept": false, - "reason": "Registering position_ids as a buffer causing read errors is unrelated to processor.save_pretrained missing files." + "reason": "Reading a buffer-backed position_ids tensor and missing processor save files are unrelated save/load problems." }, { "left": "issue:44514", "right": "issue:45103", "accept": false, - "reason": "apply_chat_template batching crash is unrelated to auto_docstring failing under future annotations." + "reason": "One is a Qwen2_5_VL chat-template batching crash; the other is an auto-docstring annotation crash. Different subsystems." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "Layer norm epsilon precision mismatch is not the same as old InternVL2 remote-code checkpoints failing meta initialization." + "reason": "MLA layernorm epsilon handling and InternVL2 meta-init compatibility are distinct multimodal issues." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Different multimodal Qwen/InternVL save/load problems; one is meta-init incompatibility, the other is wrong visual encoder keys on save." + "reason": "Both are multimodal-model regressions, but one is meta initialization for remote-code checkpoints and the other is incorrect visual encoder keys on save." }, { "left": "issue:43723", "right": "issue:44871", "accept": false, - "reason": "Both are tokenizer-related, but one is generic loading failure and the other is eos_token_id config inconsistency for a specific model." + "reason": "Both involve tokenizer/model loading, but one is a generic AutoTokenizer load regression and the other is an eos_token_id config mismatch." }, { "left": "issue:43723", "right": "issue:44991", "accept": false, - "reason": "Both involve AutoTokenizer loading, but the affected models and failure mode are not shown to be the same bug." + "reason": "Both are tokenizer load failures, but the affected model and likely root cause are different; too broad to merge as duplicates." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence Transformers loading hang and SigLIP2 zero probabilities are different model/runtime issues." + "reason": "Sentence-Transformers loading hang and zero-probability outputs are unrelated symptoms and code paths." }, { "left": "issue:43723", "right": "issue:44442", "accept": false, - "reason": "FastSpeech2ConformerTokenizer load failure is a separate model-specific tokenizer issue, not clearly the same as the v5 AutoTokenizer regression." + "reason": "FastSpeech2ConformerTokenizer load failure is a model-specific tokenizer registration problem, not the same as the v5 AutoTokenizer regression." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Import errors for different symbols do not indicate the same underlying export bug." + "reason": "Both are import errors, but they concern different missing symbols from different model families." }, { "left": "issue:43479", "right": "issue:44913", "accept": false, - "reason": "Different config bugs: multimodal default config initialization vs GPTNeoX rotary_pct not persisting on reload." + "reason": "Default multimodal config initialization and GPTNeoX rotary_pct reload behavior are separate config bugs." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "FineGrainedFP8 accuracy issues are unrelated to Qwen3.5 MoE from_pretrained errors." + "reason": "FineGrainedFP8 accuracy and Qwen3.5 Moe from_pretrained errors are unrelated regressions." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "Different DSA/position_ids problems with no shared concrete code-path." + "reason": "Indexer split-return behavior and a docstring issue are not the same functional bug." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Tokenizer offset_mapping regression and missing BOS/EOS insertion are different tokenizer behaviors." + "reason": "Offset mapping correctness and add_special_tokens BOS/EOS behavior are different tokenizer behaviors." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "Different problems: FP8 accuracy vs Qwen3-VL-Moe bug in v5." + "reason": "An FP8 accuracy issue and a Qwen3-VL-Moe bug are unrelated." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "A single huggingface_hub import symbol issue is not the same as a broad report of many import errors after upgrade." + "reason": "A specific missing import from huggingface_hub is not the same as a broad set of import errors after upgrade." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Offline tokenizer cache loading and quantized model saving are unrelated." + "reason": "Offline cache loading for tokenizers and quantized model saving are different storage-path failures." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation kwargs handling after sync_gpus is unrelated to D-FINE auxiliary-loss behavior." + "reason": "Generation-state handling after sync_gpus and D-FINE auxiliary-loss logic are unrelated model behaviors." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Smollm3 RoPE-layer mismatch and Qwen3-Next tied-weights warning are different model/config issues." + "reason": "Smollm3 RoPE-layer dropping and Qwen3-Next tied-weights warnings are different architecture-specific issues." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "CLI env command breakage and sentence-transformers import NameError are unrelated." + "reason": "Broken transformers env CLI and a sentence-transformers import NameError are separate failures." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "FP8 accuracy issues are unrelated to Qwen3_5 infinite-loop behavior on a specific image." + "reason": "FP8 accuracy regression and Qwen3_5 infinite-loop behavior are unrelated." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline cache loading and wrong tokenizer decoder type are different tokenizer failures." + "reason": "Offline tokenizer cache loading and wrong decoder type in v5 are different tokenizer/config problems." }, { "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "Different multimodal/tokenization regressions: changed tokenization across versions vs all-zero assistant masks." + "reason": "Tokenization changes across versions and multimodal assistant_mask handling are different bugs." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 API guidance and a DSA torch.split issue are unrelated." + "reason": "TF32 API guidance and DSA indexer split behavior do not share a concrete code-path problem." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "onnx import failure and extra space after added token are unrelated issues." + "reason": "Missing transformers.onnx module and an extra-space tokenizer bug are unrelated." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "Tokenizer mapping returning None and AutoProcessor not forwarding kwargs to cached_file are separate loader bugs." + "reason": "A tokenizer-mapping None bug and passing kwargs to cached_file in AutoProcessor are separate load-path issues." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Sentence Transformers loading hang and MacOS bus error with CLIP are different failures." + "reason": "A loading hang and a MacOS bus error are different runtime failures." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Buffer access for position_ids and tokenizer_config class mismatch are unrelated serialization bugs." + }, + { + "left": "issue:43232", + "right": "issue:44743", + "accept": false, + "reason": "Both involve generation/cache behavior, but one resets recurrent states and the other is a separate modular_qwen3_5 cache bug." } ] }, @@ -12278,13 +12243,14 @@ "issue:29127", "issue:30990", "issue:36296", + "issue:38617", "issue:39401", "issue:41628", "issue:42757", "issue:42907", + "issue:43066", "issue:43116", "issue:43122", - "issue:43232", "issue:43278", "issue:43408", "issue:43452", @@ -12309,7 +12275,6 @@ "issue:44521", "issue:44625", "issue:44704", - "issue:44743", "issue:44779", "issue:44829", "issue:44945", @@ -12325,15 +12290,13 @@ "issue:45753" ], "soft_pairs": [ - "issue:44295|issue:44297", - "issue:43232|issue:44743", "issue:44704|issue:45290", "issue:42907|issue:44945", "issue:43122|issue:44514", - "issue:43976|issue:44945", "issue:41628|issue:45278", - "issue:39401|issue:44488", + "issue:43976|issue:44945", "issue:43784|issue:45600", + "issue:39401|issue:44488", "issue:44261|issue:44373", "issue:43408|issue:43475", "issue:42907|issue:43937", @@ -12357,234 +12320,236 @@ "issue:42907|issue:44222", "issue:43122|issue:44977", "issue:43992|issue:44704", - "issue:43784|issue:45446" + "issue:43784|issue:45446", + "issue:43066|issue:44779", + "issue:38617|issue:45278" ], - "prepared_review_unit_hash": "8b1a2a1e18838fa8da6d21b0b8896a7972f98c962a1ae00a2c50ae875afa4ffe", + "prepared_review_unit_hash": "2552135a8a8c9ea209aa57914601d3c7c03b1a822959f2133beb4abd25a3c167", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37225, - "estimated_input_tokens": 9307, - "estimated_eval_tokens": 18870 + "serialized_chars": 37179, + "estimated_input_tokens": 9295, + "estimated_eval_tokens": 18846 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft pairs look like true duplicates; the set is mostly unrelated issue reports across different models, import paths, tokenizer behavior, and generation bugs. The most central issue is the open multimodal apply_chat_template report with the highest activity and references.", - "confidence": 0.91, - "canonical_issue_reason": "issue:44521 is the most active and referenced report, and it has a concrete, reproducible bug description in a currently open area.", + "summary": "The cluster is highly heterogeneous: most items are distinct model-, tokenizer-, import-, or training-path bugs, not duplicates. I did not find any soft pair that looks like the same underlying issue or change.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44779 is the most self-contained, actionable regression report in the set, with a clear version boundary and concrete tokenizer misbehavior.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the best cluster anchor because it has the strongest discussion/inbound-reference signal and a clearly scoped user-facing failure mode.", + "best_issue_reason": "issue:44779 is the best representative issue for triage because it is specific, reproducible, and clearly scoped.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44295", - "right": "issue:44297", - "accept": false, - "reason": "Both are tokenizer/model serialization issues, but one is about reading registered buffer position_ids and the other is about tokenizer_class metadata in tokenizer_config.json; different failure modes and code paths." - }, - { - "left": "issue:43232", - "right": "issue:44743", - "accept": false, - "reason": "Both involve generation/cache state, but one is about _update_model_kwargs_for_generation after sync_gpus and the other about recurrent state reset in modular_qwen3_5.py; not the same bug." - }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "AutoProcessor kwargs forwarding and apply_chat_template crashes with tool-call assistant messages are distinct processor/template bugs." + "reason": "Different code paths: cached_file kwarg forwarding vs chat-template tokenization crash with tool-call messages." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and incorrect pipeline-parallel LLM output are unrelated." + "reason": "Unrelated failures: dequantized model saving vs incorrect outputs under pipeline parallelism." }, { "left": "issue:43122", "right": "issue:44514", "accept": false, - "reason": "Tokenizer behavior changes across versions and a batched apply_chat_template padding crash are different issues." - }, - { - "left": "issue:43976", - "right": "issue:44945", - "accept": false, - "reason": "Python version compatibility is unrelated to pipeline parallelism output corruption." + "reason": "Both are tokenizer/chat-template complaints, but one is a tokenization regression and the other is a batched processor padding crash." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "A missing AutoImageProcessor import and a broad wave of import errors after upgrade may be thematically similar, but they are not the same concrete broken symbol or code path." + "reason": "One is a single missing export; the other is a broad upgrade-time import breakage across many symbols." }, { - "left": "issue:39401", - "right": "issue:44488", + "left": "issue:43976", + "right": "issue:44945", "accept": false, - "reason": "Qwen3 offset_mapping correctness and failure to load a specific model are different problems." + "reason": "Python-version compatibility problem vs pipeline-parallel inference correctness are not the same bug." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "sentence-transformers import NameError and stale auto_mappings references to removed configs are separate import-time regressions." + "reason": "Import NameError in sentence-transformers integration vs stale auto_mappings references to removed configs." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Tokenizer offset-mapping bug for Qwen3 is unrelated to loading a specific model repository." }, { "left": "issue:44261", "right": "issue:44373", "accept": false, - "reason": "One is a numerical precision/config issue in MLA q_a_layernorm; the other is a wrong docstring for position_ids." + "reason": "Runtime precision/config issue vs a docstring mismatch; not the same underlying defect." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "A model-type mismatch warning and a missing fpn_position_embeddings attribute in SAM 3 video are not the same defect." + "reason": "A warning about mismatched model types is different from an attribute error in SAM3 vision encoder output." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Dequantized save failures and invalid GenerationConfig for GLM-5 are unrelated." + "reason": "Different subsystems: save/dequantization failure vs invalid generation config handling." }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "DSA indexer activation logic and GLM-5 RoPE implementation are entirely different areas." + "reason": "Unrelated ML internals: DSA indexer activation vs GLM-5 RoPE implementation." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "num_labels propagation in Qwen3.5 config and flash_attention_3 degenerate training are different behaviors." + "reason": "Config propagation bug in Qwen3.5 is unrelated to flash_attention_3 causing degenerate training." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Tokenizer offset_mapping correctness and multimodal assistant_masks being all-zero are different apply_chat_template/tokenizer bugs." + "reason": "Both involve tokenization-like behavior, but offset mapping and assistant mask generation are different defects." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Empty multi-label classification outputs in an example script and Gemma 4 mm_token_type_ids defaults are unrelated." + "reason": "Example-script multi-label output bug vs Gemma 4 multimodal token-type defaulting are not the same path." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "A Qwen3-VL-Moe load bug and tokenizer_class mismatch on save are different code paths." + "reason": "Qwen3-VL-Moe loading bug vs tokenizer save metadata mismatch; different operations and symptoms." }, { "left": "issue:29127", "right": "issue:30990", "accept": false, - "reason": "LayoutLMv3 error messaging and Sentence Transformers loading hang are unrelated." + "reason": "LayoutLMv3 error-message clarity vs Sentence Transformers loading hang are unrelated." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Saving dequantized Ministrals/Devstrals and loading Phi-4 multimodal instruct are different model-specific failures." + "reason": "Dequantized save failure vs loading Phi-4 multimodal instruct are separate model-loading/saving issues." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both concern tokenization regressions, but for different model families and different observable failures." + "reason": "Both are tokenization regressions, but they affect different models and different tokenization behaviors." }, { "left": "issue:44038", "right": "issue:44295", "accept": false, - "reason": "Qwen3-VL-Moe loading and position_ids buffer access are unrelated." + "reason": "Qwen3-VL-Moe bug vs position_ids buffer read error; no shared code path is evident." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor-parallel training bugs and multi-label classification empty outputs are different domains." + "reason": "Tensor-parallel training bug is unrelated to the multi-label classification example script returning empty results." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Dequantized save issues and gguf_file failures in from_pretrained are separate loading/saving problems." + "reason": "Saving a dequantized model and gguf_file loading failures are different serialization/loading problems." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Both are missing imports, but they are different symbols from different subsystems and not one shared underlying bug." + "reason": "Two import errors, but for different symbols and likely different missing-export causes." }, { "left": "issue:44038", "right": "issue:44222", "accept": false, - "reason": "Qwen3-VL-Moe load failure and FP8 save_pretrained for moe are different change points." + "reason": "Qwen3-VL-Moe loading issue vs FP8 save_pretrained failure are distinct save/load paths." }, { "left": "issue:45216", "right": "issue:45478", "accept": false, - "reason": "Both mention Qwen3.5 and newer versions, but one is about save_pretrained checkpoints and the other about from_pretrained loading errors; not the same concrete failure." + "reason": "Both mention Qwen3.5 and newer versions, but one is save_pretrained checkpoint corruption and the other is from_pretrained loading failure." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Missing is_offline_mode from huggingface_hub and missing HybridCache from transformers are unrelated import errors." + "reason": "Different packages and symbols: huggingface_hub import failure vs transformers import failure." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "Embedding dtype drift between train/eval and incorrect pipeline-parallel output are different bugs." + "reason": "Training/evaluate dtype drift is unrelated to incorrect outputs under pipeline parallelism." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "SigLIP2 implementation mismatch and Qwen3_5 infinite loop on a specific image are unrelated multimodal problems." + "reason": "SigLIP2 implementation discrepancy and Qwen3_5 image-specific infinite loop are different model-specific bugs." }, { "left": "issue:42907", "right": "issue:44222", "accept": false, - "reason": "Both touch MoE/saving, but one is dequantized model save failure and the other is FP8 save_pretrained behavior; not the same issue." + "reason": "Saving dequantized models vs FP8 moe save_pretrained are similar only at a very broad level, not the same bug." }, { "left": "issue:43122", "right": "issue:44977", "accept": false, - "reason": "Tokenizer differences across versions and Qwen3.5 flash-attention generation failure are separate regressions." + "reason": "Both are tokenizer regressions in v5, but they concern different models and likely different fixes." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder and cached_file kwargs forwarding are unrelated model-loading issues." + "reason": "Missing embed_tokens.weight in UMT5Encoder loading is unrelated to AutoProcessor kwarg forwarding." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "Import NameError from latest transformers vs a PyTorch version check bug in flex_attention are unrelated." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Wrong tokenizer decoder type and incorrect Deepseek tokenization are both tokenizer-related, but not the same concrete defect." }, { - "left": "issue:43784", - "right": "issue:45446", + "left": "issue:38617", + "right": "issue:45278", "accept": false, - "reason": "sentence-transformers import NameError and a PyTorch version check bug in flex_attention are different import-time problems." + "reason": "A single missing import symbol and a broad post-upgrade import breakage are not sufficiently the same root cause." } ] }, @@ -12618,7 +12583,6 @@ "issue:42371", "issue:42673", "issue:43064", - "issue:43066", "issue:43116", "issue:43421", "issue:43441", @@ -12640,13 +12604,14 @@ "issue:44484", "issue:44485", "issue:44779", + "issue:44829", "issue:44936", "issue:44977", "issue:45071", "issue:45072", "issue:45081", "issue:45083", - "issue:45278", + "issue:45216", "issue:45310", "issue:45341", "issue:45357", @@ -12658,8 +12623,6 @@ "issue:45588" ], "soft_pairs": [ - "issue:43066|issue:44779", - "issue:38617|issue:45278", "issue:42673|issue:45072", "issue:43937|issue:44485", "issue:22355|issue:29127", @@ -12672,8 +12635,8 @@ "issue:43116|issue:43825", "issue:45357|issue:45478", "issue:44038|issue:45478", - "issue:43937|issue:45563", "issue:45083|issue:45446", + "issue:43937|issue:45563", "issue:44117|issue:45588", "issue:33453|issue:39401", "issue:39401|issue:44779", @@ -12688,222 +12651,217 @@ "issue:38617|issue:41628", "issue:39401|issue:43421", "issue:44410|issue:45310", + "issue:44829|issue:45216", "issue:44484|issue:45310" ], - "prepared_review_unit_hash": "3b8a61bb52959381db3effb138e257402a71e8dc3ab1e052e1fc8711992f6000", + "prepared_review_unit_hash": "8328568362319c77935459c38a2f7d7aadf6c1159643df8afd754179cd15a817", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36739, - "estimated_input_tokens": 9185, - "estimated_eval_tokens": 18626 + "soft_pair_count": 30, + "serialized_chars": 36606, + "estimated_input_tokens": 9152, + "estimated_eval_tokens": 18560 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is not a true duplicate set: the issues span many unrelated bugs (tokenizers, training, vision, imports, docs, version checks). I would not merge any of the proposed soft pairs.", - "confidence": 0.95, - "canonical_issue_reason": "If a single representative is required, issue:45561 is the most concrete and actively discussed report, but the cluster is too heterogeneous for a real canonical issue.", + "summary": "This cluster is mostly heterogeneous; the only noticeable subtheme is a small set of Qwen3.5 save/load regressions, but the listed soft pairs are not strong duplicates. I would not merge any of the soft-edge candidates.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45310 is the closest representative of the small Qwen3.5 serialization/from_pretrained subtheme, but the cluster overall is too mixed to treat it as a true duplicate bucket.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45561 is the clearest actionable bug report with the strongest current activity, though it should not be treated as a global duplicate target.", + "best_issue_reason": "issue:45310 is the best single representative by breadth and relevance to the only recurring model-specific regression theme in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43066", - "right": "issue:44779", - "accept": false, - "reason": "Both are tokenizer regressions in v5, but they describe different models and different failure modes; no evidence of the same code-path bug." - }, - { - "left": "issue:38617", - "right": "issue:45278", - "accept": false, - "reason": "45278 is a broad collection of import errors, while 38617 is one specific missing symbol; too general to be the same issue." - }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in threaded inference and dtype mismatch in bfloat16 inference are unrelated bugs." + "reason": "Different failures: VRAM leak in dataloader-threaded generation vs dtype mismatch in bfloat16 inference." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "A GenerationConfig validation error and a RoPE implementation discussion are different problems." + "reason": "Both touch GLM-5, but one is a GenerationConfig validation error and the other is a RoPE implementation discussion." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "ONNX import failure and LayoutLMv3 box-validation messaging are unrelated." + "reason": "Unrelated issues: missing transformers.onnx module vs clearer layoutlmv3 box error messaging." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API guidance and a missing image-patch helper are different subsystems and different bugs." + "reason": "TF32 API warning and missing image-patch method are different code paths and different bugs." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version compatibility and a PyTorch version check bug are not the same underlying issue." + "reason": "Python-version compatibility regression is unrelated to the flex_attention AuxRequest version check." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "Wrong optimizer state handling in FSDP/PEFT and missing GenerationMixin cache in chunked prefill are distinct training/generation bugs." + "reason": "Wrong optimizer state assignment in FSDP2/PEFT is not the same as a missing GenerationMixin cache in chunked prefill." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Example-script multi-label output handling and SigLIP2 HF/JAX mismatch are unrelated." + "reason": "Example-script multi-label output bug and SigLIP2 implementation discrepancy are separate model/task issues." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "VitPose model breakage and Ministral flash-attention failure affect different models and code paths." + "reason": "VitPose model breakage is unrelated to Ministral FlashAttention failures." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "SGD args being ignored and train/evaluate failing after train are different trainer bugs." + "reason": "SGD argument handling and trainer evaluate-after-train failure are different trainer bugs." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "A classification example bug and a pipeline error-message issue are unrelated." + "reason": "Empty multi-label results in an example script is not the same as a pipeline() error-message/doc-support issue." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5, but one is save_pretrained key serialization and the other is from_pretrained loading; not the same defect." + "reason": "Same model family, but one is a save_pretrained key regression and the other is a from_pretrained load error; too different to treat as one bug." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Different Qwen variants and different failure symptoms; too broad to merge as one bug." + "reason": "Qwen3-VL-Moe/Transformers 5.0 issue and Qwen3.5 MoE from_pretrained error are distinct regressions." }, { - "left": "issue:43937", - "right": "issue:45563", + "left": "issue:45083", + "right": "issue:45446", "accept": false, - "reason": "GenerationConfig invalidity and a stale warning about num_return_sequences are separate issues." + "reason": "Unexpected output-length helper behavior is unrelated to the PyTorch version check in flex_attention." }, { - "left": "issue:45083", - "right": "issue:45446", + "left": "issue:43937", + "right": "issue:45563", "accept": false, - "reason": "Feature-extraction length behavior and a version-check import guard are unrelated." + "reason": "GenerationConfig invalidity and a stale num_return_sequences warning are different symptoms and different fixes." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping fallback behavior and a flash-attention crash on s_aux=None are different problems." + "reason": "Tokenizer mapping None and flash_attention s_aux handling are unrelated subsystems." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Both are tokenizer regressions, but one is loading regression and the other is wrong offset mapping; not the same bug." + "reason": "Tokenizer-loading regression and Qwen3 offset_mapping bug are not the same concrete failure." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Different tokenizer models and different symptoms; similar theme is not enough to treat as duplicates." + "reason": "Different models and different tokenizer bugs: offset_mapping vs incorrect Deepseek results in v5." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Documentation lag around pipeline removal and a PyTorch version check bug are unrelated." + "reason": "Docs still referencing pipeline() is unrelated to the AuxRequest version check." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "GenerationConfig validation and flash-attention generation failure are distinct code paths." + "reason": "GLM-5 generation config validation and Qwen3.5 flash-attention generation breakage are different bugs." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "No concrete shared bug is visible; 43906 is only an isolated reproduction of another issue, while 42673 is a VRAM leak report." + "reason": "VRAM leak in threaded dataloader use is unrelated to an isolated reproduction of another issue." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "init_empty_weights argument handling and PreTrainedConfig type-checking are separate compatibility regressions." + "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig type-checking breakage are different initialization/type issues." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Batched video processor shape handling and missing torchvision crash are different video-processing issues." + "reason": "Batched video-processor shape bug is different from missing-torchvision crash on AutoProcessor load." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "PEFT minimum-version bump and an AuxRequest PyTorch version check are unrelated." + "reason": "PEFT minimum-version bump and PyTorch version check are unrelated release/compatibility issues." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Both mention testing_utils, but one is a general bug and the other is a specific xdist race; not the same fix." + "reason": "A vague testing_utils bug and a specific xdist race on captured_info.txt are not clearly the same defect." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Tokenizer mapping assumptions and a Mistral regex patch crash are unrelated tokenizer bugs." + "reason": "Tokenizer mapping fallback and Mistral regex patch crashing on backend_tokenizer are separate tokenizer-load failures." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they concern different missing symbols and likely different causes." + "reason": "Both are import errors, but they involve different missing symbols and different code paths." }, { "left": "issue:39401", "right": "issue:43421", "accept": false, - "reason": "Wrong offset mapping and runtime post-processor updates for special tokens are different tokenizer behaviors." + "reason": "Wrong offset_mapping and runtime post-processor updates are distinct tokenizer behaviors." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "qwen3next missing projections and Qwen3.5 MoE from_pretrained errors are different model-specific bugs." + "reason": "Missing qwen3next projections and Qwen3.5 from_pretrained errors are unrelated bugs." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Degenerate training with flash_attention_3 is unrelated to incorrect Qwen3.5 save_pretrained checkpoints." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "A max_shard_size question and a from_pretrained regression are not the same underlying issue." + "reason": "save_pretrained shard-size question and Qwen3.5 from_pretrained failure are unrelated." } ] }, @@ -12967,7 +12925,6 @@ "issue:44933", "issue:45003", "issue:45071", - "issue:45216", "issue:45278", "issue:45341", "issue:45459", @@ -12977,7 +12934,6 @@ "issue:45656" ], "soft_pairs": [ - "issue:44829|issue:45216", "issue:42371|issue:44261", "issue:42673|issue:43582", "issue:43408|issue:43742", @@ -12987,8 +12943,8 @@ "issue:42490|issue:42673", "issue:43937|issue:44496", "issue:42175|issue:42222", - "issue:44483|issue:45464", "issue:38617|issue:44220", + "issue:44483|issue:45464", "issue:42371|issue:42831", "issue:43066|issue:43122", "issue:44075|issue:45656", @@ -12999,8 +12955,8 @@ "issue:44410|issue:45478", "issue:43097|issue:44493", "issue:44263|issue:44485", - "issue:43012|issue:43408", "issue:41628|issue:45071", + "issue:43012|issue:43408", "issue:42222|issue:45459", "issue:43441|issue:44829", "issue:45071|issue:45542", @@ -13011,232 +12967,226 @@ "issue:43097|issue:43452", "issue:43986|issue:45341" ], - "prepared_review_unit_hash": "5d5b20b2dd186fe1ac3bebc18bd70b3355b8002ca634968dcaad4a0ff423c170", + "prepared_review_unit_hash": "51d59883150a9aeae12a1b118910608d697f7ab75456d02927eadb72d2aac40f", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 37204, - "estimated_input_tokens": 9301, - "estimated_eval_tokens": 18858 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "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 mostly noisy and heterogeneous: the issues cover unrelated Transformers v5 regressions, model-loading failures, tokenizer changes, backend packaging, and optimizer/runtime bugs. I don\u2019t see any soft pair that looks like the same underlying bug or change.", - "confidence": 0.95, - "canonical_issue_reason": "issue:45278 is the broadest representative item, summarizing a general v5 import-regression theme that overlaps with several other reports more than any single narrow bug does.", + "summary": "This cluster is mostly a loose collection of unrelated Transformers issues, with only weak thematic overlap around v5 import/API regressions. None of the soft-pair candidates look like the same concrete bug, so I would not merge any of them as duplicates.", + "confidence": 0.44, + "canonical_issue_reason": "issue:45278 is the broadest umbrella for the import-regression subset and best represents the recurring v5 breakage reports in this set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best cluster representative because it is the most umbrella-like report and closest to the shared import/regression flavor among otherwise distinct issues.", + "best_issue_reason": "issue:45278 is the most general, user-facing report here and the best anchor for duplicate triage among the issue candidates.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44829", - "right": "issue:45216", - "accept": false, - "reason": "FlashAttention-3 training degeneration and a Qwen3.5 save_pretrained checkpoint regression are different bugs with different failure modes." - }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 API guidance and an MLA rms_norm_eps precision issue are unrelated." + "reason": "TF32 API guidance vs MLA layernorm precision bug; different subsystems and failure modes." }, { "left": "issue:42673", "right": "issue:43582", "accept": false, - "reason": "VRAM leakage across dataloader threads is not the same as an Apple Silicon TypeError in caching_allocator_warmup." + "reason": "VRAM leak in multi-threaded Qwen3 loading vs Apple Silicon warmup TypeError; unrelated causes." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "A sam3 model-type warning and a MobileLLM key error are distinct loading problems." + "reason": "SAM3 tracker config warning vs MobileLLM key error; not the same code path." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "A 3d_parallel.py runtime failure and a testing_utils bug do not share the same underlying issue." + "reason": "3d_parallel.py runtime issue vs a testing_utils bug; no overlap in concrete failure." }, { "left": "issue:44933", "right": "issue:45278", "accept": false, - "reason": "The first is a specific missing import from image_utils; the second is a broad v5 import-regression report." + "reason": "Both are import-related, but one is a missing symbol and the other is a broad upgrade regression; not the same bug." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "A broken vitpose model and a missing-TensorFlow backend packaging issue are unrelated." + "reason": "vitpose model breakage vs TensorFlow backend/package-selection error; different areas." }, { "left": "issue:42490", "right": "issue:42673", "accept": false, - "reason": "Local-path import warning behavior and a VRAM leak in dataloader threads are different code paths." + "reason": "Local-path import warning vs VRAM leak under dataloader threads; unrelated." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "Invalid GenerationConfig and unrecognized model/config.json errors are different model-loading failures." + "reason": "GenerationConfig validation failure vs unknown-model loading error; distinct config/loading problems." }, { "left": "issue:42175", "right": "issue:42222", "accept": false, - "reason": "TensorFlow not being included in a torch extra install is unrelated to vitpose model breakage." + "reason": "Backend packaging issue vs vitpose model failure; not the same underlying defect." }, { - "left": "issue:44483", - "right": "issue:45464", + "left": "issue:38617", + "right": "issue:44220", "accept": false, - "reason": "Both mention chat/completions, but one is a general request rejection and the other is a streaming-specific Qwen3.5 failure." + "reason": "Missing import symbol vs audio feature extraction issue; different code paths." }, { - "left": "issue:38617", - "right": "issue:44220", + "left": "issue:44483", + "right": "issue:45464", "accept": false, - "reason": "An import error from configuration_utils and an audio feature-extraction issue are unrelated." + "reason": "Both touch chat APIs, but one is request acceptance and the other streaming inference; not a single concrete bug." }, { "left": "issue:42371", "right": "issue:42831", "accept": false, - "reason": "TF32 API behavior and FineGrainedFP8 accuracy are different precision-related problems." + "reason": "TF32 settings warning vs FineGrainedFP8 accuracy regression; unrelated precision features." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "A wrong tokenizer decoder type and changed tokenization output are related broadly to tokenization, but not the same concrete bug." + "reason": "Tokenizer decoder type mismatch vs tokenization change across versions; related domain, but not the same bug." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "Unused SGD arguments and double optimizer stepping under deepspeed are different optimizer bugs." + "reason": "SGD args ignored vs deepspeed stepping twice; different optimizer bugs." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Generation cache missing during chunked_prefill and a testing_utils bug are unrelated." + "reason": "Generation cache regression vs testing utility bug; no shared failure path." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "A PyTorch warning during bfloat16 compilation and BLIP2 dtype fields staying float32 are different dtype issues." + "reason": "PyTorch bfloat16 compile warning vs Blip2 dtype staying float32; different symptoms and causes." }, { "left": "issue:43874", "right": "issue:44261", "accept": false, - "reason": "A missing image-patch count method and an rms_norm_eps precision mismatch are unrelated multimodal/model bugs." + "reason": "Missing multimodal patch-count method vs q_a_layernorm epsilon precision bug; unrelated." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "A save_pretrained shard-size question and a Qwen3.5 MoE from_pretrained error do not describe the same bug." + "reason": "save_pretrained shard-size question vs Qwen3.5-Moe loading error; not the same issue." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Missing layer projections in qwen3next and a Qwen3.5 MoE loading error are different model-specific failures." + "reason": "Missing projections in qwen3next vs Qwen3.5-Moe from_pretrained error; different model families and failures." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Removed tie_embeddings_and_encoder_decoder and unexpected position-id keys are different regressions." + "reason": "Removed config option vs unexpected key warnings in many models; both version-related, but not the same concrete regression." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "A torch.split return-value issue in GlmMoeDsaIndexer is not the same as a GLM-5 RoPE implementation problem." + "reason": "torch.split return handling bug vs GLM-5 RoPE implementation discussion; not the same path." }, { - "left": "issue:43012", - "right": "issue:43408", + "left": "issue:41628", + "right": "issue:45071", "accept": false, - "reason": "These concern unrelated dtype-warning behavior and sam3 model-type warnings." + "reason": "AutoImageProcessor import error vs PreTrainedConfig type checking regression; different API surfaces." }, { - "left": "issue:41628", - "right": "issue:45071", + "left": "issue:43012", + "right": "issue:43408", "accept": false, - "reason": "An AutoImageProcessor import failure and a PretrainedConfig type-checking regression are separate API breakages." + "reason": "bfloat16 compile warning vs SAM3 model-type warning; unrelated." }, { "left": "issue:42222", "right": "issue:45459", "accept": false, - "reason": "A broken vitpose model and tokenizer error handling when protobuf is absent are unrelated." + "reason": "vitpose model breakage vs protobuf-related tokenizer error handling; different layers." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Ministral FlashAttention failure and FlashAttention-3 training degeneration are not the same concrete bug." + "reason": "FlashAttention inference failure vs FlashAttention-3 training degeneration; same broad feature, but not one concrete bug." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "PretrainedConfig type checking and a missing-TensorFlow backend error are unrelated." + "reason": "Type-checking regression vs undefined TensorFlow backend error; unrelated." }, { "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and qwen3next missing projections are distinct model-loading issues." + "reason": "Qwen3-VL weight-shape mismatch vs qwen3next missing projections; distinct model-loading defects." }, { "left": "issue:43335", "right": "issue:45071", "accept": false, - "reason": "A SwitchTransformersConfig sparse-layer bug and a PretrainedConfig type-check regression are different failures." + "reason": "SwitchTransformers sparse-layer config bug vs config type-checking regression; unrelated." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "Qwen3-VL shape mismatch and Qwen3-VL-Moe bug reports are different model-specific problems." + "reason": "Both involve Qwen3-VL/Transformers 5.0, but one is a specific checkpoint shape mismatch and the other is a generic bug report." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils and a testing_utils bug are unrelated." + "reason": "sys.modules access bug vs testing_utils bug; no shared underlying code path." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "A removed config field and gguf_file breaking from_pretrained are different regressions." + "reason": "Removed tie_embeddings option vs gguf_file from_pretrained breakage; different APIs." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "A missing-torchvision crash in AutoProcessor and a testing_utils bug are unrelated." + "reason": "AutoProcessor video-model crash vs testing_utils bug; unrelated." } ] }, @@ -13321,8 +13271,8 @@ "issue:38617|issue:42757", "issue:44261|issue:44485", "issue:43064|issue:44315", - "issue:44410|issue:45375", "issue:38617|issue:44351", + "issue:44410|issue:45375", "issue:44062|issue:44117", "issue:43097|issue:45588", "issue:44038|issue:44410", @@ -13343,14 +13293,14 @@ "issue:42907|issue:45753", "issue:42907|issue:44655" ], - "prepared_review_unit_hash": "ff9adebd2232d78a24639231f15017a3a66bf272c20ce0d0cd022bb43ea08a10", + "prepared_review_unit_hash": "f53b3a577ca6ffaf1b094da4a506afbd44ae2159029b07cbc6c0e4cfe942f66f", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, @@ -13365,204 +13315,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of unrelated Transformers issues. The soft-similarity pairs generally share a subsystem or symptom class, but not the same concrete bug, so I would not collapse any of them as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "No single true canonical issue stands out because the set is heterogeneous. If one representative is needed, issue 43873 is the strongest anchor: it is open, has substantial discussion, and describes a concrete runtime problem with quantization/offloading.", + "summary": "These issues are mostly unrelated Transformers bug reports spanning imports, model configs, training, and docs. None of the soft pairs look like the same underlying bug/change, so all should be rejected. The most suitable anchor issue is the active quantization/offloading report.", + "confidence": 0.93, + "canonical_issue_reason": "issue:43873 is the most active and broadly scoped report in the set, with the highest discussion/inbound reference signal, so it is the best canonical anchor.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the most suitable representative because it is active, detailed, and has the clearest broad user impact among this mostly unrelated set.", + "best_issue_reason": "issue:43873 stands out as the most representative and triage-worthy issue here because it is open, has substantial discussion, and concerns a broadly impactful runtime behavior.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Both involve loading/tokenizer resolution, but 44117 is a generic mapping bug while 44987 is a model-specific load failure; not the same underlying code path." + "reason": "Both are tokenizer/loading failures, but one is a tokenizer mapping assumption and the other is a specific model-loading regression; not the same bug." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "One is an AutoProcessor/video-model crash without torchvision; the other is about save_pretrained shard sizing. Different features and failure modes." + "reason": "Video AutoProcessor/torchvision crash vs save_pretrained shard-size question; unrelated code paths." }, { "left": "issue:43408", "right": "issue:44368", "accept": false, - "reason": "These concern unrelated model/config warnings in different architectures; they do not describe the same bug." + "reason": "A model-type mismatch warning and a tie_word_embeddings warning are different config-validation issues." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Both are GLM-related, but one is an image-processor multimodal token AttributeError and the other is a torch.split indexing issue in a different component." + "reason": "Missing image-patch helper in GLM46V and torch.split return handling in GlmMoeDsaIndexer are separate model bugs." }, { "left": "issue:43976", "right": "issue:45071", "accept": false, - "reason": "A Python-version compatibility regression is not the same as a PreTrainedConfig type-checking breakage." + "reason": "Both mention version breakage, but one is Python compatibility and the other is PretrainedConfig type checking; different failures." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Both are processor-loading failures, but caused by different missing dependencies/symbols and different model families." + "reason": "Different crashes in different loading stacks: missing torchvision vs missing ReasoningEffort import from mistral_common." }, { "left": "issue:44829", "right": "issue:45245", "accept": false, - "reason": "These are unrelated training/runtime problems: flash_attention_3 degeneracy versus a categorical-cardinality RuntimeError." + "reason": "FlashAttention-3 degenerate training and category-count overflow are unrelated problems." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and a testing_utils bug are unrelated." + "reason": "A Qwen3-MoE sliding_window bug and a testing_utils bug do not share the same underlying cause." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols from different modules; not the same concrete breakage." + "reason": "Two separate ImportErrors from different symbols/modules; same symptom type, but not the same bug." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "One is an eps/config precision mismatch in MLA layernorm; the other is a RoPE implementation question. Different code paths." + "reason": "RMS norm epsilon precision/config issue vs GLM-5 RoPE implementation issue; different model internals." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption is unrelated to Liger Kernel application with model_init." + "reason": "Optimizer-state corruption under FSDP2/PEFT is unrelated to Liger Kernel application with model_init." }, { - "left": "issue:44410", - "right": "issue:45375", + "left": "issue:38617", + "right": "issue:44351", "accept": false, - "reason": "Both are Qwen-family model/config issues, but one is missing attention projections and the other is a dropped vision config field." + "reason": "Different missing imports: configuration_utils.layer_type_validation vs transformers.HybridCache." }, { - "left": "issue:38617", - "right": "issue:44351", + "left": "issue:44410", + "right": "issue:45375", "accept": false, - "reason": "Both are import failures, but they involve different missing exports and different modules." + "reason": "One is missing attention projections in qwen3next; the other is a vision config field being dropped by @strict." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "An AddedToken constructor argument conflict is not the same issue as TOKENIZER_MAPPING_NAMES returning None." + "reason": "AddedToken keyword duplication and TOKENIZER_MAPPING_NAMES returning None are different tokenizer-path failures." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "A removed config API and a flash_attention runtime AttributeError are unrelated." + "reason": "A removed config flag and a flash-attention sink-less crash are unrelated issues." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "These are different Qwen-related bugs; one is a broader v5 compatibility issue, the other is a specific missing-layer-projection defect." + "reason": "Both concern Qwen variants, but one is a broad v5 compatibility issue and the other is a specific missing-projection bug." }, { "left": "issue:44704", "right": "issue:45103", "accept": false, - "reason": "Passing kwargs to cached_file and an auto_docstring annotations crash are separate utility-layer bugs." + "reason": "Passing kwargs to cached_file and auto_docstring crashing on future annotations are separate utility-layer bugs." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "local_files_only network leakage and a missing embed_tokens.weight load path are distinct from each other." + "reason": "local_files_only still making API requests is unrelated to UMT5Encoder missing embed_tokens.weight." }, { "left": "issue:43986", "right": "issue:44117", "accept": false, - "reason": "A video-model processor crash without torchvision and a tokenizer mapping None bug are different subsystems and failures." + "reason": "Video model loading without torchvision is unrelated to tokenizer mapping returning None." }, { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "Removal of a config attribute and quantization/offloading failure are not the same underlying bug." + "reason": "A removed embedding-tie config flag and quantization offloading behavior are different problems." }, { "left": "issue:43824", "right": "issue:43986", "accept": false, - "reason": "One is a missing class export/import error; the other is a processor crash due to missing torchvision. Different root causes." + "reason": "Different loading failures: missing Qwen2_5_VL export vs missing torchvision for a video model." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "Wrong tokenizer decoder type in v5 and a reproduction of another issue are not the same defect." + "reason": "Wrong tokenizer decoder type and an isolated reproduction of another issue are not enough to establish the same bug." }, { "left": "issue:43097", "right": "issue:43901", "accept": false, - "reason": "A config API removal and a docs/behavior mismatch for return_all_scores are unrelated." + "reason": "Config flag removal and outdated pipeline docs about return_all_scores are unrelated." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "These both touch pipeline behavior, but one is an error-message issue and the other is save_pretrained support; not mergeable as one fix." + "reason": "Pipeline task-support messaging and save_pretrained on Pipeline objects are different features and failure modes." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing encoder weights on load and save_pretrained shard sizing are separate save/load concerns." + "reason": "Missing UMT5Encoder weights and the max_shard_size default question do not share an underlying code-path issue." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training blockers and Bamba SDPA/torch.compile failures are different model-specific problems." + "reason": "xLSTM training blockers and Bamba torch.compile/SDPA failure are unrelated model-specific bugs." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "A float32 warning on bfloat16 compilation and a tie_word_embeddings warning are unrelated warnings with different triggers." - }, - { - "left": "issue:43012", - "right": "issue:43550", - "accept": false, - "reason": "These are different failure modes: a precision warning versus a torch.compile backend bug." + "reason": "A bfloat16 compile warning and a tie_word_embeddings warning are different warning sources and behaviors." }, { "left": "issue:42886", "right": "issue:43381", "accept": false, - "reason": "Offline cache loading and gradient checkpointing in eval mode are unrelated." + "reason": "Offline cache loading failure and eval-mode gradient checkpointing incompatibility are separate runtime issues." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "Both mention attention backends, but one is explicit FA2 support gating and the other is GPT-2 attention scaling behavior; not the same bug." + "reason": "Unsupported Flash Attention 2 for Nemotron and GPT2 attention-scaling being ignored are distinct backend/model bugs." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Dequantized model saving and a Qwen3.5 image infinite loop are unrelated." + "reason": "Saving dequantized models and Qwen3_5 infinite-image-loop behavior are unrelated." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Saving dequantized models and saving Pipeline objects are different save_pretrained problems." + "reason": "Both mention saving, but one is dequantized model persistence and the other is Pipeline.save_pretrained; different artifacts and code paths." } ] }, @@ -13591,6 +13535,7 @@ "issue:38617", "issue:42490", "issue:42757", + "issue:42898", "issue:42994", "issue:43012", "issue:43064", @@ -13614,7 +13559,6 @@ "issue:43976", "issue:44038", "issue:44077", - "issue:44188", "issue:44222", "issue:44230", "issue:44242", @@ -13644,279 +13588,272 @@ "issue:43208|issue:43526", "issue:44222|issue:45478", "issue:42490|issue:43012", - "issue:43618|issue:43761", "issue:43012|issue:44683", + "issue:43618|issue:43761", "issue:42757|issue:43824", "issue:43618|issue:44380", "issue:38617|issue:43502", "issue:44483|issue:45290", "issue:44038|issue:44230", "issue:43299|issue:43867", - "issue:44483|issue:45381", "issue:43756|issue:44485", + "issue:44483|issue:45381", "issue:43901|issue:44655", "issue:43299|issue:45753", "issue:36683|issue:43475", "issue:42994|issue:44655", "issue:43352|issue:43937", - "issue:43867|issue:45216", "issue:44077|issue:44380", + "issue:43867|issue:45216", "issue:36010|issue:42757", + "issue:42898|issue:43066", "issue:36683|issue:44077", - "issue:43066|issue:45381", "issue:43611|issue:44655", "issue:43976|issue:44938", "issue:36683|issue:43525", "issue:43352|issue:45593", "issue:44263|issue:45440", "issue:43618|issue:45216", - "issue:36683|issue:44242", - "issue:44188|issue:44380" + "issue:36683|issue:44242" ], - "prepared_review_unit_hash": "b13199a9087816e3b325d254134c5600ef1f368e73461cecb6258910e21ccea5", + "prepared_review_unit_hash": "9cb4dc6e8e88f0eba9c6a4ddc7da89b95297253b4625bbed65bc727b9c6871ec", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 36, - "serialized_chars": 38176, - "estimated_input_tokens": 9544, - "estimated_eval_tokens": 19344 + "soft_pair_count": 35, + "serialized_chars": 37832, + "estimated_input_tokens": 9458, + "estimated_eval_tokens": 19172 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous set of Transformers regression reports, import errors, and model-specific bugs. The items are not strong duplicates overall; they share broad v5-regression language but usually hit different code paths or models.", - "confidence": 0.83, - "canonical_issue_reason": "Issue 43611 is the broadest and most representative regression report in the set: a general Transformers 5.0 model-loading break tied to base_model_prefix handling.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bug reports and regressions. I did not find any soft pair that convincingly describes the same underlying issue or code change, so no merges are warranted from the provided evidence.", + "confidence": 0.34, + "canonical_issue_reason": "issue:43611 is the broadest, most central regression report in the set: a Transformers 5.0 loading breakage affecting models with `base_model_prefix`, which is more representative of the cluster\u2019s general v5-regression theme than the narrower model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43611 is the clearest, most actionable cluster representative because it describes a concrete, user-facing regression with a broad impact across models.", + "best_issue_reason": "issue:43611 is the best single issue to represent this cluster because it is a clear, high-impact version-regression report with wide applicability and strong triage value.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "Different subsystems: attention scaling backend behavior vs sequence-classification loss computation." + "reason": "Different failures: GPT-2 attention scaling under SDPA/FlashAttention vs. degenerate zero loss in sequence classification with `num_labels=1`." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Unrelated problems: FlashAttention support for a model vs Python-version compatibility." + "reason": "One is a model/backend support limitation, the other is a Python version compatibility failure; no shared code-path bug." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Distributed training optimizer-state bug is unrelated to GPT-2 attention scaling." + "reason": "Trainer/FSDP2/PEFT optimizer-state corruption is unrelated to GPT-2 attention scaling behavior." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "Both involve chat/template paths, but one is request acceptance and the other is batched chat-template crashing." + "reason": "An API request validation issue is unrelated to a batched `apply_chat_template` crash in a VL processor." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Different model/config attribute error vs Apple Silicon allocator TypeError." + "reason": "Gemma3Config missing `vocab_size` and Apple Silicon allocator warmup TypeError are distinct bugs in different areas." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "xLSTM training bugs are unrelated to BeitImageProcessorFast label reduction." + "reason": "xLSTM training blockers and BeitImageProcessorFast label reduction are unrelated model/processor bugs." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "Save-pretrained FP8 MoE bug and from-pretrained Qwen3.5 MoE load error are not the same concrete failure." + "reason": "FP8 `save_pretrained` on MoE and Qwen3.5 MoE `from_pretrained` load failure are different directions of failure and not the same concrete bug." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Trailing-slash import warning and bf16 compiler warning are unrelated issues." + "reason": "Local-path import warning and a bfloat16 compilation precision warning are unrelated." }, { - "left": "issue:43618", - "right": "issue:43761", + "left": "issue:43012", + "right": "issue:44683", "accept": false, - "reason": "Both are CLIP regressions, but the titles point to different missing outputs and no shared concrete failure is shown." + "reason": "A PyTorch warning during bfloat16 compilation is not the same as flex_attention failing on torch >= 2.9." }, { - "left": "issue:43012", - "right": "issue:44683", + "left": "issue:43618", + "right": "issue:43761", "accept": false, - "reason": "PyTorch precision warning is unrelated to compiled flex_attention failure on newer torch." + "reason": "Both involve CLIP, but one is missing attentions and the other missing hidden_states; these look like separate output-field regressions." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Different missing imports from different packages and different symbols." + "reason": "Importing `is_offline_mode` from huggingface_hub and importing `Qwen2_5_VLForConditionalGeneration` from transformers are different import-path issues." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "CLIP output-field regression is unrelated to GPT-2 attention scaling config being ignored." + "reason": "CLIP output field assignment and GPT-2 attention scaling under specific backends are unrelated." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "Import error in configuration_utils is unrelated to local_files_only network requests." + "reason": "A missing internal import and unwanted network requests despite `local_files_only=True` are different failure modes." }, { "left": "issue:44483", "right": "issue:45290", "accept": false, - "reason": "Chat-completions request rejection and tool-call template crashing are different code paths." + "reason": "A chat-completions API rejection is unrelated to tool-call assistant message handling in `apply_chat_template`." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Generic Qwen3-VL-Moe breakage and FP8 inference support are not the same bug." + "reason": "Qwen3-VL-Moe loading breakage and FP8 inference support are not the same bug." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Model-loading failure for Qwen3VL MoE differs from a sorted-state_dict load error." + "reason": "Qwen3VL MoE loading regression and a state-dict sorting load error are distinct loading problems." }, { - "left": "issue:44483", - "right": "issue:45381", + "left": "issue:43756", + "right": "issue:44485", "accept": false, - "reason": "API request validation and Qwen2.5-VL video position-id issues are unrelated." + "reason": "Smollm3 RoPE-layer mismatch and GLM-5 RoPE implementation discussion are not the same underlying issue." }, { - "left": "issue:43756", - "right": "issue:44485", + "left": "issue:44483", + "right": "issue:45381", "accept": false, - "reason": "Smollm3 RoPE-layer mismatch is unrelated to GLM-5 RoPE implementation discussion." + "reason": "API request acceptance and Qwen2.5-VL video position-id correctness are unrelated." }, { "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to Pipeline save_pretrained failure." + "reason": "Documentation drift for `return_all_scores` and inability to save Pipeline objects are different issues." }, { "left": "issue:43299", "right": "issue:45753", "accept": false, - "reason": "Qwen3VL MoE loading failure and a specific-image infinite loop are different defects." + "reason": "A Qwen3VL MoE model-loading regression is not the same as an infinite-loop generation bug on a specific image." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Gemma3Config missing vocab_size and SAM3 missing fpn_position_embeddings are distinct model attribute bugs." + "reason": "Gemma3Config missing `vocab_size` and SAM 3 video missing `fpn_position_embeddings` are separate missing-attribute bugs in different models." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Quantized model saving failure is not the same as Pipeline save_pretrained failure." + "reason": "Quantized model saving failure and saving Pipeline objects are different save-path bugs." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "FlashAttention unsupported-model issue is unrelated to GLM-5 generation config validation." + "reason": "FlashAttention 2 unsupported for Nemotron and GLM-5 generation config invalid are unrelated model-specific issues." }, { - "left": "issue:43867", - "right": "issue:45216", + "left": "issue:44077", + "right": "issue:44380", "accept": false, - "reason": "State-dict ordering load error and Qwen3.5 checkpoint save regression are different save/load paths." + "reason": "`post_init` allowance in patchtsmixer is unrelated to GPT-2 attention scaling being ignored." }, { - "left": "issue:44077", - "right": "issue:44380", + "left": "issue:43867", + "right": "issue:45216", "accept": false, - "reason": "post_init contract issue in patchtsmixer is unrelated to GPT-2 attention scaling." + "reason": "State-dict sorting load error and Qwen3.5 checkpoint saving regression are different persistence bugs." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Different missing imports from different modules and symbols." + "reason": "Both are import errors, but for different symbols and different upstream modules; not the same missing export." }, { - "left": "issue:36683", - "right": "issue:44077", + "left": "issue:42898", + "right": "issue:43066", "accept": false, - "reason": "Gemma3 config attribute missing is unrelated to patchtsmixer post_init policy." + "reason": "Tokenizer decoder-type behavior change and wrong tokenizer decoder type are related only superficially; the reported problems and likely fixes differ." }, { - "left": "issue:43066", - "right": "issue:45381", + "left": "issue:36683", + "right": "issue:44077", "accept": false, - "reason": "Tokenizer decoder type bug is unrelated to Qwen2.5-VL vision_position_ids." + "reason": "A config attribute missing on Gemma3 and a `post_init` constraint issue in patchtsmixer are unrelated." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "Base-model-prefix loading regression is unrelated to Pipeline serialization." + "reason": "Model loading via `base_model_prefix` and Pipeline serialization are separate APIs with different breakages." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Different Python compatibility failures affecting different versions and environments." + "reason": "Python version support failure in Transformers 5.1.0 and a Python 3.14 load failure are not the same issue." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Different missing config attributes on different model families." + "reason": "Gemma3Config missing `vocab_size` and Llama4Config missing `pad_token_id` are different missing-config-field regressions." }, { "left": "issue:43352", "right": "issue:45593", "accept": false, - "reason": "FlashAttention support issue is unrelated to D-FINE auxiliary-loss behavior." + "reason": "Unsupported FlashAttention 2 for Nemotron and D-FINE auxiliary-loss omission are unrelated model bugs." }, { "left": "issue:44263", "right": "issue:45440", "accept": false, - "reason": "torch.split indexing behavior is unrelated to DeepSeekV3 implementation divergence." + "reason": "A `torch.split()` return-value issue in GlmMoeDsaIndexer and DeepseekV3MoE divergence from remote implementation do not describe the same bug." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIP output regression is unrelated to Qwen3.5 save_pretrained checkpoint corruption." + "reason": "Missing CLIP attentions and incorrect Qwen3.5 checkpoint saving are unrelated regressions." }, { "left": "issue:36683", "right": "issue:44242", "accept": false, - "reason": "Gemma3 config attribute error is unrelated to MoE load-balancing loss omission." - }, - { - "left": "issue:44188", - "right": "issue:44380", - "accept": false, - "reason": "torch.compile branching divergence is unrelated to GPT-2 attention scaling config handling." + "reason": "A missing config attribute and missing load-balancing loss when `output_router_logits=False` are different defects." } ] }, @@ -13962,11 +13899,11 @@ "issue:44075", "issue:44077", "issue:44112", - "issue:44222", + "issue:44188", "issue:44246", "issue:44355", "issue:44360", - "issue:44387", + "issue:44380", "issue:44484", "issue:44492", "issue:44556", @@ -13976,7 +13913,6 @@ "issue:44964", "issue:44977", "issue:45071", - "issue:45125", "issue:45278", "issue:45310", "issue:45341", @@ -13990,6 +13926,7 @@ "issue:45850" ], "soft_pairs": [ + "issue:44188|issue:44380", "issue:43064|issue:44492", "issue:43352|issue:43931", "issue:43992|issue:45310", @@ -14018,200 +13955,222 @@ "issue:42890|issue:45600", "issue:43502|issue:44062", "issue:43723|issue:44556", - "issue:42890|issue:45341", - "issue:44222|issue:45125", - "issue:44387|issue:44556" + "issue:42890|issue:45341" ], - "prepared_review_unit_hash": "7a999fb156dbf30a17af3bfa72a43e0fc6598f1906ad207ce0301f2c3282bf69", + "prepared_review_unit_hash": "7f99eedff8b3aa07e8bc0f6313560b6bb3f3c152b31f3d0a5e700c78358ac974", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36452, - "estimated_input_tokens": 9113, - "estimated_eval_tokens": 18482 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35639, + "estimated_input_tokens": 8910, + "estimated_eval_tokens": 18076 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of Transformer v5 regression reports, especially model-loading/import/tokenizer failures, with several unrelated training, quantization, and test issues mixed in. The soft similarity links are largely false positives.", - "confidence": 0.71, - "canonical_issue_reason": "issue:45278 is the best umbrella representative: it explicitly captures the broad post-upgrade import/regression theme that most of the cluster superficially resembles, while still being a concrete user report.", + "summary": "The cluster is mostly heterogeneous. One soft duplicate is clear: 45310 and 45478 are the same Qwen3.5 MoE from_pretrained regression. Most other pairs only share broad themes like loading/import failures or training issues, but not the same concrete bug.", + "confidence": 0.78, + "canonical_issue_reason": "issue:45310 is the cleanest anchor for the only clear duplicate pair in this set: it is the earlier, slightly more specific report of the Qwen3.5 MoE from_pretrained failure, and issue:45478 is essentially the same title later on.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most representative issue for this cluster because it is broad, recent, and aligned with the dominant failure mode seen across the grouped issues (v5 upgrade regressions and import/load breakage).", + "best_issue_reason": "issue:45310 is the best overall issue to keep as the representative duplicate target because it is concrete, reproducible, and has an exact duplicate; the rest of the cluster is mostly unrelated or too broad to serve as a better canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44188", + "right": "issue:44380", + "accept": false, + "reason": "Both are attention/backend bugs, but they describe different code paths and failures (compile branching vs SDPA/FlashAttention scaling)." + }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Different problems: distributed training optimizer-state corruption vs a cache-strategy typo." + "reason": "One is a distributed training/optimizer-state bug; the other is a cache-strategy typo. No shared underlying fix." }, { "left": "issue:43352", "right": "issue:43931", "accept": false, - "reason": "Different models and failure modes: Flash Attention support error vs Qwen3-VL weight-shape mismatch." + "reason": "Different model-specific load failures: unsupported Flash Attention vs weight-shape mismatch." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "Both are loading errors, but they affect different models and concrete code paths; too broad to be the same bug." + "reason": "UMT5 encoder missing embed_tokens.weight is unrelated to the Qwen3.5 MoE from_pretrained regression." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "One is about adding @torch.no_grad to cache update methods; the other is a broad import-regression report. Not the same change/bug." + "reason": "Both mention imports/regressions, but one is about decorating cache update with no_grad and the other is broad import errors after upgrade; not the same bug." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "State-dict loading order issue vs missing config field in Qwen3.5 Vision config; distinct root causes." + "reason": "State-dict sorting/load error is not the same as a missing config field silently dropped by strict config handling." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Both mention import-time failures, but one is compiled Python files and the other is a Python 3.13 torch.jit parsing/IndentationError issue." + "reason": "Both are import-time problems, but the root causes differ (compiled Python files vs torch.jit.script parsing on Python 3.13)." }, { "left": "issue:41669", "right": "issue:44246", "accept": false, - "reason": "Both concern import performance, but one is a specific import-* cleanup change and the other is a generic slow-import report; not the same concrete bug." + "reason": "Both concern import performance, but 41669 has a specific import-star root cause while 44246 is only a vague slowdown report." }, { "left": "issue:36683", "right": "issue:43064", "accept": false, - "reason": "Gemma3Config missing vocab_size is unrelated to the FSDP2+PEFT optimizer-state bug." + "reason": "Gemma3Config vocab_size missing and distributed optimizer-state corruption are unrelated." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Test flakiness from missing set_seed() is different from an xdist race on captured_info.txt." + "reason": "A missing set_seed in integration tests is not the same as an xdist race on captured_info.txt." }, { "left": "issue:44938", "right": "issue:45071", "accept": false, - "reason": "Python 3.14 import failure and PretrainedConfig type-checking breakage are different regressions." + "reason": "Python 3.14 import failure and PretrainedConfig type-checking breakage are different compatibility issues." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Different Qwen3.5 issues: generic state_dict loading error vs incorrect visual encoder keys saved by save_pretrained." + "reason": "Load failure with sorted state_dict is not the same as save_pretrained writing incorrect visual encoder keys." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "One is a specific no_grad decoration bug; the other is a vague complaint about many broken models." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "The UMT5 encoder weight-missing issue is unrelated to the Qwen3.5 MoE from_pretrained regression." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache update no_grad and Python-version compatibility for Transformers 5.1.0 are unrelated." + }, + { + "left": "issue:44938", + "right": "issue:45381", + "accept": false, + "reason": "General Python 3.14 loading failure and qwen2.5-vl video position-id bug are different problems." }, { "left": "issue:43618", "right": "issue:44077", "accept": false, - "reason": "CLIPOutput attentions regression and patchtsmixer post_init validation are unrelated." + "reason": "CLIPOutput attentions disappearing and patchtsmixer post_init constraints are unrelated model/API issues." }, { "left": "issue:42490", "right": "issue:44075", "accept": false, - "reason": "Local-files-only path warning and unused SGD args are unrelated." + "reason": "Local-files-only path warning and SGD optimizer args being ignored are unrelated." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Generic tensor-parallel training bug and OOM when loading a specific Qwen2 model are not the same underlying issue." + "reason": "Tensor-parallel training bug and Qwen2-57B load OOM are not the same failure mode." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer/ReLU discussion is unrelated to DeepseekV3MoE implementation divergence." + "reason": "A missing ReLU in the DSA indexer is unrelated to DeepseekV3MoE divergence from remote implementation." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "PyTorch missing dependency error and tokenizer-loading regression are unrelated." + "reason": "PyTorch-missing AutoModel ImportError and tokenizer-loading regression are different issues." }, { "left": "issue:43867", "right": "issue:44484", "accept": false, - "reason": "Model-loading failure vs save_pretrained shard-size question; not the same bug." + "reason": "State-dict sorting/load error is unrelated to max_shard_size being 50GB." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to LwDetrImageLoss AMP/CUDA crash." + "reason": "CLIPOutput attentions and LwDetrImageLoss AMP/CUDA crash are unrelated." }, { "left": "issue:43618", "right": "issue:44977", "accept": false, - "reason": "Different failures: CLIPOutput attentions assignment vs Qwen3.5 flash-attention generation bug." + "reason": "CLIPOutput attentions issue is unrelated to Qwen3.5 flash-attention generation failure." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Test seeding flake is unrelated to Phi-4-multimodal model loading failure." + "reason": "A flaky integration test and a model-loading regression are not the same bug." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "A stale GraniteSpeech CI test and a missing set_seed() issue are different test problems." + "reason": "A test seeding issue is unrelated to a stale device override CI failure." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "Patched testing-utils xdist race and auto_mappings CI breakage are unrelated." + "reason": "A flaky test case and an auto_mappings config reference breakage are unrelated." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "local_files_only network leakage and AddedToken keyword-argument conflict are different tokenizer/download bugs." + "reason": "local_files_only network leakage and AddedToken keyword duplication are unrelated loader/tokenizer bugs." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Tokenizer loading regression in v5 and checkpoint reload incompatibility after version upgrade are related only at a very broad level, but not the same concrete bug." + "reason": "Tokenizer loading in v5 and checkpoint reload failures after upgrade are too broad to be the same concrete bug." }, { "left": "issue:42890", "right": "issue:45341", "accept": false, - "reason": "Missing set_seed() test flakiness is unrelated to the testing_utils typo bug." - }, - { - "left": "issue:44222", - "right": "issue:45125", - "accept": false, - "reason": "FP8 save_pretrained issue and missing _tp_plan for tensor parallelism are different Qwen-related bugs." + "reason": "A test seeding problem and a testing_utils bug are different CI/test issues." }, { - "left": "issue:44387", - "right": "issue:44556", - "accept": false, - "reason": "Int4 quantization memory regression and checkpoint reload failures are unrelated." + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "These are the same Qwen3.5 MoE from_pretrained error with nearly identical titles; 45478 is a later duplicate of the same concrete bug." } ] }, @@ -14265,6 +14224,7 @@ "issue:44351", "issue:44355", "issue:44360", + "issue:44387", "issue:44485", "issue:44492", "issue:44556", @@ -14284,6 +14244,8 @@ "issue:45850" ], "soft_pairs": [ + "issue:44222|issue:45125", + "issue:44387|issue:44556", "issue:42548|issue:44351", "issue:43010|issue:43065", "issue:43502|issue:44485", @@ -14291,11 +14253,11 @@ "issue:45125|issue:45478", "issue:30990|issue:31795", "issue:43010|issue:43232", - "issue:45125|issue:45375", "issue:30333|issue:33453", + "issue:45125|issue:45375", "issue:43299|issue:44556", - "issue:42831|issue:43874", "issue:43502|issue:44222", + "issue:42831|issue:43874", "issue:44315|issue:44360", "issue:43502|issue:44961", "issue:44355|issue:45071", @@ -14306,8 +14268,8 @@ "issue:33453|issue:34689", "issue:42757|issue:43576", "issue:33453|issue:36296", - "issue:43576|issue:43824", "issue:43576|issue:45071", + "issue:43576|issue:43824", "issue:31795|issue:38617", "issue:44961|issue:45468", "issue:41628|issue:42548", @@ -14323,262 +14285,274 @@ "issue:44261|issue:44360", "issue:44623|issue:45230" ], - "prepared_review_unit_hash": "c5e31dea24d7b009784b14d779c4b9571f31b849064f06e35b8ce17b2048351f", + "prepared_review_unit_hash": "ba0b530ff6a1f979552383221100b55dece16a967e5aa249decadb3a1ac55bba", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 38, - "serialized_chars": 37735, - "estimated_input_tokens": 9434, - "estimated_eval_tokens": 19124 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 40, + "serialized_chars": 38925, + "estimated_input_tokens": 9732, + "estimated_eval_tokens": 19720 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly false positives: the set mixes unrelated Transformers bugs, import failures, docs issues, and version-regression reports. No soft pair looks like the same concrete underlying bug or fix.", + "summary": "The cluster is a loose set of Transformers/HF regressions and docs/import issues, not a near-duplicate set. No soft edge looks like the same underlying bug/change, so all candidate pairs should be rejected.", "confidence": 0.93, - "canonical_issue_reason": "issue:45850 is the broadest umbrella ('Since 5.0 version it breaks too many models') and best matches the cluster\u2019s repeated v5 regression theme, though it is very generic.", + "canonical_issue_reason": "Issue 44556 is the best broad representative of the cluster\u2019s main theme: version-upgrade regressions that break model reload/loading across Transformers v5.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 is the most suitable cluster representative because it can absorb many of the v5-breakage complaints; it is still too vague to be an ideal canonical issue.", + "best_issue_reason": "Issue 44556 is the clearest, most generally representative bug report in the set; it captures the repeated v5 loading regression pattern better than the more model-specific or documentation-only issues.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "Both are FP8/tensor-parallel themed, but one is about FP8 save_pretrained while the other is about a missing tensor-parallel plan for Qwen3.5-MoE; different failure modes." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "One reports CUDA reserved-memory OOM under int4 quantization, the other is a reload failure after upgrading versions; not the same bug." + }, { "left": "issue:42548", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different symbols and likely different breakages." + "reason": "Both are import errors, but they involve different symbols and likely different missing exports; too broad to treat as duplicates." }, { "left": "issue:43010", "right": "issue:43065", "accept": false, - "reason": "Different problems: cache update no_grad behavior vs a dummy Conv2d in Sam3PixelDecoder." + "reason": "One is about no_grad decoration for cache/layer update methods, the other is a dummy Conv2d in Sam3PixelDecoder; unrelated code paths." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Unrelated code paths: local_files_only network requests vs GLM-5 RoPE implementation." + "reason": "local_files_only still making API requests is unrelated to a GLM-5 RoPE implementation discussion." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "One is compiled Python file errors; the other is a data filtering slowdown after upgrade." + "reason": "Both mention compiled/v5 behavior, but one is about compiled Python files failing and the other about a Chinese-described slowdown after upgrade; no clear same bug." }, { "left": "issue:45125", "right": "issue:45478", "accept": false, - "reason": "Both mention Qwen3.5 MoE, but one is tensor-parallel planning and the other is from_pretrained loading." + "reason": "Both involve Qwen3.5 MoE, but one is tensor-parallel metadata and the other is a from_pretrained loading error in v5.4.0; not enough evidence of one underlying defect." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Sentence Transformers loading hang is unrelated to documentation confusion about forward arguments." + "reason": "Sentence Transformers loading hangs versus confusing model.forward documentation; one is a runtime bug, the other is docs." }, { "left": "issue:43010", "right": "issue:43232", "accept": false, - "reason": "Cache no_grad decoration and generation kwargs after sync_gpus are different generation bugs." + "reason": "Cache update no_grad issue versus generation kwargs after sync_gpus; different generation internals and symptoms." }, { - "left": "issue:45125", - "right": "issue:45375", + "left": "issue:30333", + "right": "issue:33453", "accept": false, - "reason": "Both are Qwen3.5 MoE-related, but one is tensor parallelism and the other is a missing config field." + "reason": "MLFlow reporting on training failure is unrelated to tokenizer loading regression." }, { - "left": "issue:30333", - "right": "issue:33453", + "left": "issue:45125", + "right": "issue:45375", "accept": false, - "reason": "MLFlow job status reporting and tokenizer loading regression are different failures." + "reason": "Both mention Qwen3.5 MoE, but one is tensor-parallel planning and the other is a missing config field silently dropped by strict validation." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Different version regressions affecting different model-loading paths." + "reason": "Qwen3VL MoE loading breakage in v5.0.0.dev0 is model-specific and separate from the general v4.57-to-v5.3 reload regression." }, { - "left": "issue:42831", - "right": "issue:43874", + "left": "issue:43502", + "right": "issue:44222", "accept": false, - "reason": "FineGrainedFP8 accuracy and missing image-patch counting are unrelated bugs." + "reason": "API requests despite local_files_only=True is unrelated to FP8 save_pretrained behavior." }, { - "left": "issue:43502", - "right": "issue:44222", + "left": "issue:42831", + "right": "issue:43874", "accept": false, - "reason": "Both involve saving/loading behavior, but one is offline API requests and the other is FP8 moe save_pretrained." + "reason": "FineGrainedFP8 accuracy issue and missing get_number_of_image_patches are different FP8/multimodal problems." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "Liger kernel application in model_init is unrelated to a missing ReLU in the DSA indexer." + "reason": "Liger Kernel not applied with model_init is unrelated to the DSA indexer lacking a ReLU." }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "local_files_only networking bug and a standalone 'racoon' issue are not the same problem." + "reason": "local_files_only network leakage and a nonsense-titled issue are unrelated." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled Python file errors are unrelated to PretrainedConfig type checking in v5.4.0." + "reason": "Compiled file errors are not the same as PretrainedConfig type-checking regressions." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLFlow reporting on training failure and tensor parallel training are different subsystems." + "reason": "MLFlow failure reporting and tensor-parallel training bugs affect different subsystems and failure modes." }, { "left": "issue:44355", "right": "issue:44655", "accept": false, - "reason": "Compiled file execution errors are unrelated to Pipeline save_pretrained support." + "reason": "Compiled Python file failures are unrelated to inability to save Pipeline objects." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs still mentioning pipeline() and an unrelated 'racoon' issue share no underlying bug." + "reason": "Docs still referencing pipeline() are unrelated to the unrelated placeholder issue." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "Both are FP8/moe themed, but one is save_pretrained and the other is inference support." + "reason": "FP8 save_pretrained and FP8 inference support are related by theme but not the same concrete bug." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Tokenizer loading regression and Llama 3.2 90B Vision model loading are not the same failure." + "reason": "Tokenizer loading regression and Llama 3.2 90B Vision model loading breakage are different model-loading failures." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Importing is_offline_mode and the transformers env command are different breakages." + "reason": "Missing is_offline_mode import and broken transformers env command are distinct CLI/API regressions." }, { "left": "issue:33453", "right": "issue:36296", "accept": false, - "reason": "Tokenizer loading and tensor parallel training bugs do not share a concrete code path." + "reason": "Tokenizer loading regression and tensor-parallel training bug are different areas and symptoms." }, { "left": "issue:43576", - "right": "issue:43824", + "right": "issue:45071", "accept": false, - "reason": "Broken env command and missing Qwen2.5-VL import are unrelated v5 issues." + "reason": "Broken env command is unrelated to PretrainedConfig type checking." }, { "left": "issue:43576", - "right": "issue:45071", + "right": "issue:43824", "accept": false, - "reason": "CLI env command failure and PretrainedConfig type checking are different features." + "reason": "Broken env command and missing Qwen2_5_VLForConditionalGeneration import are separate issues." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Docs confusion about forward arguments is unrelated to a missing layer_type_validation import." + "reason": "Docs confusion for model.forward is unrelated to the missing layer_type_validation import error." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "A placeholder 'racoon' issue is unrelated to Gemma-4 audio positional encoding." + "reason": "Placeholder/nonsense issue versus Gemma-4 audio positional encoding bug; no duplicate signal." }, { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "AutoImageProcessor import failure and PreTrainedModel import failure are different missing-export issues." + "reason": "AutoImageProcessor import failure and PreTrainedModel import failure are both imports, but not the same missing symbol or path." }, { "left": "issue:38617", "right": "issue:43576", "accept": false, - "reason": "Missing layer_type_validation import and broken env command are different regressions." + "reason": "Missing import from configuration_utils and a broken env CLI command are unrelated." }, { "left": "issue:43576", "right": "issue:44938", "accept": false, - "reason": "CLI env command breakage and Python 3.14 load failure are not the same bug." + "reason": "Broken env command and Python 3.14 load failure are different compatibility problems." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "Missing GenerationMixin cache in chunked_prefill is unrelated to a typo in cache strategies." + "reason": "Generation cache missing in chunked_prefill and a typo in cache strategies are unrelated." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs handling and an unrelated placeholder issue are different problems." + "reason": "Scheduler kwargs being ignored is unrelated to the placeholder issue." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation and an unrelated placeholder issue are not duplicates." + "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor parallel training and TextClassificationPipeline docs are unrelated." + "reason": "Tensor-parallel training bug and docs mentioning return_all_scores are unrelated." }, { "left": "issue:41628", "right": "issue:44355", "accept": false, - "reason": "Missing AutoImageProcessor export and compiled Python file errors are different issues." + "reason": "Importing AutoImageProcessor fails versus compiled Python files erroring; different problems." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor parallel training and BeitImageProcessorFast label reduction are unrelated bugs." + "reason": "Tensor-parallel training bug and BeitImageProcessorFast reduce_labels bug are different subsystems." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "Qwen3.5 chat crash and the broad 'many models break' report are related only by general version complaints, not the same concrete bug." + "reason": "Qwen3.5-35B chat crash and the broad 'too many models break' complaint are not a single concrete bug." }, { "left": "issue:44261", "right": "issue:44360", "accept": false, - "reason": "RMS norm epsilon precision discussion and DSA indexer missing ReLU are different code paths." + "reason": "Layernorm config precision error and DSA indexer ReLU issue are unrelated." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "Missing processor save_pretrained files and a generic bug report are not the same underlying issue." + "reason": "Missing files on processor.save_pretrained and a generic bug report are not the same underlying issue." } ] }, @@ -14655,8 +14629,8 @@ "issue:44273|issue:45542", "issue:44273|issue:45003", "issue:44393|issue:45753", - "issue:44393|issue:45071", "issue:44038|issue:45125", + "issue:44393|issue:45071", "issue:43576|issue:44556", "issue:43525|issue:45850", "issue:41669|issue:43673", @@ -14694,14 +14668,14 @@ "issue:43519|issue:44908", "issue:41669|issue:43704" ], - "prepared_review_unit_hash": "dedf8843e9a9926d998c7c17091d0ffd9ca24dd19469a177b3829b70a447d4e3", + "prepared_review_unit_hash": "d98dfe4a6ac6e40dea166701353f7e49b023fdea0bbce10aff582cf635e5fbea", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 48, @@ -14716,258 +14690,258 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose grab-bag of unrelated Transformers v5 regressions, model-specific bugs, and a few docs/perf complaints. The titles are superficially similar, but the concrete failure modes do not line up closely enough for duplicate merging.", - "confidence": 0.9, - "canonical_issue_reason": "No single issue is a reliable canonical for this cluster; it spans many distinct bugs and regressions rather than one underlying problem.", + "summary": "Mostly false-positive similarity: these issues span unrelated bugs, v5 regression complaints, model-specific errors, docs, and performance/import topics. No soft edge is safe to merge as the same underlying bug/change.", + "confidence": 0.91, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "If forced to pick the closest umbrella, issue:44273 is the broadest open item, but it is still too vague to serve as a true duplicate anchor for the rest.", + "best_issue_reason": "issue:44273 is the closest broad anchor because it is open and about loading/lazy-loading behavior, which overlaps the import-time regressions in the cluster. It is still only an umbrella, not a true duplicate canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Generic lazy-loading complaint vs TensorBoard-without-TF backend error; different subsystems and root causes." + "reason": "Both mention loading/import behavior, but one is lazy loading and the other is a TensorFlow backend detection problem; not the same bug." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Lazy loading failure vs unsafe sys.modules access in modeling_utils; not the same bug." + "reason": "Import slowdown and unsafe sys.modules access are related themes, but they are distinct failure modes with different fixes." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "Qwen3-VL 2D box hallucination/error vs infinite loop on one image; distinct symptoms and likely causes." + "reason": "Both are Qwen3-VL/Qwen3_5 user-facing failures, but one is 2D box output corruption and the other is an infinite-loop/image-specific hang." }, { - "left": "issue:44393", - "right": "issue:45071", + "left": "issue:44038", + "right": "issue:45125", "accept": false, - "reason": "Model output bug vs PretrainedConfig type-checking regression; unrelated code paths." + "reason": "Both involve Qwen3 MoE models, but one is a generic transformers 5.0 bug report and the other is a missing tensor-parallel plan attribute." }, { - "left": "issue:44038", - "right": "issue:45125", + "left": "issue:44393", + "right": "issue:45071", "accept": false, - "reason": "Qwen3-VL-Moe bug vs missing _tp_plan for Qwen3_5Moe tensor parallelism; separate model features." + "reason": "A Qwen3-VL output bug and a PretrainedConfig type-checking regression are unrelated code paths." }, { "left": "issue:43576", "right": "issue:44556", "accept": false, - "reason": "Broken `transformers env` command vs checkpoint reload failure after upgrade; not the same issue." + "reason": "Both are v5 upgrade regressions, but one breaks the env command and the other breaks checkpoint reload; different issues." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "Llama4 pad_token_id AttributeError vs broad complaint that many models break in v5." + "reason": "A missing pad_token_id on Llama4Config is a concrete config bug; the other is only a generic complaint about many models breaking." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Import-* slowdown/explosion vs GenerationMixin cache missing during chunked_prefill; different failure modes." + "reason": "Import-star removal and missing GenerationMixin cache in chunked prefill are different regressions." }, { "left": "issue:44556", "right": "issue:45092", "accept": false, - "reason": "Both upgrade-related, but one is checkpoint reload regression and the other is meta-init incompatibility for old remote-code checkpoints." + "reason": "Both concern v5 compatibility, but one is checkpoint reload after upgrade and the other is remote-code/meta-init incompatibility." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "Cache-strategy typo vs generic v5 breakage report; not the same concrete problem." + "reason": "A cache-strategy typo is not the same underlying problem as a broad v5 breakage complaint." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig vs vague 'many models break' complaint." + "reason": "StableLmConfig missing pad_token_idx is a specific config regression, not the same as a generic 'too many models break' report." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 API-setting warning vs cache-strategy typo; unrelated topics." + "reason": "TF32 API guidance and a cache-strategy typo are unrelated." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "AutoProcessor kwargs not forwarded to cached_file vs inverse_sqrt scheduler kwargs ignored; different components." + "reason": "Processor kwargs forwarding and scheduler kwargs handling affect different subsystems and code paths." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "Missing set_seed in SamHQ tests vs scheduler kwargs bug; no shared underlying defect." + "reason": "Missing set_seed in tests and inverse_sqrt scheduler kwargs are unrelated." }, { "left": "issue:45278", "right": "issue:45850", "accept": false, - "reason": "Many import errors after upgrade vs generic v5 breakage complaint; too broad to be duplicates." + "reason": "A specific import-error report after upgrade is not the same underlying bug as a vague general complaint." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Importing transformers is slow sometimes vs lazy loading not working properly; related theme, but not the same bug." + "reason": "Both involve import behavior, but one is intermittent import slowness and the other is lazy-loading malfunction; not the same concrete defect." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "sam3_video/sam3_tracker model-type warning vs Liger Kernel not applied with model_init; different code paths." + "reason": "Model-type warning for sam3_tracker and Liger Kernel not applied with model_init are unrelated." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python version compatibility issue vs Qwen3-VL 2D bbox output bug; unrelated." + "reason": "Python-version compatibility failure and Qwen3-VL hallucination/output error are unrelated." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash Attention 2 unsupported for Nemotron vs Qwen3-VL hallucination/error; different models and failures." + "reason": "Flash Attention 2 support missing for NemotronH vs a Qwen3-VL output bug are different model-specific problems." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer_class mismatch on save_pretrained vs a vague generic bug report; no shared concrete issue." + "reason": "Tokenizer save metadata mismatch and a generic bug report are not the same issue." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "`transformers env` command broken vs scheduler kwargs ignored; unrelated functionality." + "reason": "An env CLI breakage and a scheduler kwarg bug are different components." }, { "left": "issue:44295", "right": "issue:45230", "accept": false, - "reason": "position_ids buffer AttributeError vs generic bug report; not the same problem." + "reason": "Position_ids buffer access and a generic bug report do not match." }, { "left": "issue:44908", "right": "issue:45071", "accept": false, - "reason": "Scheduler kwarg omission vs PretrainedConfig type checking regression; separate bugs." + "reason": "Scheduler kwargs and PretrainedConfig type checking are unrelated." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-* slowdown vs cannot import HybridCache; both import-related but clearly different causes." + "reason": "Import-star cleanup and missing HybridCache export are different import-surface issues." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Import-* slowdown vs a small testing_utils bug; unrelated." + "reason": "A model import-performance issue and a testing_utils bug are unrelated." }, { "left": "issue:44492", "right": "issue:44945", "accept": false, - "reason": "Cache-strategy typo vs incorrect LLM output under pipeline parallelism; different areas." + "reason": "A cache-strategy typo and incorrect pipeline-parallel output are different bugs." }, { "left": "issue:43874", "right": "issue:45850", "accept": false, - "reason": "GLM46V processor missing method vs generic v5 breakage complaint; not duplicates." + "reason": "A missing multimodal-token helper and a generic v5 complaint are not the same change or bug." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Qwen3-VL bbox issue vs Qwen3.5 Moe from_pretrained error; separate models and failures." + "reason": "Two Qwen3.5/Qwen3-VL reports, but one is bbox output hallucination and the other is a from_pretrained error for Qwen3.5 MoE." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "patchtsmixer post_init policy vs _tied_weight_keys list AttributeError; unrelated model utilities." + "reason": "Optional post_init handling and tied-weight-key AttributeError are different implementation bugs." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Docs still reference pipeline() vs scheduler kwargs ignored; docs issue vs runtime bug." + "reason": "Docs still referencing pipeline() and scheduler kwargs handling are unrelated." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion in model.forward vs tokenizer loading regression; unrelated issue types." + "reason": "Documentation confusion about forward args and tokenizer-loading regression are not the same issue." }, { "left": "issue:41669", "right": "issue:44908", "accept": false, - "reason": "Import-* slowdown vs scheduler kwargs bug; no shared underlying code path." + "reason": "Import-time model loading changes and scheduler kwargs are unrelated." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel application with model_init vs degenerate training under flash_attention_3; different behaviors." + "reason": "Liger Kernel application with model_init and flash_attention_3 training degeneration are different problems." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken `transformers env` command vs lazy loading malfunction; different features." + "reason": "v5 env CLI breakage and lazy-loading malfunction are separate code paths." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Saving Pipeline objects vs tied weight keys AttributeError; unrelated internals." + "reason": "Pipeline save_pretrained failure and tied-weight-key AttributeError are unrelated." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "DSA indexer/ReLU discussion vs Gemma-4 audio positional encoding bug; not the same bug." + "reason": "DSA indexer/ReLU discussion and Gemma4AudioRelPositionalEncoding are completely different areas." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "Generic bug report vs specific Qwen3.5 Moe from_pretrained error; no matching concrete failure." + "reason": "Both are generic bug reports around Qwen3.5 Moe, but the descriptions point to different failure modes and no shared concrete fix path." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading problem vs inverse_sqrt scheduler kwargs bug; separate concerns." + "reason": "Lazy-loading/import behavior and scheduler kwargs are different subsystems." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Qwen3VL timestamp calculation bug vs GLM-5 RoPE implementation discussion; distinct models/issues." + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are separate model math bugs." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Qwen3VL timestamp bug vs scheduler kwargs bug; unrelated." + "reason": "A processor timestamp bug and a scheduler kwargs bug are unrelated." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import-* slowdown vs VRAM leak across dataloader threads; different symptom and subsystem." + "reason": "Import-star removal/performance and VRAM leakage in multi-threaded dataloaders are different issues." } ] }, @@ -15058,208 +15032,208 @@ "issue:31515|issue:33453", "issue:43298|issue:44230" ], - "prepared_review_unit_hash": "98f6e14948d3aa984537c0b0a85be964cf212a28dada1f1bf72c0d6904e73731", + "prepared_review_unit_hash": "9d58d72e9dd4d0bea4f322d9024f6ba304d633bf070d0057703d98c5c7bed327", "component_budget": { "node_count": 402, "item_count": 402, - "soft_pair_count": 1503, - "serialized_chars": 775573, - "estimated_input_tokens": 193894, - "estimated_eval_tokens": 388044 + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 }, "budget": { "node_count": 35, "item_count": 35, "soft_pair_count": 29, - "serialized_chars": 28342, - "estimated_input_tokens": 7086, - "estimated_eval_tokens": 14428 + "serialized_chars": 28339, + "estimated_input_tokens": 7085, + "estimated_eval_tokens": 14426 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous; the soft-similarity links are mostly superficial and none look like true duplicates.", - "confidence": 0.96, - "canonical_issue_reason": "issue:45092 is the clearest representative of the v5/model-loading regression subset, but the overall cluster is too mixed to be treated as one duplicate set.", + "summary": "This cluster is mostly a loose collection of Transformers issues around loading, tokenizer/model regressions, and v5 compatibility. The soft links look like weak topical similarity rather than true duplicates; I would not merge any of the candidate pairs.", + "confidence": 0.78, + "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set and best matches the cluster\u2019s general theme of v5 regressions breaking many models, though it is too vague to serve as a precise duplicate target.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45092 is the most concrete, user-facing regression among the mixed reports, so it is the best single issue to anchor the cluster if one must be chosen.", + "best_issue_reason": "issue:45850 is the most generally representative issue for this cluster because it describes widespread breakage, not a single narrow symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Documentation confusion about forward args vs a specific model-loading regression; different problems." + "reason": "Both involve model/loading docs or regressions, but one is a documentation confusion issue and the other is a specific Llama 3.2 model-loading failure." }, { "left": "issue:31795", "right": "issue:37428", "accept": false, - "reason": "Docs wording issue vs flash-attention import failure; unrelated code paths." + "reason": "Different problems: documentation ambiguity vs an import error for a flash-attention helper." }, { "left": "issue:44315", "right": "issue:45092", "accept": false, - "reason": "Liger Kernel not applied with model_init vs remote-code/meta-init incompatibility; not the same bug." + "reason": "Both touch model initialization, but one is about Liger Kernel not being applied and the other is remote-code checkpoint incompatibility with v5 meta init." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "GLM-5 RoPE implementation vs Gemma4 audio positional encoding; different model families and failures." + "reason": "Both are architecture-specific math/positioning bugs, but they affect different models and different implementations." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM threading vs GLM-5 RoPE math; unrelated." + "reason": "VRAM leakage in multi-threaded inference is unrelated to RoPE implementation issues." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API migration notice vs a memory leak bug; no shared code-path." + "reason": "TF32 API guidance and VRAM leakage are unrelated issues." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "local_files_only still triggers requests vs timestamp calculation in a processor; distinct behaviors." + "reason": "One is unwanted network access during local loading; the other is a timestamp calculation bug in a processor." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp bug vs missing save_pretrained files; different processor issues." + "reason": "Processor timestamp math and missing saved files are different code paths and symptoms." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Processor timestamp calculation vs VRAM leak in causal LM usage; unrelated." + "reason": "A tokenizer/processor timestamp bug is not the same as a dataloader-thread VRAM leak." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return handling in an indexer vs audio positional encoding; different components." + "reason": "Different models and different failure modes: torch.split return handling vs audio positional encoding." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch vs audio positional encoding; unrelated failure modes." + "reason": "CPU offload device mismatch is unrelated to Gemma audio positional encoding." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "Layernorm epsilon precision issue vs audio positional encoding; not the same underlying bug." + "reason": "Precision error from missing rms_norm_eps is not the same as an audio positional encoding bug." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both concern tokenizer changes in v5, but one is general behavioral drift and the other is a decoder-type bug; too broad to merge." + "reason": "Both concern tokenizer behavior in v5, but one is a broad behavior change and the other is a specific wrong decoder type issue." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs bug vs an off-topic/invalid issue; no duplicate relationship." + "reason": "Scheduler kwargs handling is unrelated to the unrelated 'Unemployment' issue." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Tied-weights key crash vs chat template crash with tool calls; different code paths." + "reason": "A tied-weights helper crash and a chat template crash are distinct failures in different utilities." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "Cache-strategy typo vs an unrelated issue; not duplicates." + "reason": "A cache-strategy typo is unrelated to the unrelated 'Unemployment' issue." }, { "left": "issue:44998", "right": "issue:45850", "accept": false, - "reason": "Unrelated issue text vs broad v5 regression complaint; no shared concrete bug." + "reason": "The second is a broad v5 breakage report; the first is not a technical bug report and is clearly unrelated." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression vs unrelated issue; no overlap." + "reason": "CLIPOutput attentions regression is unrelated to the unrelated 'Unemployment' issue." }, { "left": "issue:44998", "right": "issue:45341", "accept": false, - "reason": "Unrelated issue text vs testing_utils bug; different artifacts." + "reason": "A testing_utils bug and an unrelated non-technical issue are not duplicates." }, { "left": "issue:43525", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_id in Llama4Config vs tied-weights key crash; different subsystems." + "reason": "Missing pad_token_id in Llama4Config and a tied-weights list crash are different config/utility bugs." }, { "left": "issue:43572", "right": "issue:44861", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig vs tied-weights key crash; unrelated." + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to _get_tied_weight_keys crashing on list input." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Slow checkpoint loading vs import error for layer_type_validation; different symptoms and code paths." + "reason": "Both are import/loading-related, but one is slow checkpoint loading and the other is a missing import symbol." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "fp8 inference support request vs processor save_pretrained file omission; unrelated." + "reason": "FP8 support for Qwen models and missing processor save files are different feature/bug areas." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLflow status reporting on failed training vs slow from_pretrained loading; different areas." + "reason": "MLflow reporting on failed training jobs is unrelated to slow checkpoint loading." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Slow checkpoint loading vs tensor-parallel training bug; not the same bug." + "reason": "Slow checkpoint loading and tensor-parallel training bugs are different problems." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Loading performance regression vs documentation confusion; unrelated." + "reason": "Loading slowness is not the same as confusing forward-argument documentation." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "fp8 support for qwen3-vl/qwen3.5 moe vs from_pretrained error in transformers>=5.4.0; different problems." + "reason": "Both mention Qwen and v5-era errors, but one is FP8 inference support and the other is a from_pretrained failure." }, { "left": "issue:31515", "right": "issue:33453", "accept": false, - "reason": "Slow checkpoint loading vs tokenizer loading regression; related topic but not the same concrete failure." + "reason": "Slow from_pretrained checkpoint loading and tokenizer-loading regression are related only at a high level, not the same bug." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Continuous batching support for audio models vs fp8 inference support; different feature gaps." + "reason": "Continuous batching support for audio models is a different capability gap from Qwen FP8 inference support." } ] }, @@ -15291,7 +15265,7 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "e88daaae9626e68719478e7d4036d6089bc3864f97b3c13bdf192f369e1f34ee", + "prepared_review_unit_hash": "8a6c846e734d045b44600385112d43b3f53d7b57a0c7532082b0b6f99aff1343", "component_budget": { "node_count": 3, "item_count": 3, @@ -15313,24 +15287,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These three issues are not true duplicates. Two are about model-loading/tokenization behavior in unrelated contexts, but the concrete problems differ: one is Tapas tokenization mismatch, one is a private-model auth/error-message problem, and one is a generic failure to load bert-base-uncased.", - "confidence": 0.85, - "canonical_issue_reason": "Issue #16618 is the broadest and most central model-loading complaint in the cluster, so it is the best single issue to represent the cluster if one must be chosen.", + "summary": "These three issues do not look like duplicates: one is about Tapas tokenization mismatching TensorFlow, one is about an unhelpful auth error when downloading a private model, and one is about failing to load bert-base-uncased. They share only broad Transformers/model-loading context.", + "confidence": 0.89, + "canonical_issue_reason": "issue #16618 is the most concrete and user-facing of the set, with a specific model-load failure tied to bert-base-uncased.", "canonical_pr_reason": null, - "best_issue_reason": "#16618 is the most generally framed and likely to attract related model-loading reports, unlike the more specific Tapas tokenization and private-model login issues.", + "best_issue_reason": "#16618 is the best representative issue because it has the clearest reproducible failure and a specific affected model.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tapas tokenization mismatch is a specific preprocessing/implementation issue, while #16618 is a generic model-loading failure for bert-base-uncased; they do not describe the same bug." + "reason": "Different symptoms and layers: Tapas tokenization parity vs. a model loading failure for bert-base-uncased." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "#15980 is specifically about the error message when downloading a private model while logged out, whereas #16618 concerns failure to load bert-base-uncased; the underlying failures are different." + "reason": "Auth/download error for private models is unrelated to a load failure for a public model." } ] }, @@ -15360,7 +15334,7 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "368ce75a4fa55bb21be0d293db4a24e26bc300daf8b31ce38440df972765776e", + "prepared_review_unit_hash": "13f17d72a3aed6f0289d0cadad0118b386fab32c1f56fe21683d43065ddf5517", "component_budget": { "node_count": 2, "item_count": 2, @@ -15382,18 +15356,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one asks for Flash Attention 2 support across more architectures, the other asks for GGUF support across more architectures. They share a generic framing but target different features/code paths.", - "confidence": 0.97, + "summary": "These two issues are superficially similar \u201csupport for more architectures\u201d requests, but they target different features: Flash Attention 2 vs GGUF. They should not be treated as duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:26350 is the older, more actively discussed request, so it is the better representative issue if one must be chosen, but it is not a duplicate of issue:33260.", + "best_issue_reason": "Neither issue is a duplicate of the other; they request support for different capabilities and should remain separate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Different requested capabilities (Flash Attention 2 vs GGUF) with no indication of the same underlying bug or change." + "reason": "Same broad framing, but different underlying work: Flash Attention 2 support is unrelated to GGUF support. They would not be merged into one issue." } ] }, @@ -15435,12 +15409,12 @@ "issue:43834|issue:44365", "issue:44365|issue:44370" ], - "prepared_review_unit_hash": "d4d14e48950bf859b7e0938466fbc5370276ac7ca0b4183957eb0711a5a7feff", + "prepared_review_unit_hash": "a6c85be2455c5216a65b3503c14ecc827bf8e64f8ae98a75d392a2a78ee8f3e2", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 10, - "serialized_chars": 6441, + "serialized_chars": 6443, "estimated_input_tokens": 1611, "estimated_eval_tokens": 3478 }, @@ -15448,7 +15422,7 @@ "node_count": 5, "item_count": 5, "soft_pair_count": 10, - "serialized_chars": 6441, + "serialized_chars": 6443, "estimated_input_tokens": 1611, "estimated_eval_tokens": 3478 }, @@ -15457,82 +15431,79 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster splits into one near-duplicate trio of generic i18n translation issues and two standalone language-specific requests (Turkish, Spanish). Only the three placeholder-titled issues look like the same underlying issue; the language-specific ones are separate requests.", - "confidence": 0.89, - "canonical_issue_reason": "issue:43834 is the earliest of the near-identical generic translation issues and serves as the cleanest representative of that duplicate set.", + "summary": "These are all documentation-translation requests, but they point to different target languages or separate placeholder instances, so they should not be deduplicated into one underlying issue.", + "confidence": 0.86, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43834 is the best overall issue representative because it is the oldest member of the duplicate trio and matches the shared generic translation request most cleanly.", + "best_issue_reason": "issue:43834 is the cleanest representative of the generic i18n translation template, but the cluster does not contain a true duplicate set because the requests appear language-specific.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Different language targets: Turkish vs generic placeholder translation request. Same broad i18n theme, but not the same concrete issue." + "reason": "Turkish translation request vs a separate placeholder translation issue; different underlying language target." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Turkish docs translation request is distinct from the generic placeholder issue; no evidence they track the same language or task." + "reason": "Turkish translation request vs a separate placeholder translation issue; not the same language or change." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Different language-specific request versus generic placeholder issue; too broad to call duplicates." + "reason": "Specific Turkish request and a generic placeholder translation issue are not the same underlying request." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Turkish and Spanish translations are separate language requests, not the same underlying issue." + "reason": "Turkish vs Spanish docs translation; same broad category but clearly different underlying requests." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish request is distinct from the generic placeholder translation issue; same subsystem, different concrete target." + "reason": "Spanish request vs a separate generic placeholder translation issue; different language target." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Spanish docs translation request does not match the generic placeholder issue at the concrete level." + "reason": "Spanish request vs a separate placeholder translation issue; not the same concrete translation task." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Spanish versus generic placeholder translation request; not the same underlying issue." + "reason": "Spanish request vs a separate placeholder translation issue; different underlying language target." }, { "left": "issue:43834", "right": "issue:44370", - "accept": true, - "reason": "Near-identical titles and extremely high body similarity indicate the same generic translation issue." + "accept": false, + "reason": "Both are template-like i18n issues, but there is no evidence they refer to the same language or exact request." }, { "left": "issue:43834", "right": "issue:44365", - "accept": true, - "reason": "These look like duplicate copies of the same placeholder translation issue with only minor wording/metadata differences." + "accept": false, + "reason": "Near-identical template text is not enough; likely distinct language-specific translation requests." }, { "left": "issue:44365", "right": "issue:44370", - "accept": true, - "reason": "The two issues are effectively identical generic translation requests and plausibly one duplicate set." + "accept": false, + "reason": "Two placeholder translation issues, but not enough evidence they represent the same concrete request." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The clustering is well grounded in the packet: the three placeholder-titled issues have near-identical titles and very high similarity, while the Turkish and Spanish issues are meaningfully different, language-specific requests. The soft-edge verdicts are conservative and consistent with the 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 }, @@ -15602,244 +15573,256 @@ "issue:37015|issue:45027", "issue:43837|issue:45177" ], - "prepared_review_unit_hash": "6aab5a3bbc399cb4ee5ba9918efd769ea68ea55641896e4a3ca26192ce33d491", + "prepared_review_unit_hash": "80b5173ee9ceb62227c9dd5007d0f25731782ad32b1cdd73cca3d95126edbf61", "component_budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 35, - "serialized_chars": 20710, - "estimated_input_tokens": 5178, - "estimated_eval_tokens": 10612 + "serialized_chars": 20708, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 }, "budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 35, - "serialized_chars": 20710, - "estimated_input_tokens": 5178, - "estimated_eval_tokens": 10612 + "serialized_chars": 20708, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are separate model-support requests for different architectures and modalities (e.g., EuroBert, RF-DETR, NeoBERT, MiniCPM3, Qwen3-ASR/TTS, Voxtral, EUPE). The soft links reflect topical similarity only; none look like the same underlying bug or change.", - "confidence": 0.97, - "canonical_issue_reason": "No canonical issue: the items are distinct model requests, not duplicates of one underlying report or implementation.", + "summary": "A loose cluster of model-request issues spanning unrelated models and modalities (text, vision, audio, OCR). They are thematically similar but do not look like duplicate reports of the same underlying request.", + "confidence": 0.94, + "canonical_issue_reason": "issue:43671 is the most active representative of the cluster, but it is still a distinct Qwen3-TTS support request rather than a duplicate of the neighboring ASR/audio/model requests.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good global representative because the cluster spans unrelated model additions with different code paths and model families.", + "best_issue_reason": "issue:43671 is the best single issue to anchor this loose cluster because it is well-discussed and representative of the broader audio-model request subgroup, even though the items are not duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Different model requests: EuroBert config support vs DEIMv2. Same broad task type, not the same change." + "reason": "Different requested models: EuroBert vs DEIMv2; same pattern of model request, not the same change." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert config support and Qwen3-TTS support are unrelated model families and code paths." + "reason": "EuroBert text model request vs Qwen3-TTS audio support; different code paths and artifacts." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert and RF-DETR are distinct models; not the same implementation request." + "reason": "EuroBert config addition is unrelated to RF-DETR model support." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are different vision models, so this is only category-level similarity." + "reason": "RF-DETR and DEIMv2 are distinct model additions with different architectures." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR and LFM2.5 Audio 1.5B are unrelated modalities and changes." + "reason": "RF-DETR detection model vs LFM2.5 Audio support; unrelated requests." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR and a weight-sparse transformer request are different model additions." + "reason": "RF-DETR and the weight-sparse transformer request are different concrete models." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR vs Qwen3-ASR support: different model families and tasks." + "reason": "RF-DETR vision model request is not the same as Qwen3-ASR audio support." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR vs Qwen3-TTS support are distinct requests, not one bug." + "reason": "RF-DETR and Qwen3-TTS target different model families and functionality." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR are unrelated architectures and code paths." + "reason": "NeoBERT and Qwen3-ASR are unrelated model requests." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT support and a sparse-transformer request do not describe the same change." + "reason": "NeoBERT text-model support vs a sparse-transformer request; not the same bug/change." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT and LFM2.5 Audio 1.5B are different model support requests." + "reason": "NeoBERT and LFM2.5 Audio are different models/modalities." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are different model families; no duplicate evidence beyond similarity." + "reason": "RF-DETR and NeoBERT are different model additions." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are separate model additions, not one underlying fix." + "reason": "NeoBERT and ModernVBERT are distinct model requests, not duplicates." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture support and Qwen3-ASR support are different tasks." + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are different concrete requests." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 and the sparse-transformer request target different model architectures." + "reason": "MiniCPM3 and the OpenAI sparse transformer request are unrelated." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are distinct model support requests." + "reason": "MiniCPM3 and ModernVBERT are separate model additions." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 vs Voxtral TTS support are unrelated model families and modalities." + "reason": "MiniCPM3 and Voxtral TTS support are different model families and modalities." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are different model requests, not the same code change." + "reason": "DEIMv2 vision request and Qwen3-TTS audio request are unrelated." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are separate model additions." + "reason": "ModernVBERT and Qwen3-ASR are different support requests." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are unrelated model requests." + "reason": "ModernVBERT is a BERT variant request; Qwen3-TTS is an audio-model request." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT models and BERT-like models with RoPE are only loosely related by BERT naming." + "reason": "ModernVBERT and the generic RoPE BERT-like request are related in theme but not the same underlying issue." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like RoPE support and Voxtral TTS support are different model behaviors." + "reason": "BERT-like models with RoPE vs Voxtral TTS; different model classes and code paths." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are sister models, but they are separate support requests and code paths." + "reason": "Qwen3-TTS and Qwen3-ASR are related family requests but distinct tasks and implementations." }, { "left": "issue:43671", - "right": "issue:45027", + "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and Voxtral TTS are different TTS models with different implementations." + "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are different audio model requests." }, { "left": "issue:43671", - "right": "issue:45177", + "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated modalities and model families." + "reason": "Qwen3-TTS support and the sparse-transformer request are unrelated." }, { "left": "issue:43671", - "right": "issue:44121", + "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS support and sparse-transformer support are unrelated requests." + "reason": "Qwen3-TTS and Voxtral TTS are separate model integrations." }, { "left": "issue:43671", - "right": "issue:43909", + "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are different audio-model additions." + "reason": "Qwen3-TTS support and DeepSeek-OCR2 are different modalities and requests." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are separate audio model requests." + "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are distinct audio model support requests." }, { "left": "issue:43837", - "right": "issue:45322", + "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR and EUPE are different model families and tasks." + "reason": "Qwen3-ASR and the sparse-transformer request are unrelated." }, { "left": "issue:43837", - "right": "issue:44121", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are different audio tasks/models." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are not the same underlying change." + }, + { + "left": "issue:43837", + "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR support and sparse-transformer support are not the same underlying change." + "reason": "Qwen3-ASR and EUPE are distinct model requests." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio 1.5B and SAM3-LiteText are unrelated models/modalities." + "reason": "LFM2.5 Audio 1.5B and SAM3-LiteText are different model additions." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Sparse-transformer support and EUPE support are different model implementations." + "reason": "Sparse transformer request vs EUPE request; no concrete overlap in the underlying model change." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral-4B-TTS are different speech models and fixes." + "reason": "Qwen3-ASR and Voxtral TTS are separate audio-model support requests." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral-4B-TTS target completely different code paths." + "reason": "NeoBERT and Voxtral TTS are unrelated model requests." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR support and DeepSeek-OCR2 support are separate model additions." + "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model families and tasks." } ] }, @@ -15880,12 +15863,12 @@ "issue:43668|issue:44031", "issue:43668|issue:43813" ], - "prepared_review_unit_hash": "6ef6af7df13263c80867dbcaae899bd61e99222a39b689e9a93adce905da9fd6", + "prepared_review_unit_hash": "c731ec4ecbc851041bc7f48f15f2dae9639e74c4fdb30083f95937f2f3ac51ad", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 8, - "serialized_chars": 6294, + "serialized_chars": 6293, "estimated_input_tokens": 1574, "estimated_eval_tokens": 3404 }, @@ -15893,7 +15876,7 @@ "node_count": 6, "item_count": 6, "soft_pair_count": 8, - "serialized_chars": 6294, + "serialized_chars": 6293, "estimated_input_tokens": 1574, "estimated_eval_tokens": 3404 }, @@ -15902,60 +15885,60 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is highly heterogeneous: normalization implementation, Mask2Former loss plumbing, generic kwargs propagation, a config type-hint typo, a quantization typo, and a tokenizer warning regression. The only commonality is loose text similarity, not the same underlying bug.", - "confidence": 0.97, - "canonical_issue_reason": "No clear canonical issue: the items describe distinct bugs across unrelated modules and change types, so there is no single duplicate target.", + "summary": "These issues are only loosely text-similar; they span unrelated areas (normalization behavior, Mask2Former loss plumbing, config typing, quantization typos, and tokenizer warnings) and do not look like duplicates of the same underlying bug.", + "confidence": 0.92, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "If forced to name a representative issue, #43240 is the broadest concrete bug report, but it still does not unify the rest of the cluster.", + "best_issue_reason": "issue:44031 is the broadest user-facing regression in the set (tokenizers warning everywhere after a version bump), so it is the best representative issue, though it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Different problems: a model norm implementation question vs. an incorrect type hint in a config class." + "reason": "Both mention normalization, but one is about Llama4TextL2Norm semantics and the other is a ModernBERTConfig type hint; different code paths and fixes." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Both involve losses, but one is Mask2Former ignore_value handling and the other is generic kwargs plumbing; not the same bug." + "reason": "Mask2Former ignore_value usage and kwargs not being forwarded to loss are different bugs in different parts of the loss pipeline." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Unrelated: Mask2Former loss config vs. tokenizer regex warning regression." + "reason": "Mask2Former loss handling and tokenizer regex warnings are unrelated subsystems with no shared concrete bug." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Mask2Former loss behavior and a quantization typo are different code paths and fix types." + "reason": "A Mask2Former loss/configuration issue is not the same as a quantization typo in a different feature area." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Different components and bug classes: model loss handling vs. config annotation typo." + "reason": "These concern unrelated models and defect types: unused ignore_value versus an incorrect type hint." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Generic loss kwargs bug does not overlap with a ModernBERT config type-hint issue." + "reason": "Passing kwargs to a loss function is a different issue from a config annotation problem." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "Config typing issue and tokenizer warning regression are unrelated." + "reason": "ModernBERT config typing and tokenizer regex warnings are unrelated; no shared underlying change or bug." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "A type-hint correction and a typo in a different symbol are not the same underlying change." + "reason": "Incorrect type hint in a config class and a specific typo in a quantization field are distinct defects." } ] }, @@ -15985,7 +15968,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "b40c999af3fb40266822aff110e744baba9b5d23b3e5e7ef2ac378b888296405", + "prepared_review_unit_hash": "25defb6e84f9477312afafff52ed668cd4994c028a143d65f45ad21450aa6ce2", "component_budget": { "node_count": 2, "item_count": 2, @@ -16007,18 +15990,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are not duplicates: one is a vague request about block decoding, the other is a model-specific feature request for DeepSeek v3.2 support.", + "summary": "The two issues are not duplicates: one requests block decoding behavior, the other requests DeepSeek v3.2 model support. They describe different user-facing changes and different code paths.", "confidence": 0.97, - "canonical_issue_reason": "No canonical duplicate issue can be chosen because the titles and likely scope are unrelated.", + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 43037 is the more concrete and actionable request; 41084 is too broad/vague to serve as a clear representative.", + "best_issue_reason": "Neither issue is a good canonical representative for the other; they are unrelated. If forced, 43037 is more specific, while 41084 is too vague to serve as a cluster anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different underlying concerns: block decoding implementation vs support for a specific model version. They do not appear to describe the same bug or change." + "reason": "Different underlying requests: block decoding configuration vs adding support for a specific model version. No shared bug or change target." } ] }, @@ -16048,7 +16031,7 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "b1eb4e98536b0066491fd13618064228aa8c4526283c94443a20555ff5795d91", + "prepared_review_unit_hash": "5290df8cbd033c3775db66ee7ffa8f9bc4be36475fed9a3c237d30ada5cf0904", "component_budget": { "node_count": 2, "item_count": 2, @@ -16070,18 +16053,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are related only at a very high level (custom Trainer behavior) but they describe different problems and different code paths, so they should not be treated as duplicates.", - "confidence": 0.89, - "canonical_issue_reason": "Issue 43598 is the more concrete Trainer customization request; issue 43600 is a separate FLOPS-estimation concern.", + "summary": "These two closed issues are related only at the broad Trainer subsystem level, but they describe different problems: one is about `remove_unused_column` behavior/customizability, the other about FLOPS estimation for custom trainers. They should not be deduplicated.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #43598 is the more concrete, narrowly scoped Trainer behavior change; #43600 is a separate FLOPS-accounting concern and not the same underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "43598 is the better representative issue because it is specific and directly about a single Trainer behavior change, while 43600 is a distinct performance/accounting topic.", + "best_issue_reason": "#43598 is the better representative issue because it has a specific actionable target (`remove_unused_column`) and a clearer expected fix path.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Reject: the titles and likely fixes target different behaviors in Trainer\u2014unused-column handling vs FLOPS estimation for custom trainers\u2014so they are not the same underlying bug." + "reason": "Different code paths and outcomes: one concerns column pruning customization in Trainer, the other concerns FLOPS estimation for custom trainers. They share the Trainer area but not the same underlying bug/change." } ] }, @@ -16111,12 +16094,12 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "f799de03a450a9b17304c04cb1762353586a62cfa6b66ed255fddf6a58fb3778", + "prepared_review_unit_hash": "3066966d528d6f060e5d7a3f64cdc0801355f436c79c94fc5d92707c4eca7a61", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1612, + "serialized_chars": 1611, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -16124,7 +16107,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1612, + "serialized_chars": 1611, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -16133,18 +16116,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a very broad AutoModel-support level, but they request different capabilities for different model/config combinations. They should not be treated as duplicates.", - "confidence": 0.94, - "canonical_issue_reason": "Issue 43684 is the broader AutoModel registration request and is the better representative of the cluster\u2019s auto-class support theme.", + "summary": "These are related only at a very broad Transformers/model-registration level, but they are not the same issue: one asks for Qwen3-Omni auto-class registration, the other asks for sequence-classification support for Qwen3.5.", + "confidence": 0.95, + "canonical_issue_reason": "No canonical duplicate: issue 43684 is a broad AutoModel registration request for Qwen3-Omni, while 44405 is a separate task-specific request for AutoModelForSequenceClassification support on Qwen3.5.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43684 is the more general feature request; 44405 is a separate AutoModelForSequenceClassification request for a different config.", + "best_issue_reason": "43684 is the broader, more central issue if a single issue must represent the topic, but it should not subsume 44405 because the model family and requested capability differ.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying changes: Qwen3-Omni registration for AutoModel/AutoModelForConditionalGeneration vs Qwen3.5 sequence-classification support. They touch different model families and different auto heads." + "reason": "Reject: different underlying requests. 43684 is about adding Qwen3-Omni registration to auto classes; 44405 is about enabling sequence classification for Qwen3.5. They target different model/config paths and are not the same bug/change." } ] }, @@ -16181,80 +16164,80 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "4528debffbb114be8ae3eb249f5137032d095023a38c451210b571731d45a982", + "prepared_review_unit_hash": "6cd681caaebece3f205458182afd319b3f6f08aeacde9d178eeed33d0e704de9", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4263, - "estimated_input_tokens": 1066, - "estimated_eval_tokens": 2388 + "serialized_chars": 4267, + "estimated_input_tokens": 1067, + "estimated_eval_tokens": 2390 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4263, - "estimated_input_tokens": 1066, - "estimated_eval_tokens": 2388 + "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 appear to be near-identical duplicates of the vague \"huggingface \" report; the \"Fett\" issue looks unrelated. No PRs are present.", - "confidence": 0.76, - "canonical_issue_reason": "Issue 43845 is the earliest of the three identical \"huggingface \" issues and is the safest representative for the duplicate set.", + "summary": "This cluster contains one clearly different issue ('Fett') and three issues with the same 'huggingface' title. The available evidence is limited to title similarity, timestamps, and body-length differences, so the huggingface items are best treated as related-looking but not proven duplicates. I\u2019d keep 43845 as the reference issue only because it is the earliest of that trio.", + "confidence": 0.61, + "canonical_issue_reason": "issue:43845 is the earliest of the three huggingface-titled issues and is the most reasonable anchor for that small cluster, but the evidence here is not strong enough to prove the others are duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43845 is the best single issue to keep because it matches the repeated title, is earliest, and has no sign of being distinct from the other two huggingface reports.", + "best_issue_reason": "issue:43845 is the best overall issue candidate because it predates the other huggingface reports and appears to be the most representative of that subgroup.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Different titles and no evidence they describe the same problem; \"Fett\" does not align with the huggingface reports." + "reason": "Different titles and no body text or other evidence tying them to the same underlying issue." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Different titles and no shared concrete bug details; likely unrelated." + "reason": "Different titles and no concrete evidence beyond a weak soft similarity signal." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Different titles and no shared concrete bug details; likely unrelated." + "reason": "Different titles; the packet does not show enough to conclude duplication." }, { "left": "issue:43845", "right": "issue:43846", - "accept": true, - "reason": "Same title and no distinguishing details; they look like duplicate reports of the same underlying issue." + "accept": false, + "reason": "Same title and close timestamps, but the packet only shows similarity signals and body-length differences, not enough to confirm the same issue." }, { "left": "issue:43845", "right": "issue:43859", - "accept": true, - "reason": "Same title and no distinguishing details; they look like duplicate reports of the same underlying issue." + "accept": false, + "reason": "Same title, but only weak observable evidence is available here; body text is missing, so duplication is not established." }, { "left": "issue:43846", "right": "issue:43859", - "accept": true, - "reason": "Same title and no distinguishing details; they look like duplicate reports of the same underlying issue." + "accept": false, + "reason": "Same title alone is insufficient; the packet does not provide enough detail to accept this as a duplicate." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: three issues share the same title, 43845 is the earliest of that group, and 43835 has a clearly different title. The soft-edge calls are conservative enough overall given the limited evidence; no PR-related claim is made." + "accept": false, + "feedback": "The summary is mostly grounded, but the soft-edge decisions are too strong for the packet. The data only shows title similarity, close timestamps for 43845/43846, and body-length differences; it does not show the actual body text or enough evidence to conclude the three 'huggingface' issues are duplicates with high confidence. In particular, the claims about 'placeholder content' and 'same vague subject matter' overstate what is observable here. A more conservative verdict would keep these as unaccepted soft matches." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -16288,88 +16271,88 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "576b3721d664a7b07420969dc1e0a2395b30b70e5c1b78cfb051ff44f9b92488", + "prepared_review_unit_hash": "70723412500245f4618c7719cc67f9f9cffd1ca94d517cf1dbf068c1ce5b36a9", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5791, - "estimated_input_tokens": 1448, - "estimated_eval_tokens": 3152 + "serialized_chars": 5794, + "estimated_input_tokens": 1449, + "estimated_eval_tokens": 3154 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5791, - "estimated_input_tokens": 1448, - "estimated_eval_tokens": 3152 + "serialized_chars": 5794, + "estimated_input_tokens": 1449, + "estimated_eval_tokens": 3154 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These issues do not look like duplicates of a single underlying bug or change. Most are spam-like placeholder issues, while one is a vague documentation suggestion; there is no concrete shared problem to merge.", + "summary": "This cluster is mostly low-signal spam-like issues plus one generic documentation suggestion; there is no clear underlying bug/change shared across them.", "confidence": 0.96, - "canonical_issue_reason": "issue:44418 is the only non-spam, somewhat descriptive issue, so it is the best available anchor, though it is still too vague to represent a real duplicate cluster.", + "canonical_issue_reason": "No canonical issue: the items are unrelated and too vague/spammy to represent one duplicated problem.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44418 is the most substantive and least spammy item in the set; the others are generic placeholders with no actionable content.", + "best_issue_reason": "issue:44418 is the least spam-like and most issue-like title, but it is still far too generic to be a strong canonical target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are spam placeholders with no concrete bug or change; not the same underlying issue." + "reason": "Both are just '' placeholders; no concrete shared bug or change." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both are spam placeholders with no actionable description; no duplicate signal beyond generic similarity." + "reason": "Both are just '' placeholders; no concrete shared bug or change." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam and the other is a vague documentation suggestion; they do not describe the same problem." + "reason": "Spam placeholder versus generic documentation request; not the same underlying issue." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus a specific titled issue; no shared underlying bug or request." + "reason": "Spam placeholder versus another spam-like title; no shared concrete problem." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "A generic documentation suggestion is not the same as a spam placeholder." + "reason": "Generic docs suggestion versus spam placeholder; not a duplicate pair." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "A generic documentation suggestion is not the same as a spam placeholder." + "reason": "Generic docs suggestion versus spam placeholder; not a duplicate pair." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are spam placeholders, but there is no evidence they refer to the same concrete issue." + "reason": "Identical spam-like titles, but there is no real underlying bug/change to unify." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus a specific titled issue; no duplicate relationship." + "reason": "Spam placeholder versus unrelated spam-like title; no concrete shared issue." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus a specific titled issue; no duplicate relationship." + "reason": "Spam placeholder versus unrelated spam-like title; no concrete shared issue." } ] }, @@ -16399,7 +16382,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "a38699bc6b3ca7b72b39d510cc1cbc9bd205066dcdf10244bef420481c0b7222", + "prepared_review_unit_hash": "5258bfbf88b33c7d05662c4a9169cb35009b0e5a59469f8e7b8b242bbb938e4e", "component_budget": { "node_count": 2, "item_count": 2, @@ -16421,28 +16404,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues have identical titles and very similar scope, so they look closely related, but the evidence is only soft similarity with no explicit links or shared concrete targets/files. This supports a likely-duplicate reading, not a definitive duplicate conclusion.", - "confidence": 0.73, - "canonical_issue_reason": "If one must be kept as the representative report, #45145 is the better canonical issue because it is earlier and has the same apparent scope as #45151, but the duplication signal is not definitive.", + "summary": "The two closed issues have an exact title match and very similar body sizes, so they are likely the same request, but the bodies themselves are not visible here, so this should be treated as a probable duplicate rather than a certain one.", + "confidence": 0.88, + "canonical_issue_reason": "issue #45145 is the better canonical thread because it was created first and already has an inbound reference, making it the more established issue.", "canonical_pr_reason": null, - "best_issue_reason": "#45145 is the best issue representative because it is earlier and matches the same title/scope, though the evidence only supports a likely duplicate rather than a confirmed one.", + "best_issue_reason": "issue #45145 is the best representative issue because it is earlier, has slightly more reference activity, and appears to be the original thread for the same topic.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": false, - "reason": "The titles match and the topics are highly similar, but there is no explicit linkage or concrete shared implementation target; with only soft-similarity evidence and deterministic false, this is not a confirmed duplicate pair." + "accept": true, + "reason": "Exact title match plus very similar body length strongly suggests the same underlying issue, but without the bodies this should be read as a likely/probable duplicate rather than a definite one." } ] }, "evaluator_result": { "accept": false, - "feedback": "The packet supports a strong similarity signal, but not a definitive duplicate conclusion. The pair only has soft-similarity evidence, and the deterministic verdict is false. Marking it as a \"clear duplicate pair\" and setting confidence to 0.98 overstates the evidence; a more conservative verdict would be \"likely duplicate\" or \"possibly duplicate.\"" + "feedback": "The packet supports strong similarity and an exact title match, but it does not expose the issue bodies, so claiming \u201cstrongly overlapping content\u201d or a definite duplicate is a bit too strong. The soft-edge verdict should stay conservative (e.g. likely duplicate / probable same issue), and the confidence should be lower than 0.99 given the limited evidence." }, "evaluator_used": true, "retried": true, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -16519,12 +16502,12 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "b74f884d6ff229fec405c7e7cfe2e7e16172a6f73047dd5b682aec67dc47df9d", + "prepared_review_unit_hash": "5d2a879d76386fd537b9b0f047977585d1167b218b6f1c41c429fa3fad4f1674", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1668, + "serialized_chars": 1667, "estimated_input_tokens": 417, "estimated_eval_tokens": 1090 }, @@ -16532,7 +16515,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1668, + "serialized_chars": 1667, "estimated_input_tokens": 417, "estimated_eval_tokens": 1090 }, @@ -16541,18 +16524,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a very broad level (4-bit inference), but they describe different bugs with different triggers and failure modes, so they should not be merged as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "No canonical duplicate here: #45672 is about a Gemma4 audio-module uint8/torch.finfo TypeError in NF4 inference, while #45674 is about llm_int8_skip_modules unintentionally removing the default lm_head exclusion and causing an AssertionError.", + "summary": "These are two separate 4-bit inference bugs: one in Gemma4/audio modules causing a uint8 `torch.finfo()` TypeError, and one in `BitsAndBytesConfig` where `llm_int8_skip_modules` drops the default `lm_head` exclusion and triggers an AssertionError. They share only a broad theme, not the same underlying failure.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 45674 is the better canonical issue for the cluster because it states a clearer, more general regression in 4-bit inference configuration and has a concrete, actionable failure mode. Issue 45672 is a separate Gemma4/audio-module-specific type error.", "canonical_pr_reason": null, - "best_issue_reason": "#45674 is the cleaner standalone report: it names the exact BitsAndBytesConfig behavior, the affected default exclusion, and the concrete assertion failure path.", + "best_issue_reason": "45674 is the strongest representative issue: it is broader, more self-contained, and describes a configuration bug with an explicit causal chain and reproducible assertion failure.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Reject: both concern 4-bit inference, but the root causes differ. One is a Gemma4/audio-module uint8 torch.finfo TypeError; the other is a BitsAndBytesConfig skip-modules regression that clears the default lm_head exclusion and trips an AssertionError." + "reason": "Both are 4-bit inference problems, but the code paths and failures differ: Gemma4/audio `torch.finfo()` on uint8 weights versus `BitsAndBytesConfig` skip-module handling and `lm_head` exclusion. They do not look like the same bug." } ] }, @@ -16582,7 +16565,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "f05fd103dff83b5b4af396ecb5a9f99c9a17f1bbf180d709a52b9d9bd3c43426", + "prepared_review_unit_hash": "ecea8af619346c989f5b900b9772e6679e44788b4c0501647c8dfaca0dd8fb7b", "component_budget": { "node_count": 2, "item_count": 2, @@ -16604,18 +16587,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two closed issues are just short name/date entries with no bug report content; they do not describe the same underlying problem and should not be deduplicated.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "The two closed issues appear unrelated and spam-like, each containing only a different person\u2019s name and date of birth. They do not describe the same bug or request, so they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the items are unrelated and lack any shared problem statement or technical context.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong canonical candidate because neither contains actionable problem details; if one must be chosen, the earlier issue is only a weak placeholder.", + "best_issue_reason": "Neither issue is a good canonical choice; both look like standalone, non-technical/spam entries rather than duplicate reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Both are bare personal-name/date posts with no shared technical problem, code path, or issue context; not a duplicate bug report." + "reason": "Different names and dates of birth with no shared bug, feature, or code context; not the same underlying issue." } ] }, @@ -16645,40 +16628,40 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "9a7bb77aface30f009b30197d707ed45f241792493ebd0cd28df5af10811b0dc", + "prepared_review_unit_hash": "9e00e8d85a689e610ece8633b33399b2de1d6ef9148503043f8a5ef9a91619c7", "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": "The two closed issues appear to be unrelated placeholder/spam entries with different personal names and no shared technical content, so they should not be clustered as duplicates.", + "summary": "The two closed issues are body-less and their titles are just different personal-name/date strings; they do not appear to describe the same bug or request, so this is not a duplicate cluster.", "confidence": 0.98, - "canonical_issue_reason": null, + "canonical_issue_reason": "No true canonical issue: both entries are content-free and unrelated, so there is no shared underlying problem to canonicalize.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "Issue 45764 is only marginally cleaner as a representative because its title is slightly more formatted, but it is not a meaningful duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "Different names, different dates, and no body or issue-link evidence indicate they are not the same underlying issue." + "reason": "Reject: the titles are different personal-name/date strings with no body, code path, or problem statement indicating the same underlying issue." } ] }, @@ -16718,7 +16701,7 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "a206431ebe894adc7e441b69c1b009883a5b9bbb7b373d5a95cd70aa518c1f83", + "prepared_review_unit_hash": "b3add61107fbb46a35dc68775a64dd282d1d15f7503d0c9760d7ed0c2c855311", "component_budget": { "node_count": 5, "item_count": 5, @@ -16740,66 +16723,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster splits into two separate exact-duplicate pairs: 45870/45871 and 45872/45873. Issue 45869 is only loosely finance-related and should not be merged with either pair.", - "confidence": 0.96, - "canonical_issue_reason": "45870 is the earliest copy in the exact-duplicate pair 45870/45871 and is a reasonable representative for that subcluster.", + "summary": "The set contains three unrelated issue topics: one dataset listing, one duplicate example request pair, and one duplicate scraping-service request pair. Only the exact-topic pairs should be treated as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits the whole set because the items split into separate topics; if forced to pick a representative, issue 45872 best captures the concrete feature request and has an exact duplicate in 45873.", "canonical_pr_reason": null, - "best_issue_reason": "45870 is the cleanest representative of one real duplicate pair, but the full set is not one single issue family; 45872 is the representative of a separate duplicate pair.", + "best_issue_reason": "Issue 45872 is the strongest representative issue overall: it is specific, actionable, and its duplicate 45873 is clearly the same request. The other issues are either a different topic or a less distinctive example request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Different artifacts: a dataset listing vs an example issue. Same finance theme, but not the same underlying request/bug." + "reason": "Different topics: a dataset bundle listing versus an example request; they are not the same underlying issue." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Different artifacts: a dataset listing vs an example issue. Only broad topic overlap, not a duplicate." + "reason": "Different topics: dataset listing versus example request; no evidence they describe the same bug or change." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Same title and matching metadata/body length; these look like exact duplicates of the same example issue." + "reason": "Same title and apparent same example-request content; these look like exact duplicates." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both are DataForge/finance-related, but one is an example prediction issue and the other is a scraping-service issue; not the same underlying request." + "reason": "Both mention financial time series, but one is an example request and the other adds a scraping service; different underlying requests." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Same as the 45870/45872 pair: related topic, but different concrete issue content." + "reason": "Example-request issue versus scraping-service request; not the same concrete change." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Different concrete requests; shared finance/DataForge wording is too broad to call them duplicates." + "reason": "Same as above: example request and scraping-service request are distinct issues." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Different concrete requests; only broad subsystem/topic overlap." + "reason": "Same as above: not the same underlying request." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Same title and matching metadata/body length; these look like exact duplicates of the same scraping-service issue." + "reason": "Same title and same feature request wording; these look like exact duplicates." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 are the only clearly tight pairs (identical titles and body lengths), while 45869 is only loosely related by theme and should not be merged. The soft-edge verdicts are conservative overall, and there is no PR-mergeability issue here." + "feedback": "Grounded and conservative overall. The packet supports two exact duplicate pairs (45870/45871 and 45872/45873) plus the distinct dataset-listing issue 45869. The soft-edge verdicts are appropriately cautious, and the summary does not overclaim beyond the titles/body metadata provided." }, "evaluator_used": true, "retried": false, @@ -16810,7 +16793,7 @@ { "label": "pull_request", "component_index": 1, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -16818,7 +16801,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-39895-15", + "cluster_id": "cluster-39895-14", "nodes": [ "pull_request:39895", "pull_request:41855", @@ -16833,7 +16816,6 @@ "pull_request:45213", "pull_request:45655", "pull_request:45795", - "pull_request:45919", "pull_request:45922" ], "soft_pairs": [ @@ -16843,14 +16825,13 @@ "pull_request:42668|pull_request:43823", "pull_request:43251|pull_request:43823", "pull_request:43251|pull_request:44827", - "pull_request:43424|pull_request:45919", "pull_request:43424|pull_request:44827", "pull_request:43424|pull_request:43823", "pull_request:43823|pull_request:45655", "pull_request:43823|pull_request:45922", "pull_request:43823|pull_request:43995", - "pull_request:43995|pull_request:45213", "pull_request:43995|pull_request:44827", + "pull_request:43995|pull_request:45213", "pull_request:44827|pull_request:45170", "pull_request:44827|pull_request:45795", "pull_request:44827|pull_request:45655", @@ -16860,175 +16841,175 @@ "pull_request:44827|pull_request:45922", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "811fad3f2ed86f4f1666fd01ae6514a9c274f9716c0cfa68ca0afe1da94176ca", + "prepared_review_unit_hash": "01eef6c2206db6eadceab27e75fd7f54e2654f196a3ced41e1047993a61e9a64", "component_budget": { - "node_count": 15, - "item_count": 15, - "soft_pair_count": 22, - "serialized_chars": 42216, - "estimated_input_tokens": 10554, - "estimated_eval_tokens": 21364 + "node_count": 14, + "item_count": 14, + "soft_pair_count": 21, + "serialized_chars": 39196, + "estimated_input_tokens": 9799, + "estimated_eval_tokens": 19854 }, "budget": { - "node_count": 15, - "item_count": 15, - "soft_pair_count": 22, - "serialized_chars": 42216, - "estimated_input_tokens": 10554, - "estimated_eval_tokens": 21364 + "node_count": 14, + "item_count": 14, + "soft_pair_count": 21, + "serialized_chars": 39196, + "estimated_input_tokens": 9799, + "estimated_eval_tokens": 19854 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly unrelated PRs spanning model additions, docs, refactors, and separate bugfixes. The only true duplicate-like pair is the two `fixed_cross_entropy` PRs, which implement the same `nn.functional.cross_entropy` kwargs passthrough in `loss_utils.py`.", - "confidence": 0.97, + "summary": "Most PRs are unrelated model/docs/tooling changes. The only accepted duplicate pair is the two `fixed_cross_entropy` patches, which both target the same `loss_utils.py` kwargs-support bug and issue 43240.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43251 is the best canonical PR for the loss fix: it directly targets issue #43240, changes the same code path as 43254, and is the more explicit implementation of the cross-entropy kwargs passthrough.", + "canonical_pr_reason": "PR 43251 is a solid canonical representative for the `fixed_cross_entropy` duplicate pair: it updates `loss_utils.py` to forward supported `cross_entropy` kwargs and explicitly targets issue 43240.", "best_issue_reason": null, - "best_pr_reason": "PR 43251 is the best overall representative of the only duplicated change in this cluster because it is issue-linked and fixes the exact same `fixed_cross_entropy` behavior as 43254.", + "best_pr_reason": "PR 43251 is the strongest overall representative in this packet because it is the clearest example of the accepted `fixed_cross_entropy` fix cluster and has the more developed patch/review trail of the two near-duplicates.", "soft_edge_verdicts": [ { "left": "pull_request:39895", "right": "pull_request:43424", "accept": false, - "reason": "Unrelated PRs: VideoPrism model addition vs Executorch dynamic-shape test." + "reason": "VideoPrism model addition vs Executorch dynamic-shape export test; unrelated changes." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Different changes: Mistral tokenizer methods vs adding a new MobileLLM model." + "reason": "Mistral tokenizer helper methods vs adding a new MobileLLM model; different workstreams." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Tokenizer API fix and Executorch export test are unrelated." + "reason": "Tokenizer API fix vs Executorch export test; not the same underlying bug." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor auto-loading robustness vs new MobileLLM model; different code paths." + "reason": "Processor robustness changes vs new MobileLLM model implementation; distinct scopes." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Loss utility kwargs passthrough is unrelated to adding MobileLLM." + "reason": "Cross-entropy helper bugfix vs new MobileLLM model addition; unrelated." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "Both touch model/runtime internals but fix different problems: cross-entropy kwargs vs Mistral4 tests/model plumbing." - }, - { - "left": "pull_request:43424", - "right": "pull_request:45919", - "accept": false, - "reason": "Executorch export testing and Sapiens2 model addition are unrelated." + "reason": "`fixed_cross_entropy` kwargs fix and Mistral4 model/test changes address different code paths." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Dynamic export test vs Mistral4 test/model fix; not the same bug or change." + "reason": "Executorch export testing is unrelated to Mistral4 fixes." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Executorch test PR and MobileLLM model addition do not address the same issue." + "reason": "Dynamic-shape export test vs new model addition; no common defect." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "New MobileLLM model and config-resolution fix are unrelated." + "reason": "MobileLLM model code vs config-resolution ordering fix; different problems." }, { "left": "pull_request:43823", "right": "pull_request:45922", "accept": false, - "reason": "MobileLLM model addition and vision-model memory leak fix are unrelated." + "reason": "New model addition vs vision-model lru memory leak fix; unrelated." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "Model addition vs Falcon refactor; different concrete changes." + "reason": "MobileLLM model addition and Falcon output-interface refactor are different changes." }, { "left": "pull_request:43995", - "right": "pull_request:45213", + "right": "pull_request:44827", "accept": false, - "reason": "Falcon output-interface refactor is unrelated to the model-creation skill PR." + "reason": "Falcon output-collection refactor vs Mistral4 fixes; not the same bug or patch." }, { "left": "pull_request:43995", - "right": "pull_request:44827", + "right": "pull_request:45213", "accept": false, - "reason": "Falcon refactor and Mistral4 fixes target different models and issues." + "reason": "Falcon refactor is unrelated to the model-creation skill PR." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test fix and `layrnorm` rename are unrelated bugfixes." + "reason": "Mistral4 test/model fixes are unrelated to the `layrnorm` -> `layernorm` rename cleanup." }, { "left": "pull_request:44827", "right": "pull_request:45795", "accept": false, - "reason": "Mistral4 fixes are unrelated to docs for audio/video processors." + "reason": "Mistral4 changes and audio/video processor docs do not address the same problem." }, { "left": "pull_request:44827", "right": "pull_request:45655", "accept": false, - "reason": "Mistral4 test/model work and config resolution ordering fix are different changes." + "reason": "Mistral4 fixes are unrelated to the config-resolution ordering change." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 changes are unrelated to the model-creation skill PR." + "reason": "Mistral4 test/model work does not match the model-creation skill PR." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45922", + "accept": false, + "reason": "Mistral4 fixes and BEiT lru/memory-leak fix are unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs patch `loss_utils.py` for the same `fixed_cross_entropy` kwargs-support bug and both target issue 43240." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "New model implementation and the model-creation skill are not the same underlying change." + "reason": "New MobileLLM model implementation is unrelated to the model-creation skill repository changes." }, { "left": "pull_request:43823", "right": "pull_request:45795", "accept": false, - "reason": "MobileLLM model addition and docs-only processor guide are unrelated." + "reason": "MobileLLM model code vs docs for audio/video processors; not the same underlying change." }, { "left": "pull_request:44827", "right": "pull_request:45922", "accept": false, - "reason": "Mistral4 test fix and vision-model memory leak fix are unrelated." - }, - { - "left": "pull_request:43251", - "right": "pull_request:43254", - "accept": true, - "reason": "Both PRs fix the same `fixed_cross_entropy` code path by adding supported kwargs (`weight`/`label_smoothing`) to `nn.functional.cross_entropy`; they are direct alternate implementations of the same bugfix." + "reason": "Mistral4 test/model fixes and BEiT cache/memory-leak fix are unrelated issues." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet: the non-loss PRs are supported as unrelated by titles, filenames, and diffs, and the 43251/43254 pair is a clear same-file, same-issue, same-bugfix duplicate. The soft-edge verdicts stay conservative rather than over-connecting the unrelated model/addition/docs/refactor PRs." + "accept": false, + "feedback": "Reject: the output overstates evidence. `pull_request:43251` is not the only item with an explicit issue link; `pull_request:43254` also targets issue 43240. The duplicate-pair judgment for 43251/43254 is fine and conservative, but the representative/canonical rationale should avoid that claim and avoid implying validation/completeness beyond the packet." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -17036,7 +17017,7 @@ { "label": "pull_request", "component_index": 2, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17052,7 +17033,7 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "e32fe2d0efa89b942eb87e67005ea7f74b9bb1f98c136655465dee25f75975c0", + "prepared_review_unit_hash": "bf294f34623077bc9928368be50e94ba6f523a33509ac9fc336b370e643f949f", "component_budget": { "node_count": 2, "item_count": 2, @@ -17074,18 +17055,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related VibeVoice PRs, but they are not the same concrete change: one implements the base VibeVoice integration, while the other adds VibeVoice Realtime plus the acoustic tokenizer and generation tweaks. Shared filenames alone are not enough to merge them as duplicates.", - "confidence": 0.93, + "summary": "Both PRs introduce VibeVoice support, but they are not duplicates: #40546 is the original VibeVoice implementation, while #42781 adds a realtime variant plus the acoustic tokenizer and related generation changes. They overlap on shared model plumbing, but they do not fix the same concrete code-path problem.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 40546 is the better canonical PR because it represents the core VibeVoice implementation and broader initial integration, while PR 42781 is a later, narrower follow-up for realtime/acoustic-tokenizer additions.", + "canonical_pr_reason": "#40546 is the better canonical PR for the cluster because it is the core VibeVoice implementation and already has substantial review activity; #42781 is a later, draft extension focused on realtime support rather than the same base change.", "best_issue_reason": null, - "best_pr_reason": "PR 40546 is the best single representative PR for the cluster since it covers the main VibeVoice feature landing with substantial review activity and a cleaner, more general scope than the draft realtime follow-up.", + "best_pr_reason": "#40546 is the best representative PR for VibeVoice because it contains the primary model bring-up, docs, auto-registry wiring, and conversion/modeling pieces that define the feature.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "Different feature scopes: 40546 implements VibeVoice, while 42781 adds VibeVoice Realtime and acoustic-tokenizer-related changes. They overlap on model scaffolding, but not the same underlying code-path change, so they should not be merged as duplicates." + "reason": "Shared filenames indicate common VibeVoice infrastructure, but #42781 is not the same change: it adds realtime and acoustic-tokenizer work on top of the base implementation instead of duplicating the exact code-path fix in #40546." } ] }, @@ -17099,7 +17080,7 @@ { "label": "pull_request", "component_index": 3, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17115,12 +17096,12 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "29d1cdf2d71b399711c80a999569144467bf584170a2b6fdfe84696a407dca35", + "prepared_review_unit_hash": "d39afd3876776a5ba5206f21e5011ec5ba54b47e91f9af9d9e18d6bbc62bd975", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4674, + "serialized_chars": 4675, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -17128,7 +17109,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4674, + "serialized_chars": 4675, "estimated_input_tokens": 1169, "estimated_eval_tokens": 2594 }, @@ -17137,18 +17118,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests are unrelated feature additions in different areas: TimesFM covariate forecasting vs DINOv3 image classification support. They do not look like duplicate fixes or mergeable alternatives to the same change.", + "summary": "The two PRs are unrelated feature additions in different subsystems: TimesFM covariate forecasting vs. DINOv3 image classification support. They should not be clustered as duplicates or merged into one change.", "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the PRs target different model families and different code paths, so there is no duplicate relationship to collapse.", + "canonical_pr_reason": "No canonical PR: these PRs address different models and different code paths, so there is no shared underlying change to deduplicate around.", "best_issue_reason": null, - "best_pr_reason": "No single best PR for deduplication: neither PR is a substitute for the other, and both are independent feature work.", + "best_pr_reason": "Neither PR is a good global representative for the cluster because the cluster is mixed and the PRs are about unrelated features rather than the same fix.", "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Reject: the titles, touched files, and diff previews show different features in different subsystems, not the same underlying bug or change." + "reason": "Reject: TimesFM covariate forecasting and DINOv3 image classification support are unrelated model features, with different files, APIs, and code paths." } ] }, @@ -17162,7 +17143,7 @@ { "label": "pull_request", "component_index": 4, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17225,7 +17206,7 @@ { "label": "pull_request", "component_index": 5, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17243,46 +17224,46 @@ "pull_request:41797|pull_request:45186", "pull_request:45186|pull_request:45630" ], - "prepared_review_unit_hash": "a170705dd49da2d94868b0d1a40cf84380c28e0848cbf1e1960162ab75289711", + "prepared_review_unit_hash": "4ab6116155ca22b10c5467ed2536705a1cc299cfec60c7bdba04f5a145669b32", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 10409, - "estimated_input_tokens": 2603, - "estimated_eval_tokens": 5462 + "serialized_chars": 10407, + "estimated_input_tokens": 2602, + "estimated_eval_tokens": 5460 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 10409, - "estimated_input_tokens": 2603, - "estimated_eval_tokens": 5462 + "serialized_chars": 10407, + "estimated_input_tokens": 2602, + "estimated_eval_tokens": 5460 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition PRs (DeepSeek OCR, Isaac, Kimi2-6) that share boilerplate integration files but target different models and code paths. They are not duplicates.", - "confidence": 0.98, + "summary": "All three items are separate model-addition PRs (DeepSeek OCR, Isaac, Kimi2-6) that share Transformers scaffolding but target different models and code paths. They are not duplicates of one another.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45630 is the most substantial/established of the three and the best representative of the cluster, but it is not a duplicate of the others because it implements Kimi2-6-specific model, processor, and auto-mapping support.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR #45630 is the strongest standalone PR in the set: it has the highest review activity and a broad, model-specific integration. Still, it should not be merged with the other two because each PR introduces a different model family.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:41797", "right": "pull_request:45186", "accept": false, - "reason": "Both are model-addition PRs and overlap on generic Transformers integration files, but DeepSeek OCR and Isaac are different models with different architectures and documentation/tests; they are not the same underlying change." + "reason": "Both add different models and implementations; overlap is only in shared registration/docs infrastructure, not the same underlying feature or bug fix." }, { "left": "pull_request:45186", "right": "pull_request:45630", "accept": false, - "reason": "Isaac and Kimi2-6 are distinct new model implementations. The shared auto-registration/docs patterns are boilerplate, not evidence of a single fix or mergeable duplicate PR." + "reason": "These are distinct new-model integrations for different architectures. Shared files are generic auto-mapping/documentation hooks, which is not enough to treat them as one change." } ] }, @@ -17296,7 +17277,7 @@ { "label": "pull_request", "component_index": 6, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17312,12 +17293,12 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "faede77314437db271c11fac60eab3b66c4fa5b38624ff1f8fe524a7de36017b", + "prepared_review_unit_hash": "5e5eef70db3b291541671002a8a63c7f3a40a1be97d264052b272a3f1ecbfe06", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5934, + "serialized_chars": 5935, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -17325,7 +17306,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5934, + "serialized_chars": 5935, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -17334,18 +17315,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are broad GitHub Actions dependency bumps, but they target different actions and different failure modes: one is for GitHub Actions runtime/node24 compatibility, the other is for docker/build-push-action version updates. They are related in theme but not the same underlying change.", + "summary": "These are related maintenance PRs for GitHub Actions upgrades, but they target different actions and different concrete workflow changes, so they are not duplicates.", "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 42975 is the broader and more representative workflow-maintenance change, updating multiple core GitHub Actions used across many workflows for Node 24 compatibility.", + "canonical_pr_reason": "PR #42975 is the better representative because it is a broader, clearly scoped workflow-action compatibility upgrade centered on Node 24 support.", "best_issue_reason": null, - "best_pr_reason": "PR 42975 is the best single representative of the cluster because it spans more workflows and the change is a clearer platform-wide maintenance update than the narrower Docker action bump in 42976.", + "best_pr_reason": "PR #42975 is the strongest cluster representative: it has the clearest overarching purpose and the wider set of workflow action updates, while #42976 is a separate action-version bump for Docker build steps.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Reject: these are separate dependency upgrades affecting different actions and code paths. 42975 updates checkout/cache/upload-artifact for workflow runtime compatibility, while 42976 only bumps docker/build-push-action in container-image workflows; they could not plausibly be merged as one duplicate PR." + "reason": "Both are GitHub Actions dependency bumps, but they update different action families and fix different compatibility paths (checkout/cache/upload-artifact vs docker/build-push-action). They are related maintenance changes, not the same underlying bug or mergeable single PR." } ] }, @@ -17359,7 +17340,7 @@ { "label": "pull_request", "component_index": 7, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17388,88 +17369,88 @@ "pull_request:44601|pull_request:45838", "pull_request:45490|pull_request:45838" ], - "prepared_review_unit_hash": "919fa37cb19082776953bd7f18e1e3ac6405962eec55d3e1bdf5b61281402bd9", + "prepared_review_unit_hash": "84d19cbe1bd8ce14c0de1c125c10059fd4052729f4a20a02b66f571241a74ea9", "component_budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 9, - "serialized_chars": 23542, - "estimated_input_tokens": 5886, - "estimated_eval_tokens": 12028 + "serialized_chars": 23536, + "estimated_input_tokens": 5884, + "estimated_eval_tokens": 12024 }, "budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 9, - "serialized_chars": 23542, - "estimated_input_tokens": 5886, - "estimated_eval_tokens": 12028 + "serialized_chars": 23536, + "estimated_input_tokens": 5884, + "estimated_eval_tokens": 12024 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is a set of unrelated pull requests: several distinct model-addition PRs, one distributed pipeline-parallel infrastructure PR, and one PP-OCRv6 text-recognition PR. They share superficial patterns like docs/toctree and auto-registry edits, but not the same concrete code-path or bug fix.", + "summary": "No true duplicate cluster here: the PRs are separate model additions or one distributed-infra change, and the shared files are mostly generic docs/auto-registry plumbing. All soft edges should be rejected.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the cluster does not contain duplicate changes. The PRs target different models or a separate distributed loading feature, so they should remain separate.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 43451 is the strongest standalone representative by scope and review activity, but it is still not a duplicate of the others; it is a distinct Molmo2 model addition with broad model/processor/test coverage.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo. Shared docs/auto-registry edits are incidental, not the same underlying change." + "reason": "Different models (ViT NEPA vs Molmo). Similarity is only from shared docs and auto-registry edits, not the same code path." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo2. No shared concrete bug or code path." + "reason": "Different model families (ViT NEPA vs Molmo2). Shared auto/model registry churn is superficial." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different model additions: ViT NEPA vs CTSM. Similar scaffolding only, not duplicate work." + "reason": "ViT NEPA model add vs CTSM model add; only broad Transformers registry/docs overlap." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are related by naming, but they are distinct model integrations with different files and coverage, not one mergeable fix." + "reason": "Molmo vs Molmo2 are related names but distinct model integrations with different files, configs, and processors." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo vs CTSM are separate model additions with different implementations and targets." + "reason": "Molmo and CTSM are unrelated model additions; overlap is limited to generic model-registration files." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 and CTSM are different model integrations; no evidence of the same bug or feature." + "reason": "Molmo2 vs CTSM are different model implementations and docs, not the same underlying change." }, { "left": "pull_request:43973", "right": "pull_request:45838", "accept": false, - "reason": "Lfm2 Audio support and PP-OCRv6 text recognition are unrelated model additions, despite both touching docs and auto mappings." + "reason": "Lfm2 audio model add vs PP-OCRv6 text-recognition support; different modalities and code paths." }, { "left": "pull_request:44601", "right": "pull_request:45838", "accept": false, - "reason": "Pipeline-parallel loading infrastructure is unrelated to PP-OCRv6 model support; they fix different layers and cannot plausibly be merged as one PR." + "reason": "Pipeline-parallel loading infrastructure is not the same as OCR model support; no concrete shared bug/change." }, { "left": "pull_request:45490", "right": "pull_request:45838", "accept": false, - "reason": "CTSM model support and PP-OCRv6 text recognition are distinct model additions with different code paths and tests." + "reason": "CTSM model add and PP-OCRv6 model support are separate model integrations with only generic registry/docs overlap." } ] }, @@ -17483,7 +17464,7 @@ { "label": "pull_request", "component_index": 8, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17491,85 +17472,63 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-43291-4", + "cluster_id": "cluster-43291-3", "nodes": [ "pull_request:43291", "pull_request:43665", - "pull_request:44815", - "pull_request:45981" + "pull_request:44815" ], "soft_pairs": [ "pull_request:43291|pull_request:43665", "pull_request:43291|pull_request:44815", - "pull_request:43291|pull_request:45981", - "pull_request:43665|pull_request:44815", - "pull_request:43665|pull_request:45981", - "pull_request:44815|pull_request:45981" + "pull_request:43665|pull_request:44815" ], - "prepared_review_unit_hash": "1a167d54829c67ead3fec374a62639f9a65109221ac405b4288b5814acbd5689", + "prepared_review_unit_hash": "8e8f13cc6741964f0695d9d77355e69b4b3f063f1463c46f9011e61b9e5f0238", "component_budget": { - "node_count": 4, - "item_count": 4, - "soft_pair_count": 6, - "serialized_chars": 9639, - "estimated_input_tokens": 2410, - "estimated_eval_tokens": 5076 + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7091, + "estimated_input_tokens": 1773, + "estimated_eval_tokens": 3802 }, "budget": { - "node_count": 4, - "item_count": 4, - "soft_pair_count": 6, - "serialized_chars": 9639, - "estimated_input_tokens": 2410, - "estimated_eval_tokens": 5076 + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7091, + "estimated_input_tokens": 1773, + "estimated_eval_tokens": 3802 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is not a duplicate set: it mixes unrelated PRs across Whisper tokenization, CLIP/ViT test edits, FP8 dequant/loading logic, and a ColQwen2 test fix. The only substantial code-path fix is the FP8 dequant PR; the others are test-only or model-specific changes with no shared underlying bug.", - "confidence": 0.92, + "summary": "The three PRs are not duplicates: one fixes Whisper tokenizer decoding, one is a CLIP/ViT test-only change, and one is an FP8/dequantization fix across loading and quantization code. The shared similarity is only superficial/title-level.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44815 is the strongest representative because it contains a real implementation fix in core loading / FP8 integration, while the other PRs are narrow test tweaks or different model areas.", + "canonical_pr_reason": "PR 44815 is the strongest representative because it makes a concrete runtime bugfix in loading/FP8 code, whereas 43291 is Whisper-specific and 43665 is just test-file churn with no shared underlying change.", "best_issue_reason": null, - "best_pr_reason": "PR 44815 is the best standalone PR in the set; it addresses a concrete runtime bug with actual code changes and is more substantive than the test-only PRs.", + "best_pr_reason": "PR 44815 is the best standalone PR in the set: it has the clearest technical scope and the most substantive code-path fix.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Both are test-oriented, but they target different model stacks and different issues: Whisper tokenizer decode behavior vs. CLIP/ViT test edits. No shared underlying bug or mergeable single fix." + "reason": "Whisper tokenizer decoding fix vs CLIP/ViT test edits; different models, different code paths, no shared underlying bug." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper tokenization/test changes are unrelated to the FP8 dequant/core loading fix. Different subsystems and different code paths." - }, - { - "left": "pull_request:43291", - "right": "pull_request:45981", - "accept": false, - "reason": "Whisper decode logic and ColQwen2 test-device placement are unrelated model-specific changes." + "reason": "Whisper timestamp/batch decode changes are unrelated to FP8 dequantization/loading fixes." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "PR 43665 is effectively test noise in CLIP/ViT, while PR 44815 fixes FP8 quantization/loading behavior. They do not fix the same concrete problem." - }, - { - "left": "pull_request:43665", - "right": "pull_request:45981", - "accept": false, - "reason": "These are separate test changes for different models (CLIP/ViT vs. ColQwen2), not the same bug or change." - }, - { - "left": "pull_request:44815", - "right": "pull_request:45981", - "accept": false, - "reason": "FP8 dequant/core-loading logic is unrelated to a ColQwen2 test tensor device fix; different code paths and different failure modes." + "reason": "Test-file edits for vision models are unrelated to model loading and FP8 quantization changes." } ] }, @@ -17583,7 +17542,7 @@ { "label": "pull_request", "component_index": 9, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17599,12 +17558,12 @@ "soft_pairs": [ "pull_request:43488|pull_request:43532" ], - "prepared_review_unit_hash": "c028e0b0e92be0ba3c9dbd914d97cc8daa7406f0910c9f5e2e26c59557ba26fb", + "prepared_review_unit_hash": "a7fd27b2d7792d89eba54cd608c6c695574b148dada4ef3eef33af1a260b3f1a", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4563, + "serialized_chars": 4562, "estimated_input_tokens": 1141, "estimated_eval_tokens": 2538 }, @@ -17612,7 +17571,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4563, + "serialized_chars": 4562, "estimated_input_tokens": 1141, "estimated_eval_tokens": 2538 }, @@ -17621,18 +17580,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are not duplicates: one is a synthetic repo-bot/check PR around setup/versioning and a Vit test, while the other is a workflow/infrastructure PR spanning GitHub Actions, testing utilities, and notifications. The shared Vit test filename appears incidental.", - "confidence": 0.96, + "summary": "These are two unrelated synthetic PRs: one checks repo-bot/version formatting in setup.py, the other alters CI/workflow and test utilities. They only share a Vit test filename, not the same underlying change or bug.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43532 is the better canonical representative because it is a coherent, substantive change across workflow and test infrastructure, whereas #43488 looks like a small synthetic/bot-validation PR.", + "canonical_pr_reason": "PR #43532 is the more substantial and representative change in the cluster, with multiple workflow and test utility files touched; #43488 is a narrower bot/formatting check.", "best_issue_reason": null, - "best_pr_reason": "PR #43532 is the strongest stand-alone PR in the cluster: it has a broader, internally consistent diff and appears to target a real codebase change rather than a placeholder check.", + "best_pr_reason": "PR #43532 is the best cluster representative, but it is still a non-mergeable synthetic PR rather than a real duplicate target.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "Different underlying changes: #43488 edits setup/versioning and a Vit test for a repo-bot check, while #43532 changes CI workflows plus testing/notification code. Shared test filename is not enough to indicate the same fix." + "reason": "Different intents and code paths: #43488 is a setup/version-format repo-bot check, while #43532 rewires CI/workflow logic and test helpers. Shared Vit test file is incidental." } ] }, @@ -17646,7 +17605,7 @@ { "label": "pull_request", "component_index": 10, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17662,7 +17621,7 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "3f7c96661bd42bb001690a584d06a538a127d50c5aa66fc74af77e85275c54a3", + "prepared_review_unit_hash": "a2aa1081108b6c8030fc2aa455ae0a498969ab14120c7c5646c39ff0e7f037ff", "component_budget": { "node_count": 2, "item_count": 2, @@ -17684,18 +17643,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related by theme but are not duplicates: they add two different promptable segmentation pipelines with different targets, code paths, and model/test coverage.", - "confidence": 0.95, + "summary": "These are two similar-looking but distinct PRs: one adds a Promptable Concept Segmentation pipeline, the other adds a Promptable Visual Segmentation pipeline. They share scaffolding and docs updates, but they target different pipelines/models and are not duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR for duplicate triage; 43612 and 43613 implement different features (concept segmentation vs visual segmentation) rather than the same change.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "If a representative PR is needed, 43613 is the broader one, but it still is not a substitute for 43612 because it covers a different pipeline and different underlying models.", + "best_pr_reason": "Neither PR is a duplicate of the other; both are standalone feature additions for different segmentation tasks. If forced to pick a representative, 43613 is slightly broader, but it should not be treated as canonical for this cluster.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Reject: despite overlapping docs and shared plumbing, these are distinct feature PRs for different pipelines and different underlying code/model paths, so they could not plausibly be merged into one duplicate change." + "reason": "Same general segmentation theme, but different concrete features: concept segmentation vs visual segmentation, with different pipeline names, task docs, and model/test coverage. They are not the same code-path problem and would not plausibly merge into one PR." } ] }, @@ -17709,7 +17668,7 @@ { "label": "pull_request", "component_index": 11, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17725,7 +17684,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "3c4c88639e1f40c8087035c24fdebabbae14b7758ca831ddf0898ef85c19eeae", + "prepared_review_unit_hash": "127267880010f0dd1d488cbb18d2bb69cd34a7e5dfeefb1f80aa9c6dc54b282e", "component_budget": { "node_count": 2, "item_count": 2, @@ -17747,18 +17706,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two distinct trainer.py PRs: one adds a hook for gradient-accumulation loss scaling, the other adds a hook for signature-column filtering. They share a file and theme, but not the same underlying bug or change.", - "confidence": 0.96, + "summary": "Two open PRs touch the same file but implement unrelated Trainer extensibility changes: one adjusts gradient-accumulation loss scaling, the other refactors signature-column filtering for custom trainers. They are not duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43651 is the clearer representative of a concrete trainer behavior change, with a focused code-path update in training_step and a new hook for loss scaling.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 43651 is the strongest standalone PR in the cluster because it addresses a specific training-time bug/behavior and is easier to reason about as a representative change.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Both touch transformers/trainer.py, but they fix different trainer extension points: loss scaling during gradient accumulation vs. signature-column filtering for model inputs. They are not the same concrete bug or mergeable duplicate change." + "reason": "Different code paths and goals: 43651 adds _loss_is_scaled_for_ga to control loss scaling during training, while 43663 adds _get_signature_columns to customize input-column filtering. Sharing trainer.py and a broad subsystem is not enough to treat them as the same change." } ] }, @@ -17772,7 +17731,7 @@ { "label": "pull_request", "component_index": 12, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17790,12 +17749,12 @@ "pull_request:43656|pull_request:43842", "pull_request:43836|pull_request:43842" ], - "prepared_review_unit_hash": "4dea8268748b64739dc6f6ba9f690dc4f4487426af8849ad38cd0bdf8543ce9f", + "prepared_review_unit_hash": "a5657bf8bd8971f77abacf83534a8b51201261eae1e5a051307ddd631435c9c1", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 5991, + "serialized_chars": 5992, "estimated_input_tokens": 1498, "estimated_eval_tokens": 3252 }, @@ -17803,7 +17762,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 5991, + "serialized_chars": 5992, "estimated_input_tokens": 1498, "estimated_eval_tokens": 3252 }, @@ -17812,41 +17771,41 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs cluster around the same CLI `serve.py` annotation/runtime `TypeAdapter` NameError. PR 43842 is the cleanest narrowly-scoped fix; 43656 is related but broader, and 43836 is more refactor-heavy.", - "confidence": 0.77, + "summary": "The cluster centers on `src/transformers/cli/serve.py` and TypeAdapter-related NameErrors, but the PRs are not close enough to treat as the same mergeable fix. PR 43842 is the most focused patch; 43656 is a related but different typing/runtime cleanup, and 43836 is a broader Serve refactor with the same area touched.", + "confidence": 0.82, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:43842 is the most focused fix for the concrete `TypeAdapter` NameError in `src/transformers/cli/serve.py`, with a minimal targeted change and an explicit issue link.", + "canonical_pr_reason": "PR 43842 is the cleanest and most targeted fix for the pydantic-absent TypeAdapter NameError, with the smallest patch and a direct annotation-only change in `serve.py`.", "best_issue_reason": null, - "best_pr_reason": "pull_request:43842 best represents the underlying bug fix because it directly targets the runtime/import problem with the smallest, most specific patch.", + "best_pr_reason": "PR 43842 best represents the cluster because it addresses the concrete runtime NameError with a narrowly scoped change, unlike the other PRs that mix in broader typing or refactor work.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", - "accept": true, - "reason": "Both address the same `TypeAdapter` NameError in `serve.py`; the fixes are different implementations of the same annotation/runtime problem and could plausibly be merged into one PR." + "accept": false, + "reason": "They both touch `serve.py` and mention TypeAdapter, but the packet does not show the same concrete bugfix path or a patch that is clearly mergeable into one PR; the explicit issue targets also differ." }, { "left": "pull_request:43836", "right": "pull_request:43842", - "accept": true, - "reason": "Both target the same `serve.py` `TypeAdapter` NameError path. Although 43836 is broader, the core bug and fix area overlap enough that they look like alternate solutions to the same issue." + "accept": false, + "reason": "They share the same file and issue target, but 43836 is a broader Serve refactor and not clearly the same focused TypeAdapter NameError fix as 43842." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary stays within the packet: all three PRs touch `src/transformers/cli/serve.py` and are plausibly part of the same `TypeAdapter`/annotation runtime issue, with 43842 the narrowest fix. The soft-edge verdicts are conservative enough, especially since 43836 and 43842 share the same issue target and 43656 is framed as a plausible alternate implementation rather than a definitive duplicate." + "accept": false, + "feedback": "The cluster summary is only partly grounded, but one soft-edge verdict is too strong. For 43656 vs 43842, the packet supports that both touch `src/transformers/cli/serve.py` and both reference a TypeAdapter-related NameError, but it does not clearly show the same concrete bugfix path or that the changes are straightforwardly mergeable into one PR. The explicit issue targets also differ. The 43836 vs 43842 rejection is conservative and consistent." }, "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, { "label": "pull_request", "component_index": 13, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17912,7 +17871,7 @@ { "label": "pull_request", "component_index": 14, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 2, "status": "reviewed", @@ -17968,8 +17927,8 @@ "pull_request:44019|pull_request:44085", "pull_request:44018|pull_request:44019", "pull_request:44019|pull_request:44071", - "pull_request:44068|pull_request:44722", "pull_request:44068|pull_request:44116", + "pull_request:44068|pull_request:44722", "pull_request:44071|pull_request:44722", "pull_request:44018|pull_request:44116", "pull_request:44018|pull_request:44086", @@ -17978,8 +17937,8 @@ "pull_request:44068|pull_request:44086", "pull_request:44059|pull_request:44086", "pull_request:44085|pull_request:44086", - "pull_request:44059|pull_request:44722", "pull_request:44098|pull_request:44101", + "pull_request:44059|pull_request:44722", "pull_request:44018|pull_request:44071", "pull_request:44017|pull_request:44722", "pull_request:44017|pull_request:44161", @@ -17990,9 +17949,9 @@ "pull_request:44017|pull_request:44071", "pull_request:44073|pull_request:44101", "pull_request:44059|pull_request:44071", - "pull_request:44073|pull_request:44161", "pull_request:44056|pull_request:44101", "pull_request:44059|pull_request:44068", + "pull_request:44073|pull_request:44161", "pull_request:44018|pull_request:44059", "pull_request:44101|pull_request:44161", "pull_request:43997|pull_request:44017", @@ -18024,298 +17983,394 @@ "pull_request:44044|pull_request:44129", "pull_request:44044|pull_request:44154" ], - "prepared_review_unit_hash": "14d5da96df2e714c672a3a8ef74d6d47e009fc353ed67590fc185ffbe0003af4", + "prepared_review_unit_hash": "4b1bd05d6acc17c27375dcea736d70030435ad997b5afcb04c0461dacd366197", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84892, - "estimated_input_tokens": 21223, - "estimated_eval_tokens": 42702 + "serialized_chars": 84885, + "estimated_input_tokens": 21222, + "estimated_eval_tokens": 42700 }, "budget": { "node_count": 27, "item_count": 27, "soft_pair_count": 72, - "serialized_chars": 75087, - "estimated_input_tokens": 18772, - "estimated_eval_tokens": 37800 + "serialized_chars": 75084, + "estimated_input_tokens": 18771, + "estimated_eval_tokens": 37798 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a broad theme cluster of model-specific output-tracing refactors, not one single duplicate. The only true duplicate-like groups are the GPT-J/CodeGen PRs and the GPT-Neo PRs; the rest touch different model implementations and should stay separate.", - "confidence": 0.93, + "summary": "This cluster is mostly a set of model-specific output-tracing refactors, not true duplicates across different architectures. The only clear duplicate set is the GPT-J/CodeGen trio, which all target the same standardized output-capturing change in `modeling_gptj.py` (and, for two of them, `modeling_codegen.py`).", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44722 is the clearest representative of the repeated GPT-J/CodeGen output-tracing refactor: it targets the same standardized-decorator change, touches the same core code paths, and is the most self-contained late-stage variant in the duplicate-like subgroup.", + "canonical_pr_reason": "44066 is the best anchor: it has the explicit tracking issue, covers the shared GPT-J/CodeGen output-tracing change, and is the most complete representative of the near-duplicate GPT-J follow-ups.", "best_issue_reason": null, - "best_pr_reason": "44722 best represents the main duplicate-like subgroup because it directly refactors GPT-J output tracing with the standardized decorator approach and also includes the related CodeGen companion change.", + "best_pr_reason": "44066 is the strongest representative PR for the cluster\u2019s true duplicate set: same concrete GPT-J output-tracing refactor, explicit issue target, and broader scope than the GPT-J-only follow-up.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different model families and different code paths; same issue target alone is not enough." + "reason": "Different models and code paths (CVT/FNet vs GPT-J); same refactor theme is not enough." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "CVT/FNet refactor is unrelated to DeBERTa V2 beyond the shared output-tracing theme." + "reason": "Different architectures/models; both touch output tracing, but not the same underlying bug/change." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Both are GPT-J output-tracing refactors in the same file and implement the same concrete change." + "reason": "Same GPT-J output-tracing refactor in `modeling_gptj.py`; one PR is just a narrower slice of the same change." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "ResNet/RT-DETR ResNet versus EfficientNet are different model implementations and not the same duplicate." + "reason": "Different vision models/code paths (RegNet/ResNet vs EfficientNet); not the same fix." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "EfficientNet output tracing is unrelated to GPT-J/CodeGen output tracing." + "reason": "EfficientNet vs GPT-J/CodeGen; only the high-level refactor pattern overlaps." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "Different model code paths; only the high-level refactor theme is shared." + "reason": "GPT-J/CodeGen vs EfficientNet are unrelated concrete code paths." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa V2 are unrelated concrete changes." + "reason": "MobileNetV2 vs DeBERTa V2; same standardized-tracing theme, different fixes." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J/CodeGen versus MGP-STR are different underlying implementations." + "reason": "GPT-J/CodeGen vs MGP-STR are different model implementations." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen and MPT are separate model-specific refactors." + "reason": "GPT-J/CodeGen vs MPT; not the same code-path problem." }, { "left": "pull_request:44018", "right": "pull_request:44068", - "accept": true, - "reason": "Same GPT-Neo file and same standardized output-tracing change; one looks like a later/alternate version of the other." + "accept": false, + "reason": "GPT-Neo and GPT-J are related but distinct model code paths." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are different code paths." + "reason": "GPT-J/CodeGen vs GPT-Neo; shared refactor style, but different changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen are separate model ports." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet refactor is not the same underlying change as GPT-J/CodeGen." + "reason": "ResNet vs GPT-J/CodeGen; unrelated concrete fixes." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR output tracing is a different model-specific change." + "reason": "MGP-STR vs GPT-J/CodeGen are different model paths." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet and GPT-J are unrelated implementations." + "reason": "ResNet vs GPT-J; only the broad output-tracing theme overlaps." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are different model paths." + "reason": "GPT-Neo vs ResNet are unrelated implementations." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "ResNet and MPT are unrelated concrete changes." + "reason": "ResNet vs MPT; not the same underlying issue." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo vs Flaubert/XLM are different model families and code paths." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo is distinct from GPT-J/CodeGen; similar refactor style is not enough." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT vs GPT-J/CodeGen; different concrete changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo vs Flaubert/XLM are unrelated model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo vs MGP-STR are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo vs GPT-J; related architecture, but not the same change." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT vs MGP-STR are different models and fixes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo vs MGP-STR are unrelated code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT2 vs MGP-STR; same refactor motif, different model bug." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J vs MGP-STR are different concrete changes." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT vs Flaubert/XLM are different model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT2 vs GPT-J/CodeGen are unrelated model paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo vs MPT; only the decorator migration pattern matches." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer vs GPT-J/CodeGen are unrelated implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer vs LongT5 are different code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet vs SegFormer are distinct model refactors." }, { - "left": "pull_request:44068", - "right": "pull_request:44722", + "left": "pull_request:43997", + "right": "pull_request:44056", "accept": false, - "reason": "GPT-Neo versus GPT-J/CodeGen are different model refactors." + "reason": "RegNet vs MPNet; same output-tracing theme, different bugs." }, { "left": "pull_request:44071", - "right": "pull_request:44085", + "right": "pull_request:44076", "accept": false, - "reason": "MPT and GPT-J are different code paths." + "reason": "MPT vs ImageGPT are different model implementations." }, { - "left": "pull_request:44068", - "right": "pull_request:44086", + "left": "pull_request:44074", + "right": "pull_request:44101", "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated implementations." + "reason": "TextNet vs Flaubert/XLM are unrelated code paths." }, { - "left": "pull_request:44059", - "right": "pull_request:44086", + "left": "pull_request:44056", + "right": "pull_request:44161", "accept": false, - "reason": "GPT-2 refactor is not the same change as MGP-STR." + "reason": "MPNet vs LongT5 are different concrete changes." }, { - "left": "pull_request:44085", - "right": "pull_request:44086", + "left": "pull_request:43997", + "right": "pull_request:44018", "accept": false, - "reason": "GPT-J and MGP-STR are different model-specific changes." + "reason": "RegNet vs GPT-Neo are separate model refactors." }, { - "left": "pull_request:44059", - "right": "pull_request:44722", + "left": "pull_request:43997", + "right": "pull_request:44161", "accept": false, - "reason": "GPT-2 output tracing is not the same concrete bug/change as GPT-J/CodeGen." + "reason": "RegNet vs LongT5; not the same underlying bug/change." }, { - "left": "pull_request:44098", - "right": "pull_request:44101", + "left": "pull_request:44071", + "right": "pull_request:44085", "accept": false, - "reason": "ViLT and Flaubert/XLM are different model families and separate changes." + "reason": "MPT vs GPT-J are different model paths." }, { - "left": "pull_request:44018", + "left": "pull_request:44010", "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and MPT are unrelated model implementations." + "reason": "SqueezeBERT vs MPT are unrelated implementations." }, { - "left": "pull_request:44017", - "right": "pull_request:44722", + "left": "pull_request:44074", + "right": "pull_request:44161", "accept": false, - "reason": "SegFormer output tracing is a different code path from GPT-J/CodeGen." + "reason": "TextNet vs LongT5 are different models and fixes." }, { - "left": "pull_request:44017", - "right": "pull_request:44161", + "left": "pull_request:44018", + "right": "pull_request:44059", "accept": false, - "reason": "SegFormer and LongT5 are different model-specific refactors." + "reason": "GPT-Neo vs GPT2 are distinct model code paths." }, { - "left": "pull_request:44017", - "right": "pull_request:44073", + "left": "pull_request:43997", + "right": "pull_request:44074", "accept": false, - "reason": "SegFormer and VisualBert are unrelated implementations." + "reason": "RegNet vs TextNet; same tracing pattern, different concrete changes." }, { - "left": "pull_request:44068", - "right": "pull_request:44071", + "left": "pull_request:44010", + "right": "pull_request:44018", "accept": false, - "reason": "GPT-Neo and MPT are different code paths." + "reason": "SqueezeBERT vs GPT-Neo are unrelated model implementations." }, { - "left": "pull_request:44071", - "right": "pull_request:44086", + "left": "pull_request:44044", + "right": "pull_request:44074", "accept": false, - "reason": "MPT and MGP-STR are unrelated model changes." + "reason": "DeBERTa V2 vs TextNet are different code paths." }, { - "left": "pull_request:43997", - "right": "pull_request:44073", - "accept": false, - "reason": "RegNet and VisualBert are different model implementations." + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor; both are plausibly one combined PR." }, { - "left": "pull_request:44056", - "right": "pull_request:44101", - "accept": false, - "reason": "MPNet and Flaubert/XLM are unrelated concrete changes." + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J output-tracing change in the same file; the title mismatch does not change the code-path match." }, { - "left": "pull_request:44059", - "right": "pull_request:44071", + "left": "pull_request:44085", + "right": "pull_request:44129", "accept": false, - "reason": "GPT-2 and MPT are different code paths." + "reason": "GPT-J vs SpeechT5 are unrelated model paths." }, { - "left": "pull_request:44073", - "right": "pull_request:44161", + "left": "pull_request:43996", + "right": "pull_request:44007", "accept": false, - "reason": "VisualBert and LongT5 are different model-specific refactors." + "reason": "CVT/FNet vs RegNet/ResNet/RT-DETR ResNet are different fixes." }, { - "left": "pull_request:44056", - "right": "pull_request:44161", + "left": "pull_request:43996", + "right": "pull_request:44129", "accept": false, - "reason": "MPNet and LongT5 are unrelated implementations." + "reason": "CVT/FNet vs SpeechT5 are unrelated model implementations." }, { - "left": "pull_request:43997", - "right": "pull_request:44018", + "left": "pull_request:44044", + "right": "pull_request:44129", "accept": false, - "reason": "RegNet and GPT-Neo are not the same underlying change." + "reason": "DeBERTa V2 vs SpeechT5 are different concrete changes." }, { - "left": "pull_request:43997", - "right": "pull_request:44161", + "left": "pull_request:44044", + "right": "pull_request:44154", "accept": false, - "reason": "RegNet and LongT5 are separate model-specific changes." + "reason": "DeBERTa V2 vs Vits; same broad theme, different code paths." }, { - "left": "pull_request:44071", - "right": "pull_request:44129", + "left": "pull_request:44066", + "right": "pull_request:44086", "accept": false, - "reason": "MPT and SpeechT5 are different code paths." + "reason": "GPT-J/CodeGen vs MGP-STR are unrelated model fixes." }, { - "left": "pull_request:44010", + "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "SqueezeBERT and MPT are unrelated implementations." + "reason": "GPT-J/CodeGen vs MPT; only the standardized-output refactor is shared." }, { "left": "pull_request:44018", "right": "pull_request:44161", "accept": false, - "reason": "GPT-Neo and LongT5 are different model-specific refactors." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44044", - "accept": false, - "reason": "SqueezeBERT and DeBERTa V2 are different underlying changes." + "reason": "GPT-Neo vs LongT5 are different model families and bugs." }, { - "left": "pull_request:44044", - "right": "pull_request:44074", + "left": "pull_request:43996", + "right": "pull_request:44017", "accept": false, - "reason": "DeBERTa V2 and TextNet are unrelated model implementations." + "reason": "CVT/FNet vs SegFormer are unrelated model refactors." }, { - "left": "pull_request:44044", - "right": "pull_request:44129", + "left": "pull_request:44010", + "right": "pull_request:44044", "accept": false, - "reason": "DeBERTa V2 and SpeechT5 are different code paths." + "reason": "SqueezeBERT vs DeBERTa V2 are distinct code paths." }, { "left": "pull_request:44044", - "right": "pull_request:44154", + "right": "pull_request:44101", "accept": false, - "reason": "DeBERTa V2 and Vits are separate model-specific changes." + "reason": "DeBERTa V2 vs Flaubert/XLM are different models." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster summary is grounded in the packet and stays conservative about duplicate-like relationships. The two accepted soft pairs are plausible same-file/same-bugfix variants, and the rejected pairs are appropriately split by model family or code path rather than overstating similarity." + "feedback": "Grounded overall. The summary stays conservative about cross-architecture similarity, and the GPT-J/CodeGen duplicate set is supported by the packet: 44066 and 44722 clearly share the GPT-J/CodeGen output-tracing refactor, while 44085 is a narrower GPT-J-only slice of the same file-level change. I don\u2019t see an overclaim that would require rejection." }, "evaluator_used": true, "retried": false, @@ -18326,7 +18381,7 @@ { "label": "pull_request", "component_index": 14, - "component_count": 30, + "component_count": 27, "review_unit_index": 2, "review_unit_count": 2, "status": "reviewed", @@ -18353,70 +18408,70 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "16be0a6daa058864c1b77950a1d65475c700d1dd49b3392866d821a2ab020ffd", + "prepared_review_unit_hash": "6ea512a0d315d20d314bae5e2db51ffff5e7c9eeb8995bdf8bf2479785e51921", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84892, - "estimated_input_tokens": 21223, - "estimated_eval_tokens": 42702 + "serialized_chars": 84885, + "estimated_input_tokens": 21222, + "estimated_eval_tokens": 42700 }, "budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 6, - "serialized_chars": 17126, - "estimated_input_tokens": 4282, - "estimated_eval_tokens": 8820 + "serialized_chars": 17122, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All items are model-specific PRs for standardized output tracing / capture_outputs refactors. They share a theme, but they target different architectures and code paths, so they are related changes rather than duplicates.", - "confidence": 0.92, + "summary": "These are not duplicates; they are separate model-specific PRs applying the same output-tracing refactor pattern across different code paths. The cluster is best treated as a thematic series, not one mergeable change.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the best representative PR in this set: it has an explicit issue target, a focused single-model refactor, and appears more complete than the smaller model migrations.", + "canonical_pr_reason": "PR 44056 is a clean representative of the shared refactor pattern: it adds standardized output tracing plumbing (`capture_outputs`, `_can_record_outputs`) directly in a single model implementation.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the strongest standalone candidate for triage because it is concrete, scoped, and anchored to a tracked issue; the others are similar refactor patterns but not interchangeable with it.", + "best_pr_reason": "44056 is the most representative single PR in the cluster because it targets the core standardized output-capture mechanism with a straightforward, self-contained model change.", "soft_edge_verdicts": [ { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "Same refactor theme, but FocalNet and ImageGPT are different model implementations with different output paths; not the same underlying change." + "reason": "Both are output-tracing refactors, but they affect different models (FocalNet vs ImageGPT) and different forward paths, so they are not the same change." }, { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "Both migrate to output-capture helpers, but they touch different models and code paths, so they should not be merged as one PR." + "reason": "Same refactor theme, but VisualBert and TextNet are distinct model implementations with separate code paths; not mergeable as one PR." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "MPNet vs VisualBert is a shared pattern only; the concrete fix sites and model behavior are different." + "reason": "Both standardize output capture, but MPNet and VisualBert are different model-specific implementations, so this is not a duplicate bug/change." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "These are separate model-specific output-tracing refactors, not one duplicate bug or change." + "reason": "They share the same refactor style, but MPNet and TextNet are separate model paths and could not plausibly be merged into one PR." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model code paths; similar helper migration does not make them duplicates." + "reason": "RegNet and SqueezeBert are unrelated model files; this is the same migration pattern, not the same underlying change." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "Both are standardized output refactors, but SqueezeBert and SpeechT5 address different concrete implementations and cannot plausibly be one merged PR." + "reason": "SqueezeBert and SpeechT5 modify different model code paths; both are output-capture refactors but not the same concrete fix." } ] }, @@ -18430,7 +18485,7 @@ { "label": "pull_request", "component_index": 15, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18496,274 +18551,268 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "8000c24243315a35bc317867e4a705c633c1d2f61d49fe3406528b3b12a6e428", + "prepared_review_unit_hash": "0cd4f8d8924d08edb362cab52fb23f143e1e0c103698f82a398d045350cfcd65", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40179, - "estimated_input_tokens": 10045, - "estimated_eval_tokens": 20346 + "serialized_chars": 40174, + "estimated_input_tokens": 10044, + "estimated_eval_tokens": 20344 }, "budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40179, - "estimated_input_tokens": 10045, - "estimated_eval_tokens": 20346 + "serialized_chars": 40174, + "estimated_input_tokens": 10044, + "estimated_eval_tokens": 20344 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All items are distinct model-specific output-tracing refactor PRs. They share a common pattern, but none appear to fix the same concrete bug or change, so I would not merge any pair as duplicates.", - "confidence": 0.94, + "summary": "This cluster is a set of similar-but-separate model-specific refactor PRs for output tracing / can_return_tuple cleanup. They share a common theme, but they do not look like duplicate fixes to the same underlying code path.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44003 is the broadest representative of the cluster, touching two closely related Mamba variants and showing the shared output-tracing refactor pattern most clearly.", + "canonical_pr_reason": "PR 44003 is the strongest single representative: it shows the most complete output-tracing plumbing change across closely related Mamba/FalconMamba files. The rest are separate model-specific refactors, not duplicates of one another.", "best_issue_reason": null, - "best_pr_reason": "PR 44003 is the best global representative because it spans multiple related code paths while still being a single coherent refactor.", + "best_pr_reason": "44003 is the best global representative for the cluster\u2019s theme because it is broad and concrete, but it still represents only one model family\u2019s tracing refactor rather than a duplicate of the others.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different models and forward paths; same refactor theme, not the same fix." + "reason": "Different architectures and files (timm_backbone vs upernet); same refactor theme only." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Timm backbone vs vision-text dual encoder are unrelated code paths." + "reason": "Different model families and code paths; both are output-tracing cleanup, not the same change." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "Timm backbone and UnivNet are separate model implementations with different behavior." + "reason": "timm_backbone and univnet are unrelated implementations; no shared concrete bug/change." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Different model families; both refactor output handling but not the same underlying change." + "reason": "Different model families and filenames; similar tracing refactor but not one duplicate PR." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Timm backbone and RWKV are unrelated model code paths." + "reason": "timm_backbone vs rwkv are unrelated code paths; only the meta-pattern matches." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNetV1 and UnivNet are distinct architectures; not a duplicate fix." + "reason": "MobileNetV1 and UnivNet are unrelated models; same tracing refactor theme, not the same underlying fix." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "MobileNetV1 vs UperNet; similar refactor pattern only." + "reason": "Different model families/files; no concrete shared bug or merged-able change." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "Different models and different output-tracing surfaces; not the same PR-level change." + "reason": "MobileNetV1 vs Depth Anything are different code paths; overlap is only in output-tracing cleanup." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "MobileNetV1 and VisionEncoderDecoder have separate forward implementations." + "reason": "Different architectures and files; same refactor motif only." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "MobileNetV1 and SpeechEncoderDecoder are unrelated code paths." + "reason": "MobileNetV1 and SpeechEncoderDecoder are separate model implementations; not the same PR change." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder and UnivNet fix different model outputs." + "reason": "VisionTextDualEncoder and UnivNet are unrelated model paths; similar tracing refactor theme only." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different models and different forward logic; not mergeable as one fix." + "reason": "Different model families/files; no evidence of the same underlying bug." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "Separate model families with no shared concrete bug." + "reason": "VisionTextDualEncoder vs Mamba/FalconMamba are different code paths; not a duplicate fix." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different model architectures; same refactor style but not the same code-path problem." + "reason": "Different architectures and filenames; only the output-tracing refactor motif is shared." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are unrelated implementations." + "reason": "UnivNet and UperNet are unrelated implementations; same style of refactor, not the same change." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "UnivNet vs Mamba family; no shared concrete fix." + "reason": "Different model families/files; no shared concrete code-path problem." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "Different model code paths; only the output-tracing pattern overlaps." + "reason": "UnivNet vs Depth Anything are separate model paths; not mergeable as one duplicate PR." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "UperNet and RWKV are unrelated model paths." + "reason": "UperNet and RWKV are unrelated code paths; the similarity is only in tracing cleanup." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "Different subsystems and forward contracts; not duplicates." + "reason": "Different model families and files; not the same underlying bug/change." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Mamba-family refactor vs SuperPoint feature extractor; distinct code paths." + "reason": "Mamba/FalconMamba vs SuperPoint are different code paths; similar refactor theme only." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Mamba-family tracing changes and CodeGen cache/output changes are separate fixes." + "reason": "Different implementations and filenames; both mention tracing, but they fix different model code." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "Timm backbone and CodeGen address different model internals." + "reason": "timm_backbone and codegen are unrelated models; no duplicate underlying change." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet audio generation and CodeGen attention/cache code are unrelated." + "reason": "UnivNet and CodeGen are separate code paths; only the meta-refactor overlaps." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet segmentation and CodeGen are distinct code paths." - }, - { - "left": "pull_request:44000", - "right": "pull_request:44004", - "accept": false, - "reason": "Vision-text dual encoder and CodeGen do not share the same concrete bug." + "reason": "UperNet and CodeGen are unrelated; not the same fix." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "Depth-Anything and VisionEncoderDecoder are separate model implementations." + "reason": "Depth Anything and VisionEncoderDecoder are different model families/files; no concrete duplicate change." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "Depth-Anything and SpeechEncoderDecoder are different code paths." + "reason": "Depth Anything and SpeechEncoderDecoder are unrelated code paths; same theme only." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "Depth-Anything and SuperPoint are unrelated models." + "reason": "Different models and filenames; not the same underlying bug/change." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Depth-Anything and RWKV fix different output-tracing surfaces." + "reason": "Depth Anything vs RWKV are unrelated implementations; same refactor pattern only." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "Depth-Anything and DPR are separate architectures and forward methods." + "reason": "Different model families/files; no shared concrete code-path problem." }, { "left": "pull_request:44026", "right": "pull_request:44029", "accept": false, - "reason": "VisionEncoderDecoder and RWKV are unrelated model code paths." + "reason": "VisionEncoderDecoder and RWKV are unrelated model paths; not a duplicate PR." }, { "left": "pull_request:44026", "right": "pull_request:44027", "accept": false, - "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are similar wrappers but still different concrete implementations." + "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are distinct code paths; only the tracing refactor theme matches." }, { "left": "pull_request:44026", "right": "pull_request:44028", "accept": false, - "reason": "Different model families; no single underlying bug." + "reason": "Different model families/files; same meta-change, not same bug or change." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "VisionEncoderDecoder and DPR change different code paths." + "reason": "VisionEncoderDecoder and DPR are different implementations; no duplicate underlying fix." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "SpeechEncoderDecoder and SuperPoint are unrelated." + "reason": "SpeechEncoderDecoder and SuperPoint are unrelated models; not the same change." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "SpeechEncoderDecoder and RWKV are distinct model implementations." + "reason": "Different code paths and filenames; only the output-tracing cleanup overlaps." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "SpeechEncoderDecoder and DPR do not share the same concrete change." + "reason": "SpeechEncoderDecoder and DPR are separate model implementations; not mergeable as one duplicate PR." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are unrelated code paths." + "reason": "SuperPoint and RWKV are unrelated code paths; similar refactor style only." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "SuperPoint and DPR are different model internals." + "reason": "SuperPoint and DPR are different model families/files; no shared concrete fix." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are separate implementations; same pattern only." + "reason": "RWKV and DPR are unrelated implementations; same tracing-refactor motif only." } ] }, @@ -18777,7 +18826,7 @@ { "label": "pull_request", "component_index": 16, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18796,52 +18845,52 @@ "pull_request:44662|pull_request:45075", "pull_request:45075|pull_request:45401" ], - "prepared_review_unit_hash": "62ac0d9f71d85c36648ed78c2279275a6ae09f6d2f7afbd8c9942fa72a7d4023", + "prepared_review_unit_hash": "ef6ffd227a8203a03fcd335e0c38df0e82cbeea4df50463454cf4de9b082ee5d", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10913, - "estimated_input_tokens": 2729, - "estimated_eval_tokens": 5714 + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10913, - "estimated_input_tokens": 2729, - "estimated_eval_tokens": 5714 + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition PRs for different architectures (PenguinVL, DeepSeek-OCR-2, Voxtral TTS). They share Transformers boilerplate and auto-registration changes, but they do not appear to fix or implement the same concrete code path, so they should not be clustered as duplicates.", - "confidence": 0.98, + "summary": "These are three separate model-implementation PRs (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They overlap only on shared registration/docs/auto-mapping boilerplate, not on the same underlying code change, so this cluster should not be deduplicated.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45075 is the best representative if one is needed: it is non-draft, has the most review activity, and looks like a full model integration with conversion, image processing, modeling, and tests.", + "canonical_pr_reason": "No single PR is canonical here: each one adds a different model family with distinct configs, modeling code, and tests.", "best_issue_reason": null, - "best_pr_reason": "PR 45075 is the strongest standalone candidate because it is the most mature and complete of the three integrations, with broader implementation coverage than the others.", + "best_pr_reason": "No PR is a global best duplicate candidate; they address unrelated model additions rather than one shared bug or feature.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different models and different code paths: PenguinVL vs Voxtral TTS. Shared docs/auto-registry boilerplate is not enough to treat them as the same change." + "reason": "Different models and code paths (PenguinVL vs Voxtral TTS); the overlap is only generic auto/docs wiring." }, { "left": "pull_request:44662", "right": "pull_request:45075", "accept": false, - "reason": "Different model integrations (PenguinVL vs DeepSeek-OCR-2). They both add new model packages, but not the same underlying bug or feature." + "reason": "PenguinVL and DeepSeek-OCR-2 are distinct model implementations; shared files are boilerplate registration changes only." }, { "left": "pull_request:45075", "right": "pull_request:45401", "accept": false, - "reason": "DeepSeek-OCR-2 and Voxtral TTS are unrelated model additions; overlap is limited to common Transformers scaffolding, so they should not be merged as duplicates." + "reason": "These add different model stacks (DeepSeek-OCR-2 vs Voxtral TTS) and do not look like one mergeable fix/change." } ] }, @@ -18855,7 +18904,7 @@ { "label": "pull_request", "component_index": 17, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18871,12 +18920,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "806642044db55b74098c46cfb840b289997e85cbe0386b797d5bed42638818ff", + "prepared_review_unit_hash": "33cb670797c5d4e0aee911ee4459ba26f554c099a59bd5badc4f3f568a9039db", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4407, + "serialized_chars": 4408, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -18884,7 +18933,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4407, + "serialized_chars": 4408, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -18893,18 +18942,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related to fusion infrastructure, but they change different layers and are not the same mergeable fix/change: #44979 introduces a new module fusion API, while #45363 wires kernel-config-based n-to-1 fusion into model loading and adds an example. Related theme, not a duplicate.", - "confidence": 0.91, + "summary": "These two PRs are related in the broad area of model fusion, but they are not the same change: one introduces a general Module Fusion API, while the other wires n-to-1 kernel fusion through KernelConfig and related integration points. They are likely complementary rather than duplicates.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "#44979 is the more foundational PR: it defines the standalone Module Fusion API and test coverage, whereas #45363 is a later, narrower integration built on a different concept (KernelConfig-based kernel fusion).", + "canonical_pr_reason": "PR #44979 is the more foundational change: it introduces the standalone Module Fusion API and test coverage, which can serve as the generic abstraction layer for later fusion integrations.", "best_issue_reason": null, - "best_pr_reason": "#44979 is the best representative of the cluster because it is the core API change; #45363 is an application/integration of fusion ideas rather than the same concrete code-path change.", + "best_pr_reason": "PR #45363 is the more concrete user-facing feature for kernel-based n-to-1 fusion, but it still depends on broader fusion infrastructure and is not a duplicate of the API-only PR.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Different concrete changes: #44979 adds a new module fusion framework in src/transformers/module_fusion.py, while #45363 adds KernelConfig-driven n-to-1 fusion plumbing across model loading/integrations. Same broad area, but not the same underlying bug or one PR that could simply absorb the other." + "reason": "Not the same concrete bug/change. #44979 adds a generic Module Fusion API, while #45363 adds KernelConfig-based n-to-1 fusion plumbing and an example. They touch related code, but they are not mergeable as one duplicate PR." } ] }, @@ -18918,7 +18967,7 @@ { "label": "pull_request", "component_index": 18, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18934,7 +18983,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "781c5b225f05a4980757f27344cb412f689767fe155734770ee5ee17be6d3494", + "prepared_review_unit_hash": "23673118accf8423a4094233d191412a13b3b42d0ee6202779152e7f16d74bc1", "component_budget": { "node_count": 2, "item_count": 2, @@ -18956,18 +19005,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests appear to fix the same concrete Mistral regex AttributeError in the same code path (`_patch_mistral_regex`) with essentially the same change; the later PR also adds a test, so this looks like a duplicate PR cluster.", - "confidence": 0.99, + "summary": "Both pull requests target the same Mistral regex patch in the same file and fix the same AttributeError by switching from backend_tokenizer.pre_tokenizer to tokenizer.pre_tokenizer; PR 45317 is the more complete version because it adds tests.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "`pull_request:45317` is the stronger canonical PR because it implements the same fix as 45086 and also includes a test update, making it the more complete representative of the shared change.", + "canonical_pr_reason": "PR 45317 is the more complete representative of the same fix: same code change as 45086 plus test coverage and a clearer problem statement.", "best_issue_reason": null, - "best_pr_reason": "`pull_request:45317` is the best representative overall since it addresses the same bug and adds accompanying test coverage, whereas 45086 is the narrower earlier version.", + "best_pr_reason": "PR 45317 best captures the cluster because it includes the same bug fix with additional tests and discussion, making it the most complete candidate.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same issue target, same file, and the diff shows the same AttributeError fix in `_patch_mistral_regex`; the second PR is essentially the same patch with added tests." + "reason": "Same underlying AttributeError fix in the same code path and same file; the second PR is essentially the first fix plus tests, so they are mergeable as one change." } ] }, @@ -18981,7 +19030,7 @@ { "label": "pull_request", "component_index": 19, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18997,7 +19046,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "fd0ce53f19a9e54fe304d43cfeb0103cc64be2310004f82affd40f8c43ee1199", + "prepared_review_unit_hash": "b3db35f9213f4365eeae472e69949ddc23e3efcbc373887c406757c3f7c0bb9d", "component_budget": { "node_count": 2, "item_count": 2, @@ -19019,18 +19068,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs overlap on NemotronH model files, but they address different changes: one refactors NemotronH to inherit GraniteMoeHybrid and updates conversion mapping, while the other splits NemotronH into dense/sparse variants with a backward-compat dispatcher plus docs/tests. They are related, but not the same concrete fix.", - "confidence": 0.93, + "summary": "Two NemotronH-related PRs, but they target different changes: one is a refactor/inheritance and conversion-mapping update, the other is a broader model split into dense/sparse variants with docs and tests. They are related but not duplicates.", + "confidence": 0.83, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45569 is the more comprehensive and user-facing NemotronH change, introducing the dense/sparse model split and compatibility wrapper rather than only a conversion/refactor layer.", + "canonical_pr_reason": "PR #45569 is the broader, more complete implementation of the NemotronH model split and associated docs/tests, so it best represents the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR #45569 best represents the cluster because it covers the larger architectural change and the broader set of affected files, including docs and tests.", + "best_pr_reason": "PR #45569 is the strongest standalone candidate because it covers the user-facing model reorganization and supporting documentation/testing, not just internal refactoring.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "Both touch NemotronH internals, but #45115 is a narrow GraniteMoeHybrid inheritance/conversion-mapping refactor, while #45569 restructures NemotronH into dense/sparse models and updates docs/tests. They are not the same underlying change and would not plausibly merge as one PR." + "reason": "Shared NemotronH files, but the concrete changes differ: #45115 focuses on inheritance/conversion mapping to GraniteMoeHybrid, while #45569 introduces dense/sparse model variants plus docs/tests. They are related but not the same underlying fix and would not plausibly merge into one PR." } ] }, @@ -19044,7 +19093,7 @@ { "label": "pull_request", "component_index": 20, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19060,12 +19109,12 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "2cd1977abba5636b5a1b7db895000412206fa9ab82db242a6375172f6ef33193", + "prepared_review_unit_hash": "2536324f84c89e604d3a69bb2933610a9a7e84d756d6e180dfce42f98d9af336", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5397, + "serialized_chars": 5398, "estimated_input_tokens": 1350, "estimated_eval_tokens": 2956 }, @@ -19073,7 +19122,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5397, + "serialized_chars": 5398, "estimated_input_tokens": 1350, "estimated_eval_tokens": 2956 }, @@ -19082,18 +19131,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two related but distinct PRs: one adds per-layer heterogeneous config support, the other adds broader heterogeneous model/runtime support. They overlap in shared config plumbing and tests, but they are not the same concrete change.", - "confidence": 0.87, + "summary": "These are related heterogeneity feature PRs, but they are not duplicates: one adds per-layer config plumbing, the other adds downstream model/cache/masking support on top of it.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45332 is the broader umbrella change: it includes heterogeneous config handling plus model/cache/masking support and therefore best represents the cluster.", + "canonical_pr_reason": "PR 45332 is the broader umbrella change: it includes the model/runtime changes that actually consume heterogeneous configs, so it best represents the full feature cluster.", "best_issue_reason": null, - "best_pr_reason": "PR #45332 is the best single representative because it covers the full heterogeneous-model feature set, not just the config foundation.", + "best_pr_reason": "PR 45332 is the most representative end-to-end implementation and covers the larger concrete code-path change across modeling, cache, and masking logic.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "Related feature work, but not the same underlying bug or change: #45333 is config-only plumbing, while #45332 extends into model/runtime behavior. They could be separate PRs rather than one merged fix." + "reason": "Related but distinct work: 45333 adds heterogeneous config support and helpers, while 45332 adds model/cache/masking behavior that consumes those configs. They are prerequisite/follow-on changes, not the same underlying bug or a single mergeable PR." } ] }, @@ -19107,7 +19156,7 @@ { "label": "pull_request", "component_index": 21, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19123,7 +19172,7 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "1b7a2c25de84c595dc7441ccef7b4bf54540635d967322c823f3620d2b1a5393", + "prepared_review_unit_hash": "11a56c8a04125bf52179caa118edf5248285ac1b20ce85d897ca590fdfe0c357", "component_budget": { "node_count": 2, "item_count": 2, @@ -19145,18 +19194,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs both touch typing-related files, but they are not the same underlying change: #45415 is a broad repository-wide type-checking sweep and config tweak, while #45425 adds specific typing helper protocols/aliases for model utility code. The overlap is incidental, not a duplicate fix.", - "confidence": 0.91, + "summary": "Two open PRs in the typing area, but they are not duplicates: one is a broad type-checking sweep across many Transformers modules, while the other adds typing protocols/helpers and related modeling-utils plumbing. The overlap in a few shared files looks incidental rather than the same underlying change.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "#45425 is the more focused substantive typing-infrastructure change; #45415 is a broad type-checking cleanup pass with many unrelated file edits.", + "canonical_pr_reason": "PR #45415 is the broader, more representative typing cleanup across `src/transformers/*py` and best matches the cluster\u2019s overall surface area.", "best_issue_reason": null, - "best_pr_reason": "#45425 is the best representative PR for the cluster because it introduces concrete typing abstractions used by multiple code paths, whereas #45415 is primarily a sweeping type-check enablement/change-set.", + "best_pr_reason": "PR #45415 is the best single representative for the cluster because it is the larger, more central typing sweep; PR #45425 is a narrower typing-infrastructure follow-up with different intent.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Reject: #45415 is a broad type-checking sweep/configuration update, while #45425 adds specific typing protocols and aliases. Shared files do not indicate the same bug or a single mergeable fix." + "reason": "Both are typing-related, but they address different changes: #45415 is a wide type-checking pass, while #45425 introduces specific typing protocols and helpers. Shared files alone do not make them the same underlying bug/change, and they would not naturally merge into one PR." } ] }, @@ -19170,7 +19219,7 @@ { "label": "pull_request", "component_index": 22, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19195,70 +19244,70 @@ "pull_request:45858|pull_request:45926", "pull_request:45890|pull_request:45926" ], - "prepared_review_unit_hash": "893f6287bc80aa435359d4f4c5b86a9a30e2a75dd9f845c205ef7ba91a422c1e", + "prepared_review_unit_hash": "cb0d30bcff04a3339453f2c9788f25361f9f06b820999652dbbd590388a918cd", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 6, - "serialized_chars": 14717, - "estimated_input_tokens": 3680, - "estimated_eval_tokens": 7616 + "serialized_chars": 14708, + "estimated_input_tokens": 3677, + "estimated_eval_tokens": 7610 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 6, - "serialized_chars": 14717, - "estimated_input_tokens": 3680, - "estimated_eval_tokens": 7616 + "serialized_chars": 14708, + "estimated_input_tokens": 3677, + "estimated_eval_tokens": 7610 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related only at a very broad theme level (model-parallel/XPU support), but they target different models and mostly different code paths. None of the soft edges look like true duplicates.", - "confidence": 0.95, + "summary": "These PRs are all about accelerator/model-parallel or XPU-related behavior, but they target different models and different fixes. I don\u2019t see any pair that is the same underlying bug/change, so none of the soft edges should be merged.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45817 is the strongest representative: it makes concrete code-path fixes for Gemma4 model-parallel behavior (device placement plus split-module handling) and includes tests, unlike the test-only expectation updates.", + "canonical_pr_reason": "PR #45817 is the strongest representative because it makes a real code-path fix for Gemma4 model-parallel behavior, not just test expectation updates, and it has the clearest end-to-end change set.", "best_issue_reason": null, - "best_pr_reason": "PR 45817 is the best standalone candidate because it addresses an actual runtime model-parallel bug in the implementation, not just platform-specific test expectations.", + "best_pr_reason": "PR #45817 is the best standalone PR in the set: it addresses an actual model-parallel bug in model code and includes corresponding test coverage, making it the most substantial and actionable change.", "soft_edge_verdicts": [ { "left": "pull_request:45487", "right": "pull_request:45843", "accept": false, - "reason": "Both are model-parallel fixes, but for different model families and different code paths. They are not the same underlying bug and would not plausibly be merged into one PR." + "reason": "Both are model-parallel fixes, but for different architectures (AltCLIP/ChineseCLIP vs Deimv2). The code paths and bugs are unrelated, so they should not be deduplicated." }, { "left": "pull_request:45487", "right": "pull_request:45926", "accept": false, - "reason": "AltCLIP/ChineseCLIP model-parallel device handling is unrelated to HyperCLOVAX XPU test expectations." + "reason": "These touch different models and different accelerator concerns; one is a model-parallel fix, the other is XPU test expectations. Same broad theme, not the same bug." }, { "left": "pull_request:45817", "right": "pull_request:45843", "accept": false, - "reason": "Gemma4 and Deimv2 are separate model-specific fixes; the shared theme is too broad to treat as a duplicate." + "reason": "Gemma4 and Deimv2 are unrelated model implementations; the shared phrase 'model parallel bugs' is too broad to treat as one duplicate change." }, { "left": "pull_request:45858", "right": "pull_request:45890", "accept": false, - "reason": "Both add XPU expectations, but for different models and tests. Same pattern, not the same bug or change." + "reason": "Both only add XPU expectations, but for different model test suites. This is test maintenance for separate models, not one underlying defect." }, { "left": "pull_request:45858", "right": "pull_request:45926", "accept": false, - "reason": "These are model-specific test expectation updates for different architectures, with no shared concrete code-path issue." + "reason": "Same accelerator class of change, but different models and different expected outputs. No evidence they fix the same concrete code-path problem." }, { "left": "pull_request:45890", "right": "pull_request:45926", "accept": false, - "reason": "Different model test suites and different expectation values; not a duplicate change." + "reason": "These are separate XPU expectation updates for distinct model tests; they do not appear to be the same bug or change." } ] }, @@ -19272,7 +19321,7 @@ { "label": "pull_request", "component_index": 23, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19288,7 +19337,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "3537647a7127d65aa27f750de3ecf30a027180c35d5a3ecf43406e3296389432", + "prepared_review_unit_hash": "8b9991acf91473b763654dc3318c8e370ac1dfd033614cae3daa1f6d456d81da", "component_budget": { "node_count": 2, "item_count": 2, @@ -19310,18 +19359,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs both concern expert parallelism loading, but they fix different failure modes on different code paths: one strips/recognizes distributed config for EP + ZeRO-3 loading, while the other prevents expert tensors from being overwritten during FSDP2/rank-0 broadcast handling.", - "confidence": 0.94, + "summary": "Both PRs are in the EP/distributed-loading area, but they fix different bugs: #45548 focuses on preserving distributed config / EP state during accelerate+ZeRO-3 loading, while #45662 fixes expert tensors being overwritten by rank-0 broadcast in the FSDP2 loading path. Shared filename and theme are not enough to treat them as duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "No exact duplicate here; if forced to pick one representative, PR #45662 is the stronger canonical PR because it targets a concrete overwrite bug in the loading path and includes the broader set of code-path fixes.", + "canonical_pr_reason": "PR #45662 is the better representative of the cluster theme because it targets a concrete EP loading corruption bug, whereas #45548 is a narrower serialization/config-preservation fix for a different code path.", "best_issue_reason": null, - "best_pr_reason": "PR #45662 is the best standalone PR because it addresses a specific, actionable loading corruption issue with clear runtime impact, whereas PR #45548 is narrower and solves a different EP/ZeRO-3 serialization/loading problem.", + "best_pr_reason": "PR #45662 is the strongest standalone candidate: it addresses a specific, user-visible failure mode in the distributed loading path and has the more direct fix for the underlying bug.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "They share expert-parallel/loading context, but the bugs and fixes differ: #45548 removes distributed_config from serialization and adds has_ep for EP + DeepSpeed ZeRO-3 loading, while #45662 changes post-shard wrapping to prevent expert parameters being overwritten under FSDP2/rank-0 broadcast. Not the same underlying bug or mergeable into one PR." + "reason": "Different underlying problems and code paths: #45548 removes distributed_config from serialization and adds has_ep for accelerate/ZeRO-3 loading, while #45662 prevents expert parameters from being overwritten during FSDP2/post-shard loading. Related area, but not the same change." } ] }, @@ -19335,7 +19384,7 @@ { "label": "pull_request", "component_index": 24, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19351,7 +19400,7 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "6e2bc6fe3a4f855d1614633ba1831711dbcf5532ec936c6ef80b03ea49319843", + "prepared_review_unit_hash": "e683d343d857337f7772df82e698ee948f634161cd5c0b36c062835f44855393", "component_budget": { "node_count": 2, "item_count": 2, @@ -19373,18 +19422,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target the same xdist-safe captured_info logging problem under issue #45561; PR 45645 is the broader, more complete fix and subsumes the testing_utils changes from 45639.", - "confidence": 0.92, + "summary": "Both PRs target the same xdist collision around `captured_info` debug artifacts; one is a narrower code-only fix, while the other bundles the same fix with CI log collection updates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the best canonical PR because it includes the core xdist-safe file handling change plus the CI/workflow updates needed to aggregate multiple captured_info files, making it the more complete fix.", + "canonical_pr_reason": "PR #45645 is the broader superset: it contains the xdist-safe `captured_info` handling in `testing_utils.py` plus the CI workflow update to read all `captured_info*.txt` files, so it best represents the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the strongest representative because it addresses the same underlying collision bug and also preserves/debugs CI log collection across xdist workers, whereas 45639 only fixes the library-side write path.", + "best_pr_reason": "PR #45645 is the best representative because it addresses the same concrete bug as #45639 and also updates the downstream consumer of those files, making it the more complete end-to-end fix.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Both PRs fix the same concrete captured_info xdist collision in testing_utils; 45645 is effectively an expanded version of the 45639 fix with additional workflow handling, so they are close enough to treat as the same change cluster." + "reason": "Both PRs fix the same underlying xdist collision for `captured_info` artifacts in the testing debug path; #45645 simply adds the CI-side aggregation needed for the same change and could plausibly be merged as one PR." } ] }, @@ -19398,7 +19447,7 @@ { "label": "pull_request", "component_index": 25, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19461,7 +19510,7 @@ { "label": "pull_request", "component_index": 26, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19527,73 +19576,7 @@ { "label": "pull_request", "component_index": 27, - "component_count": 30, - "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-45863-2", - "nodes": [ - "pull_request:45863", - "pull_request:45986" - ], - "soft_pairs": [ - "pull_request:45863|pull_request:45986" - ], - "prepared_review_unit_hash": "7f3114ee77134795638a4283bd1af3474d358fde534abcd721fee18249cf0286", - "component_budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 3691, - "estimated_input_tokens": 923, - "estimated_eval_tokens": 2102 - }, - "budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 3691, - "estimated_input_tokens": 923, - "estimated_eval_tokens": 2102 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": false, - "analyst_result": { - "summary": "Both pull requests address the same MLX availability bug: Transformers should opt out of MLX detection when HF_USE_MLX is disabled. PR 45863 is the broader, more complete fix with docs and tests; PR 45986 is a narrower code-only variant of the same change.", - "confidence": 0.91, - "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45863 is the better canonical representative because it implements the HF_USE_MLX opt-out and also includes documentation and test coverage, making it the more complete version of the fix.", - "best_issue_reason": null, - "best_pr_reason": "PR 45863 is the strongest overall fit for the cluster since it directly fixes the MLX detection path and provides the supporting docs/tests; PR 45986 is a partial duplicate of the same code-path change.", - "soft_edge_verdicts": [ - { - "left": "pull_request:45863", - "right": "pull_request:45986", - "accept": true, - "reason": "Both PRs fix the same concrete bug in is_mlx_available() and would plausibly be merged into one change; the second PR is a narrower variant of the same HF_USE_MLX opt-out." - } - ] - }, - "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is grounded in the packet: both PRs target the same issue target and the same MLX availability code path, and the mergeability claim is conservative. The canonical pick for 45863 is also supported by the extra docs/tests it adds. " - }, - "evaluator_used": true, - "retried": false, - "accepted_nontrivial_soft_edge": true, - "error_kind": null, - "error_message": null - }, - { - "label": "pull_request", - "component_index": 28, - "component_count": 30, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19609,7 +19592,7 @@ "soft_pairs": [ "pull_request:45975|pull_request:45977" ], - "prepared_review_unit_hash": "285d91188b4c345e9c7509db18eae662e9a523993e0d97f20fc759b3826cae18", + "prepared_review_unit_hash": "3f07da4de38f5f80d631618c18d5634fa68244a564f82c75c1cbb5e278edbb4c", "component_budget": { "node_count": 2, "item_count": 2, @@ -19631,144 +19614,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs target GGUF on Apple Silicon, but they address different layers of the stack: PR 45975 adds CLI/model-manager support and an optional kernels-community dequant fast path, while PR 45977 introduces a new GgufLinear integration with model plumbing and tests. Related subsystem, but not the same underlying change.", - "confidence": 0.91, + "summary": "Both PRs target GGUF on Apple Silicon and kernels-community integration, but they address different layers: one adds the core GGUF matmul/linear implementation, while the other adds an optional Metal dequant fast path plus CLI/model-manager wiring. Related topic, not the same underlying change.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45977 is the more substantial, self-contained feature PR: it adds a new GGUF linear integration and supporting model/quantizer/test plumbing, whereas PR 45975 is mainly an optional fast-path toggle and CLI/help wiring.", + "canonical_pr_reason": "PR 45977 is the core functional change: it introduces the new GgufLinear inference path and related quantizer/modeling support. PR 45975 is narrower configuration/docs plumbing around selecting the kernels-community path.", "best_issue_reason": null, - "best_pr_reason": "PR 45977 is the strongest representative of the cluster because it makes the core implementation change and has the broadest functional surface area; PR 45975 looks like a narrower companion optimization rather than a duplicate fix.", + "best_pr_reason": "PR 45977 is the best representative for the cluster because it contains the substantive implementation that likely motivated the broader GGUF-on-Apple-Silicon work.", "soft_edge_verdicts": [ { "left": "pull_request:45975", "right": "pull_request:45977", "accept": false, - "reason": "They both involve GGUF/Metal on Apple Silicon, but 45975 is about an optional kernels-community dequant fast path and CLI/model-manager wiring, while 45977 adds a new GgufLinear inference path. Different concrete changes, not mergeable as one PR." - } - ] - }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, - "error_kind": null, - "error_message": null - }, - { - "label": "pull_request", - "component_index": 29, - "component_count": 30, - "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-45976-2", - "nodes": [ - "pull_request:45976", - "pull_request:45989" - ], - "soft_pairs": [ - "pull_request:45976|pull_request:45989" - ], - "prepared_review_unit_hash": "055ab1d620a97c8693b657a9612c592ed664686d7e55eb1aff45ceffad3fd0be", - "component_budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 4493, - "estimated_input_tokens": 1124, - "estimated_eval_tokens": 2504 - }, - "budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 4493, - "estimated_input_tokens": 1124, - "estimated_eval_tokens": 2504 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": false, - "analyst_result": { - "summary": "These are related Llama4 MoE changes, but they address different layers of the stack: one fixes tensor-parallel expert calling/aggregation behavior, the other switches Llama4 experts to the standard expert implementation interface. They do not look like the same mergeable bugfix/change.", - "confidence": 0.79, - "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45976 is the broader, more concrete fix: it patches the tensor-parallel expert path, adds Llama4-specific model/config/test coverage, and directly addresses the crash described in the title.", - "best_issue_reason": null, - "best_pr_reason": "PR 45976 is the strongest representative of the cluster because it fixes the actual crash path and includes validation; PR 45989 is narrower and looks like an interface-adoption refactor that may complement, but not duplicate, the crash fix.", - "soft_edge_verdicts": [ - { - "left": "pull_request:45976", - "right": "pull_request:45989", - "accept": false, - "reason": "They touch the same Llama4 expert subsystem, but 45976 fixes a tensor-parallel crash for pre-weighted MoE experts while 45989 adopts the standard expert implementation decorator. Different code paths, different concrete problems." - } - ] - }, - "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": 30, - "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-45991-2", - "nodes": [ - "pull_request:45991", - "pull_request:45993" - ], - "soft_pairs": [ - "pull_request:45991|pull_request:45993" - ], - "prepared_review_unit_hash": "db9dd3522f36bd4d9a61568669869bd83deb3e403c87e2e5c51163798f4bcb31", - "component_budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 4234, - "estimated_input_tokens": 1059, - "estimated_eval_tokens": 2374 - }, - "budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 4234, - "estimated_input_tokens": 1059, - "estimated_eval_tokens": 2374 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": false, - "analyst_result": { - "summary": "These two PRs are not duplicates: one enables Flash Attention 2 for T5Gemma2, the other enables flash-attn support for EOMT. They touch different model families and different code paths.", - "confidence": 0.96, - "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: the changes target different architectures (T5Gemma2 vs EOMT) and are independent model-specific enablement patches.", - "best_issue_reason": null, - "best_pr_reason": "Neither PR is a global representative of the other; if forced, 45991 is the more substantial change, but it is not a better duplicate canonical for 45993.", - "soft_edge_verdicts": [ - { - "left": "pull_request:45991", - "right": "pull_request:45993", - "accept": false, - "reason": "Different models and different support toggles; they do not fix the same underlying bug or change and would not be merged into one PR." + "reason": "They share the GGUF/Metal subsystem, but 45977 implements the main GGUF linear matmul path while 45975 only adjusts optional dequant/serve integration and help text. They are not the same concrete bug/change and would not naturally collapse into one PR." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 39c8a6fde89d2b9727bc682bdfa113b63ea09a5d..6cf455da8655cb0a683836e6e17bd53e92b917ea 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260515t120027z", + "analysis_id": "hybrid-model-20260515t180026z", "archived_artifacts": { - "hybrid": "snapshots/20260515T120027Z/analysis-runs/hybrid-model-20260515t120027z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260515T120027Z/analysis-runs/hybrid-model-20260515t120027z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-05-15T12:09:11Z", + "published_at": "2026-05-15T18:10:41Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260515T120027Z", + "snapshot_id": "20260515T180026Z", "variant": "hybrid" } diff --git a/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.json b/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..ee69a590363c3ffd68f5a3ddd9aeb1556fe34d53 --- /dev/null +++ b/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.json @@ -0,0 +1,1791 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260515T180026Z", + "generated_at": "2026-05-15T18:10:39Z", + "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-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-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-41115-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41115.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41115, + "canonical_pr_number": 45613, + "issue_numbers": [ + 41115 + ], + "pr_numbers": [ + 41116, + 45613 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41116, + "right_pr_number": 45613, + "code_similarity": 0.805, + "size_similarity": 0.885, + "file_overlap": 0.75, + "area_overlap": 0.854, + "patch_similarity": 0.693, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "docs/source/en/model_doc/minicpm3.md", + "src/transformers/models/auto/modeling_auto.py", + "src/transformers/models/minicpm3/__init__.py", + "src/transformers/models/minicpm3/configuration_minicpm3.py", + "src/transformers/models/minicpm3/modeling_minicpm3.py", + "src/transformers/models/minicpm3/modular_minicpm3.py", + "tests/models/minicpm3/__init__.py", + "tests/models/minicpm3/test_modeling_minicpm3.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 713, + 720 + ] + ], + "right_ranges": [ + [ + 713, + 720 + ] + ] + }, + { + "filename": "docs/source/en/model_doc/minicpm3.md", + "left_ranges": [ + [ + 1, + 67 + ] + ], + "right_ranges": [ + [ + 1, + 45 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 285, + 291 + ], + [ + 700, + 706 + ], + [ + 1298, + 1304 + ] + ], + "right_ranges": [ + [ + 699, + 705 + ], + [ + 1301, + 1307 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/__init__.py", + "left_ranges": [ + [ + 1, + 29 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/configuration_minicpm3.py", + "left_ranges": [ + [ + 1, + 141 + ] + ], + "right_ranges": [ + [ + 1, + 126 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 544 + ] + ], + "right_ranges": [ + [ + 1, + 522 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modular_minicpm3.py", + "left_ranges": [ + [ + 1, + 444 + ] + ], + "right_ranges": [ + [ + 1, + 342 + ] + ] + }, + { + "filename": "tests/models/minicpm3/test_modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 109 + ] + ], + "right_ranges": [ + [ + 1, + 136 + ] + ] + } + ] + } + ] + }, + { + "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-43824-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43824.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 43824, + "canonical_pr_number": 43836, + "issue_numbers": [ + 43824 + ], + "pr_numbers": [ + 43836, + 43842 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 43836, + "right_pr_number": 43842, + "code_similarity": 0.332, + "size_similarity": 0.088, + "file_overlap": 1.0, + "area_overlap": 0.033, + "patch_similarity": 0.017, + "shared_filenames": [ + "src/transformers/cli/serve.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/cli/serve.py", + "left_ranges": [ + [ + 11, + 18 + ], + [ + 359, + 429 + ], + [ + 584, + 590 + ], + [ + 1892, + 1910 + ], + [ + 1917, + 1923 + ] + ], + "right_ranges": [ + [ + 54, + 61 + ], + [ + 587, + 593 + ] + ] + } + ] + } + ] + }, + { + "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-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": 0.532, + "size_similarity": 0.619, + "file_overlap": 0.5, + "area_overlap": 0.543, + "patch_similarity": 0.451, + "shared_filenames": [ + "src/transformers/testing_utils.py", + "tests/utils/test_testing_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/testing_utils.py", + "left_ranges": [ + [ + 3529, + 3536 + ], + [ + 3753, + 3787 + ] + ], + "right_ranges": [ + [ + 3525, + 3558 + ], + [ + 3782, + 3788 + ] + ] + }, + { + "filename": "tests/utils/test_testing_utils.py", + "left_ranges": [ + [ + 1, + 86 + ] + ], + "right_ranges": [ + [ + 1, + 114 + ] + ] + } + ] + } + ] + }, + { + "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 + ] + ] + } + ] + } + ] + } + ], + "duplicate_issues": [], + "duplicate_prs": [ + { + "cluster_id": "cluster-41115-3", + "canonical_pr_number": 45613, + "duplicate_pr_numbers": [ + 41116 + ], + "target_issue_number": 41115, + "reason": "PRs in cluster-41115-3 are treated as duplicates because they converge on issue #41115 with closing_reference evidence." + }, + { + "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-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-43824-3", + "canonical_pr_number": 43836, + "duplicate_pr_numbers": [ + 43842 + ], + "target_issue_number": 43824, + "reason": "PRs in cluster-43824-3 are treated as duplicates because they converge on issue #43824 with closing_reference 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-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." + } + ], + "best_issue": { + "cluster_id": "cluster-45647-3", + "issue_number": 45647, + "reason": "Issue #45647 is the strongest global issue candidate because it is open, belongs to a cluster with 3 artifacts, and carries 12 discussion comments plus 8 inbound references.", + "score": 162.0 + }, + "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/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..61a620d562187f8c9d10fa17486998da18bc06e4 --- /dev/null +++ b/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,19640 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260515T180026Z", + "generated_at": "2026-05-15T18:10:39Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43425", + "issue:43541", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43792", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43901", + "issue:43975", + "issue:44292", + "issue:44322", + "issue:44457", + "issue:44509", + "issue:44512", + "issue:44560", + "issue:44805", + "issue:44841", + "issue:44863", + "issue:44918", + "issue:45081", + "issue:45161", + "issue:45237", + "issue:45464", + "issue:45507", + "issue:45685", + "issue:45812", + "issue:45823", + "issue:45859", + "issue:45874", + "issue:45901", + "issue:45902", + "issue:45920" + ], + "soft_pairs": [ + "issue:44322|issue:45464", + "issue:45237|issue:45685", + "issue:43716|issue:45237", + "issue:43828|issue:45237", + "issue:45237|issue:45507", + "issue:44292|issue:45237", + "issue:30277|issue:35545", + "issue:35545|issue:42915", + "issue:35545|issue:41720", + "issue:44509|issue:44512", + "issue:35545|issue:36010", + "issue:44805|issue:45823", + "issue:43638|issue:44805", + "issue:43596|issue:44805", + "issue:44805|issue:45161", + "issue:44805|issue:44918", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:34928|issue:44805", + "issue:24643|issue:30277", + "issue:45859|issue:45902", + "issue:43638|issue:45161", + "issue:34928|issue:36331", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:39290|issue:45081", + "issue:45823|issue:45874", + "issue:43425|issue:44292", + "issue:43828|issue:45161", + "issue:34928|issue:41720", + "issue:44918|issue:45161", + "issue:43716|issue:45685", + "issue:43541|issue:43716", + "issue:43828|issue:44292", + "issue:45507|issue:45685", + "issue:43716|issue:44292", + "issue:44292|issue:44918", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43716|issue:44560", + "issue:43828|issue:45685", + "issue:44560|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:43039|issue:44805", + "issue:41762|issue:43596", + "issue:43901|issue:44512", + "issue:39290|issue:41720", + "issue:45237|issue:45901", + "issue:44841|issue:45823", + "issue:43039|issue:44560", + "issue:45812|issue:45920", + "issue:43866|issue:44863", + "issue:43854|issue:43866", + "issue:41093|issue:43638", + "issue:43792|issue:45901" + ], + "prepared_review_unit_hash": "184deb9bdd0407ffb4bdbdf8985a1fe3f9bfb0dc57d12b0b5b8ff8124647d47f", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 62, + "serialized_chars": 46368, + "estimated_input_tokens": 11592, + "estimated_eval_tokens": 23440 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous, with a few clear similarity pockets. The ZeRO-3/BERT items are very similar and likely the same bug family, but the evidence is not strong enough to call 43638 an exact duplicate of 43596; 43596 remains the best anchor for that subgroup. The tokenizer-ID regression pair is a strong duplicate candidate, and the mask-shape IndexError items look like the same family as well.", + "confidence": 0.63, + "canonical_issue_reason": "issue:43596 is the cleanest anchor for the ZeRO-3 empty-tensor/index-out-of-bounds subgroup: it is the most generic BERT + ZeRO-3 report, while 43638 and 41762 read as very similar neighbors rather than proven exact duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43596 is the best overall issue representative for the ZeRO-3 subgroup because it captures the shared failure mode with the broadest framing and sits at the center of the strongest similarity cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different failures: Qwen3.5 config attribute error vs streaming chat/completions behavior." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "AMD GPU load failure and MPS histogram Int backend error are unrelated backend-specific bugs." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Image preprocessor dtype mismatch is a different issue from the AMD GPU device-mapping failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-MoE is unrelated to the AMD GPU runtime failure." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different code paths: AMD GPU support failure vs GraniteMoEHybrid calling an invalid method." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Qwen NVFP4 runtime error is unrelated to the AMD GPU failure." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch and ModernBERT ONNX export are unrelated." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "ModernBERT export failure and Qwen3Moe FP8 failure are different bugs." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export error and Qwen3 auto-device-map cuda assert are unrelated." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs cleanup items, but they target different removed features." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ModernBERT export and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask/index shape mismatch and Gemma4 PLE device mismatch are different failures." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "ZeRO-3 empty-tensor IndexError and mask/tensor shape mismatch are different bug families." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Different root causes: ZeRO-3 loading failure vs mask-shape IndexError." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Mask-shape IndexError and GPT-OSS TP-only failure are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Unpacking embeddings with TRL SFT and the mask-shape IndexError are different bugs." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice preset and ModernBERT export are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Collective mismatch under ZeRO-3 is a different failure from the empty-tensor index-out-of-bounds bug." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": false, + "reason": "Very similar ZeRO-3/BERT IndexError reports, but not enough evidence to treat them as an exact duplicate." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross_entropy call and GPT-OSS TP-only failure are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "DeepSeek detokenization and LoRA merge/save mismatch are different bugs." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Qwen NVFP4 runtime failure and LoRA merge/save mismatch are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation-checkpointing/FSDP recompute mismatch is unrelated to the mask-shape IndexError." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "DeepSpeed weight-dimension error and collective mismatch are different failures." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Same model family, but CPU offload incompatibility and meta-to-empty uninitialized-memory are different bugs." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 loading failure and GPT-OSS TP-only failure are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "FSDP activation checkpointing mismatch and compute_loss signature change are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 loading failure and GPT-OSS TP-only failure are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Mixed-precision dtype mismatch and GraniteMoEHybrid invalid method are different bugs." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Gemma3 vLLM attribute regression and Mistral regex patch crash are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Gemma4 device-mapping mismatch and reverse embedding lookup are different issues." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Torch version incompatibility and Qwen NVFP4 runtime error are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Autocast dtype mismatch and GPT-OSS TP-only failure are different code paths." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "TRL SFT embedding unpacking failure is unrelated to the GPT-OSS TP issue." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Mixed-precision dtype mismatch and MPS histogram Int backend error are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Autocast dtype mismatch and Qwen NVFP4 runtime failure are different bugs." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Video StopIteration in Qwen3-vl-embedding is unrelated to the MPS histogram Int backend bug." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Liger Kernel cross_entropy call and autocast dtype mismatch are different issues." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and CustomTrainer compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation-checkpointing/FSDP bug and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross_entropy call and mask-shape IndexError are unrelated." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Strongly similar ZeRO-3 empty-tensor IndexError reports, but the available evidence does not prove they are the same exact bug." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Docs around return_all_scores and docs around the removed run command are separate cleanup items." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Gemma3 vLLM attribute regression and Qwen3 auto-device-map cuda assert are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "AMD GPU support failure and table-question-answering crash are different bugs." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross_entropy call and Qwen3-vl video StopIteration are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same silent v4\u2192v5 AutoTokenizer wrong-token-ID regression, reported across different model families." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Ovis2 checkpoint corruption and NemotronH checkpoint loading failure are unrelated." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Unit-test load failure and checkpoint corruption are not the same underlying bug." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": true, + "reason": "Same mask-shape IndexError template; likely the same masking/indexing bug family." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper load failure and table-question-answering crash are unrelated." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly reasonable, but it overstates the evidence for the ZeRO-3 subgroup. In the packet, 43596 and 43638 are strongly similar, but there is no explicit proof that 43638 is an exact duplicate of 43596; likewise, calling them the 'same concrete code path' is stronger than the available evidence supports. Please tone these claims down to 'very similar' or 'likely same bug family' rather than exact duplicate." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 2, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36010", + "issue:39186", + "issue:39290", + "issue:40990", + "issue:41093", + "issue:41762", + "issue:42915", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43645", + "issue:43716", + "issue:43792", + "issue:43828", + "issue:43866", + "issue:43975", + "issue:43994", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44530", + "issue:44589", + "issue:44841", + "issue:44918", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45161", + "issue:45362", + "issue:45399", + "issue:45464", + "issue:45542", + "issue:45632", + "issue:45685", + "issue:45698", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45901", + "issue:45902" + ], + "soft_pairs": [ + "issue:39290|issue:43531", + "issue:43425|issue:45070", + "issue:43596|issue:45800", + "issue:43975|issue:45901", + "issue:43792|issue:45070", + "issue:45362|issue:45464", + "issue:44292|issue:45901", + "issue:43404|issue:43866", + "issue:39290|issue:45823", + "issue:45464|issue:45685", + "issue:43541|issue:45685", + "issue:43541|issue:43828", + "issue:44841|issue:45084", + "issue:36010|issue:39290", + "issue:41093|issue:41762", + "issue:45800|issue:45823", + "issue:43296|issue:43366", + "issue:43572|issue:45070", + "issue:44387|issue:45005", + "issue:43541|issue:45161", + "issue:43828|issue:45464", + "issue:44291|issue:45070", + "issue:43366|issue:45464", + "issue:45020|issue:45800", + "issue:43531|issue:45070", + "issue:43541|issue:44589", + "issue:41093|issue:43596", + "issue:43054|issue:43994", + "issue:45542|issue:45800", + "issue:45759|issue:45902", + "issue:40990|issue:44960", + "issue:43716|issue:44918", + "issue:39186|issue:44918", + "issue:45632|issue:45698", + "issue:43404|issue:45005", + "issue:43645|issue:45399", + "issue:44530|issue:45005", + "issue:42915|issue:45399" + ], + "prepared_review_unit_hash": "affc09d5b1fde44ca16ec08628b0d3f70d3f30e30e357457030661e3bd868fcd", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 38, + "serialized_chars": 38767, + "estimated_input_tokens": 9692, + "estimated_eval_tokens": 19640 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of distinct model- and runtime-specific issues rather than true duplicates. The only clear soft duplicate here is the remote_code/custom-module loading pair; the rest share broad themes but not the same concrete bug path.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45020 is the broadest open report and the best umbrella for the cluster\u2019s strongest shared theme: recent-version regressions in remote_code model loading. The other issues are narrower or point to different failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the most suitable representative issue overall because it is broad, open, and framed around a user-visible regression that can subsume related custom-code loading problems.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both mention sliding-window behavior, but they involve different model families and different failure modes; not the same bug path." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "One is a torch-version compatibility report, the other is a pydantic/config regression; too different to merge." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "Both mention incompatibility-ish symptoms, but one is a ZeRO-3 index error and the other is torch/transformers version incompatibility." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization bugs and table-question-answering crashes are unrelated code paths and symptoms." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper runtime failure is not the same underlying issue as a PretrainedConfig/pydantic regression." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both touch Qwen3.5 chat usage, but one is a general chat crash and the other is a streaming chat/completions failure; different paths." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Qwen-3 NVFP4 loading and table QA crashes are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "A tied-weight bug in Mistral3 is not the same as a corrupted Ovis2 checkpoint report." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3 sliding-window config and Gemma4 device-map mismatch are different model issues." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Streaming inference on Qwen3.5 and an MPS histogram error on MoE models are unrelated." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Grouped_mm tracing failure and an MPS Int histogram limitation are different runtime problems." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Both mention dtype-ish symptoms, but one is torch dynamo tracing and the other is autocast mismatch; not the same bug." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor failure for Voxtral and a template-node compilation error do not share a concrete code path." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "Importing GenerationMixin and Gemma3 sliding-window config errors are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are shape/index errors, but on different tensors, models, and loading scenarios." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Torch version incompatibility and Gemma4 device mismatch are different failures." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are not the same bug." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is a different config regression from the pydantic PretrainedConfig breakage." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 memory/OOM regression and tied-weights translation-model issues are distinct." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Torch tracing failure and only-TP-not-working on GPT-OSS MoE are different implementation issues." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Autocast dtype mismatch for Phi-tiny-MoE is not the same as Qwen3.5 streaming API failure." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "init_empty_weights argument handling and pydantic field validation are different breakages." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF architecture support and chat/completions streaming errors are unrelated." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "remote_code loading regressions and a torch/transformers version incompatibility are too broad and different to treat as one bug." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window and a pydantic config regression do not share a concrete failure path." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "torch dynamo grouped_mm errors and missing Float8 storage are separate float/dtype issues." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are index errors, but one is a mask-length mismatch and the other is a zero-sized dimension during ZeRO-3 init." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Both concern SigLIP2 output quality, but one is a relative embedding-quality complaint and the other is a broader bad-results report; too imprecise to confirm same bug." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard-without-TF backend detection and torch/transformers incompatibility are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Both involve Qwen3.5/3.6 multimodal internals, but config unwrapping and uninitialized-memory materialization are different code paths." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "High perplexity on gpt-oss and a GLM5 issue are unrelated model-specific reports." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Image-preprocessor dtype mismatch and Qwen3.5 input-embedding unpacking in TRL are different failures." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP weight dimensionality errors and TRL embedding unpacking are not the same bug." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": true, + "reason": "Both describe the same remote_code/custom-module resolution bug: cache/path collision leads to the wrong custom module being loaded after save/load." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Mistral3 tied-lm_head weights and translation-model tied-weight regressions are related only at a very broad level, not the same concrete issue." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Jupyter custom-model initialization and flash-attn fallback checks are unrelated." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention failures and tied-weights translation issues are different subsystems." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "FineGrainedFP8Config failures and flash-attn fallback gating are unrelated runtime paths." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The single accepted soft pair (issue:45632 / issue:45698) is plausibly the same remote_code/custom-module loading bug, and the other rejections are appropriately cautious rather than overbroad. The representative choice of issue:45020 as the open umbrella issue is also defensible from the titles and descriptions provided." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 3, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36331", + "issue:38175", + "issue:39692", + "issue:41553", + "issue:41628", + "issue:41762", + "issue:42915", + "issue:42947", + "issue:43054", + "issue:43493", + "issue:43643", + "issue:43644", + "issue:43782", + "issue:43828", + "issue:43856", + "issue:43866", + "issue:43872", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:43950", + "issue:43975", + "issue:44292", + "issue:44360", + "issue:44387", + "issue:44403", + "issue:44451", + "issue:44488", + "issue:44492", + "issue:44512", + "issue:44534", + "issue:44589", + "issue:44863", + "issue:44910", + "issue:44918", + "issue:44928", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45237", + "issue:45507", + "issue:45520", + "issue:45663", + "issue:45701", + "issue:45800", + "issue:45859", + "issue:45902" + ], + "soft_pairs": [ + "issue:44360|issue:44512", + "issue:43873|issue:45005", + "issue:44863|issue:45005", + "issue:44492|issue:44512", + "issue:44918|issue:45507", + "issue:43975|issue:44292", + "issue:43950|issue:44534", + "issue:44292|issue:45507", + "issue:39692|issue:43054", + "issue:43644|issue:43950", + "issue:43828|issue:44928", + "issue:36331|issue:41762", + "issue:41628|issue:45800", + "issue:43866|issue:44403", + "issue:43950|issue:45902", + "issue:43872|issue:44589", + "issue:43054|issue:45701", + "issue:41553|issue:43883", + "issue:38175|issue:43054", + "issue:44910|issue:45859", + "issue:45663|issue:45902", + "issue:41553|issue:43881", + "issue:42947|issue:44387", + "issue:43054|issue:43493", + "issue:44403|issue:44451", + "issue:43643|issue:45020", + "issue:43975|issue:45520", + "issue:43975|issue:45237", + "issue:44403|issue:44488", + "issue:42947|issue:43856", + "issue:43881|issue:43883", + "issue:41762|issue:42915", + "issue:43782|issue:44960" + ], + "prepared_review_unit_hash": "99deff697ff1a1ab3d20c37c5f1316f24a69253a2a4b37287d26cbfa9ecfce4b", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 36584, + "estimated_input_tokens": 9146, + "estimated_eval_tokens": 18548 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Only one soft duplicate looks convincing: the Transformers v5 non-persistent buffer corruption regression. The other title-similar pairs appear to be different model/load-path issues and shouldn\u2019t be merged.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43950 is the broadest, most concrete report of the v5 non-persistent-buffer corruption regression; issue:44534 is the same symptom restated more generally.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43950 best represents the underlying bug because it names the regression directly and covers the fix surface more clearly than the narrower duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "Different problems: DSA indexer ReLU discussion vs a v5 docs/run-command note." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Both mention quantization/tied weights, but the concrete failures are different and not the same bug." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading failure is unrelated to tied-weights behavior in translation models." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "A typo in cache strategies is not the same issue as stale v5 docs about a removed command." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Different model/code paths: Qwen3.5 embedding unpacking vs GraniteMoEHybrid invalid method call." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Different concrete failures: detokenization bug vs NVFP4 runtime error." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same underlying v5 non-persistent buffer corruption regression; 44534 is a reworded symptom of 43950." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "No shared underlying bug: Qwen-3 NVFP4 execution error vs GraniteMoEHybrid method invocation problem." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "SigLIP2 docs-example errors are not the same as a broad embedding-quality regression." + }, + { + "left": "issue:43644", + "right": "issue:43950", + "accept": false, + "reason": "Remote-code config field loss is a different failure mode than non-persistent buffer corruption." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Autocast dtype mismatch and Qwen3.5 3D position_ids/SDPA issues are different code-path bugs." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Trainer signature mismatch vs ZeRO-3 load failure; no concrete shared root cause." + }, + { + "left": "issue:41628", + "right": "issue:45800", + "accept": false, + "reason": "Import error for AutoImageProcessor is not the same as a torch/transformers version incompatibility." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Corrupted checkpoint vs generic load noise; not the same underlying bug." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Both touch buffer initialization, but 45902 is a specific Qwen3_5Moe rotary-embedding materialization bug, not the same regression as 43950." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Different quantization/backing-storage failures with different symptoms and code paths." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 embedding quality issue is not the same as tokenizer behavior changing across versions." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Bad AutoTokenizer error for Voxtral is unrelated to Molmo tied-weights metadata." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Both concern SigLIP2, but zero probabilities and worse embeddings are not clearly the same concrete bug." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "FlashAttention illegal access from 3D position_ids is different from CPU-offload incompatibility in rotary embedding." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "FSDP2 shared_kv_states KeyError is unrelated to uninitialized rotary-embedding memory after materialization." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "AutoTokenizer loading error for Voxtral and glm-4v-9b loading failure are distinct issues." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Gradient checkpointing/LoRA ineffectiveness is not the same as int4 quantization causing higher reserved memory." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Poor SigLIP2 embeddings and HF-vs-JAX discrepancy are related themes but not the same concrete report." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic loading noise is not the same as an inability to load a specific model." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "Remote-code config field loss may be part of a broader class, but 45020 is too umbrella-like to treat as the same bug." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "DeepSeek detokenization and flash_attn import errors are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45237", + "accept": false, + "reason": "Detokenization bug and AMD GPU execution failure are different problems." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Generic load noise is not the same as a specific model failing to load." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "Both mention memory, but one is checkpointing/LoRA behavior and the other is MoE training memory usage." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Model loading failure and missing all_tied_weights_keys are different issues." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading error and Qwen3Moe FP8 failure are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL weight_only load error is not the same as a GLM5 issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The one accepted soft duplicate pair (issue:43950 / issue:44534) is well supported by the titles and clearly points to the same Transformers v5 non-persistent-buffer corruption regression. The remaining pairs are appropriately rejected based on differing bug surfaces, and the summary does not overstate the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 4, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43366", + "issue:43404", + "issue:43638", + "issue:43742", + "issue:43792", + "issue:43824", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43940", + "issue:43950", + "issue:44220", + "issue:44291", + "issue:44368", + "issue:44479", + "issue:44488", + "issue:44509", + "issue:44530", + "issue:44560", + "issue:44661", + "issue:44805", + "issue:44863", + "issue:44928", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45305", + "issue:45313", + "issue:45325", + "issue:45356", + "issue:45399", + "issue:45538", + "issue:45588", + "issue:45685", + "issue:45823", + "issue:45902", + "issue:45923" + ], + "soft_pairs": [ + "issue:45685|issue:45823", + "issue:42491|issue:44368", + "issue:43824|issue:43872", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:44991|issue:45399", + "issue:43828|issue:45823", + "issue:45081|issue:45399", + "issue:43827|issue:44509", + "issue:45020|issue:45356", + "issue:43404|issue:43950", + "issue:43742|issue:45399", + "issue:42915|issue:45020", + "issue:43638|issue:45823", + "issue:43257|issue:43866", + "issue:43792|issue:44220", + "issue:43872|issue:44291", + "issue:42915|issue:43872", + "issue:43940|issue:45313", + "issue:41553|issue:44991", + "issue:43854|issue:44863", + "issue:36331|issue:41093", + "issue:43257|issue:45305", + "issue:42947|issue:45923", + "issue:44928|issue:45588", + "issue:41093|issue:44805", + "issue:43872|issue:45538", + "issue:41553|issue:44488", + "issue:44530|issue:44863", + "issue:44479|issue:44560", + "issue:24643|issue:30064", + "issue:45325|issue:45902", + "issue:43366|issue:43828" + ], + "prepared_review_unit_hash": "335d8be8f0c5c2f691e3e1e26decf1fc959508aa0dd29081abbce0ae0be141f0", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 38069, + "estimated_input_tokens": 9518, + "estimated_eval_tokens": 19292 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a collection of unrelated model-specific regressions and docs issues. Only a few soft pairs look like true duplicates: the removed-pipeline docs references, the `_is_hf_initialized` TypeError, and the repeated mask-shape IndexError.", + "confidence": 0.81, + "canonical_issue_reason": "Issue 45020 is the broadest and most representative umbrella bug in the set: recent transformers versions breaking `remote_code` models across multiple concrete failures. It is a better anchor than the narrower model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45020 is the best global issue candidate because it captures the widest recurring regression pattern and is the most suitable umbrella triage target among mostly narrow one-off bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "Different failures: an MPS histogram dtype limitation in MoE code vs a Gemma4 PLE forward device-mismatch bug." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "Both mention Qwen3.5/Qwen3 MoE, but one is a loadability regression across hf4.x/hf5.x and the other is only a warning about `tie_word_embeddings`." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Import error for a missing Qwen2.5-VL class is unrelated to the bitsandbytes `_is_hf_initialized` constructor incompatibility." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "`remote_code` breakage is a broad loader regression, while Gemma4 PLE device placement is a model-specific forward-path issue." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "Tokenizer-mapping edge case for `add-new-model-like` is a different code path from remote-code model loading failures." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading for `est-roberta` is unrelated to flash-attn fallback gating checks." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch in Phi-tiny-MoE is not the same bug as Gemma4 PLE device mismatch." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Mistral regex patch crashing in tokenizer setup is separate from flash-attn fallback being blocked by other checks." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Both report the same doc cleanup needed after v5 pipeline removals; same outdated pipeline-task references." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "General remote_code regressions are not the same as Kimi-K2.5 tokenizer codec-handling and warning behavior." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Untied `lm_head` weights in Mistral3 are unrelated to non-persistent buffer corruption in `from_pretrained()`." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "A model-specific key error while loading MobileLLM-125M is different from flash-attn fallback selection logic." + }, + { + "left": "issue:42915", + "right": "issue:45020", + "accept": false, + "reason": "LoRA compatibility across hf4.x/hf5.x is a specific Qwen3 MoE issue, not the broader remote_code regression." + }, + { + "left": "issue:43638", + "right": "issue:45823", + "accept": false, + "reason": "DeepSpeed ZeRO-3 indexing failure on a non-pretrained Bert model is unrelated to Gemma4 PLE device placement." + }, + { + "left": "issue:43257", + "right": "issue:43866", + "accept": false, + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed and a corrupted Ovis2 checkpoint are distinct load failures." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper not running is too broad to merge with a specific `_torch_extract_fbank_features()` bug without clearer same-path evidence." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Both center on the same `_is_hf_initialized` TypeError during model materialization/loading, just from different entry points." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "Qwen3 MoE LoRA compatibility and the `_is_hf_initialized` constructor error are different regressions." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both are ZeRO-3 load problems on Qwen variants, but they appear to be model-specific missing-weight failures rather than one concrete duplicate bug." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "A bad AutoTokenizer error message for Voxtral is not the same as failing to load EMBEDDIA/est-roberta." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "GLM-4.7-Flash unit-test loading failure and NemotronH checkpoint loading failure are separate model-specific loader bugs." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Unexpected `num_items_in_batch` in custom trainer loss is unrelated to the mask/tensor shape mismatch IndexError." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Qwen3 MoE weight conversion with deepspeed is not the same as gradient averaging missing under GAS/model_accepts_loss_kwargs." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "LoRA gradient checkpointing being ineffective is not the same issue as a missing `supports_gradient_checkpointing` flag on a trust_remote_code model." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "RLHF NaN explosions from Qwen3.5 SDPA/BF16 behavior are unrelated to flash_attention.py crashing on `s_aux=None`." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": true, + "reason": "These are the same mask-shape mismatch IndexError, with only the reported sizes/context differing." + }, + { + "left": "issue:43872", + "right": "issue:45538", + "accept": false, + "reason": "The `_is_hf_initialized` loading bug is unrelated to CLIPTokenizer's `model_max_length` sentinel value." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Both are tokenizer-loading complaints, but the Voxtral error message issue and the sleng-bert load failure are not the same bug." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3.5 PagedAttentionCache group-type handling and NemotronH checkpoint loading are different subsystems and failures." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Both are video-related, but one is a broad v5.3.0 regression across several Qwen models and the other is a specific StopIteration on Qwen3-vl-embedding." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "DeepSpeed training weight-dimension error and void segmentation-map processing are unrelated." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Qwen2.5-VL rope-index scaling and Qwen3.5 MoE rotary-embedding uninitialized-memory issues are distinct model internals." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "GGUF gpt-oss architecture support and Phi-tiny-MoE autocast dtype mismatch are separate problems." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet, and the soft-edge decisions are conservative. The accepted pairs are plausible same-bug duplicates/docs-cleanup matches, while the rejected pairs avoid overclaiming broader similarity. The choice of issue 45020 as the umbrella issue is also supported by its broad \u2018remote_code\u2019 regression scope." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 5, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36010", + "issue:38175", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42617", + "issue:42915", + "issue:43317", + "issue:43366", + "issue:43531", + "issue:43550", + "issue:43596", + "issue:43632", + "issue:43638", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43720", + "issue:43824", + "issue:43854", + "issue:43931", + "issue:43950", + "issue:43957", + "issue:43994", + "issue:44075", + "issue:44155", + "issue:44451", + "issue:44488", + "issue:44534", + "issue:44560", + "issue:44589", + "issue:44661", + "issue:44756", + "issue:44805", + "issue:44877", + "issue:44910", + "issue:44918", + "issue:44960", + "issue:44977", + "issue:45030", + "issue:45276", + "issue:45313", + "issue:45357", + "issue:45584", + "issue:45663", + "issue:45823", + "issue:45834" + ], + "soft_pairs": [ + "issue:45663|issue:45823", + "issue:44155|issue:45584", + "issue:43366|issue:43531", + "issue:43950|issue:45834", + "issue:43550|issue:44910", + "issue:43824|issue:43957", + "issue:43645|issue:43646", + "issue:43366|issue:44589", + "issue:42915|issue:43957", + "issue:44560|issue:44918", + "issue:42617|issue:43366", + "issue:36010|issue:42915", + "issue:44661|issue:45823", + "issue:38175|issue:43994", + "issue:43632|issue:45834", + "issue:44560|issue:44805", + "issue:44877|issue:45030", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:44661|issue:44805", + "issue:43317|issue:44756", + "issue:43644|issue:44534", + "issue:38175|issue:39692", + "issue:44960|issue:44977", + "issue:42491|issue:43720", + "issue:45313|issue:45357", + "issue:41720|issue:42915", + "issue:44451|issue:44488", + "issue:42491|issue:45276", + "issue:42915|issue:44756", + "issue:43708|issue:44075", + "issue:43854|issue:43931" + ], + "prepared_review_unit_hash": "ef2ac56870a0ff8745e39c714ac80192bfb5c59702f7e9a3cc3030108184900d", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37193, + "estimated_input_tokens": 9299, + "estimated_eval_tokens": 18854 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a mixed cluster of unrelated Transformers regressions, with a few clear duplicate pairs around the same bug family. The strongest duplicate subgroup is the non-persistent buffer corruption in Transformers 5.x.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 43950 is the clearest, most technical statement of the broad 5.x regression and covers the same non-persistent-buffer corruption described in nearby duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43950 is the best canonical issue because it names the exact failing behavior, has broad applicability, and matches the duplicate subgroup more precisely than the shorter title variants.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4 issues, but one is an FSDP2 shared_kv_states bug and the other is a device_map=\"auto\" PLE device-mismatch bug; different code paths." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "Different models and failures: AudioFlamingo3 track leakage vs Whisper empty-transcription handling." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "Unrelated problems: GGUF gpt-oss support versus Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Different subsystems entirely: non-persistent buffer corruption vs Kosmos2.5 long-OCR index error." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Both touch attention/compile paths, but the model, trigger, and failure mode are different; not the same concrete bug." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Import/export issue for Qwen2.5-VL class vs meta-device loading regression; not the same underlying defect." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Same custom-model initialization regression in Transformers 5.0.0, with nearly identical symptoms and scope." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF gpt-oss support and Float8 storage lookup errors are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Qwen3Moe FP8 failure and meta-device loading are different bugs." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Both involve Qwen3.5-family features, but one is video/StopIteration and the other is TRL embedding unpacking; not the same bug." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "3d_parallel.py runtime failure is unrelated to GGUF architecture support." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "Importing GenerationMixin from transformers.generation is unrelated to Qwen3Moe FP8 behavior." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping/model-registration failure and Gemma4 device mismatch are different issues." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Same SigLIP2 model producing bad outputs on the same checkpoint, with closely aligned symptoms." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized breakage is unrelated to Kosmos2.5 OCR index errors." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different failure modes and model contexts; no shared concrete code path." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both are config-validation complaints, but they affect different model families and appear to be separate regressions." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "DeepSpeed ZeRO-3 Bert initialization and add-new-model-like tokenizer mapping are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Zero3/BertModel index error is a different path from the tokenizer-mapping registration bug." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer-mapping registration and mask-shape IndexError are not the same bug." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Both relate to loading/memory behavior, but one is gpu+cpu offload dequantized loading and the other is Strix Halo mmap OOM; too different to merge." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "These are the same Transformers v5 non-persistent-buffer corruption regression, just phrased differently." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same model family, but the reported problems differ: wrong probabilities vs documentation example/model-processor mismatch and quantization failures." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "GLM5 and Qwen3.5 flash-attention failures are unrelated." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Qwen3-MoE LoRA compatibility on hf5.x and BitNet packed-weight loading are different bugs." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5 issues, but one is ZeRO-3 weight loading and the other is save_pretrained key serialization; distinct code paths." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Auto device mapping cudaErrorAssert and FineGrainedFP8Config failure are different Qwen3-family bugs." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are model-load failures for different BERT checkpoints, but the available evidence only shows a broad symptom match, not a concrete shared root cause." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "LoRA hf5.x compatibility and Gemma4 resize_token_embeddings behavior are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Strix Halo mmap OOM are separate issues." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Trainer resume/max_steps calculation and SGD argument handling are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative about the mixed cluster. The canonical issue choice is well supported by the non-persistent-buffer regression subgroup, and the accepted soft-edge pairs all look like plausible same-bug duplicates/mergeable fix targets. No obvious overstatement in the negative verdicts." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 6, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43335", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43644", + "issue:43645", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44488", + "issue:44743", + "issue:44756", + "issue:44910", + "issue:44912", + "issue:44960", + "issue:44991", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:45488|issue:45812", + "issue:39692|issue:43994", + "issue:43257|issue:43854", + "issue:44488|issue:44991", + "issue:42491|issue:43931", + "issue:43856|issue:43873", + "issue:43644|issue:43645", + "issue:45362|issue:45406", + "issue:43856|issue:44387", + "issue:44910|issue:45588", + "issue:42491|issue:45292", + "issue:45276|issue:45335", + "issue:44164|issue:44291", + "issue:36010|issue:39404", + "issue:43404|issue:43454", + "issue:41720|issue:44155", + "issue:44402|issue:44488", + "issue:39692|issue:44960", + "issue:43425|issue:43957", + "issue:42947|issue:45704", + "issue:45663|issue:45759", + "issue:43335|issue:45663", + "issue:43873|issue:44387", + "issue:45663|issue:45859", + "issue:44291|issue:44756", + "issue:44155|issue:45823", + "issue:44292|issue:44912", + "issue:44387|issue:44756", + "issue:43854|issue:45216", + "issue:44743|issue:44960", + "issue:43957|issue:44756", + "issue:45276|issue:45292", + "issue:45684|issue:45698" + ], + "prepared_review_unit_hash": "91a222db6499e3e33eaded5c536a325a04a9e077cc97554153984314deec5032", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37771, + "estimated_input_tokens": 9443, + "estimated_eval_tokens": 19142 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a loose collection of unrelated Transformers bug reports spanning model loading, tokenization, quantization, and model-specific regressions. Only one clear duplicate pair stands out: the resize_token_embeddings/output_embeddings bug.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45292 is the broader duplicate target: it describes the generic resize_token_embeddings/output_embeddings failure, while issue:45276 is a Gemma4-specific manifestation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the best representative because it is the most general statement of the duplicate resize_token_embeddings bug and subsumes the narrower Gemma4 report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but one is DeepSeek/LlamaTokenizer pre-tokenizer behavior and the other is Granite token IDs; different failures and likely different fixes." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both involve SigLIP2, but one is a docs/example mismatch with quantization errors and the other is incorrect inference results in AutoModel/pipeline." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different models and different loading problems: Qwen3 MoE weight conversion under accelerate/deepspeed vs GLM-4.7-Flash loading in unit tests." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer loading failures, but they concern different models and no shared concrete bug is evident." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Unrelated issues: Qwen3 MoE LoRA compatibility across branches vs Qwen3-VL weight shape mismatch." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Both mention memory, but one is Qwen3 MoE training inefficiency and the other is quantization/offloading behavior." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Both are Transformers 5.0 regressions, but one is junk-filled non-persistent buffers and the other is custom-model initialization in Jupyter." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different model families and different runtime failures: Qwen3.5 chat crash vs Gemma4Processor missing _tokenizer in serve." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both are memory-related, but the causes differ: MoE training inefficiency vs int4 quantization reserved-memory OOM." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Different flash-attention bugs: Qwen3.5 position_ids misinterpretation vs s_aux=None AttributeError on sink-less models." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "LoRA branch compatibility is unrelated to the resize_token_embeddings/output_embeddings bug." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": false, + "reason": "Both are resize_token_embeddings issues, but they hit different model implementations and different target modules, so the underlying bug is not clearly the same." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Different save/load paths: extra_state handling vs init_empty_weights argument mismatch." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "Unrelated import error vs Whisper pipeline return_language regression." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both are weight-tying bugs, but they are reported for different multimodal model classes and are not clearly the same concrete code-path." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Different subsystems and symptoms: Qwen3 device-map CUDA assert vs AudioFlamingo3 batched inference token leak." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "One is a tokenizer vocab-size mismatch, the other is a tokenizer load failure; related theme but not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Both touch model families with similar naming, but the bugs are about docs/example errors vs a GLM5-specific issue." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Different issues: Torch 2.10 compatibility vs meta-device model loading regressions." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing with LoRA is unrelated to the apex.FusedRMSNorm memory leak in T5." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Different bugs: Gemma4 FSDP2 shared_kv_states failure vs Qwen 3.5/3.6 multimodal config unwrapping." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "SwitchTransformers sparse-layer configuration is unrelated to Gemma4 FSDP2 attention state handling." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Different memory regressions with different triggers: quantization offloading vs int4 CUDA reserved-memory growth." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Gemma4 FSDP2 attention state bug is unrelated to Qwen3.5 MoE CPU offload incompatibility." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "TypeError from init_empty_weights is unrelated to disabling mmap for Strix Halo OOM." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo3 track leakage and Gemma4 PLE device mismatch are different code paths." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Different quantization problems: NVFP4 loading error vs MXFP4 fallback to bf16." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Both are OOM-related, but one is int4 quantization memory growth and the other is mmap-specific Strix Halo OOM." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Different regressions: GLM-4.7-Flash loading in tests vs Qwen3.5 save_pretrained checkpoint corruption." + }, + { + "left": "issue:44743", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3 modular recurrent-state caching bug is unrelated to the generic GLM5 issue." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading failures and Strix Halo mmap OOM are unrelated." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same underlying resize_token_embeddings bug: both report output embeddings not being updated, with 45276 being the Gemma4-specific reproduction and 45292 the broader generic report." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both concern save/load of custom models, but one is read-only file permissions from save_pretrained and the other is loading the wrong custom module after save_pretrained." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only accepted soft-edge pair, issue:45276 vs issue:45292, is well supported by the nearly identical resize_token_embeddings/output_embeddings wording. The other pairwise rejections are cautious and the narrative does not overclaim broader duplication." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 7, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36032", + "issue:39404", + "issue:42947", + "issue:43257", + "issue:43299", + "issue:43381", + "issue:43425", + "issue:43452", + "issue:43596", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43901", + "issue:43906", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44671", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44877", + "issue:44898", + "issue:44912", + "issue:44918", + "issue:44928", + "issue:44991", + "issue:45084", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45701", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:42947|issue:43381", + "issue:44479|issue:44877", + "issue:43646|issue:43708", + "issue:45588|issue:45663", + "issue:45292|issue:45335", + "issue:43645|issue:44164", + "issue:44402|issue:44451", + "issue:43257|issue:44863", + "issue:43828|issue:43957", + "issue:43425|issue:43828", + "issue:44589|issue:44756", + "issue:44683|issue:45823", + "issue:39404|issue:45701", + "issue:43931|issue:44863", + "issue:43906|issue:44530", + "issue:43425|issue:45823", + "issue:43611|issue:43645", + "issue:43425|issue:43716", + "issue:44530|issue:45084", + "issue:39404|issue:44912", + "issue:43901|issue:44509", + "issue:44265|issue:44898", + "issue:44479|issue:44805", + "issue:43994|issue:44079", + "issue:43299|issue:43931", + "issue:43425|issue:44918", + "issue:39404|issue:44671", + "issue:44589|issue:45084", + "issue:44164|issue:45759", + "issue:36032|issue:43452", + "issue:43596|issue:44683", + "issue:44671|issue:44991" + ], + "prepared_review_unit_hash": "d0f3842138cdd2f3be79f405fd829a1a71bf305138a745cd48aa01b7914effe0", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37972, + "estimated_input_tokens": 9493, + "estimated_eval_tokens": 19242 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is dominated by broad but mostly unrelated Transformers regressions (model loading, dtype/device mismatches, docs, training edge cases). Only one pair looks like a true duplicate: the resize_token_embeddings propagation bug.", + "confidence": 0.83, + "canonical_issue_reason": "issue:45292 is the broader, model-agnostic report of the resize_token_embeddings bug, while issue:45335 is the T5Gemma-specific manifestation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the clearest and most general representative of the one confirmed duplicate pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Different bugs: gradient checkpointing in eval mode vs Qwen3.5 RLHF NaN/SDPA fallback instability." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated failures: extra_state save/load handling vs mmap/OOM behavior on Strix Halo." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is LoRA inefficiency and the other is eval-mode incompatibility." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Different model/config regressions: Qwen VL video-input path vs strict config blocking granite_speech loading." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model init in notebooks is unrelated to Trainer resume/max_steps calculation." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Different code paths and symptoms: flash_attention s_aux handling vs Gemma4 FSDP2 shared_kv_states rebuild." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same underlying bug: resize_token_embeddings does not propagate to the active embedding/output embedding; 45335 is the T5Gemma-specific case." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Notebook custom-model initialization is unrelated to extra_state save/from_pretrained handling." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer vocab-size issue for one model is not the same as a general load failure for another model." + }, + { + "left": "issue:43257", + "right": "issue:44863", + "accept": false, + "reason": "Both are loading problems, but they affect different architectures and failure mechanisms (Qwen3 MoE conversion vs NemotronH checkpoint loading)." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Autocast dtype mismatch is a different problem from meta-device loading failures." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version incompatibility is not the same as a Phi-tiny-MoE autocast dtype mismatch." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Float8 storage lookup failure is unrelated to Strix Halo mmap/OOM behavior." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Different subsystems: compiled flex_attention on torch>=2.9 vs Gemma4 PLE device placement." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper return_language pipeline regression is not the same as general tokenization changes." + }, + { + "left": "issue:43931", + "right": "issue:44863", + "accept": false, + "reason": "Different model families and symptoms: Qwen3-VL shape mismatch vs NemotronH checkpoint loading." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "Isolated reproduction of an earlier issue is not the same as the Qwen3.5 PagedAttentionCache crash." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Torch compatibility issue does not match Gemma4 forward-device mismatch." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Base_model_prefix loading regression is different from notebook-defined custom model initialization." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch version incompatibility is unrelated to Mistral-3 image preprocessor/model dtype mismatch." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttentionCache group-type crash is unrelated to template-node compilation errors." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline language return regression is unrelated to MXFP4 quantization fallback." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs issues, but they cover different obsolete pipeline/task mentions." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to Perceiver non-default-resolution inference." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Different regressions: Qwen VL video-input path vs generic mask/index shape mismatch." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Nonsensical model outputs are not the same bug as incorrect ModelOutput key assignment." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both concern Qwen3-VL loading, but one is a MoE/version regression and the other is a specific weight-shape mismatch." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch compatibility is unrelated to Qwen3.5 embedding unpacking with trl SFT trainer." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline behavior and CamemBERT masked-LM predictions are different code paths." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage deserialization failure is unrelated to non-template-node compilation." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state handling and AutoModelForCausalLM.from_config text_config unwrapping are different loading bugs." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "Tokenizer special-token conflict is unrelated to gguf_file loading behavior." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Deepspeed ZeRO3/BertModel index error is unrelated to compiled flex_attention on newer torch." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT masked-LM predictions and tokenizer load failure for EMBEDDIA/est-roberta are different issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The judgments are well grounded in the packet and stay conservative. The only accepted pair, issue:45292 / issue:45335, is supported by the titles as the same resize_token_embeddings propagation bug with a model-specific manifestation. The rest of the soft-edge rejections are cautious and not overstated, and the cluster summary matches the mostly heterogeneous 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": 17, + "review_unit_index": 8, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42915", + "issue:43089", + "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:43856", + "issue:43883", + "issue:44164", + "issue:44206", + "issue:44220", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44530", + "issue:44661", + "issue:44683", + "issue:44737", + "issue:44756", + "issue:44811", + "issue:44849", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45507", + "issue:45584", + "issue:45859", + "issue:45987" + ], + "soft_pairs": [ + "issue:32090|issue:39404", + "issue:43856|issue:44756", + "issue:44991|issue:45081", + "issue:44737|issue:45859", + "issue:44164|issue:44479", + "issue:43582|issue:43632", + "issue:44488|issue:45081", + "issue:44661|issue:44683", + "issue:40990|issue:43421", + "issue:44488|issue:44987", + "issue:43611|issue:43644", + "issue:44206|issue:44220", + "issue:43716|issue:45084", + "issue:43883|issue:44488", + "issue:43643|issue:44530", + "issue:36032|issue:42915", + "issue:44811|issue:45584", + "issue:34567|issue:40444", + "issue:44464|issue:45987", + "issue:43644|issue:43646", + "issue:44451|issue:44987", + "issue:43452|issue:43482", + "issue:43541|issue:45084", + "issue:42915|issue:44530", + "issue:43317|issue:43482", + "issue:44849|issue:45325", + "issue:36032|issue:43645", + "issue:44987|issue:44991", + "issue:43883|issue:44991", + "issue:43404|issue:44466", + "issue:43089|issue:45987", + "issue:45084|issue:45507" + ], + "prepared_review_unit_hash": "8664bfbb92e3c2987d4422374eafded9829395d2986d47fe25538558739bc4ab", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37267, + "estimated_input_tokens": 9317, + "estimated_eval_tokens": 18890 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous and the soft evidence does not justify any duplicate merges. In particular, the two most tempting pairs, 44206/44220 and 43452/43482, are only thematically similar; the packet marks both as non-duplicates, so they should not be treated as confirmed dupes.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43611 is the broadest representative of the v5 loading/init regression subgroup, though it does not capture the whole mixed cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best single representative because it sits in the largest coherent subgroup of v5-related loading regressions.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Different subsystems: Trainer GPU broadcast TypeError vs Whisper pipeline return_language regression." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated problems: Qwen3 MoE training memory usage vs Strix Halo mmap OOM avoidance." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-loading failures, but the models and failure modes differ." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both involve device/offload themes, but the actual code paths are different." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to the Qwen video-input regression." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon caching allocator warmup TypeError is unrelated to the _is_hf_initialized regression." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-load bugs, but for different models and different root causes." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Model-mapping logic and compiled flex_attention are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity on gpt-oss-20b is unrelated to runtime post-processor updates in TokenizersBackend." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are model-load failures, but there is no concrete shared bug or code path." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but base_model_prefix loading and non-persistent buffer junk are distinct bugs." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "Same audio feature-extraction area, but the packet does not support a single shared underlying bug." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Mistral image-preprocessor dtype mismatch is unrelated to the template compilation error." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Missing all_tied_weights_keys in Molmo is unrelated to a BERT tokenizer load failure." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "AutoConfig trust_remote_code field loss is unrelated to Qwen3.5 PagedAttentionCache linear_attention failures." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "T5 tokenizer special-token conflict is unrelated to Qwen3 MoE FineGrainedFP8Config failures." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both are Whisper-related, but they describe different bugs and code paths." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counting and Qwen2.5-VL multi-image finetuning are unrelated." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Chunked generation inconsistency and StaticCache.get_seq_length return-type mismatch are not the same issue." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 initialization regressions, but the concrete failures differ." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Different model-loading regressions for different checkpoints; no shared underlying code path is evident." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both are GGUF/loading regressions in v5, but the titles point to different failure modes and models." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Grouped_mm tracing failure and non-template node compilation error are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Different Qwen runtime issues: FP8 training config vs PagedAttentionCache linear_attention crash." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Dequantized model offload failure is unrelated to GGUF loading regression." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Qwen3.5 hidden_states behavior and Qwen2.5-VL rope-index scaling are different model-specific bugs." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer special-token conflict is unrelated to Jupyter custom-model initialization." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both fail on model loading, but they concern different models and different failure modes." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Molmo tied-weights attribute error is unrelated to loading EMBEDDIA/est-roberta." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Mistral3 lm_head tying and device-dependent serialization are related to weights, but not the same concrete bug." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation overhead from sync/dispatch and StaticCache seq-length type mismatch are different issues." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Template compilation TypeError and GraniteMoEHybrid invalid method call are unrelated." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary overstates the evidence. In the packet, every soft-pair verdict is deterministic_accept=false, including the two pairs the summary calls true duplicates (`44206/44220` and `43452/43482`). That makes the claim that these are duplicates unsupported. The rest of the summary is broadly conservative, but this specific conclusion is too strong for the provided evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 9, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43262", + "issue:43296", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43540", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43819", + "issue:43994", + "issue:44062", + "issue:44079", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44479", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44811", + "issue:44877", + "issue:44987", + "issue:45081", + "issue:45084", + "issue:45127", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45542", + "issue:45663" + ], + "soft_pairs": [ + "issue:43262|issue:43377", + "issue:43994|issue:44610", + "issue:43646|issue:44534", + "issue:43322|issue:43540", + "issue:43645|issue:44534", + "issue:44403|issue:45081", + "issue:43377|issue:43819", + "issue:42915|issue:45084", + "issue:44265|issue:44610", + "issue:44451|issue:45081", + "issue:45216|issue:45357", + "issue:44987|issue:45081", + "issue:44220|issue:44683", + "issue:43404|issue:45127", + "issue:42915|issue:44291", + "issue:43632|issue:44756", + "issue:44062|issue:44987", + "issue:44479|issue:45325", + "issue:36032|issue:43742", + "issue:42915|issue:44589", + "issue:43296|issue:43531", + "issue:40444|issue:45325", + "issue:43531|issue:44877", + "issue:43611|issue:44534", + "issue:36032|issue:41720", + "issue:43425|issue:44403", + "issue:41553|issue:43582", + "issue:44079|issue:45663", + "issue:40990|issue:44811", + "issue:43632|issue:44534", + "issue:42175|issue:45542" + ], + "prepared_review_unit_hash": "e5590337e6b1dce4771794e143091280b210af4af60bf9b4cfa7ded89fab51cc", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36391, + "estimated_input_tokens": 9098, + "estimated_eval_tokens": 18452 + }, + "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 regressions, not one duplicate family. I only see two strong duplicate-like pairs: the Qwen3.5 save_pretrained regression and the Qwen video/temporal-position regression; the rest are too semantically distant to merge.", + "confidence": 0.71, + "canonical_issue_reason": "If one issue must anchor the cluster, issue 44479 is the broadest concrete regression report and best representative of the Qwen video/position-id bug subset. It is still not a good umbrella for the many unrelated loading/init/backend issues in the set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44479 is the best overall issue candidate because it describes a specific regression affecting multiple Qwen variants and is a plausible duplicate target for the narrower Qwen2.5-VL position-id report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Both are audio-related, but one is a chat-template sampling-rate default bug and the other is a padding-mask/ batching correctness bug in MIMI; different code paths." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different models and failures: SigLIP2 pipeline output quality vs OmDet processor image-size mismatch." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Custom model initialization failure is not the same underlying bug as non-persistent buffers being filled with junk." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Different multimodal model families and different symptoms: Llava Next load segfault vs Qwen3OmniMoe video ValueError." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Notebook/custom-model init regression is unrelated to the non-persistent buffer initialization bug." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Generic loading noise vs a Mistral tokenizer regex crash; not the same bug." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "MIMI batching/padding-mask bug and DAC from_latents STE mismatch are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3MoE FP8 config failure and compile-time template-node error are different failures." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export with torch_compilable_check and OmDet processor image sizing are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "ScandiBERT loading failure and Mistral regex patch crash are different model-specific bugs." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same Qwen3.5 save_pretrained regression: 45357 is a later report of incorrect visual encoder keys from the same underlying bug." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Physical-intelligence/fast loading failure is not the same as the Mistral tokenizer backend_tokenizer crash." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction issue and compiled flex_attention failure are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Tied lm_head weights in Mistral3ForConditionalGeneration is not the same as LoRA merge collapse with extended vocab." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Different root causes: Qwen3MoE FP8 config vs init_empty_weights/_is_hf_initialized argument handling." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "_is_hf_initialized flag regression and mmap/OOM on Strix Halo are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken special-argument duplication and loading physical-intelligence/fast are different bugs." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": true, + "reason": "45325 is a narrower manifestation of the same Qwen video/temporal-position regression described broadly in 44479." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "T5Tokenizer add_special_tokens conflict and MobileLLM-125M key error are unrelated loading problems." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3MoE FineGrainedFP8Config failure is unrelated to missing Float8 storage support." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL load failure and Qwen3-MoE sliding_window issue are different model-specific regressions." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "IterableDataset multi-image fine-tuning failure is not the same as the Qwen2.5-VL temporal-position bug." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and granite_speech strict config loading are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "base_model_prefix loading breakage is a different issue from non-persistent buffer corruption." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer add_special_tokens conflict and Qwen3 device-mapping cudaErrorAssert are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch 2.10 incompatibility is not the same as generic loading noise." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Bad AutoTokenizer error for Voxtral and AppleSilicon caching_allocator_warmup TypeError are unrelated." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment and Gemma4 FSDP2 shared_kv_states KeyError are different code paths." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "High perplexity on gpt-oss-20b and Whisper batch_decode skip_special_tokens are unrelated." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "These are both Transformers v5 regressions, but they affect different mechanisms and are not the same underlying bug." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both mention TensorFlow packaging/backend issues, but one is missing tf in the torch extra and the other is a tensorboard-only runtime backend error; too different to treat as duplicates." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The chosen anchor issue 44479 is a reasonable broad representative for the Qwen video/position-id subset, and the only accepted duplicate-like pair (45216/45357) is plausibly the same save_pretrained regression. The rejections mostly rest on clearly different model families or bug mechanisms, so I don\u2019t see an overstatement." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 10, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:41720", + "issue:42175", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43381", + "issue:43421", + "issue:43454", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43646", + "issue:43749", + "issue:43792", + "issue:43818", + "issue:43856", + "issue:43994", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44466", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44683", + "issue:44829", + "issue:44898", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45812", + "issue:45834", + "issue:45923" + ], + "soft_pairs": [ + "issue:43421|issue:44568", + "issue:43792|issue:44589", + "issue:43818|issue:45357", + "issue:44683|issue:45020", + "issue:44403|issue:44991", + "issue:44462|issue:45812", + "issue:33666|issue:40444", + "issue:44829|issue:45479", + "issue:43317|issue:43856", + "issue:43482|issue:45538", + "issue:43381|issue:45923", + "issue:35707|issue:40444", + "issue:44079|issue:44534", + "issue:44291|issue:44589", + "issue:43482|issue:43525", + "issue:43377|issue:45491", + "issue:43749|issue:44079", + "issue:40444|issue:45381", + "issue:43531|issue:44291", + "issue:41720|issue:45081", + "issue:43644|issue:44079", + "issue:44265|issue:45072", + "issue:45325|issue:45381", + "issue:43632|issue:45072", + "issue:43994|issue:44898", + "issue:43611|issue:43646", + "issue:42175|issue:43296", + "issue:43454|issue:44466", + "issue:45325|issue:45834" + ], + "prepared_review_unit_hash": "349dad4f140efb7bc3ff5a08bc0a839e2ee5e9ef8db2f482929c436ab23ee9fd", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36580, + "estimated_input_tokens": 9145, + "estimated_eval_tokens": 18546 + }, + "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 v5 regressions, with the strongest umbrella being the open remote_code loading breakage. The Qwen2.5-VL position-id pair should stay rejected: the titles point to related model family bugs, but one is still-image temporal position_id scaling and the other is video vision_position_ids, with no explicit shared code path in the packet.", + "confidence": 0.76, + "canonical_issue_reason": "issue:45020 is the best umbrella issue: it is broad, open, and explicitly covers recent-version regressions breaking model loading with remote_code, which matches the cluster's overall theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the strongest representative for triage because it is the broadest open report and the most plausible place for cross-cutting follow-up links or fixes.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Related tokenizer/special-token area, but one is runtime post-processor refresh and the other is add_special_tokens behavior; different bugs." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Different failure modes and code paths: Whisper loading versus a Float8 storage lookup error." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Different models and bugs: Video-LLaVA tower/weight sharing versus Qwen3.5 visual-encoder save_pretrained keys." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Torch flex_attention compatibility issue, not the remote_code loading regression." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic loading noise versus a specific tokenizer loading regression for EMBEDDIA/est-roberta." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "Both are AutoTokenizer regressions, but they affect different model families and concrete failure modes." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are Qwen-VL training issues, but multi-GPU training and multiple-images-per-prompt are distinct bugs." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both affect classification training, but flash_attention_3 behavior and num_labels=1 loss handling are different paths." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Device_map/offload loading failure versus MoE training memory efficiency; not the same bug." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "Different areas: GGUF loading with v5 versus CLIPTokenizer model_max_length behavior." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Related to gradient checkpointing, but eval-mode usage and a missing capability flag on a remote-code model are different issues." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL multi-image finetuning." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment and non-persistent buffer serialization are separate internal regressions." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected _is_hf_initialized init arg and Float8 storage lookup are different v5 breakages." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "GGUF loading failure and missing pad_token_id on Llama4Config are unrelated." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Padding-mask correctness in MIMI encoder versus Gemma3 NaN embeddings from sliding-window attention." + }, + { + "left": "issue:43749", + "right": "issue:44079", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and ModelOutput key assignment are distinct regressions." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Related model family, but the packet only shows different paths: multi-image training versus video vision_position_ids." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Sliding_window behavior in Qwen3-MoE is unrelated to init_empty_weights/_is_hf_initialized." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Auto device-mapping cudaErrorAssert and a Mistral tokenizer regex crash are unrelated." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Non-persistent buffers filled with junk and ModelOutput key assignment are different internal bugs." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export with torch_compilable_check is unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Related Qwen2.5-VL bugs, but the packet suggests different code paths: still-image temporal position_ids scaling versus video vision_position_ids." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "_is_hf_initialized breaks model init/loading, while this is a dtype mismatch in inference." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Both are vision-model correctness issues, but they involve different models and failure mechanisms." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 loading/init regressions, but one is about base_model_prefix loading and the other custom model initialization." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "Package/backend composition issue versus a specific PaddleOCR-VL loading failure." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both involve lm_head weights, but one is a missing tie in a model config and the other is device-dependent serialization inconsistency." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Different model families and symptoms: Qwen2.5-VL position IDs versus Kosmos2.5 OCR long-input index errors." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but one soft-edge verdict is too strong: the acceptance of issue:45325 \u2194 issue:45381 goes beyond the packet evidence. The titles suggest related Qwen2.5-VL position-id bugs, but they refer to different paths (still-image temporal position_ids scaling vs. video vision_position_ids), and the packet provides no explicit linkage or shared codepath. Please keep that pair rejected or qualify it more conservatively." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 11, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43262", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43701", + "issue:43746", + "issue:43792", + "issue:43819", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:43957", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44387", + "issue:44479", + "issue:44488", + "issue:44568", + "issue:44610", + "issue:44617", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45584", + "issue:45701", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45920", + "issue:45923" + ], + "soft_pairs": [ + "issue:43792|issue:43957", + "issue:42915|issue:43482", + "issue:43296|issue:45800", + "issue:44387|issue:44617", + "issue:43701|issue:45325", + "issue:44479|issue:45823", + "issue:43881|issue:44488", + "issue:43606|issue:44265", + "issue:44291|issue:44877", + "issue:43262|issue:43746", + "issue:44568|issue:45920", + "issue:40990|issue:45584", + "issue:34634|issue:35707", + "issue:43873|issue:45701", + "issue:44186|issue:44265", + "issue:35707|issue:41720", + "issue:43606|issue:43819", + "issue:41720|issue:45823", + "issue:43454|issue:43550", + "issue:43377|issue:44610", + "issue:43317|issue:43873", + "issue:39692|issue:40444", + "issue:43572|issue:45759", + "issue:43454|issue:45072", + "issue:43377|issue:44898", + "issue:43257|issue:45923", + "issue:43526|issue:43883", + "issue:40990|issue:42947" + ], + "prepared_review_unit_hash": "75297cbafcc2eb0ea32309ab413ad86c06c538f209d3d94d2606202133b39171", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35603, + "estimated_input_tokens": 8901, + "estimated_eval_tokens": 18058 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: the items span audio, vision, tokenization, training, checkpointing, and multiple unrelated loading/offload regressions. I don\u2019t see any soft pair that clearly shares the same concrete bug path, so all candidate duplicate links should be rejected.", + "confidence": 0.34, + "canonical_issue_reason": "issue:43873 is the broadest and most active open report here, and it best represents the loose loading/offloading/quantization subtheme among the items.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the strongest single issue to keep as the representative bug report: it is open, discussion-heavy, and describes a broadly scoped regression rather than a one-off model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "Whisper runtime failure and meta-device loading breakage are different code paths and different failure modes." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Different models, different formats, and different root causes: FP8 config failure vs GGUF loading failure." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "Both mention incompatibility/loading, but one is a PaddleOCR-VL/vLLM model-load issue and the other is a torch/transformers version mismatch; not the same bug." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both can end in OOM, but one is an int4 quantization memory regression and the other is a large video model running out of memory." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint resume key mismatch and Qwen2.5-VL rope-index scaling are unrelated training vs multimodal-positioning bugs." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Both involve multimodal models, but video-input regression and Gemma4 device mismatch are distinct failures." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Two separate loading failures on different models with no shared concrete code path." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU-offload device mismatch and torch.export failing on torch_compilable_check are unrelated subsystems." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "init_empty_weights argument handling and strict config loading are different initialization/config bugs." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sample-rate handling and PEFT checkpoint loading are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "A specific special-tokens regression for one tokenizer is not the same concrete bug as a broad AutoTokenizer ID regression across multiple models." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "Perplexity evaluation issue and Whisper empty-transcription generation failure are unrelated model behaviors." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "BarkProcessor voice_preset and progressive generation with inputs_embeds/past_key_values do not share the same bug." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Quantization/offloading regression and tokenizer-version behavior change are unrelated." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer crash on NER/padding is not the same as torch.export failing on compilability checks." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation and Qwen3 auto device mapping/cudaErrorAssert are different code paths." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "CPU offload device mismatch and DAC latent/STE mismatch are distinct audio-model bugs." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both mention device mismatch, but they affect different models and different execution paths." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "lm_head tying and torch.compile SDPA are unrelated model setup vs compilation issues." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "Missing padding-mask support in MIMI and OmDet-Turbo processor size mismatch are different preprocessing/model-shape bugs." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both are about offloading/quantization, but one is a dequantized-model load failure and the other is a broader offloading behavior regression; not enough to call the same bug." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "Documentation/example errors and finetuning with multiple images per prompt are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx and text_config unwrapping are separate config-compatibility bugs." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "Weight tying and bfloat16 dtype mismatch are different issues." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "Batched-padding mask behavior in MIMI and Perceiver interpolate_pos_encoding failure are unrelated." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "MoE weight conversion under accelerate/deepspeed and a missing supports_gradient_checkpointing flag are different fixes." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "Label reduction in BEiT image processing and missing all_tied_weights_keys in Molmo are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "High perplexity on one model and ineffective gradient checkpointing with LoRA 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": 17, + "review_unit_index": 12, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43296", + "issue:43316", + "issue:43317", + "issue:43322", + "issue:43404", + "issue:43450", + "issue:43454", + "issue:43526", + "issue:43550", + "issue:43577", + "issue:43606", + "issue:43708", + "issue:43720", + "issue:43749", + "issue:43873", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44387", + "issue:44423", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44610", + "issue:44671", + "issue:44743", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:44918", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45823" + ], + "soft_pairs": [ + "issue:44466|issue:45357", + "issue:43881|issue:44991", + "issue:43749|issue:44534", + "issue:43454|issue:43720", + "issue:41720|issue:43322", + "issue:43296|issue:44877", + "issue:41720|issue:44977", + "issue:44008|issue:45823", + "issue:42617|issue:42915", + "issue:43404|issue:45357", + "issue:40990|issue:45701", + "issue:44186|issue:45072", + "issue:45632|issue:45684", + "issue:44423|issue:45406", + "issue:43577|issue:43708", + "issue:44387|issue:45704", + "issue:43873|issue:44756", + "issue:36331|issue:42617", + "issue:44877|issue:44991", + "issue:43550|issue:43720", + "issue:36331|issue:42915", + "issue:43450|issue:45701", + "issue:44610|issue:44898", + "issue:44186|issue:44610", + "issue:44479|issue:44918", + "issue:43550|issue:43606", + "issue:41720|issue:42617", + "issue:43316|issue:44913", + "issue:42915|issue:43317", + "issue:43295|issue:45701", + "issue:43526|issue:43881", + "issue:44671|issue:44743", + "issue:41720|issue:45161" + ], + "prepared_review_unit_hash": "7dde2dea9dab4da0c95b0fc3caf13477864ac3fb3a357f5e112b74f8d1d59e90", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37656, + "estimated_input_tokens": 9414, + "estimated_eval_tokens": 19084 + }, + "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 across loading, tokenization, multimodal processing, quantization, and trainer behavior. No soft pair looks like the same underlying defect, so all candidate duplicate links should be rejected.", + "confidence": 0.87, + "canonical_issue_reason": "Issue 43749 is the strongest anchor: it has the highest discussion/inbound reference signal and describes a concrete, user-facing loading/offloading regression that could plausibly attract related reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43749 is the best issue to use as the cluster representative because it is well-scoped, high-traffic, and clearly about a core broken code path rather than a narrow model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Different bugs: one is lm_head weight tying/serialization, the other is incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Unrelated failures: model loading for glm-4v-9b versus tokenizer loading for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "One is FSDP CPU RAM efficient loading, the other is non-persistent buffers being filled with junk; different code paths." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Different model and failure mode: AyaVision lm_head tying vs BitNet packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Both are crashes, but on different models and causes: Qwen3 auto device mapping cuda assert vs Llava Next segmentation fault." + }, + { + "left": "issue:43296", + "right": "issue:44877", + "accept": false, + "reason": "A model load failure and a strict-config compatibility issue are not the same underlying bug." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Device-mapping cuda assert on Qwen3 is distinct from a flash-attention generation regression in Qwen3.5." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Different model families and failures: Gemma3n variable collision vs Gemma4 PLE device mismatch." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "Completely different problems: a script execution failure versus Qwen3Moe failing with FineGrainedFP8Config." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "One is tied lm_head weights in Mistral3, the other is incorrect visual encoder key serialization in Qwen3.5." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "High perplexity on one model is not the same as a tokenizer-behavior change across versions." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer crash in NER/padding is unrelated to dtype mismatches during bfloat16 inference." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Cache-path collisions for trust_remote_code are unrelated to permission propagation in save_pretrained." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Different multimodal serve crashes: continuous batching with Qwen3.5 versus Gemma4Processor missing _tokenizer." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "One is dtype loading for BLIP2, the other is resuming training and recomputing max_steps." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 quantization OOM and a T5 Apex RMSNorm memory leak are separate issues." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Quantized offloading behavior and a Strix Halo mmap OOM workaround are not the same bug." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "CustomTrainer compute_loss signature mismatch is unrelated to the 3d_parallel.py execution failure." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading for granite_speech and tokenizer loading for est-roberta are distinct." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Bamba torch.compile with SDPA and BitNet packed-weight unpacking are different failure modes." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "A trainer API signature regression is unrelated to Qwen3Moe FP8 loading." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor batch-shape bug does not match a version-dependent tokenization change." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are image-size related, but one is a processor/model size mismatch and the other is a non-default resolution failure in Perceiver." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer crashes and OmDet-Turbo image sizing issues affect different subsystems." + }, + { + "left": "issue:44479", + "right": "issue:44918", + "accept": false, + "reason": "Qwen video-input regression and Qwen3.5 embedding unpacking with TRL are separate problems." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is unrelated to Bark-small CPU offload device mismatch." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 auto device mapping and 3d_parallel.py startup failure are not the same underlying issue." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Gemma3TextConfig API discrepancy and GPTNeoXConfig rotary_pct reload behavior are different config bugs." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure is unrelated to dequantized model loading with device_map=auto and CPU offload." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Processor/tokenizer regression is not the same as a general tokenization change across versions." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and glm-4v-9b loading 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": 17, + "review_unit_index": 13, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36010", + "issue:41720", + "issue:42490", + "issue:42617", + "issue:42915", + "issue:43322", + "issue:43329", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:43927", + "issue:44186", + "issue:44361", + "issue:44368", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44514", + "issue:44545", + "issue:44625", + "issue:44898", + "issue:44991", + "issue:45059", + "issue:45072", + "issue:45081", + "issue:45357", + "issue:45381", + "issue:45406", + "issue:45464", + "issue:45479", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45812" + ], + "soft_pairs": [ + "issue:45081|issue:45812", + "issue:43531|issue:43572", + "issue:43540|issue:43782", + "issue:36010|issue:42617", + "issue:43421|issue:45812", + "issue:44186|issue:44361", + "issue:42915|issue:43322", + "issue:43388|issue:45059", + "issue:45406|issue:45464", + "issue:33357|issue:35707", + "issue:43421|issue:45701", + "issue:43606|issue:43720", + "issue:41720|issue:43746", + "issue:30064|issue:34634", + "issue:44492|issue:44509", + "issue:43550|issue:43653", + "issue:43873|issue:45704", + "issue:44514|issue:44545", + "issue:44625|issue:45479", + "issue:43606|issue:43653", + "issue:43531|issue:45759", + "issue:42490|issue:43329", + "issue:34634|issue:41720", + "issue:34567|issue:35707", + "issue:43421|issue:43927", + "issue:44368|issue:45357", + "issue:43782|issue:45381", + "issue:44898|issue:45072", + "issue:44451|issue:44991" + ], + "prepared_review_unit_hash": "f30c2c8bf4f5de9e851e3c32572edf06262e4d9386afbf94fe5ac5ccaa14a54b", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36656, + "estimated_input_tokens": 9164, + "estimated_eval_tokens": 18584 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous. Only the two Qwen2_5_VLProcessor apply_chat_template reports appear to be the same underlying bug; the rest are broad subsystem/model similarities, not true duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "issue:44514 and issue:44545 describe the same batched apply_chat_template crash with padding=False; 44514 is the earlier, more detailed report.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44514 is the best representative because it is the earliest duplicate report and contains the fuller reproduction context.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer-related, but they concern different models and different failure modes (Mistral regex patch crash vs Granite token IDs regression)." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Unrelated bugs: Qwen3-MoE sliding_window behavior vs StableLmConfig pad_token_idx regression." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Different Qwen multimodal paths and different errors; not the same bug." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "GenerationMixin import error and a 3d_parallel.py runtime failure are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:45812", + "accept": false, + "reason": "Both touch tokenization, but one is runtime post-processor updates and the other is wrong token IDs for Granite; different underlying problems." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Both are tokenizer crashes, but on different tokenizer classes and different code paths." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Llava Next segmentation fault are unrelated." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Different subsystems: metric gathering for tuple labels vs SAM3 text/bbox behavior." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer and Qwen3.5 streaming inference failure are not the same bug." + }, + { + "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:43421", + "right": "issue:45701", + "accept": false, + "reason": "Too broad a match; one is a specific special-token post-processor bug, the other is a general tokenization version-change report." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "Different loading/offload issues in different model paths; not mergeable as one fix." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "CUDA assert during auto device mapping and GraniteSpeech PEFT local checkpoint loading are unrelated." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "ImageProcessor void segmentation maps and BarkProcessor voice_preset are unrelated." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "Both are documentation/cleanup flavored, but they are different edits in different areas and not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "Different issues: torch.compile/SDPA failure vs tokenizer special-token registration." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization offloading behavior and T5 apex RMSNorm memory leak are unrelated." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same crash report: Qwen2_5_VLProcessor.apply_chat_template fails on batched input when padding=False." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Both concern classification config, but one is missing num_labels propagation in Qwen3.5 and the other is a broader zero-loss issue across multiple models." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "CPU-offload device mismatch and BigBirdTokenizer decode output are different problems." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Different Qwen config/model bugs: sliding_window behavior vs text_config unwrapping for AutoModelForCausalLM.from_config." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Local-path trailing slash warning and undefined variables in multimodal token counting are unrelated." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping are unrelated." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token counter not updating and progressive generation with past_key_values are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Runtime special-token post-processor updates vs save/load losing custom token IDs are different issues." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "Tie-word-embeddings warning and incorrect visual encoder keys on save_pretrained are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen multimodal bugs: weight_only loading vs video vision_position_ids." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver resolution/interpolation failure and bfloat16 dtype mismatch are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are model/tokenizer loading regressions, but for different models and with no evidence of the same underlying failure." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is well grounded in the packet: only issue:44514 and issue:44545 have clearly matching titles and dates support the choice of 44514 as the representative. The remaining pair verdicts are conservative and avoid overclaiming duplicates when the evidence is only broad subsystem/model similarity." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 14, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:40444", + "issue:42886", + "issue:42915", + "issue:43381", + "issue:43421", + "issue:43525", + "issue:43531", + "issue:43540", + "issue:43596", + "issue:43618", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43746", + "issue:43819", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44077", + "issue:44186", + "issue:44220", + "issue:44451", + "issue:44589", + "issue:44610", + "issue:44625", + "issue:44661", + "issue:44671", + "issue:44843", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:45042", + "issue:45081", + "issue:45183", + "issue:45356", + "issue:45479", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:45907", + "issue:45923" + ], + "soft_pairs": [ + "issue:44077|issue:44625", + "issue:44987|issue:45042", + "issue:45759|issue:45859", + "issue:43531|issue:44589", + "issue:44062|issue:44451", + "issue:45704|issue:45923", + "issue:39692|issue:44610", + "issue:43421|issue:43653", + "issue:44918|issue:44936", + "issue:45663|issue:45923", + "issue:45588|issue:45759", + "issue:42886|issue:44843", + "issue:43381|issue:43844", + "issue:43421|issue:44186", + "issue:43994|issue:44671", + "issue:43525|issue:43531", + "issue:45704|issue:45859", + "issue:45081|issue:45356", + "issue:45356|issue:45907", + "issue:39692|issue:43873", + "issue:43746|issue:43873", + "issue:43596|issue:44220", + "issue:40444|issue:43746", + "issue:42915|issue:43746", + "issue:43720|issue:43819", + "issue:43540|issue:44008", + "issue:44220|issue:44661", + "issue:45183|issue:45479", + "issue:30064|issue:36010", + "issue:43618|issue:43697", + "issue:43653|issue:43720" + ], + "prepared_review_unit_hash": "fce444d8daf69ff3b5d28c9d2bf79c5e895cc735e6b610540967f22c19e39f48", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37340, + "estimated_input_tokens": 9335, + "estimated_eval_tokens": 18926 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: it mixes unrelated tokenizer, vision, multimodal, training, and loading/offline bugs. None of the soft pairs look like the same concrete fix target, so I rejected all merges.", + "confidence": 0.93, + "canonical_issue_reason": "issue:43421 is the closest thing to a hub here because it sits in the densest tokenizer-special-token subtheme, but the cluster is too mixed to treat it as a true duplicate canonical.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43421 is the best representative of the most coherent subcluster (runtime tokenizer/post-processor behavior), but it should not be used as a global stand-in for the whole set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "Different problems: optional post_init on patchtsmixer vs missing num_labels propagation in Qwen3.5 config." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Unrelated failures: loading physical-intelligence/fast vs PIL image processors wrongly requiring torchvision." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Both involve Qwen 3.5, but one is composite config unwrapping and the other is CPU-offload rotary embedding behavior; not the same bug." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior is unrelated to the Float8 storage load error." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken keyword-collision constructor bug is not the same as ScandiBERT loading regression." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "T5 RMSNorm memory leak and missing supports_gradient_checkpointing are different code paths." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "SigLIP2 doc/example errors and OmDet-Turbo processor size mismatch are separate vision issues." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Both are tokenizer/special-token related, but one is post-processor refresh on runtime edits and the other is BigBird mask-token registration/empty decode; not the same concrete bug." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Qwen3.5 embedding unpacking in TRL SFT is unrelated to trainer.evaluate() failing after train()." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Gemma-4 FSDP2 shared_kv_states bug is unrelated to the missing gradient_checkpointing flag." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "flash_attention s_aux=None crash is not the same as Qwen 3.5 multimodal config unwrapping." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Both touch offline/tokenizer loading, but one is cache loading with HF_HUB_OFFLINE and the other is an unconditional model_info() call in _patch_mistral_regex." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Eval-mode gradient checkpointing restriction is unrelated to exploding gradients under HfDeepSpeedConfig + ZeRO-3." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Special-token post-processor update bug is not the same as LayoutLMv2Tokenizer crashing on NER/padding/truncation." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different models and symptoms: SigLIP2 nonsensical outputs vs CamemBERT masked-LM prediction regression." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config is unrelated to Qwen3-MoE sliding_window handling." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "Apex RMSNorm memory leak and Qwen3_5MoeTextRotaryEmbedding CPU offload compatibility are different defects." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both mention fix_mistral_regex, but one is a tokenizer crash and the other is a Kimi-K2.5 codec/regression/warning issue; not the same fix target." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Kimi tokenizer regression is unrelated to list_repo_files not forwarding revision." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "SigLIP2 example/doc issues are unrelated to quantization offloading behavior." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Local checkpoint PEFT loading and quantized offloading are different loading problems." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "DeepSpeed ZeRO-3/BertModel index error is unrelated to _torch_extract_fbank_features()." + }, + { + "left": "issue:40444", + "right": "issue:43746", + "accept": false, + "reason": "IterableDataset with multiple images per prompt is unrelated to GraniteSpeech PEFT local checkpoint loading." + }, + { + "left": "issue:42915", + "right": "issue:43746", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure is unrelated to GraniteSpeech local-checkpoint loading." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "BitNet packed-weight loading and DAC.from_latents/forward mismatch are unrelated model bugs." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Both are multimodal, but one is Qwen3OmniMoe video-input processing and the other is Gemma 3n audio tensor name collision; not the same bug." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio fbank extraction failure is unrelated to add-new-model-like failing inside TOKENIZER_MAPPING_NAMES." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Input validation XOR messaging and single-label-classification zero-loss behavior are different issues." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Image segmentation-map processing and GenerationMixin import error are unrelated." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "CLIPOutput attentions assignment and RTDetrV2 output differences in v5 are not the same code-path bug." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Tokenizer special-token registration and BitNet accelerate loading are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 15, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:30064", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43295", + "issue:43322", + "issue:43335", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43643", + "issue:43653", + "issue:43716", + "issue:43756", + "issue:43782", + "issue:43818", + "issue:43819", + "issue:43825", + "issue:43950", + "issue:43994", + "issue:44008", + "issue:44062", + "issue:44186", + "issue:44448", + "issue:44485", + "issue:44488", + "issue:44509", + "issue:44561", + "issue:44671", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45381", + "issue:45584", + "issue:45685", + "issue:45834", + "issue:45987" + ], + "soft_pairs": [ + "issue:43335|issue:43756", + "issue:43295|issue:45834", + "issue:43716|issue:44008", + "issue:43550|issue:45072", + "issue:43550|issue:43819", + "issue:44062|issue:44488", + "issue:43322|issue:45685", + "issue:43950|issue:45987", + "issue:44186|issue:44898", + "issue:44843|issue:45081", + "issue:43531|issue:43643", + "issue:43653|issue:43819", + "issue:43643|issue:44877", + "issue:43421|issue:44811", + "issue:43577|issue:43818", + "issue:43540|issue:45381", + "issue:43782|issue:44977", + "issue:44448|issue:44671", + "issue:45584|issue:45834", + "issue:43825|issue:43994", + "issue:44561|issue:45020", + "issue:39692|issue:45834", + "issue:44821|issue:45084", + "issue:30064|issue:36331", + "issue:44485|issue:44509", + "issue:41720|issue:44991" + ], + "prepared_review_unit_hash": "5c958e05d6399e8e4e1240da7182bbb34aa49c7bfd0185bff7b9fb787f45d314", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 35170, + "estimated_input_tokens": 8793, + "estimated_eval_tokens": 17842 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: most items are distinct regressions across different models, tokenizers, and APIs. None of the soft edges look like the same underlying bug closely enough to merge.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45020 is the broadest umbrella report ('Recent transformers versions break models using `remote_code`'), so it is the most plausible single anchor among otherwise distinct issues.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the best representative by scope and potential to subsume related remote_code/config-loading regressions, though the cluster is not truly duplicate-like.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Different bugs: sparse-layer creation in SwitchTransformers vs RoPE layer-dropping in Smollm3." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal, but one is a processor/tokenizer regression and the other is a long OCR input index error in Kosmos2.5." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Unrelated failures: image preprocessor/model dtype mismatch vs Gemma3n audio variable-name collision." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Different code paths: torch.compile+SDPA failure in Bamba-9B-v2 vs bfloat16 dtype mismatch in SwitchTransformers/TimmWrapperModel." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Different underlying problems: Bamba-9B-v2 compile/SDPA issue vs DAC.from_latents not matching forward pass." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "Tokenizer-related but not the same bug: duplicate AddedToken kwargs vs loading a specific tokenizer model." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Completely different subsystems and failures: Llava Next load segfault vs MPS histogram int support in MoE." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "One is from_pretrained buffer corruption; the other is a StaticCache return-type contract bug." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Different areas: LayoutLMv2 tokenization/padding crash vs Perceiver image classification failing at non-default resolution." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both involve Mistral regex patching, but one is offline Hub access and the other is an AttributeError on a different tokenizer object shape." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Distinct issues: Qwen3-MoE sliding_window behavior vs AutoConfig trust_remote_code missing fields." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "BigBird special-token registration bug is unrelated to DAC.from_latents forward-pass mismatch." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both are config-loading problems, but one is missing fields from trust_remote_code and the other is strict config rejection for granite_speech." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "One is runtime post-processor updates for special tokens; the other is Whisper batch_decode ignoring skip_special_tokens." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Different model bugs: BLIP-2 dtype propagation vs Video-LLaVA missing temporal attention/shared weights." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Both are video-input issues, but they affect different Qwen model families and different internal failures." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3VL loading with weight_only=True is not the same as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are v5 output regressions, but for different models and different code paths." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription generation bug is unrelated to Kosmos2.5 long-OCR index errors." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "One is an error-message/docs mismatch in pipeline(); the other is an inference correctness issue for SigLIP2." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Related theme, but not the same concrete bug: a specific missing helper breaks trust_remote_code models, while the other is a broad umbrella report." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "Different models and failures: SigLIP2 docs/example errors vs Kosmos2.5 OCR index error." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading and template-node compilation are unrelated failure modes." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Unrelated bugs: void segmentation map processing vs CustomTrainer.compute_loss signature regression." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion is unrelated to v5 pipeline task documentation leftovers." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Different failures in different components: Qwen3 device-mapping CUDA assert vs tokenizer loading for est-roberta." + } + ] + }, + "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": 17, + "review_unit_index": 16, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:34634", + "issue:36010", + "issue:36331", + "issue:42175", + "issue:43262", + "issue:43329", + "issue:43526", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43638", + "issue:43644", + "issue:43650", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:44060", + "issue:44186", + "issue:44291", + "issue:44368", + "issue:44442", + "issue:44479", + "issue:44554", + "issue:44610", + "issue:44671", + "issue:44805", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45406", + "issue:45459", + "issue:45538", + "issue:45600", + "issue:45701", + "issue:45803", + "issue:45834" + ], + "soft_pairs": [ + "issue:43644|issue:43749", + "issue:44186|issue:44442", + "issue:44479|issue:45834", + "issue:43577|issue:43582", + "issue:43606|issue:45072", + "issue:43825|issue:43827", + "issue:44805|issue:44936", + "issue:44671|issue:45701", + "issue:34634|issue:36010", + "issue:44610|issue:45072", + "issue:43782|issue:45325", + "issue:43720|issue:45072", + "issue:42175|issue:45081", + "issue:45081|issue:45406", + "issue:43749|issue:44291", + "issue:43650|issue:43906", + "issue:44060|issue:44368", + "issue:43582|issue:44857", + "issue:43262|issue:43819", + "issue:43526|issue:43697", + "issue:43526|issue:44554", + "issue:45459|issue:45803", + "issue:34634|issue:36331", + "issue:45084|issue:45538", + "issue:43742|issue:45020", + "issue:43653|issue:44912", + "issue:43329|issue:45600", + "issue:43638|issue:44936" + ], + "prepared_review_unit_hash": "3a95c784ed6dc7a6c761780fde3572c79ce68b114f166ca271787979b60ee35a", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36471, + "estimated_input_tokens": 9118, + "estimated_eval_tokens": 18492 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose set of unrelated bug reports that only share broad themes like loading, tokenizers, multimodal models, or v5 regressions. None of the soft pairs look like the same underlying bug/change, so all soft edges should be rejected. No PRs are present; the most representative issue is the broad v5/regression report around remote_code loading.", + "confidence": 0.59, + "canonical_issue_reason": "issue:45020 is the broadest, most central issue in the set: it describes recent transformers versions breaking remote_code models, which matches the general regression-focused nature of the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is open, broad, and impacts multiple models/paths, making it the best representative issue for the cluster even though the rest are mostly unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Both are loading/runtime regressions, but one is about non-persistent buffers and the other about FSDP CPU RAM efficient loading; different failure modes and code paths." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Both are tokenizer issues, but they affect different tokenizers and different bugs (NER/batched padding vs AutoTokenizer load failure)." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal input issues, but one is a v5.3.0 video regression in Qwen VL models and the other is a long OCR index error in Kosmos2.5." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Blip2 dtype handling and Apple Silicon allocator warmup are unrelated bugs." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "Both mention inference/loading problems, but one is a Bark CPU offload device mismatch and the other is a dtype mismatch in different models." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "The first is an incorrect pipeline error message; the second is outdated docs. Related topic, but not the same underlying bug." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Different failures: a mask/tensor shape IndexError versus trainer.evaluate breaking after train()." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "One is incorrect masked LM predictions for CamemBERT; the other is a generic tokenization-change report." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "Bark voice_preset and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "OmDet-Turbo image size mismatch and dtype mismatches in other models are different issues." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen VL-related, but one is a weight_only load error and the other is a rope-index temporal scaling bug." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "BitNet packed-weight loading failure and bfloat16 dtype mismatches are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "An installation/backend packaging issue is not the same as a tokenizer regex patch crash." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Both are AttributeErrors during loading, but they occur in different components and for different missing attributes." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "FSDP CPU RAM loading and init_empty_weights/_is_hf_initialized are separate loading bugs." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "A placeholder 'ADD THE DATA' issue is unrelated to an isolated reproduction report." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both mention tied-weights warnings, but one is a concrete wrong tie in Qwen3-Next and the other is a finetuning warning about config cleanup." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon allocator warmup and CUDA AMP loss crashes are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio processor chat-template sampling-rate handling and DAC latent reconstruction are different audio bugs." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BEiT label reduction and RTDetrV2 output drift are different vision bugs." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "Image label reduction and an MPS attention correctness issue do not describe the same problem." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both are error-handling bugs, but they affect different code paths and different swallowed errors." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice preset and CustomTrainer.compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "A template compilation TypeError is unrelated to CLIPTokenizer model_max_length behavior." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "A key error loading one model is not the same underlying issue as broad remote_code breakage." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "A multimodal token counting bug and stale auto_mappings references are different regressions." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "DeepSpeed zero3 indexing failure and trainer.evaluate-after-train failure are different training bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 17, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:34567", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42490", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43454", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43606", + "issue:43643", + "issue:43645", + "issue:43696", + "issue:43697", + "issue:43720", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43854", + "issue:43927", + "issue:43931", + "issue:44291", + "issue:44479", + "issue:44568", + "issue:44589", + "issue:44610", + "issue:44756", + "issue:44811", + "issue:44849", + "issue:44912", + "issue:44977", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45356", + "issue:45357", + "issue:45488", + "issue:45538", + "issue:45701", + "issue:45759", + "issue:45859" + ], + "soft_pairs": [ + "issue:43452|issue:44991", + "issue:43746|issue:44811", + "issue:44589|issue:45042", + "issue:38175|issue:44610", + "issue:43854|issue:44977", + "issue:43819|issue:43927", + "issue:39692|issue:44912", + "issue:43257|issue:43329", + "issue:36010|issue:43645", + "issue:43550|issue:44912", + "issue:43720|issue:44610", + "issue:43257|issue:43696", + "issue:43454|issue:43577", + "issue:43329|issue:44479", + "issue:43531|issue:45042", + "issue:43262|issue:44811", + "issue:42490|issue:43257", + "issue:36010|issue:36331", + "issue:44568|issue:45356", + "issue:43749|issue:44756", + "issue:43643|issue:45759", + "issue:44568|issue:45488", + "issue:34567|issue:40990", + "issue:43329|issue:43450", + "issue:43257|issue:43931", + "issue:43606|issue:45859", + "issue:44849|issue:45701", + "issue:43697|issue:45357", + "issue:44291|issue:45042", + "issue:44987|issue:45538" + ], + "prepared_review_unit_hash": "716667f9d1bcdc8f0b26c981c067718a21d3c1308281a83bc139da5e3d9daa70", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36979, + "estimated_input_tokens": 9245, + "estimated_eval_tokens": 18746 + }, + "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 regressions that only share broad vocabulary like tokenization, loading, video, or offload; the soft pairs do not look like true duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:45701 is the broadest, most referenced representative of the loose tokenization-regression subset; most other issues here are narrower or unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45701 is the best umbrella representative because it is open, has the most inbound references, and its title matches the cluster\u2019s broad tokenization theme better than the more model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Both involve loading tokenizers/models, but the failure modes and models differ." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "PEFT local-checkpoint loading and Whisper batch_decode are unrelated bugs." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Float8 storage lookup and PIL image processor torchvision requirements are different issues." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Both are vision-related, but one is output/probability behavior and the other is processor image size mismatch." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Different models and different failure modes; not the same code-path bug." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC forward mismatch and DiaConfig save/load token IDs are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "Docs/example errors and MXFP4 quantization fallback are separate problems." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE loading/conversion and multimodal video token counting are different bugs." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "Both are version-regression style reports, but they hit different APIs and causes." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "torch.compile SDPA failure and quantization fallback are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "BitNet accelerate loading and OmDet-Turbo processor sizing do not share a concrete bug." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "MoE weight conversion and OOM are distinct issues." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "lm_head tying failure and dtype staying float32 are different model-loading bugs." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related, but one is undefined helper usage and the other is a broader regression." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Qwen3-MoE sliding_window and PIL/torchvision image processing are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sampling-rate handling and Whisper batch_decode special-token handling differ." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Trailing-slash path warning and Qwen3 MoE weight conversion are unrelated." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "GenerationMixin import failure and custom_trainer loss signature breakage are separate API regressions." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Tokenizer special-token insertion and Kimi codec handling are different tokenizer bugs." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP RAM-efficient loading and mmap OOM on Strix Halo are different loading/resource issues." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Missing config fields with trust_remote_code and text_config unwrapping are not the same defect." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "BOS/EOS insertion failure and pre-tokenizer override are different tokenization behaviors." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "Trainer token-count tracking and GPT-oss perplexity are unrelated." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Video token counting undefined helpers and batched video processor shape are different bugs." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3 MoE loading conversion and Qwen3-VL weight-shape mismatch are separate model-loading issues." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but the model paths and concrete failures differ." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Hidden-states output bug and tokenization version change are unrelated." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetrV2 output differences and Qwen3.5 save_pretrained key issues do not match." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized and PIL/torchvision are different regressions." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Model loading failure and CLIPTokenizer max_length bug are unrelated." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Tokenizer special-token insertion and ByteLevel-to-Metaspace override are different defects." + }, + { + "left": "issue:44991", + "right": "issue:45538", + "accept": false, + "reason": "Different tokenizers and different symptoms; not the same bug." + }, + { + "left": "issue:45488", + "right": "issue:45701", + "accept": false, + "reason": "Broad tokenization regression vs specific LlamaTokenizer pre-tokenizer override; too imprecise to merge." + } + ] + }, + "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": 17, + "review_unit_index": 18, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:28282", + "issue:29127", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:38175", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43404", + "issue:43452", + "issue:43526", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43782", + "issue:43824", + "issue:43906", + "issue:43927", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44737", + "issue:44821", + "issue:44898", + "issue:44912", + "issue:44938", + "issue:44987", + "issue:44991", + "issue:45072", + "issue:45137", + "issue:45325", + "issue:45381", + "issue:45405", + "issue:45479", + "issue:45538", + "issue:45588", + "issue:45823", + "issue:45907", + "issue:45987" + ], + "soft_pairs": [ + "issue:42915|issue:43742", + "issue:44479|issue:45381", + "issue:43257|issue:43756", + "issue:41720|issue:43540", + "issue:43452|issue:43582", + "issue:42915|issue:44821", + "issue:33357|issue:34567", + "issue:44987|issue:45907", + "issue:42915|issue:45538", + "issue:43927|issue:44912", + "issue:43317|issue:45538", + "issue:42175|issue:43824", + "issue:43540|issue:44479", + "issue:43540|issue:44560", + "issue:44062|issue:44991", + "issue:43577|issue:45072", + "issue:42175|issue:44938", + "issue:38175|issue:44898", + "issue:43404|issue:45987", + "issue:42175|issue:44488", + "issue:45137|issue:45405", + "issue:43906|issue:45381", + "issue:28282|issue:34689", + "issue:43742|issue:44291", + "issue:42915|issue:43749", + "issue:44737|issue:45588", + "issue:43540|issue:45823", + "issue:43526|issue:45479", + "issue:43582|issue:43782", + "issue:43756|issue:45325", + "issue:28282|issue:29127" + ], + "prepared_review_unit_hash": "417844dc8116336f94f910c4d7ed4ee7d432c5b9e9013adb5f16c39e72c79e1e", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36534, + "estimated_input_tokens": 9134, + "estimated_eval_tokens": 18524 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous, with a loose concentration around Transformers model-loading and Qwen/VL video-input regressions, but no pair is convincingly the same underlying bug/change.", + "confidence": 0.31, + "canonical_issue_reason": "issue:44479 is the broadest representative of the recurring Qwen/VL video-input regression theme, with a clear concrete failure surface and the widest scope among the related reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44479 is the most representative issue for this cluster\u2019s strongest recurring theme (video-input regressions in Qwen/VL models).", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Different models and failures: Qwen3MoE FP8 config vs MobileLLM key error; same loading area only." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": false, + "reason": "Both involve Qwen/VL video input, but one is a broad multi-model regression and the other is a specific vision_position_ids bug; not clearly the same root cause." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Different subsystems: MoE weight conversion with accelerate/deepspeed vs Smollm3 RoPE layer handling." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Different Qwen3 failures: auto device mapping cudaErrorAssert vs video-input ValueError in Qwen3OmniMoe." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "Unrelated issues: gguf tokenizer/model loading vs Apple Silicon allocator warmup TypeError." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Different problem areas: Qwen3MoE FP8 loading vs AutoImageProcessor URL loading." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "Different bugs: MacOS CLIP bus error vs TrainerState token counter not updating." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Different failures: model loading regression vs list_repo_files needing revision forwarding." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "Different areas: Qwen3MoE FP8 loading vs CLIPTokenizer max-length constant." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "Different concrete bugs: DiaConfig save/load token IDs vs MXFP4 quantization fallback behavior." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Different failures: device_map/offload load path vs tokenizer max_length constant." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Different issues: missing TensorFlow backend in torch install vs missing Qwen2_5_VL import." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both touch video input, but they target different model families and distinct failure modes." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Different Qwen video bugs: Qwen3OmniMoe video ValueError vs Qwen3-vl-embedding StopIteration." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "Different tokenizer problems: AddedToken keyword collision vs tokenizer loading failure for EMBIDDIA/est-roberta." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both are dtype-related, but one is BLIP2 loading dtype propagation and the other is inference dtype mismatch in different models." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Different compatibility bugs: install extras/backend packaging vs Python 3.14 import/load failure." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Different vision-model bugs: SigLIP2 zero probabilities vs Perceiver non-default resolution handling." + }, + { + "left": "issue:43404", + "right": "issue:45987", + "accept": false, + "reason": "Different code paths: tied lm_head weights in Mistral3 vs StaticCache return type contract." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Different failures: backend packaging omission vs model loading failure for cjvt/sleng-bert." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "Different bugs: DeepSpeed ZeRO3 deque error vs PEFT version pin mismatch." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "No evidence they share the same root cause; isolated reproduction of another issue vs Qwen2.5-VL video position_ids bug." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Different import/load problems: missing PyTorch dependency vs a specific Llama 3.2 Vision loading regression." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "Both involve model loading, but one is a KeyError on a specific model and the other is a generic init_empty_weights argument mismatch." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Different failures: Qwen3MoE FP8 config vs FSDP CPU RAM efficient loading." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "Different bugs in different areas: XLNet CPU tensor creation vs flash_attention sink handling." + }, + { + "left": "issue:43540", + "right": "issue:45823", + "accept": false, + "reason": "Different model families and symptoms: Qwen3OmniMoe video processing vs Gemma4 device mismatch." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "Different semantics: BeitImageProcessorFast label reduction vs sequence-classification loss degeneracy." + }, + { + "left": "issue:43582", + "right": "issue:43782", + "accept": false, + "reason": "Unrelated: Apple Silicon warmup TypeError vs Qwen3VL from_pretrained weight_only error." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Different RoPE-related issues in different models; no evidence of a shared underlying bug." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "Different issues: missing PyTorch ImportError vs LayoutLMv3 box-data error message clarity." + } + ] + }, + "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": 17, + "review_unit_index": 19, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:24643", + "issue:29127", + "issue:34567", + "issue:36331", + "issue:41762", + "issue:43295", + "issue:43452", + "issue:43493", + "issue:43531", + "issue:43577", + "issue:43643", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43819", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44466", + "issue:44484", + "issue:44488", + "issue:44514", + "issue:44589", + "issue:44625", + "issue:44704", + "issue:44792", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44936", + "issue:44964", + "issue:45005", + "issue:45042", + "issue:45081", + "issue:45216", + "issue:45290", + "issue:45356", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45405", + "issue:45600", + "issue:45823", + "issue:45920" + ], + "soft_pairs": [ + "issue:43493|issue:43994", + "issue:34567|issue:36331", + "issue:44062|issue:45081", + "issue:44291|issue:44821", + "issue:45357|issue:45920", + "issue:43643|issue:45375", + "issue:43452|issue:44488", + "issue:44792|issue:44964", + "issue:45042|issue:45081", + "issue:43742|issue:44589", + "issue:43531|issue:44821", + "issue:45405|issue:45600", + "issue:43697|issue:45216", + "issue:43688|issue:43749", + "issue:44514|issue:45381", + "issue:43720|issue:44898", + "issue:44484|issue:44704", + "issue:41762|issue:44936", + "issue:24643|issue:29127", + "issue:44625|issue:44877", + "issue:44514|issue:45290", + "issue:43452|issue:43577", + "issue:43452|issue:44843", + "issue:44466|issue:45005", + "issue:43295|issue:45356", + "issue:43819|issue:44811", + "issue:44936|issue:45823" + ], + "prepared_review_unit_hash": "0607cf2ab6e3a17269883f36ec5062013a5ccf58df0a03db3cb60eee18423ff9", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 36186, + "estimated_input_tokens": 9047, + "estimated_eval_tokens": 18350 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mix of unrelated issues. Based on titles alone, only one pair looks tentatively like the same underlying SigLIP2 correctness problem; the rest are better treated as separate bugs or too broad to merge safely.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 43493 is the best representative of the only plausible duplicate family here: it describes a broader SigLIP2 discrepancy, while 43994 appears to be a more specific symptom involving a particular checkpoint/model path.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43493 is the strongest issue anchor for triage because it is the broadest SigLIP2 report in the set and most plausibly covers the related model-specific complaint in 43994.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "These both point to SigLIP2 producing incorrect results; one is a general implementation-vs-JAX discrepancy and the other is a model-specific bad-output report, so they tentatively look like the same underlying bug." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "Different Trainer problems: token counting not updating versus a custom compute_loss signature mismatch." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Both involve tokenizer loading, but the failure modes are different and do not look like the same bug." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "Different loading paths and different error shapes; no strong sign they are the same underlying issue." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Both are Qwen-related, but the titles indicate different regression surfaces: save_pretrained serialization versus tokenizer ID generation." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both are config-related, but they describe different config-handling failures and do not appear to be duplicates." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "Different from_pretrained failures with different root symptoms; too little evidence that they are the same bug." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Both are multimodal/model-loading related, but the titles suggest different model families and different failures." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "They share a tokenizer-adjacent area, but the specific issues are different enough that merging them would be unsafe." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "Different low-level errors: a key lookup problem versus a missing storage object type." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Unrelated issue surfaces: Qwen3-MoE sliding window behavior versus AutoImageProcessor URL loading." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "Maintenance/config change versus stale mapping references; not the same bug." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "Different model families and different regressions; no concrete overlap beyond both being correctness issues." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Separate bugs affecting different training/loading paths." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Same broader Qwen2.5-VL area, but the titles indicate different failure modes." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "Different model-specific problems with no clear shared code-path from the titles alone." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "A save_pretrained size question is unrelated to AutoProcessor passing kwargs to cached_file." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Different Trainer/DeepSpeed behaviors with distinct symptoms." + }, + { + "left": "issue:24643", + "right": "issue:29127", + "accept": false, + "reason": "DeepSpeed training runtime error versus LayoutLMv3 box-related error messaging; unrelated." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Both involve config handling, but the underlying failures are different." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Same API surface, but the batched padding crash and the assistant tool-call/content crash look distinct." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Different loading issues: gguf_file handling versus dtype propagation in BLIP2." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Tokenizer/loading-related, but the specific failure modes are different." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Related to tied weights, but the titles suggest different manifestations and too broad a scope to treat as one bug." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Different regression surfaces: processor/tokenizer access versus tokenizer codec/regex handling." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "Different bugs in different model areas; no sign of a shared underlying defect." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer lifecycle failure versus a device mismatch in Gemma4 forward pass." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The triage is mostly conservative, but it still overstates what is grounded in the packet. In particular, \"clearly reproducible\" and \"captures the underlying SigLIP2 mismatch\" are not supported by the packet itself (we only have titles/body lengths here), so they read as stronger than the evidence allows. Also, the summary says \"a couple of pairs\" look like duplicates even though only one pair is actually marked accept=true. Please soften those claims and keep the duplicate judgment explicitly tentative." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 20, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:28282", + "issue:30064", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:34567", + "issue:35141", + "issue:36010", + "issue:38175", + "issue:41720", + "issue:43122", + "issue:43278", + "issue:43295", + "issue:43452", + "issue:43504", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43906", + "issue:43927", + "issue:44361", + "issue:44442", + "issue:44451", + "issue:44464", + "issue:44479", + "issue:44554", + "issue:44561", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44743", + "issue:44821", + "issue:44843", + "issue:44977", + "issue:44991", + "issue:45440", + "issue:45588", + "issue:45701", + "issue:45704" + ], + "soft_pairs": [ + "issue:38175|issue:43295", + "issue:43653|issue:44568", + "issue:43653|issue:44361", + "issue:44361|issue:44442", + "issue:30064|issue:33666", + "issue:43643|issue:44561", + "issue:44464|issue:44743", + "issue:41720|issue:43906", + "issue:43688|issue:45440", + "issue:43688|issue:45704", + "issue:28282|issue:32090", + "issue:28282|issue:33357", + "issue:43577|issue:43697", + "issue:28282|issue:35141", + "issue:44843|issue:44991", + "issue:43646|issue:43749", + "issue:43688|issue:45701", + "issue:43653|issue:43927", + "issue:43531|issue:43742", + "issue:43550|issue:43927", + "issue:44464|issue:44977", + "issue:43550|issue:44361", + "issue:44479|issue:44625", + "issue:43122|issue:43906", + "issue:43577|issue:44554", + "issue:43577|issue:45588", + "issue:43504|issue:43720", + "issue:43504|issue:43550", + "issue:32090|issue:34567", + "issue:41720|issue:44464", + "issue:43504|issue:44610", + "issue:32090|issue:33666", + "issue:43452|issue:44451", + "issue:43742|issue:44821", + "issue:43278|issue:43688", + "issue:33666|issue:36010" + ], + "prepared_review_unit_hash": "3fd1233f82948697ab3814f1d84c4efa37218912a27db3db22e0b14b02110161", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 36, + "serialized_chars": 37895, + "estimated_input_tokens": 9474, + "estimated_eval_tokens": 19204 + }, + "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, with a few loose tokenizer/model-loading regressions but no true duplicate pair among the soft edges. I would keep one concrete tokenizer regression as the representative issue, and there are no PRs to consider.", + "confidence": 0.84, + "canonical_issue_reason": "issue:43122 is a concrete, user-visible tokenizer regression with version-specific repro and inbound references, making it the cleanest representative issue in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43122 is the strongest issue candidate because it is specific, already closed, and describes a reproducible version-to-version behavior change rather than a vague umbrella symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Different failures: zero probabilities in SigLIP2 vs processor/tokenizer access regression in custom model code." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer-special-token bugs, but they affect different models and different symptoms/code paths." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird special-token decode issue vs MLukeTokenizer AttributeError; not the same bug." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Tokenizer AttributeError on tasks vs AutoTokenizer failing to load a specific tokenizer class; different root causes." + }, + { + "left": "issue:30064", + "right": "issue:33666", + "accept": false, + "reason": "Image processor segmentation-map handling and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both touch trust_remote_code/v5 behavior, but one is missing config fields and the other is a removed utility breaking models." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Chunked generation with compiled forward is a different code path from recurrent-state reset in modular Qwen3.5." + }, + { + "left": "issue:41720", + "right": "issue:43906", + "accept": false, + "reason": "Different CUDA/Qwen3 device-mapping crash versus an isolated reproduction of another issue; no concrete same defect." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Both involve MoE models, but one is auxiliary-loss normalization and the other is a remote/native implementation divergence." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Auxiliary-loss normalization and an apex RMSNorm memory leak are unrelated bugs." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "Missing PyTorch import vs NoneType passed to gpu broadcast are distinct runtime errors." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "PyTorch-not-found import error and MacOS bus error on CLIP are different failures." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "BLIP-2 dtype propagation and RTDetrV2 output drift are different model-specific issues." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "ImportError from missing torch and embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline hub lookup bug and tokenizer load failure for a specific model are different problems." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Custom model initialization breakage is not the same as FSDP RAM-efficient loading failure." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Auxiliary-loss normalization and broad tokenization changes are not the same underlying bug." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "BigBird special-token registration and DiaConfig custom token-ID loss are both tokenizer/config issues, but they are separate code paths." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and MobileLLM loading key error are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is unrelated to DiaConfig token-ID persistence." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Compiled-forward chunked generation and flash-attention generation regression are different execution paths." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "Model compile/SDPA failure and tokenizer AttributeError are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression in Qwen VL models and num_labels propagation are different bugs." + }, + { + "left": "issue:43122", + "right": "issue:43906", + "accept": false, + "reason": "Tokenizer behavior change across versions is not the same as an isolated reproduction of another issue." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "BLIP-2 dtype mismatch and MPS attention correctness are different issues." + }, + { + "left": "issue:43577", + "right": "issue:45588", + "accept": false, + "reason": "BLIP-2 dtype propagation and flash_attention crashing on s_aux=None are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "BEiT legacy-field loading and BitNet packed-weight loading are different loaders and bugs." + }, + { + "left": "issue:32090", + "right": "issue:34567", + "accept": false, + "reason": "gpu_broadcast NoneType failure and TrainerState token counter not updating are unrelated trainer issues." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Qwen3 auto device-mapping CUDA assert and compiled-forward generation inconsistency are different failures." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "BEiT preset loading bug and OmDet processor size mismatch are unrelated." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "NoneType broadcast crash and Qwen2-VL multi-GPU training are not the same bug." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "gguf_file loading breakage and ScandiBERT tokenizer loading failure are different model-loading problems." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "MobileLLM key error and AutoImageProcessor URL loading failure are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype drift during evaluate and auxiliary-loss normalization in MoE models are different bugs." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 21, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:28218", + "issue:33357", + "issue:33666", + "issue:36331", + "issue:39692", + "issue:43257", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43299", + "issue:43334", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43540", + "issue:43572", + "issue:43606", + "issue:43646", + "issue:43720", + "issue:43756", + "issue:43824", + "issue:43825", + "issue:43844", + "issue:43927", + "issue:44291", + "issue:44361", + "issue:44466", + "issue:44488", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44843", + "issue:44849", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45357", + "issue:45405", + "issue:45520", + "issue:45584", + "issue:45759", + "issue:45803", + "issue:45812", + "issue:45823", + "issue:45987" + ], + "soft_pairs": [ + "issue:43493|issue:43825", + "issue:44849|issue:45357", + "issue:43606|issue:43927", + "issue:43720|issue:43927", + "issue:44568|issue:45812", + "issue:43525|issue:43572", + "issue:43262|issue:43295", + "issue:43756|issue:44913", + "issue:43720|issue:44361", + "issue:28218|issue:33357", + "issue:33666|issue:36331", + "issue:39692|issue:43493", + "issue:43540|issue:44977", + "issue:43493|issue:43844", + "issue:44964|issue:45405", + "issue:43334|issue:43572", + "issue:45803|issue:45987", + "issue:43278|issue:43844", + "issue:44625|issue:45584", + "issue:43756|issue:45083", + "issue:43257|issue:43299", + "issue:43334|issue:45759", + "issue:45200|issue:45229", + "issue:44488|issue:44843", + "issue:43824|issue:44589", + "issue:39692|issue:43295", + "issue:43646|issue:44466", + "issue:45229|issue:45823", + "issue:43504|issue:43606", + "issue:43572|issue:44291", + "issue:44589|issue:45520", + "issue:33357|issue:33666", + "issue:43927|issue:44361" + ], + "prepared_review_unit_hash": "70615c1120c9439e055843c6214f1a47823326b4fc5660c43e8199c3ffe3980e", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37873, + "estimated_input_tokens": 9469, + "estimated_eval_tokens": 19194 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous issue set: many items share tokenizer/config/model-loading language, but the underlying bugs differ by model, subsystem, and failure mode. I found no convincing duplicate clusters, so the best anchor is a broad, user-visible tokenizer regression rather than a narrow model-specific report.", + "confidence": 0.22, + "canonical_issue_reason": "Issue 45812 is the broadest and most cross-cutting regression in this set: it affects all Granite models, is user-visible, and has multiple inbound references, making it the best single anchor if one issue must represent the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "45812 is the strongest representative issue because it is broadly scoped, current, and tied to a concrete tokenizer behavior regression rather than a niche model-specific edge case.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "Different bugs: SigLIP2 implementation mismatch vs a pipeline error message about translation tasks." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Different Qwen3.5 problems: hidden-states behavior vs incorrect save_pretrained visual encoder keys." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "Device-mismatch/offload loading bug vs config losing custom token IDs; unrelated code paths." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Packed-weight conversion during accelerate loading is unrelated to DiaConfig token ID persistence." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer bugs, but they affect different models and different behaviors (special tokens vs Granite token IDs)." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Similar symptom name, but different model configs and different missing fields (`pad_token_id` vs `pad_token_idx`)." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug vs processor.tokenizer/image-handling regression." + }, + { + "left": "issue:43756", + "right": "issue:44913", + "accept": false, + "reason": "Different config/architecture issues: RoPE layer dropping vs rotary_pct not persisting on reload." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "Different tokenizer/loading failures: packed-weight unpacking vs MLukeTokenizer AttributeError on tasks." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer spacing bug vs Mac bus error on a CLIP model; no shared underlying defect." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Multi-GPU training support issue vs CustomTrainer compute_loss signature mismatch." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Docs/example errors are not the same bug as the SigLIP2 HF-vs-JAX implementation discrepancy." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Video-input processing error in Qwen3OmniMoe vs flash-attention generation bug in Qwen3.5." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 implementation discrepancy is unrelated to the DeepSpeed gradient growth report." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Model-loading failure for Phi-4 multimodal is unrelated to the PEFT version bump issue." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both mention missing pad token fields, but in different model families and different config attributes." + }, + { + "left": "issue:45803", + "right": "issue:45987", + "accept": false, + "reason": "Bare except swallowing interrupts is unrelated to StaticCache returning a tensor instead of an int." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype drift in eval is not the same as gradient explosion under ZeRO-3." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Qwen3.5 num_labels propagation bug vs Whisper empty-transcription generation failure; different paths." + }, + { + "left": "issue:43756", + "right": "issue:45083", + "accept": false, + "reason": "RoPE-layer omission in Smollm3 is unrelated to the qwen3_omni_moe feature-length helper behavior." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both are Qwen/MoE loading regressions, but they hit different model variants and different failure causes." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Pad-token-id missing in Qwen3-VL config is unrelated to from_config not unwrapping text_config." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Text-only mm_token_type_ids defaulting bug vs multi-GPU inference OOM; distinct issues." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Loading a specific model failing is not the same as AutoTokenizer calling model_info() offline." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "ImportError for Qwen2.5-VL symbol vs Float8 storage lookup TypeError; unrelated failures." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Documentation example mistakes are unrelated to the processor.tokenizer regression." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage is different from inconsistent lm_head.weight serialization across devices." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "CUDA OOM in Gemma4 inference vs device mismatch in Gemma4 PLE forward; different concrete bugs." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Legacy-field load failure in Beit semantic segmentation is unrelated to CPU offload device mismatch." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to unexpected _is_hf_initialized argument in init_empty_weights." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage import error vs flash_attn KeyError on Python 3.13; different dependency issues." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "Mac bus error with a CLIP model vs Qwen2-VL multi-GPU training; no overlap in root cause." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig token ID loss and MLukeTokenizer task AttributeError are unrelated 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": 17, + "review_unit_index": 22, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29942", + "issue:30333", + "issue:34567", + "issue:37428", + "issue:38175", + "issue:42490", + "issue:43232", + "issue:43278", + "issue:43295", + "issue:43493", + "issue:43504", + "issue:43540", + "issue:43575", + "issue:43582", + "issue:43606", + "issue:43653", + "issue:43696", + "issue:43701", + "issue:43844", + "issue:43957", + "issue:43994", + "issue:44112", + "issue:44315", + "issue:44423", + "issue:44464", + "issue:44493", + "issue:44521", + "issue:44554", + "issue:44610", + "issue:44734", + "issue:44792", + "issue:44849", + "issue:44964", + "issue:44977", + "issue:45072", + "issue:45084", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45356", + "issue:45405", + "issue:45446", + "issue:45520", + "issue:45584", + "issue:45823" + ], + "soft_pairs": [ + "issue:44423|issue:44734", + "issue:43504|issue:43653", + "issue:44464|issue:45584", + "issue:43606|issue:44112", + "issue:43994|issue:44493", + "issue:43575|issue:45405", + "issue:43278|issue:43701", + "issue:44792|issue:45405", + "issue:44521|issue:45290", + "issue:29942|issue:30333", + "issue:42490|issue:43696", + "issue:43844|issue:44964", + "issue:28218|issue:34567", + "issue:44554|issue:45446", + "issue:45084|issue:45520", + "issue:43582|issue:44977", + "issue:43540|issue:45200", + "issue:15354|issue:38175", + "issue:43232|issue:44849", + "issue:45072|issue:45198", + "issue:37428|issue:43957", + "issue:45200|issue:45823", + "issue:44315|issue:44849", + "issue:43493|issue:43701", + "issue:43295|issue:44610", + "issue:44554|issue:45356" + ], + "prepared_review_unit_hash": "d06800b383e02227a032c9c1c493ff61edc5f94c3d92becc84103a06ba8228c0", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34964, + "estimated_input_tokens": 8741, + "estimated_eval_tokens": 17738 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is heterogeneous: the soft pairs mostly share only broad subsystem vocabulary (generation, loading, tokenizers, CI, multimodal), but not the same concrete bug or change. I would not merge any of the proposed soft edges, and there is no single canonical issue/PR for the whole set.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serving-related crashes, but one is a multimodal `'str'.to` error and the other is KV-cache continuation indexing; different failure modes and code paths." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Both mention loading/token handling, but BEiT legacy-field loading and BigBirdTokenizer mask-token registration are unrelated bugs." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Both involve generation, but chunked generation with compiled forward and empty-transcription whisper failure after token alignment are different issues." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI failures, but CPU offload device mismatch in bark-small and a stale GraniteSpeech device-override test are not the same bug." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 nonsensical outputs and generic 'unexpected key with position id' regressions are too different in symptom and likely root cause." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Qwen2 TP OOM and a PEFT version bump/release issue are unrelated; one is memory pressure, the other dependency metadata." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "One is embedding dtype changing between train/eval, the other is a resume_from_checkpoint key mismatch; no shared underlying defect." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "A Janus image-generation test failure and a missing released PEFT version are unrelated." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both concern chat/template processing, but all-zero assistant masks for multimodal inputs and tool-call assistant messages with no content are distinct edge cases." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 test failures and MLflow not being reported after job failure are different CI/training issues with different surfaces." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Local-path import warnings and GPT-oss CUDA OOM are unrelated; one is path handling, the other memory exhaustion." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "Gradient growth with HfDeepSpeedConfig/ZeRO-3 and inability to load Phi-4 multimodal are different model/runtime problems." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer added-space behavior and TrainerState token counting are both token-related but not the same bug." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "MPS attention correctness when head dims differ and a PyTorch version check for AuxRequest import are unrelated." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Non-template node compilation error and flash_attn import key error on Python 3.13 are different failure modes." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon warmup TypeError and Qwen3.5 flash-attention generation bugs are separate platform/model-specific issues." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Qwen3OmniMoe video-processing ValueError and Gemma 4 mm_token_type_ids defaulting are different multimodal input problems." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "TorchScript export GeneratorExp limitation and SigLIP2 zero probabilities are unrelated model/runtime bugs." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and Qwen3.5 hidden_states output bug are distinct generation path issues." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "bfloat16 dtype mismatches in inference and Wav2Vec2 save/tokenization failures do not look like one underlying bug." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "Missing flash_supports_window_size import and meta-device loading regressions are both loading-related but clearly different problems." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Gemma 4 token type ids defaulting and Gemma4 PLE device mismatch are separate issues in the same model family, not the same code-path defect." + }, + { + "left": "issue:44315", + "right": "issue:44849", + "accept": false, + "reason": "Liger Kernel not applied with model_init and Qwen3.5 hidden_states output bug are unrelated training/model behavior issues." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and resume checkpoint key mismatch are different classes of bug with no clear shared root cause." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "processor.tokenizer regression with images and OmDet-Turbo input-size mismatch are both multimodal, but not the same defect." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "MPS attention correctness and Kimi-K2.5 tokenizer regression are unrelated subsystems." + } + ] + }, + "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": 17, + "review_unit_index": 23, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43388", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43526", + "issue:43572", + "issue:43575", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43827", + "issue:43844", + "issue:43856", + "issue:43873", + "issue:43901", + "issue:43957", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44442", + "issue:44466", + "issue:44514", + "issue:44521", + "issue:44610", + "issue:44792", + "issue:44857", + "issue:44898", + "issue:44936", + "issue:44987", + "issue:45083", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45245", + "issue:45325", + "issue:45412", + "issue:45563", + "issue:45584" + ], + "soft_pairs": [ + "issue:44898|issue:45412", + "issue:44466|issue:45127", + "issue:44792|issue:45563", + "issue:43572|issue:44987", + "issue:37428|issue:43824", + "issue:43572|issue:43957", + "issue:45200|issue:45325", + "issue:43844|issue:44936", + "issue:43295|issue:43450", + "issue:43334|issue:43525", + "issue:43824|issue:43975", + "issue:43295|issue:43873", + "issue:43388|issue:43526", + "issue:38175|issue:45412", + "issue:42915|issue:43824", + "issue:38175|issue:43493", + "issue:44514|issue:44521", + "issue:44442|issue:45198", + "issue:28218|issue:35141", + "issue:43278|issue:43856", + "issue:42490|issue:43493", + "issue:43653|issue:44112", + "issue:43827|issue:43901", + "issue:43450|issue:43575", + "issue:44206|issue:45584", + "issue:45127|issue:45245", + "issue:43819|issue:45083", + "issue:43819|issue:44112", + "issue:43262|issue:43450", + "issue:44610|issue:44857" + ], + "prepared_review_unit_hash": "ebb89661c4e59e1121fbb5348d46376b9147c84bfbc0bc01d4fa6d0fd137f7e3", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36948, + "estimated_input_tokens": 9237, + "estimated_eval_tokens": 18730 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are largely unrelated issue reports across different models/subsystems. I don\u2019t see any soft-edge pair that clearly represents the same underlying bug or change, so none should be merged as duplicates.", + "confidence": 0.28, + "canonical_issue_reason": "No single underlying bug clearly ties the set together; if forced to pick a representative, issue 44521 is the most central because it is open, recently active, and has the strongest reference activity in the processor/chat-template area.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44521 is the best overall issue candidate by activity and recency, but this cluster is too heterogeneous to treat it as a true canonical duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Different bugs: Perceiver image-classification resolution failure vs RT-DETR memory not being released." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both touch tied weights, but one is serialization/device-dependent while the other is LoRA merge collapse with extended vocab; not the same bug." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "Janus image-generation test failure and a stale generate() warning are unrelated symptoms." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is not the same as loading a specific model failing in v5.1.0." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "Both are ImportErrors, but they concern different missing symbols in different code paths." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "Config attribute regression vs meta-device model-loading failure; different problems." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Gemma 4 token-type defaults and Qwen2.5-VL rope-index scaling are distinct multimodal bugs." + }, + { + "left": "issue:43844", + "right": "issue:44936", + "accept": false, + "reason": "Gradient explosion under ZeRO-3 is unrelated to trainer.evaluate() failing after trainer.train()." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Processor.tokenizer regression is different from incorrect batched video output shapes." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Same missing-attribute class of failure, but for different model configs and likely separate fixes." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "An import error for a VL class is unrelated to a detokenization regression in deepseek-coder." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Custom processor API regression and quantization/offloading behavior are not the same code-path problem." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both involve labels, but one drops tuple elements in gather_for_metrics while the other truncates Beit labels." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 zero-probability outputs and RT-DETR memory leaks are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "FineGrainedFP8Config failure for Qwen3Moe is unrelated to a missing Qwen2.5-VL import." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "Both mention SigLIP2, but one is zero probabilities and the other is HF-vs-JAX implementation discrepancy; not clearly the same bug." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both are about apply_chat_template, but one is a batching/padding crash and the other is all-zero assistant masks; too different to treat as one fix." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "Different speech-model failures: tokenizer loading vs save_pretrained/tokenization." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Added-token spacing and post_init reinitializing output embeddings are separate tokenizer/model-weight bugs." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Training/eval dtype drift is unrelated to Qwen3 MoE memory usage." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "Local-path import warning and SigLIP2 implementation discrepancy are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "Tokenizer special-token registration bug is not the same as a stale CI test failure." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs issues around pipelines, but they cover different deprecations and behaviors." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Batched video shape bug and tensor-parallel OOM are unrelated." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Unsupported feature-extractor arg crash and Whisper empty-transcription generation failure are different issues." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/tied-weights collapse is unrelated to a hard category-count limit." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC latent reconstruction mismatch and qwen3_omni_moe feature-length behavior are different code paths." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "No shared bug: DAC latent mismatch vs GraniteSpeech CI test instability." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sample-rate defaulting and video batched-shape handling are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output resolution mismatch and AMP/CUDA loss crash 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": 17, + "review_unit_index": 24, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:32090", + "issue:37428", + "issue:38175", + "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:43931", + "issue:44112", + "issue:44206", + "issue:44292", + "issue:44351", + "issue:44360", + "issue:44448", + "issue:44462", + "issue:44493", + "issue:44561", + "issue:44743", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45290", + "issue:45310", + "issue:45381", + "issue:45478", + "issue:45520", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:37428|issue:43425", + "issue:43335|issue:43550", + "issue:43606|issue:45083", + "issue:45290|issue:45381", + "issue:44448|issue:44743", + "issue:44857|issue:45593", + "issue:45200|issue:45588", + "issue:44360|issue:44561", + "issue:43278|issue:43825", + "issue:37428|issue:44351", + "issue:29942|issue:32090", + "issue:43611|issue:43931", + "issue:44206|issue:45823", + "issue:39692|issue:43526", + "issue:43335|issue:43653", + "issue:43819|issue:44493", + "issue:43550|issue:44112", + "issue:43335|issue:43504", + "issue:44855|issue:45083", + "issue:45310|issue:45478", + "issue:43335|issue:45083", + "issue:29127|issue:32090", + "issue:45588|issue:45859", + "issue:39692|issue:44493", + "issue:44292|issue:45520", + "issue:44448|issue:44779", + "issue:38175|issue:44493", + "issue:43335|issue:45072", + "issue:44462|issue:44561", + "issue:43720|issue:44112", + "issue:44561|issue:44779", + "issue:43335|issue:44112", + "issue:44977|issue:45229", + "issue:42490|issue:43854" + ], + "prepared_review_unit_hash": "f82bb94aa980ddac020333cb79f8fa90141d7603f2ea2d472a11c38b12ea8eb4", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37857, + "estimated_input_tokens": 9465, + "estimated_eval_tokens": 19186 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is still mostly heterogeneous. The only clear duplicate pair is the Qwen3.5 MoE `from_pretrained` report: issue:45310 and issue:45478 share the same title and underlying failure mode, with 45310 being the earlier, more detailed report and 45478 a later repost.", + "confidence": 0.94, + "canonical_issue_reason": "issue:45310 is the better canonical issue because it is earlier and has the longer, more detailed report for the same Qwen3.5 MoE `from_pretrained` error; issue:45478 appears to be a later repost with the same title.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the strongest issue candidate overall for the duplicate cluster because it most clearly captures the underlying bug and predates issue:45478, which looks like a shorter follow-up repost.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "Both are runtime/compatibility failures, but they point to different missing-symbol vs. torch-version problems." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "SwitchTransformers sparse-layer initialization and Bamba torch.compile/SDPA failures are different code paths." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "Bark CPU-offload device mismatch is unrelated to the qwen3_omni_moe feature-length helper bug." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat-template tool-call handling and Qwen2.5-VL video position IDs are distinct issues." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Pegasus output drift and Qwen3.5 recurrent-state reset are separate model regressions." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "LwDetr AMP/CUDA crash is not the same as D-FINE's missing auxiliary losses when denoising is off." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Gemma 4 token-type defaults and flash_attention `s_aux=None` crashes are different failures." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA indexer activation behavior and removal of `is_torch_fx_available` breaking remote-code models are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype regression in evaluate is not the same as a pipeline error-message issue." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different compatibility breaks." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures and `_gpu_broadcast_one` NoneType handling are separate problems." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Base-model-prefix loading breakage and Qwen3-VL weight-shape mismatch are not the same bug." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Unsupported `center` argument in a feature extractor is unrelated to Gemma4 device mismatch during forward." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "SigLIP2 docs example errors and BeitImageProcessorFast `reduce_labels` behavior are different artifacts." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and BigBirdTokenizer special-token registration are unrelated." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC latent-forward mismatch and unexpected position-id keys are different regressions." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Bamba SDPA/torch.compile behavior and GraniteSpeech CI stale-device testing are not the same bug." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "SwitchTransformers sparse-layer initialization and Beit legacy-field loading failure are unrelated." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 `IndentationError` from scripted code is unrelated to the qwen3_omni_moe helper issue." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same title, same model family, and same `from_pretrained` error; 45478 looks like a later repost of 45310." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and qwen3_omni_moe helper behavior are different issues." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMv3 message clarity and Trainer `_gpu_broadcast_one` NoneType failure are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Both touch Qwen3.5/MoE/offload, but they describe different failure modes and code paths." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "Docs example failures are not the same as runtime unexpected key/position-id behavior." + }, + { + "left": "issue:44292", + "right": "issue:45520", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime failure and Python 3.13 flash-attn import KeyError are unrelated." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Pegasus output differences and Deepseek tokenizer regressions are different bugs." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 and unexpected position-id keys are different symptoms and subsystems." + }, + { + "left": "issue:43335", + "right": "issue:45072", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and bfloat16 dtype mismatches are unrelated." + }, + { + "left": "issue:44462", + "right": "issue:44561", + "accept": false, + "reason": "Tokenizer repository-file selection and removal of a helper function are different problems." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "BitNet packed-weight loading and a GraniteSpeech CI device-override test do not describe the same bug." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "A removed helper breaking remote-code models and a Deepseek tokenizer regression are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and GraniteSpeech CI device-override failure are different issues." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug and Gemma4 multi-GPU OOM are not the same underlying problem." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Trailing-slash import warning and GLM-4.7-Flash unit-test load failure are different loading issues." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The duplicate call on issue:45310/45478 is plausible, but the justification overstates the evidence: 45478 actually has more discussion activity than 45310, so it is not grounded to say 45310 has 'slightly more discussion/context.' A safer statement would be that 45310 is earlier and slightly longer, while 45478 appears to be a later repost with the same title. The rest of the soft-edge rejections look conservative." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 25, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:28218", + "issue:29127", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:40444", + "issue:41720", + "issue:42490", + "issue:43278", + "issue:43329", + "issue:43412", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43701", + "issue:43723", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:44448", + "issue:44493", + "issue:44554", + "issue:44568", + "issue:44857", + "issue:44977", + "issue:44987", + "issue:45072", + "issue:45083", + "issue:45198", + "issue:45229", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45712", + "issue:45803" + ], + "soft_pairs": [ + "issue:40444|issue:43526", + "issue:34689|issue:35141", + "issue:44448|issue:44977", + "issue:43606|issue:44857", + "issue:43723|issue:44554", + "issue:44857|issue:45072", + "issue:44977|issue:45563", + "issue:28218|issue:32090", + "issue:43927|issue:45698", + "issue:44568|issue:45491", + "issue:43701|issue:45083", + "issue:43550|issue:45083", + "issue:45198|issue:45803", + "issue:37428|issue:44987", + "issue:29942|issue:35141", + "issue:45372|issue:45588", + "issue:45459|issue:45712", + "issue:43873|issue:44493", + "issue:32090|issue:33357", + "issue:43873|issue:45229", + "issue:45245|issue:45563", + "issue:43412|issue:45588", + "issue:41720|issue:45229", + "issue:43278|issue:43329", + "issue:10000|issue:15354", + "issue:43976|issue:45440", + "issue:45372|issue:45406", + "issue:42490|issue:43575", + "issue:28218|issue:29127" + ], + "prepared_review_unit_hash": "520a581f8117e9e5129cd400c489b7d0c44e16462206dc8a13ed6c861c941775", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36076, + "estimated_input_tokens": 9019, + "estimated_eval_tokens": 18294 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated Transformer bugs that only share broad subsystem similarity (tokenizers, multimodal, loading, attention, offload). I did not find any soft pair that looks like the same underlying defect closely enough to merge.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43873 has the highest engagement and inbound references, making it the most defensible representative issue if a single canonical item is needed.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the strongest standalone issue in the set: high discussion activity, multiple inbound references, and a broad runtime/infrastructure impact.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "Both are multimodal/processor-related, but one is an IterableDataset multi-image failure and the other is a BeitImageProcessorFast label-reduction bug; different code paths and symptoms." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Model-loading failure for Llama 3.2 vision is not the same defect as token embedding reinitialization after resize when tie_word_embeddings is false." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Different models and failure modes: Pegasus output drift across versions versus Qwen3.5 flash-attention generation problems." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU offload device mismatch for bark-small is unrelated to the float16 AMP crash in LwDetrImageLoss." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "Tokenizer loading in v5 and an MPS attention correctness issue are separate code paths and bugs." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype/precision, but one is a loss crash under AMP and the other is a bfloat16 inference mismatch in different models." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Flash-attention generation failure and a stale warning about num_return_sequences are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing behavior and Trainer GPU broadcast NoneType errors are not the same underlying issue." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save/load behavior, but one loses custom token IDs in config while the other loads the wrong custom module after save_pretrained." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Missing BOS/EOS insertion in a tokenizer is unrelated to Gemma3 NaN embeddings from sliding-window attention." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint key mismatch and a qwen3_omni_moe helper-function bug are different defects." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba-9B-v2 does not match the qwen3_omni_moe length helper issue." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure and FuyuBatchFeature's bare except swallowing errors are related to tokenization but not the same concrete bug." + }, + { + "left": "issue:37428", + "right": "issue:44987", + "accept": false, + "reason": "Both involve loading/import issues, but there is no evidence they share the same failing code path or root cause." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Failing Flash Attention 2 tests are not the same defect as embedding reinitialization after resize." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 processor loading breaks because of a missing mistral_common symbol; flash_attention.py crashes on s_aux=None in sink-less models. Different dependency and runtime bugs." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer exception masking from protobuf handling is unrelated to leftover dummy class cleanup and dir(transformers) leakage." + }, + { + "left": "issue:43873", + "right": "issue:44493", + "accept": false, + "reason": "Quantization/offloading behavior and unexpected position-id key warnings are different problems." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer _gpu_broadcast_one NoneType errors and MacOS bus errors on a CLIP model are separate runtime failures." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Both touch memory/offload concerns, but one is a quantization offloading bug and the other is a Gemma4 multi-GPU OOM; not the same concrete issue." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "CUDA assert with auto device mapping on Qwen3 is a different failure mode from multi-GPU OOM on Gemma4 31B." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype regression in eval is unrelated to an undefined-variable bug in the video multimodal token counting path." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBertModel issue and GeneratorExp unsupported by torch.jit.script are unrelated legacy bugs." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility and DeepseekV3MoE divergence are not the same issue." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Both concern Gemma4 processor loading, but one fails on an import from mistral_common and the other crashes in serve due to a missing _tokenizer attribute; insufficient evidence of a shared root cause." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Local-path import warning and Qwen2-57B-A14B-Instruct tensor-parallel OOM are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:29127", + "accept": false, + "reason": "Tokenizer spacing behavior and LayoutLMv3 box validation error messages 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": 17, + "review_unit_index": 26, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43257", + "issue:43278", + "issue:43381", + "issue:43421", + "issue:43653", + "issue:43756", + "issue:43824", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44279", + "issue:44361", + "issue:44448", + "issue:44451", + "issue:44462", + "issue:44466", + "issue:44479", + "issue:44610", + "issue:44743", + "issue:44857", + "issue:44871", + "issue:44977", + "issue:45198", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45520", + "issue:45588", + "issue:45593", + "issue:45823" + ], + "soft_pairs": [ + "issue:43065|issue:43994", + "issue:43065|issue:45412", + "issue:45372|issue:45459", + "issue:44857|issue:45588", + "issue:44448|issue:44466", + "issue:43653|issue:44743", + "issue:45588|issue:45823", + "issue:43278|issue:43381", + "issue:44206|issue:44479", + "issue:25251|issue:32090", + "issue:43994|issue:45412", + "issue:44186|issue:45198", + "issue:43066|issue:44462", + "issue:44451|issue:45520", + "issue:44361|issue:45198", + "issue:42915|issue:42994", + "issue:44871|issue:44977", + "issue:44265|issue:45198", + "issue:43756|issue:45440", + "issue:29942|issue:30064", + "issue:43421|issue:45491", + "issue:43257|issue:43278", + "issue:33357|issue:39692", + "issue:29127|issue:34567", + "issue:44610|issue:45412", + "issue:32090|issue:33290", + "issue:42490|issue:43873", + "issue:45440|issue:45593", + "issue:43824|issue:44279" + ], + "prepared_review_unit_hash": "b6020c409e02a6276ce1ebe25d854f04d4b09696ef67431b779dbff2ed0d9124", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36352, + "estimated_input_tokens": 9088, + "estimated_eval_tokens": 18432 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: the soft pairs mostly share only broad subsystem labels (tokenizers, quantization, trainer, vision, etc.) but not the same concrete failure mode or fix. I would not merge any of the proposed pairs as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "If one issue must anchor the cluster, issue #43873 has the strongest engagement (highest inbound refs/discussion) and is open, but it is not a true duplicate representative of the rest.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43873 is the best single anchor by activity and references, though the cluster does not contain a clear canonical duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Different models and failures: Sam3PixelDecoder dummy Conv2d vs SigLIP2 incorrect outputs." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Dummy layer implementation issue vs RT-DETR memory not released; unrelated bug classes." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma 4 processor import breakage due to missing ReasoningEffort vs tokenizer error masking when protobuf is absent." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "AMP/CUDA loss crash vs flash_attention AttributeError on sink-less models; different code paths." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Both are v5 regressions, but one is generation-output difference and the other is serialization of tied weights." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "Tokenizer special-token registration bug vs Qwen3.5 recurrent-state reset bug; unrelated." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Flash-attention aux-state crash vs Gemma4 device_map forward mismatch; not the same defect." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Eval-mode dtype drift vs gradient checkpointing forbidden in eval; different symptoms and fixes." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Unsupported feature argument in a feature extractor vs video-input regression for Qwen VL models." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting behavior vs trainer broadcast NoneType error; unrelated." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 incorrect inference outputs vs RT-DETR memory leak; no shared underlying bug." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "LayoutLMv2Tokenizer NER/padding crash vs Wav2Vec2 save_pretrained/tokenization failure." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer decoder type in v5 vs AutoTokenizer ignoring tokenizer.json; both tokenizer-related but different failures." + }, + { + "left": "issue:44451", + "right": "issue:45520", + "accept": false, + "reason": "Model loading failure for ScandiBERT vs Python 3.13 flash_attn import mapping issue." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLukeTokenizer task-time AttributeError vs Wav2Vec2 save/tokenization failure." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Quantization-related, but one is FP8 loading/runtime failure and the other is quantized model saving." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma3 eos token config mismatch vs Qwen3.5 flash-attention generation regression." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure with torch_compilable_check vs Wav2Vec2 tokenization/save_pretrained issues." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "RoPE layer count mismatch in Smollm3 vs DeepseekV3MoE implementation divergence; different models and defects." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "FlashAttention test failures vs void segmentation map preprocessing failure; unrelated." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer post-processor not refreshed after special token changes vs NaN embeddings with sliding-window attention." + }, + { + "left": "issue:43257", + "right": "issue:43278", + "accept": false, + "reason": "Qwen3 MoE weight conversion problem vs embedding dtype mismatch during eval; different subsystems." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "MacOS bus error on community CLIP model vs SigLIP2 documentation/example problems." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 error messaging issue vs TrainerState token counter not updating; unrelated." + }, + { + "left": "issue:44610", + "right": "issue:45412", + "accept": false, + "reason": "Processor image-size mismatch vs model memory leak; no common bug." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer broadcast NoneType error vs DeepSpeed Adafactor OOM; both training-related but not the same issue." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Trailing-slash local model import warning vs quantization offloading not working; different code paths." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepseekV3MoE implementation mismatch vs D-FINE auxiliary-loss bug; unrelated model families." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Specific missing Qwen2_5_VL import vs generic dependency issue; too broad 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": 17, + "review_unit_index": 27, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:33357", + "issue:35141", + "issue:36010", + "issue:42490", + "issue:42915", + "issue:43232", + "issue:43262", + "issue:43334", + "issue:43381", + "issue:43388", + "issue:43441", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43653", + "issue:43701", + "issue:43723", + "issue:43761", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44186", + "issue:44188", + "issue:44206", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44977", + "issue:45003", + "issue:45072", + "issue:45103", + "issue:45245", + "issue:45372", + "issue:45440", + "issue:45479", + "issue:45491", + "issue:45588", + "issue:45803", + "issue:45834", + "issue:45859", + "issue:45907" + ], + "soft_pairs": [ + "issue:43653|issue:44442", + "issue:43761|issue:45859", + "issue:43526|issue:45834", + "issue:43381|issue:43701", + "issue:45103|issue:45372", + "issue:29942|issue:36010", + "issue:43526|issue:44186", + "issue:29127|issue:33357", + "issue:43334|issue:43531", + "issue:44977|issue:45245", + "issue:29127|issue:35141", + "issue:44464|issue:45491", + "issue:44442|issue:45072", + "issue:42915|issue:43931", + "issue:43262|issue:44206", + "issue:43526|issue:44857", + "issue:42490|issue:43381", + "issue:44062|issue:45907", + "issue:43441|issue:45588", + "issue:44743|issue:44811", + "issue:43232|issue:43388", + "issue:43575|issue:43701", + "issue:43646|issue:44077", + "issue:43761|issue:44079", + "issue:43575|issue:44387", + "issue:45003|issue:45440", + "issue:43723|issue:45479", + "issue:44442|issue:45803", + "issue:43761|issue:44188" + ], + "prepared_review_unit_hash": "c045f41cb130e5f3765fff2ec841f8f08dfa8aaef1c59f226f3200b926418f7e", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36586, + "estimated_input_tokens": 9147, + "estimated_eval_tokens": 18550 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is heterogeneous: these are not duplicate reports, but unrelated regressions and edge cases across tokenizers, generation, model loading, attention kernels, and multimodal processors. No soft pair looks safe to merge as the same underlying bug.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 43761 is the best anchor if one is needed: it is a clear, self-contained regression with a precise API, expected behavior, and reproducible symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "43761 is the most actionable and well-scoped issue in the set; the bug is concrete and isolated enough to serve as a representative anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "Both involve tokenizers, but one is a special-token decode bug and the other is an AutoTokenizer loading failure; different code paths." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "A CLIPVision hidden-state regression and a Qwen3 MoE rotary/cpu-offload issue are unrelated model-specific bugs." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "One is a BEiT Fast image label reduction bug; the other is a Kosmos2.5 long-OCR index error. No same underlying failure." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Gradient checkpointing in eval mode and checkpoint resume key mismatch are distinct training-state issues." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "Auto-docstring annotation parsing and Gemma processor loading/import dependency issues are different subsystems." + }, + { + "left": "issue:29942", + "right": "issue:36010", + "accept": false, + "reason": "Flash Attention 2 test failures and GenerationMixin import errors are unrelated regression classes." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Both touch layout/image tokenization, but the BEiT label reduction bug and LayoutLMv2 NER/padding crash are not the same defect." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 box-content validation and a macOS bus error loading CLIP are completely different failures." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id in Qwen3-VL text config and sliding_window behavior in Qwen3-MoE are separate model bugs." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention generation corruption and category-count overflow are unrelated; no shared code path." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "One is an error-message clarity issue for LayoutLMv3; the other is embedding reinitialization during post_init. Different bugs." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Compiled chunked generation inconsistency and Gemma3 mixed-length NaNs stem from different attention/generation paths." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure and bfloat16 dtype mismatch in inference are different classes of problems." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Fine-grained FP8 config failure for Qwen3Moe and Qwen3-VL weight-shape mismatch are separate model-loading bugs." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate handling and feature-extractor unsupported-arg crash are unrelated processor issues." + }, + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "A label-reduction bug in BEiT fast processing does not match an AMP/CUDA image-loss crash." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Local-path trailing-slash import warning and gradient-checkpointing eval-mode restriction are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken constructor argument duplication and list_repo_files revision forwarding are different API bugs." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Ministral FlashAttention failure and sink-less flash_attention.py s_aux=None crash are distinct attention implementations/fixes." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3.5 recurrent-state reset and Whisper batch_decode skip_special_tokens handling are not the same underlying issue." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs synchronization after sync_gpus and metric-gather label truncation are different training/generation paths." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Tensor-parallel OOM on model load and resume_from_checkpoint key mismatch are unrelated." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Transformers 5 custom initialization breakage and patchtsmixer post_init allowance are not the same code-path problem." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "Hidden_states missing from CLIPVision forward and ModelOutput key assignment when a value was previously None are separate bugs." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Qwen2-57B tp OOM on load and increased reserved memory under int4 quantization are different memory-pressure issues." + }, + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "Unsafe sys.modules access in modeling_utils and DeepseekV3MoE divergence from remote implementation are unrelated." + }, + { + "left": "issue:43723", + "right": "issue:45479", + "accept": false, + "reason": "Tokenizer loading in v5 and degenerate zero-loss sequence classification are different APIs and symptoms." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "AutoTokenizer loading failure and FuyuBatchFeature exception swallowing are unrelated tokenizer/feature-processing bugs." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVision hidden_states regression and attention-kernel divergence under torch.compile are separate model execution 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": 17, + "review_unit_index": 28, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:25251", + "issue:28282", + "issue:33290", + "issue:34567", + "issue:34689", + "issue:35141", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41628", + "issue:42175", + "issue:42490", + "issue:42907", + "issue:43065", + "issue:43316", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43493", + "issue:43525", + "issue:43540", + "issue:43611", + "issue:43643", + "issue:43720", + "issue:43749", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44077", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44617", + "issue:44743", + "issue:44855", + "issue:44871", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:45137", + "issue:45216", + "issue:45290", + "issue:45372", + "issue:45491", + "issue:45698", + "issue:45759" + ], + "soft_pairs": [ + "issue:44479|issue:45290", + "issue:43316|issue:43525", + "issue:44442|issue:44898", + "issue:42907|issue:45698", + "issue:42490|issue:44743", + "issue:33290|issue:34567", + "issue:34689|issue:43611", + "issue:33290|issue:35141", + "issue:33290|issue:34689", + "issue:39692|issue:45491", + "issue:43381|issue:43873", + "issue:43931|issue:45216", + "issue:43493|issue:43827", + "issue:44077|issue:44479", + "issue:43316|issue:45759", + "issue:43720|issue:44442", + "issue:41628|issue:45372", + "issue:43643|issue:44913", + "issue:43749|issue:43931", + "issue:34689|issue:36010", + "issue:43873|issue:44871", + "issue:33290|issue:36331", + "issue:44190|issue:44496", + "issue:44855|issue:45137", + "issue:43540|issue:44514", + "issue:43316|issue:44877", + "issue:42175|issue:43316", + "issue:43065|issue:44617", + "issue:43404|issue:43441", + "issue:25251|issue:28282" + ], + "prepared_review_unit_hash": "3fc865d5a30c3dee3013d22b6a262ec7c7df624f6f4817c4603faf4b29296f50", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36864, + "estimated_input_tokens": 9216, + "estimated_eval_tokens": 18688 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are not duplicates as a cluster: the items cover many unrelated issues across loading, config, training, docs, quantization, and multimodal/video paths. All soft-similarity pairs are superficial and should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No canonical issue: the set is heterogeneous and does not represent one underlying bug or change.", + "canonical_pr_reason": "No PRs are present in this cluster, so no canonical PR can be selected.", + "best_issue_reason": "No single best issue for deduping; none of the issues is clearly central enough to serve as a representative for the others.", + "best_pr_reason": "No PRs are present in this cluster, so no best PR exists.", + "soft_edge_verdicts": [ + { + "left": "issue:44479", + "right": "issue:45290", + "accept": false, + "reason": "Both involve video/chat-related failures, but one is a video-input regression and the other is a chat-template crash on tool-call messages; different code paths and symptoms." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both are config/API discrepancies, but they concern different models and different missing fields; not the same bug." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading failure vs image-classification interpolation bug; no shared underlying code path." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Both touch save/load behavior, but one is about saving dequantized weights and the other about loading the wrong custom module after save_pretrained; distinct failures." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local-path import warning and recurrent-state reset in Qwen3.5 are unrelated." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "DeepSpeed/Adafactor OOM and TrainerState token-count tracking are unrelated training issues." + }, + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both are model-loading regressions, but they involve different loading mechanisms and different root causes." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "OOM in optimizer setup vs embedding reinitialization during post_init; not the same defect." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Different failures: optimizer memory usage versus specific model-loading breakage." + }, + { + "left": "issue:39692", + "right": "issue:45491", + "accept": false, + "reason": "Docs example errors for SigLIP2 versus Gemma3 NaN embeddings; unrelated surfaces and code paths." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing in eval mode and quantization/offloading behavior are different bugs." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Weight-shape mismatch on load versus incorrect checkpoint saving; related to persistence broadly but not the same concrete issue." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and pipeline-doc removal are unrelated." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "Optional post_init enforcement for patchtsmixer and a video-input regression are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both mention config handling, but one is an API discrepancy in Gemma3TextConfig and the other is unwrapping text_config for composite Qwen configs; different fixes." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "Accelerate loading of packed BitNet weights versus tokenizer loading failure; no overlap beyond generic model loading." + }, + { + "left": "issue:41628", + "right": "issue:45372", + "accept": false, + "reason": "Different import errors for different components and dependencies; not the same underlying bug." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Remote-code config missing fields on load versus a specific GPTNeoX config value not persisting; distinct serialization issues." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "FSDP CPU RAM-efficient loading breakage and a Qwen3-VL weight mismatch are different loading problems." + }, + { + "left": "issue:34689", + "right": "issue:36010", + "accept": false, + "reason": "Both are import/load regressions, but they affect different symbols and code paths." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior and Gemma-3 eos_token_id inconsistency are unrelated." + }, + { + "left": "issue:33290", + "right": "issue:36331", + "accept": false, + "reason": "OOM with Adafactor/DeepSpeed is unrelated to the CustomTrainer compute_loss signature break." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Dataset-loading issue in an example script versus unrecognized model type in config.json; different layers entirely." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 import parsing error and DeepSpeed deque underflow are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Both are multimodal/chat-related, but one is video-input validation in Qwen3OmniMoe and the other is apply_chat_template crashing on tool-call messages; not the same bug." + }, + { + "left": "issue:43316", + "right": "issue:44877", + "accept": false, + "reason": "Both involve config handling, but the specific models and failure modes differ completely." + }, + { + "left": "issue:42175", + "right": "issue:43316", + "accept": false, + "reason": "Packaging/backend dependency omission and Gemma3 config API mismatch are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Same model family, but dummy Conv2d in Sam3PixelDecoder and Sam3Video CUDA OOM are different issues." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Related to Mistral/Ministral, but one is a missing lm_head tie and the other is a FlashAttention failure; not the same code-path problem." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k nesting bug and AutoModel PyTorch import error are entirely 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": 17, + "review_unit_index": 29, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:32090", + "issue:35141", + "issue:41950", + "issue:42617", + "issue:43065", + "issue:43295", + "issue:43388", + "issue:43452", + "issue:43475", + "issue:43504", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43673", + "issue:43688", + "issue:43701", + "issue:43742", + "issue:43761", + "issue:43825", + "issue:43931", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44242", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44521", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44743", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45381", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45698" + ], + "soft_pairs": [ + "issue:43673|issue:44743", + "issue:43295|issue:43825", + "issue:44162|issue:44792", + "issue:43388|issue:43688", + "issue:41950|issue:43475", + "issue:44077|issue:44534", + "issue:16998|issue:22355", + "issue:43550|issue:43701", + "issue:43606|issue:43701", + "issue:44871|issue:45584", + "issue:43761|issue:44479", + "issue:45127|issue:45200", + "issue:43295|issue:44871", + "issue:42617|issue:43575", + "issue:43452|issue:44496", + "issue:43575|issue:43742", + "issue:44734|issue:45406", + "issue:43646|issue:44913", + "issue:43504|issue:45698", + "issue:44617|issue:45584", + "issue:43931|issue:44060", + "issue:43761|issue:44242", + "issue:32090|issue:35141", + "issue:45563|issue:45584", + "issue:44464|issue:44871", + "issue:43065|issue:44610", + "issue:44792|issue:44977", + "issue:44479|issue:44734", + "issue:44521|issue:45381" + ], + "prepared_review_unit_hash": "868263f649e9596f951ca7a658eb90481e95275a28881f017c3c9a6c851ce861", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35779, + "estimated_input_tokens": 8945, + "estimated_eval_tokens": 18146 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: the soft pairs mostly share only broad terminology (generation, loading, multimodal, training) but not the same concrete bug or code path. I would not merge any of the proposed soft edges as duplicates. If a representative issue is needed, issue:44521 is the strongest activity-wise, but it is not a true canonical for the whole set.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue fits this cluster; the items span unrelated bugs across loading, generation, multimodal processing, serving, and training.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the most active/open item and has the strongest discussion footprint, but it is only representative of one narrow multimodal bug, not the cluster as a whole.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both involve cache-related generation behavior, but one is about chunked_prefill cache availability and the other about recurrent states resetting with seq_len>1; different symptoms and code paths." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "One is a regression in processor/tokenizer access and image passing; the other is an error-message mismatch about supported translation tasks. Different bug types." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 workflow breakage and a Janus image-generation test failure are unrelated models and failures." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Dropping label elements in gather_for_metrics is unrelated to auxiliary-loss normalization in expert models." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "A pipeline selecting the wrong processor type is not the same as a Sam3VisionEncoderOutput missing fpn_position_embeddings." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "Optional post_init handling for patchtsmixer and junk-filled non-persistent buffers are different initialization bugs." + }, + { + "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:43550", + "right": "issue:43701", + "accept": false, + "reason": "torch.compile/SDPA failure for Bamba and a resume_from_checkpoint key mismatch are different problems." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch in bark-small is not the same as the checkpoint key mismatch bug." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "EOS token config inconsistency in Gemma-3 and Whisper empty-transcription failure after align_special_tokens are different model-specific issues." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and Qwen VL video-input regression are separate model families and failure modes." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA merging with extended vocabulary and Gemma 4 mm_token_type_ids defaults are unrelated training/configuration bugs." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Both mention processor/tokenizer behavior, but one is a custom model regression and the other is an EOS-token config mismatch; not the same bug." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "Running 3d_parallel.py and OOM when loading Qwen2-57B with tensor parallelism are different issues." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "A gguf_file path breaking tokenizer/model loading is not the same as failing because a config.json lacks model_type." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "OOM on a very large Qwen model with tp and a key error loading MobileLLM-125M are unrelated load-time failures." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "KV-cache indexing in transformers serve and Gemma4Processor missing _tokenizer are different serving bugs." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Breaking custom model initialization in v5 and GPTNeoXConfig rotary_pct not persisting are distinct configuration/init issues." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Legacy-field loading failure in BeitForSemanticSegmentation and wrong custom module selection after save_pretrained are different pretrained-loading bugs." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Sam3Video CUDA OOM and Whisper empty transcription failure are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "A weight-shape mismatch when loading Qwen3-VL is not the same as an incorrect tied-weights warning in Qwen3-Next." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "Hidden_states=None in CLIPVisionModel and missing load-balancing loss when output_router_logits=False are unrelated." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "GPU broadcast TypeError in Trainer and embedding reinitialization after resize are separate bugs." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "A stale warning in paged generate and Whisper generation failing on empty transcription are not the same underlying issue." + }, + { + "left": "issue:44464", + "right": "issue:44871", + "accept": false, + "reason": "Compiled chunked generation inconsistency and Gemma-3 EOS configuration mismatch are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and OmDet-Turbo processor/model size mismatch are both vision-related but clearly different bugs." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Janus image-generation test failure and Qwen3.5 flash-attention generation failure are different models and code paths." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "A multimodal video-input regression and a serve KV-cache continuation indexing crash are unrelated." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both are multimodal, but one is assistant mask construction in apply_chat_template and the other is vision_position_ids for Qwen2.5-VL video input; 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": 17, + "review_unit_index": 30, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:35141", + "issue:39692", + "issue:41720", + "issue:41950", + "issue:42490", + "issue:43335", + "issue:43388", + "issue:43540", + "issue:43550", + "issue:43582", + "issue:43644", + "issue:43650", + "issue:43827", + "issue:43906", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44077", + "issue:44079", + "issue:44112", + "issue:44190", + "issue:44220", + "issue:44462", + "issue:44464", + "issue:44496", + "issue:44534", + "issue:44734", + "issue:44737", + "issue:44792", + "issue:44855", + "issue:44913", + "issue:44933", + "issue:44938", + "issue:44964", + "issue:45127", + "issue:45245", + "issue:45276", + "issue:45290", + "issue:45357", + "issue:45362", + "issue:45542", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45803", + "issue:45907" + ], + "soft_pairs": [ + "issue:43906|issue:45290", + "issue:43827|issue:43994", + "issue:44464|issue:45563", + "issue:44534|issue:44913", + "issue:44190|issue:44964", + "issue:44933|issue:45362", + "issue:41720|issue:45245", + "issue:44792|issue:45600", + "issue:44855|issue:45588", + "issue:44737|issue:45803", + "issue:45127|issue:45357", + "issue:43388|issue:43582", + "issue:35141|issue:42490", + "issue:44462|issue:45907", + "issue:44077|issue:44079", + "issue:43335|issue:44855", + "issue:44938|issue:45542", + "issue:43550|issue:44190", + "issue:43976|issue:44855", + "issue:43644|issue:44913", + "issue:43540|issue:44734", + "issue:44112|issue:45600", + "issue:44496|issue:45803", + "issue:43650|issue:44220", + "issue:43335|issue:44190", + "issue:45276|issue:45584", + "issue:39692|issue:44190", + "issue:45245|issue:45584", + "issue:41950|issue:44016" + ], + "prepared_review_unit_hash": "76fb2c28234ac7c80fb3cba196541da178f6ebf7ae23bee59945582b892e8af5", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35902, + "estimated_input_tokens": 8976, + "estimated_eval_tokens": 18208 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly heterogeneous and does not look like a tight duplicate cluster. The only clear thematic overlap is around token-embedding resizing, but the listed soft pairs are otherwise separate bugs or unrelated regressions.", + "confidence": 0.92, + "canonical_issue_reason": "issue:35141 is the clearest, most general concrete bug report in the set: it names the exact code path (`post_init` after `resize_token_embeddings`) and the failure mode (`output embedding` reinitialized when `tie_word_embedding=False`).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:35141 is the best anchor for the only plausible duplicate family here because it is specific, actionable, and directly describes the underlying embedding-resize bug more cleanly than the other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43906", + "right": "issue:45290", + "accept": false, + "reason": "Both mention chat/template-style behavior, but one is just an isolated reproduction of another issue while the other is a concrete crash on assistant tool-call messages; not enough evidence of the same bug." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Documentation still referencing `pipeline()` after v5 removals is unrelated to SigLIP2 producing bad outputs with AutoModel/pipeline." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Chunked generation inconsistency under compiled forward and a stale `num_return_sequences` warning in paged generate are different problems." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "One is about non-persistent buffers being filled with junk; the other is about a config value reverting on reload." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading in an example script and model loading for Phi-4 multimodal are separate failures." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "A bad import from `image_utils` is unrelated to Qwen3.5 chat crashes." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "CUDA assert in Qwen3 auto device mapping and a category-limit runtime error are different failure modes." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "A janus image-generation test failure and broken `auto_mappings.py` references are not the same underlying bug." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 import parsing/`torch.jit.script` issue is unrelated to flash-attention crashing on `s_aux=None`." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "Missing `device=` in `torch.arange` causing CPU work is unrelated to the bare `except:` swallowing errors." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "LoRA/extended-vocab collapse with tied embeddings and incorrect visual-encoder keys in save_pretrained are different bugs." + }, + { + "left": "issue:43388", + "right": "issue:43582", + "accept": false, + "reason": "A metrics label-truncation bug and an Apple Silicon `caching_allocator_warmup` TypeError do not match." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Embedding reinitialization after resize and a trailing-slash local-path import warning are unrelated." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Ignoring `tokenizer.json` and failing to forward `revision` in `list_repo_files` are different repository-loading issues." + }, + { + "left": "issue:44077", + "right": "issue:44079", + "accept": false, + "reason": "These are separate core-library bugs: `post_init` policy for patchtsmixer versus `ModelOutput` key assignment." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction and a Python 3.13 import parsing error are unrelated." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 import/load failure and the TensorBoard-without-TensorFlow backend error are different compatibility problems." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure is unrelated to local dataset loading in an example script." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "General Python-version compatibility and a specific DebertaV2 import indentation/parsing bug are not the same issue." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer initialization junk and `rotary_pct` not persisting on reload are separate state-handling bugs." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Qwen3OmniMoe video input processing and serve-side KV-cache tensor indexing are different code paths." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "A stale CI test and broken auto-mapping references are both CI-related but not the same underlying change." + }, + { + "left": "issue:44496", + "right": "issue:45803", + "accept": false, + "reason": "Model-type resolution failure when loading a repo and a bare exception swallowing errors are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "`ADD THE DATA` is too vague to tie to the audio feature extraction bug." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Gemma4 resize-token-embedding behavior and Whisper empty-transcription failure are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "SigLIP2 docs/model-processor mismatch and local dataset loading are different documentation/example issues." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Category-count runtime error and Whisper empty-transcription crash are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "Video-classification pipeline processor selection and a notebook syntax error do not share 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": 17, + "review_unit_index": 31, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36246", + "issue:38175", + "issue:38617", + "issue:41628", + "issue:42175", + "issue:42490", + "issue:42907", + "issue:43065", + "issue:43295", + "issue:43299", + "issue:43404", + "issue:43441", + "issue:43479", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43673", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44112", + "issue:44186", + "issue:44220", + "issue:44246", + "issue:44279", + "issue:44336", + "issue:44466", + "issue:44496", + "issue:44561", + "issue:44683", + "issue:44734", + "issue:44913", + "issue:44933", + "issue:44964", + "issue:44987", + "issue:45003", + "issue:45042", + "issue:45216", + "issue:45276", + "issue:45335", + "issue:45375", + "issue:45446", + "issue:45600", + "issue:45632" + ], + "soft_pairs": [ + "issue:43479|issue:45375", + "issue:43295|issue:45276", + "issue:36246|issue:41628", + "issue:36246|issue:38617", + "issue:44186|issue:44734", + "issue:38175|issue:43065", + "issue:42175|issue:44279", + "issue:44683|issue:45446", + "issue:44016|issue:44220", + "issue:44466|issue:44913", + "issue:43404|issue:45216", + "issue:44561|issue:45632", + "issue:44734|issue:45632", + "issue:44246|issue:45003", + "issue:42490|issue:45632", + "issue:43643|issue:44496", + "issue:44964|issue:45600", + "issue:44112|issue:44964", + "issue:43854|issue:44964", + "issue:42907|issue:43441", + "issue:44016|issue:44062", + "issue:43976|issue:44987", + "issue:43295|issue:45335", + "issue:43673|issue:44336", + "issue:43531|issue:45600", + "issue:43824|issue:44933", + "issue:44496|issue:45042", + "issue:43526|issue:43901", + "issue:43299|issue:45216" + ], + "prepared_review_unit_hash": "21a6295b2542464e886e796dc5b61e86b9f5b8dd4cfa068152c89ad033495881", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35884, + "estimated_input_tokens": 8971, + "estimated_eval_tokens": 18198 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is not a true duplicate cluster: it mixes unrelated issues across imports, model loading, configs, attention, tokenizers, docs, and CI. All soft-edge pairs should be rejected.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:44964 is the broadest user-facing regression report here (a concrete model-loading failure with recent transformers) and has relatively strong discussion activity, but it still only loosely represents this heterogeneous set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Both are multimodal-config bugs, but they affect different models and different fields/code paths; not the same underlying defect." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Both mention resize/token embedding behavior, but the regressions are on different models and symptoms; not mergeable as one fix." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Different missing imports for different symbols and modules; same broad theme, not the same bug." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Separate import errors in unrelated modules/symbols, so they are not the same underlying issue." + }, + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "Tokenizer NER/padding crash vs server KV-cache continuation indexing bug; different code paths and failures." + }, + { + "left": "issue:38175", + "right": "issue:43065", + "accept": false, + "reason": "Unrelated model behavior bug vs dummy Conv2d implementation issue; no shared defect." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "Both are dependency-related, but one is about optional backend installation and the other is a vague dependency problem; too different to deduplicate." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Both touch flex_attention, but they are distinct torch/version-check problems in different code paths." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error and fbank feature extraction bug are unrelated." + }, + { + "left": "issue:44466", + "right": "issue:44913", + "accept": false, + "reason": "Tied-weight serialization/device behavior vs rotary_pct persistence in GPTNeoXConfig; different components and fixes." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Both involve saving/tied weights, but they concern different models and different save-pretrained regressions." + }, + { + "left": "issue:44561", + "right": "issue:45632", + "accept": false, + "reason": "Removal of is_torch_fx_available breaking remote-code models is unrelated to local cache-path collisions." + }, + { + "left": "issue:44734", + "right": "issue:45632", + "accept": false, + "reason": "Server request crash and trust_remote_code cache collision are separate bugs with no shared code path." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Import-time slowdown and unsafe sys.modules access are related only loosely; they are not the same bug." + }, + { + "left": "issue:42490", + "right": "issue:45632", + "accept": false, + "reason": "Import-warning behavior for paths ending in slash is unrelated to cache collisions between local model paths." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Missing fields from trust_remote_code config loading vs unrecognized model/config.json issue; different failure modes." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Model-loading failure for one checkpoint vs CI break from removed Sam3LiteText configs; different underlying problems." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "A stale CI test failure is not the same as a real model-loading regression." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Both are load-related, but one is a specific test/model loading issue and the other is a different checkpoint load failure." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized models and FlashAttention failures are unrelated." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error vs AddedToken keyword conflict; unrelated bugs." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python-version compatibility and a specific model-loading failure are not the same defect." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Both mention resize_token_embeddings, but on different model families and different affected embeddings; not a single code-path bug." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Generation cache absence during chunked prefill vs ANSI logging behavior; unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and removed auto_mapping configs are separate regressions." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Different import errors from different modules/symbols; same class of problem, not the same bug." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrecognized-model loading error vs unnecessary torchvision dependency in PIL image processors; different issues." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "Image-processor label reduction bug vs documentation mismatch in TextClassificationPipeline; unrelated." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3-MoE loading regression and Qwen3.5 save_pretrained serialization bug are different model-specific 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": 17, + "review_unit_index": 32, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:33290", + "issue:38175", + "issue:39692", + "issue:41950", + "issue:43295", + "issue:43299", + "issue:43335", + "issue:43381", + "issue:43441", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44075", + "issue:44242", + "issue:44279", + "issue:44291", + "issue:44297", + "issue:44351", + "issue:44368", + "issue:44462", + "issue:44496", + "issue:44521", + "issue:44617", + "issue:44683", + "issue:44779", + "issue:45020", + "issue:45081", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45325", + "issue:45335", + "issue:45357", + "issue:45405", + "issue:45542", + "issue:45584", + "issue:45593", + "issue:45656", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45907" + ], + "soft_pairs": [ + "issue:43335|issue:45593", + "issue:44075|issue:44368", + "issue:45335|issue:45584", + "issue:43976|issue:44279", + "issue:43381|issue:45593", + "issue:43299|issue:45357", + "issue:43901|issue:44496", + "issue:44496|issue:45759", + "issue:44521|issue:45325", + "issue:41950|issue:45020", + "issue:44351|issue:45278", + "issue:43976|issue:45405", + "issue:44242|issue:45200", + "issue:43479|issue:43525", + "issue:44779|issue:45812", + "issue:41950|issue:45823", + "issue:38175|issue:44617", + "issue:43577|issue:45216", + "issue:44683|issue:45542", + "issue:39692|issue:43901", + "issue:43299|issue:43854", + "issue:44279|issue:45405", + "issue:43295|issue:44779", + "issue:41950|issue:43650", + "issue:44462|issue:45081", + "issue:44297|issue:44462", + "issue:45081|issue:45907", + "issue:44016|issue:44291", + "issue:33290|issue:45656", + "issue:43441|issue:45834" + ], + "prepared_review_unit_hash": "221957cd85ef245666f23dcc872c9b372578d769bcb0bfdf572d004c13eed1dc", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36740, + "estimated_input_tokens": 9185, + "estimated_eval_tokens": 18626 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: it mixes unrelated tokenizer, multimodal, config, optimizer, docs, and runtime regression reports. The soft edges are mostly superficial keyword matches, not the same underlying bug, so none should be merged.", + "confidence": 0.95, + "canonical_issue_reason": "No safe canonical issue: the items span many distinct bugs and regressions, so collapsing them into one would be misleading.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the broadest umbrella regression report here ('Recent transformers versions break models using `remote_code`') and is the closest thing to a representative issue, though it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers config construction bug vs D-FINE auxiliary-loss bug; different model families and code paths." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Optimizer SGD args ignored vs LoRA tie_word_embeddings warning; unrelated training/config behaviors." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "t5gemma embedding resize bug vs Whisper empty-transcription failure; different models and failure modes." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python version support issue vs generic dependency issue; not the same concrete bug." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing eval-mode restriction vs D-FINE auxiliary-loss omission; unrelated runtime semantics." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3VL model-loading regression vs Qwen3.5 save_pretrained key serialization bug; different code paths." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Docs mismatch for return_all_scores vs model loading/config recognition failure; unrelated concerns." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model config vs AutoModelForCausalLM text_config unwrapping; different loader failures." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "assistant_masks all-zero in chat template vs Qwen2.5-VL rope index scaling bug; both multimodal, but different subsystems." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Video-classification pipeline processor lookup vs broad remote_code breakage; not the same bug." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "Missing HybridCache import vs broad import errors after upgrade; too vague and not concrete enough to duplicate." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python support mismatch vs peft version floor bump; unrelated packaging/release issues." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Load-balancing loss omission vs Gemma 4 mm_token_type_ids defaulting; different model training paths." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Phi4Multimodal default config initialization vs Llama4Config missing pad_token_id; distinct config bugs." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Deepseek tokenizer regression vs Granite tokenizer regression; both tokenizer-related but different models and symptoms." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Video pipeline processor lookup vs Gemma4 PLE device mismatch; unrelated failures." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "SigLIP2 zero-probabilities inference issue vs Sam3Video CUDA OOM; completely different symptoms and code paths." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "BLIP2 dtype propagation bug vs Qwen3.5 save_pretrained regression; unrelated model-loading behaviors." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 vs TensorBoard/tf backend install issue; different dependency/runtime areas." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "SigLIP2 docs example errors vs TextClassificationPipeline docs mismatch; both documentation issues but not the same bug." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Qwen3VL loading regression vs GLM-4.7-Flash unit-test load failure; different models and loaders." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Generic dependency issue vs PEFT minimum version bump; packaging problems, but not the same underlying issue." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "processor.tokenizer regression in v4.57.5 vs Deepseek tokenizer incorrect results in v5; different regressions and APIs." + }, + { + "left": "issue:41950", + "right": "issue:43650", + "accept": false, + "reason": "video-classification processor lookup vs nonspecific 'ADD THE DATA' issue; no shared underlying bug." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json vs Mistral regex patch crash; different tokenizer pipeline failures." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "tokenizer_class mismatch on save_pretrained vs AutoTokenizer ignoring repository tokenizer.json; separate save/load bugs." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral regex patch crash vs list_repo_files not forwarding revision; unrelated utility functions." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error vs init_empty_weights TypeError; different failure types and code paths." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "OOM with Adafactor in DeepSpeed vs optimizer.step called twice in DeepSpeed; related subsystem but different bugs and symptoms." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Ministral FlashAttention failure vs Kosmos2.5 OCR index error; unrelated model-specific issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 33, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:42175", + "issue:42994", + "issue:43065", + "issue:43122", + "issue:43262", + "issue:43317", + "issue:43441", + "issue:43450", + "issue:43525", + "issue:43643", + "issue:43650", + "issue:43653", + "issue:43673", + "issue:43723", + "issue:43761", + "issue:43784", + "issue:43824", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44162", + "issue:44188", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44297", + "issue:44336", + "issue:44361", + "issue:44466", + "issue:44521", + "issue:44617", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:44898", + "issue:44987", + "issue:45276", + "issue:45290", + "issue:45397", + "issue:45542", + "issue:45593", + "issue:45600", + "issue:45712", + "issue:45823" + ], + "soft_pairs": [ + "issue:43450|issue:44279", + "issue:45600|issue:45712", + "issue:43065|issue:44898", + "issue:42994|issue:43317", + "issue:44466|issue:45276", + "issue:44261|issue:44263", + "issue:43650|issue:45290", + "issue:44623|issue:45397", + "issue:43441|issue:44683", + "issue:44623|issue:44749", + "issue:43653|issue:45593", + "issue:43262|issue:44521", + "issue:43723|issue:44162", + "issue:44016|issue:45823", + "issue:43673|issue:43761", + "issue:42175|issue:43976", + "issue:44361|issue:45593", + "issue:44671|issue:44779", + "issue:43784|issue:45542", + "issue:43901|issue:43994", + "issue:44987|issue:45542", + "issue:43122|issue:43650", + "issue:44297|issue:45397", + "issue:43824|issue:43976", + "issue:44188|issue:44336", + "issue:43525|issue:43881", + "issue:43525|issue:43643", + "issue:43450|issue:44617" + ], + "prepared_review_unit_hash": "395336265ca2c1e82e5d499715cca80fd8d7d76f37f7d39c192c1679a2079774", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 35943, + "estimated_input_tokens": 8986, + "estimated_eval_tokens": 18228 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Only one soft edge looks like a true duplicate pair; the rest are distinct issues across separate subsystems or different failure modes.", + "confidence": 0.72, + "canonical_issue_reason": "43643 is the broader root-cause report: `trust_remote_code=True` returns config objects with missing fields; 43525 looks like a specific `pad_token_id` symptom of that bug.", + "canonical_pr_reason": null, + "best_issue_reason": "43643 has the clearest general reproduction of the config-field-loss bug and is the best anchor for deduping the pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43450", + "right": "issue:44279", + "accept": false, + "reason": "Different bugs: batched video processor shape vs a generic quantized-model save failure." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both are cleanup/CI-related, but they target different stale artifacts and different breakages." + }, + { + "left": "issue:43065", + "right": "issue:44898", + "accept": false, + "reason": "Unrelated model issues: dummy Conv2d in Sam3PixelDecoder vs Perceiver resolution handling." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Save-failure for quantized models vs load/offload failure for dequantized models; different code paths." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Both involve Gemma4 weights, but one is serialization inconsistency and the other is resize_token_embeddings behavior." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Different GLM/MoE bugs: missing rms_norm_eps precision issue vs torch.split return handling." + }, + { + "left": "issue:43650", + "right": "issue:45290", + "accept": false, + "reason": "Generic low-signal issue vs a specific chat_template crash on tool-call messages; not the same bug." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "processor.save_pretrained missing files is separate from gemma-4 zero3 from_pretrained loading failures." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "Different attention-backend failures: FlashAttention on Ministral-3 vs compiled flex_attention on torch>=2.9." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Missing processor save files vs a tokenizer/filtering performance regression; unrelated symptoms." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer special-token registration vs D-FINE auxiliary-loss logic; different components." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Both touch apply_chat_template, but one is audio sampling-rate defaults and the other is multimodal assistant masks." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading regression vs ESM2 model breakage; no shared underlying defect." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error is unrelated to Gemma4 device-mismatch runtime behavior." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Generation cache regression and CLIP hidden_states regression are separate model-path bugs." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "TF backend packaging issue vs Python version compatibility regression; different root causes." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "MLukeTokenizer task AttributeError is unrelated to D-FINE auxiliary-loss behavior." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Different failures: CamemBERT masked-LM predictions vs Deepseek tokenizer correctness." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Sentence-transformers import NameError and TF backend packaging error are separate import/dependency issues." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Docs mismatch for return_all_scores vs a model/pipeline output quality issue; not duplicate bugs." + }, + { + "left": "issue:44987", + "right": "issue:45542", + "accept": false, + "reason": "Loading a specific model failing in v5.1.0 is not the same as the TensorFlow backend dependency problem." + }, + { + "left": "issue:43122", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer output changed across versions vs an unrelated low-signal issue." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer metadata saved incorrectly vs gemma-4 zero3 loading; separate save/load issues." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "ImportError for a missing model class vs Python 3.9 support regression; different failure modes." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention kernel branching under torch.compile is unrelated to ANSI code emission in loading_report." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Llama4Config missing pad_token_id is not the same as GLM-4V loading failure." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": true, + "reason": "Both describe missing fields on configs returned by AutoConfig/trust_remote_code; 43525 is a concrete missing-pad_token_id instance of the broader bug." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Batched video processor shape bug is unrelated to Sam3Video CUDA OOM." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative overall. The single accepted soft edge, issue:43525 with issue:43643, is a plausible duplicate based on the shared missing-config-field failure mode, and the other rejections are framed as distinct failure modes or subsystems without overstating the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 34, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36246", + "issue:41628", + "issue:41950", + "issue:42491", + "issue:42757", + "issue:43065", + "issue:43232", + "issue:43299", + "issue:43352", + "issue:43408", + "issue:43526", + "issue:43531", + "issue:43596", + "issue:43673", + "issue:43901", + "issue:43976", + "issue:44162", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44336", + "issue:44351", + "issue:44373", + "issue:44485", + "issue:44496", + "issue:44561", + "issue:44589", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:44877", + "issue:44987", + "issue:45003", + "issue:45200", + "issue:45250", + "issue:45292", + "issue:45362", + "issue:45375", + "issue:45397", + "issue:45459", + "issue:45561", + "issue:45563", + "issue:45584", + "issue:45593", + "issue:45701", + "issue:45759" + ], + "soft_pairs": [ + "issue:44589|issue:45362", + "issue:41628|issue:43976", + "issue:43901|issue:44246", + "issue:45561|issue:45563", + "issue:44336|issue:44373", + "issue:42491|issue:43299", + "issue:44242|issue:45593", + "issue:44295|issue:45397", + "issue:41950|issue:44220", + "issue:44246|issue:44496", + "issue:43531|issue:45362", + "issue:43352|issue:45250", + "issue:36246|issue:42757", + "issue:44336|issue:44485", + "issue:43526|issue:45003", + "issue:36246|issue:44351", + "issue:45003|issue:45561", + "issue:43673|issue:44561", + "issue:44246|issue:44871", + "issue:44242|issue:44857", + "issue:44162|issue:45200", + "issue:44295|issue:44749", + "issue:36246|issue:43596", + "issue:44877|issue:45375", + "issue:44779|issue:45292", + "issue:45292|issue:45584", + "issue:43901|issue:45003", + "issue:45292|issue:45701", + "issue:43232|issue:44336", + "issue:44987|issue:45459", + "issue:44496|issue:45003", + "issue:43065|issue:43408", + "issue:45375|issue:45759" + ], + "prepared_review_unit_hash": "2576f80e9e2645550051fc4e979259e8b8e963f483c9f9b342950b5e116c298c", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37279, + "estimated_input_tokens": 9320, + "estimated_eval_tokens": 18896 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous regression reports around Transformers v5, model/config loading, and unrelated utility bugs; I did not find any pair that clearly describes the same underlying fix target.", + "confidence": 0.31, + "canonical_issue_reason": "issue:44561 is the broadest, clearest compatibility regression report in the set: a concrete v5 removal breaking trust_remote_code model loading, which makes it the best umbrella issue among otherwise unrelated model-specific failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44561 is the strongest representative issue because it has a specific failure mode, clear upgrade context, and broad downstream impact compared with the more narrow or ambiguous reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Different failures: a missing Float8 storage class versus a Qwen chat crash; no shared code path is evident." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "ImportError vs Python-version compatibility; same package area, but not the same bug." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docstring mismatch for text classification is unrelated to intermittent import slowness." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "Pytest-xdist file race and a stale generation warning are unrelated behaviors." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI output leakage and a wrong docstring are separate documentation/runtime issues." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MoE, but one is LoRA loading on hf4.x/hf5.x and the other is multimodal model loading; too different to merge." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Missing load-balancing loss and missing auxiliary losses in D-FINE are different training-loss bugs." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Buffer access on position_ids and a Gemma-4 ZeRO-3 loading problem are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification pipeline processor lookup and fbank feature extraction are different pipeline/audio issues." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import performance and unrecognized model config are not the same underlying bug." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Both mention Qwen models, but one is sliding-window behavior and the other is a chat crash; different code paths." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "A model-specific Flash Attention support gap is not the same as the generic Flash Attention 2.0 issue." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Different missing symbols from different libraries; both are import errors but not one duplicate bug." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI-code output and GLM-5 RoPE implementation are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "Image label reduction logic and unsafe sys.modules access are distinct bugs." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Missing Qwen2.5 VL export and missing HybridCache export are separate symbol-export regressions." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "sys.modules access safety and captured-info file races do not share a fix target." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Both are v5 regressions, but one is generation cache/chunked prefill and the other is trust_remote_code import behavior; too broad to merge." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import slowdown and Gemma EOS-token config inconsistency are unrelated." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Load-balancing loss omission and float16 CUDA crash are different loss/runtime failures." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma-4 mm_token_type_ids defaulting are separate model issues." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer access and post-upgrade data-filter slowness are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "ImportError for a missing processor class and a zero-sized tensor IndexError are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Both involve strict config handling, but they are different model configs and different symptoms; too broad to treat as one duplicate." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer output regression and resize_token_embeddings/output_embeddings mismatch are separate tokenization/model-editing bugs." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize behavior and Whisper empty-transcription generation failure are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "TextClassificationPipeline docs and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Output-embedding resize bug and version-dependent tokenization changes are not the same issue." + }, + { + "left": "issue:43232", + "right": "issue:44336", + "accept": false, + "reason": "Generation kwargs after sync_gpus and ANSI-code emission are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "A model-loading failure on transformers>=5.1.0 and hidden tokenizer errors from protobuf handling are different bugs." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Model config recognition and sys.modules access safety are different parts of the loading stack." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "SAM3 pixel-decoder implementation and SAM3 tracker/model-type warning are different issues." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "A strict-config field omission and AutoModelForCausalLM text_config unwrapping are separate composite-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": 17, + "review_unit_index": 35, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:42371", + "issue:42886", + "issue:42994", + "issue:43232", + "issue:43257", + "issue:43262", + "issue:43475", + "issue:43482", + "issue:43540", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43673", + "issue:43931", + "issue:44038", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44246", + "issue:44297", + "issue:44361", + "issue:44368", + "issue:44466", + "issue:44623", + "issue:44734", + "issue:44749", + "issue:44869", + "issue:44933", + "issue:44977", + "issue:44995", + "issue:45092", + "issue:45137", + "issue:45200", + "issue:45229", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45335", + "issue:45356", + "issue:45381", + "issue:45397", + "issue:45459", + "issue:45563", + "issue:45588", + "issue:45600", + "issue:45656" + ], + "soft_pairs": [ + "issue:42886|issue:45229", + "issue:43262|issue:44246", + "issue:43257|issue:45137", + "issue:44995|issue:45200", + "issue:44060|issue:44077", + "issue:43475|issue:43540", + "issue:44466|issue:45335", + "issue:45397|issue:45656", + "issue:43653|issue:44162", + "issue:44995|issue:45588", + "issue:44368|issue:45563", + "issue:44368|issue:44977", + "issue:43673|issue:44734", + "issue:45137|issue:45600", + "issue:43931|issue:44368", + "issue:44933|issue:45459", + "issue:44869|issue:45356", + "issue:44297|issue:44623", + "issue:43232|issue:43673", + "issue:44361|issue:44749", + "issue:43646|issue:44162", + "issue:43673|issue:45278", + "issue:44623|issue:45656", + "issue:44749|issue:45656", + "issue:44749|issue:45310", + "issue:42994|issue:43482", + "issue:28282|issue:30990", + "issue:45230|issue:45397", + "issue:42371|issue:45092", + "issue:43475|issue:43643", + "issue:44038|issue:44623", + "issue:43475|issue:45381" + ], + "prepared_review_unit_hash": "a3b5e6c0b97ba765baaeade95f6f6192787dd2fe77833838d6eb66f5bfc010b7", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37147, + "estimated_input_tokens": 9287, + "estimated_eval_tokens": 18830 + }, + "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 regression reports, many centered on v5-era model loading/generation/tokenizer issues. I don\u2019t see any pair that clearly describes the same concrete bug or a mergeable PR-sized fix.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43646 is the broadest representative regression report: it frames a general v5.0 breakage in custom model initialization, which fits the overall upgrade-regression theme better than the more model-specific tickets.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best single issue to anchor the cluster because it is general, user-visible, and closest to the shared theme of Transformers v5 compatibility breakages.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU OOM are unrelated failures in different code paths." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio processor sampling-rate defaults and import slowdown are not the same bug." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Both involve DeepSpeed, but one is Qwen3 MOE weight conversion and the other is a ZeRO3 deque error." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "A stale second-forward cache crash is unrelated to the mm_token_type_ids defaulting issue." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Incorrect tied-weights warnings and patchtsmixer post_init allowance are different model/config behaviors." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "SAM3 video encoder missing attribute and Qwen3OmniMoe video processing ValueError are different bugs." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Serialization of tied lm_head weights is not the same as resize_token_embeddings failing to update decoder embeddings." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma4 from_pretrained with ZeRO3 and duplicated optimizer steps are different DeepSpeed problems." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "BigBirdTokenizer special-token registration and ESM2 breakage are unrelated tokenizer/model issues." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "A model cache crash and a flash-attention sinkless AttributeError are different failure modes." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "A tie_word_embeddings warning and a stale num_return_sequences warning are not the same underlying issue." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "A config warning is not the same as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Generation cache missing and KV-cache continuation indexing are both generation-related, but they are distinct code-path bugs." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "A ZeRO3 deque error and stale auto_mappings references are unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch is not the same as a tied-embeddings warning." + }, + { + "left": "issue:44933", + "right": "issue:45459", + "accept": false, + "reason": "A missing import from image_utils and suppressed tokenizer decode errors are different import/error-handling bugs." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper word-timestamp decode crash and Kimi-K2.5 codec regression are both tokenizer-adjacent, but not the same defect." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Tokenizer save_pretrained config mismatch and processor save_pretrained missing files are separate serialization issues." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "sync_gpus generation kwargs handling and chunked_prefill cache missing are different generation regressions." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "MLukeTokenizer task AttributeError and slowdown after upgrade are unrelated symptoms." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Custom model initialization breakage and ESM2 breakage are both broad regressions, but not the same concrete bug." + }, + { + "left": "issue:43673", + "right": "issue:45278", + "accept": false, + "reason": "Missing GenerationMixin cache in chunked_prefill is not the same as many import errors after upgrade." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Processor serialization missing files and duplicated optimizer steps do not share a code path." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Performance slowdown after upgrade and repeated optimizer stepping are unrelated." + }, + { + "left": "issue:44749", + "right": "issue:45310", + "accept": false, + "reason": "Data filtering slowdown and Qwen3.5 MOE from_pretrained failure are different bugs." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different directions of the serialization/load pipeline." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "PyTorch-missing ImportError and Sentence Transformers hanging on load are unrelated issues." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "The generic bug report provides no concrete overlap with Gemma4 Zero3 from_pretrained." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 settings guidance and InternVL2 meta-initialization incompatibility are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM3 video encoder attribute loss and trust_remote_code AutoConfig field loss are different failures." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe v5 issue and processor save_pretrained missing files are not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 36, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:30990", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:42994", + "issue:43232", + "issue:43377", + "issue:43475", + "issue:43479", + "issue:43504", + "issue:43525", + "issue:43526", + "issue:43673", + "issue:43704", + "issue:43784", + "issue:43824", + "issue:43873", + "issue:43976", + "issue:44038", + "issue:44079", + "issue:44188", + "issue:44222", + "issue:44242", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44661", + "issue:44792", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:44964", + "issue:45081", + "issue:45103", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45375", + "issue:45478", + "issue:45561", + "issue:45712", + "issue:45753" + ], + "soft_pairs": [ + "issue:41628|issue:44933", + "issue:45081|issue:45103", + "issue:43475|issue:44661", + "issue:42994|issue:43525", + "issue:45375|issue:45478", + "issue:43232|issue:44188", + "issue:42673|issue:43704", + "issue:45310|issue:45375", + "issue:43873|issue:44945", + "issue:41628|issue:44351", + "issue:42831|issue:44222", + "issue:44945|issue:45753", + "issue:44222|issue:45310", + "issue:43673|issue:44188", + "issue:36010|issue:41628", + "issue:42994|issue:43824", + "issue:42222|issue:45200", + "issue:42175|issue:45712", + "issue:43232|issue:44792", + "issue:44038|issue:45310", + "issue:43526|issue:45561", + "issue:43377|issue:44242", + "issue:44496|issue:45561", + "issue:43504|issue:45561", + "issue:44964|issue:45137", + "issue:42175|issue:44933", + "issue:42898|issue:44568", + "issue:30990|issue:35141", + "issue:43976|issue:44933", + "issue:43479|issue:44869", + "issue:43479|issue:44079", + "issue:44623|issue:45310", + "issue:43784|issue:43824" + ], + "prepared_review_unit_hash": "aae2b7f7a365c5119c8a224012c96782590731b73970ca4049d28b676ffa7711", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37509, + "estimated_input_tokens": 9378, + "estimated_eval_tokens": 19012 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are only loosely similar and should stay separate. The only accepted soft-edge duplicate is issue:42673 / issue:43704, which share the same title and appear to describe the same Qwen3ForCausalLM VRAM leak, with 42673 showing more discussion. No PR cluster is present.", + "confidence": 0.88, + "canonical_issue_reason": "issue:42673 is the better canonical issue because it is the earlier report and has substantially more discussion than issue:43704, while the titles indicate the same Qwen3ForCausalLM VRAM leak scenario.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the best issue overall because it is the clearest anchor for the duplicate pair and the later issue:43704 looks like a repeat report of the same bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Both are import-related, but they name different missing symbols and do not show the same concrete bug." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Tokenizer regex crash vs auto-docstring AttributeError are different code paths and different failures." + }, + { + "left": "issue:43475", + "right": "issue:44661", + "accept": false, + "reason": "SAM 3 vision encoder attribute error and tokenizer-mapping model-addition failure are unrelated." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized model saving failure and missing pad_token_id on Llama4Config are different bugs." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but the visible reports describe different failure modes and not one shared root cause." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus and attention-kernel divergence under torch.compile are different problems." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Identical title and closely matching symptom indicate the same Qwen3ForCausalLM VRAM leak bug." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is a from_pretrained error and the other is a missing config field; not enough to treat as the same bug." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization/offloading behavior and pipeline-parallel output correctness are different issues." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they refer to different missing exports and likely different fixes." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "FP8 accuracy issue versus FP8 save_pretrained bug are related thematically but not the same concrete change." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output mismatch and an infinite loop on a specific image are distinct bugs." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save/pretrained issue and Qwen3.5 MoE from_pretrained error do not point to one shared code-path problem." + }, + { + "left": "issue:43673", + "right": "issue:44188", + "accept": false, + "reason": "Generation cache missing in chunked prefill and attention-kernel branching under torch.compile are different code paths." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Both are import failures, but for different symbols and different breakages." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized model saving and missing Qwen2_5_VL import are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken vitpose models and Gemma 4 token-type-id defaults are separate model-specific bugs." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Packaging/backend dependency issue and leftover dummy classes leaking into dir(transformers) are not the same bug." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and a janus test failure do not share a concrete root cause." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Different Qwen model variants with different failure modes; too broad to merge." + }, + { + "left": "issue:43526", + "right": "issue:45561", + "accept": false, + "reason": "Processor label-reduction bug and pytest-xdist file-race bug are unrelated." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "Missing padding-mask support in MIMI encoder and missing load balancing loss are different bugs." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model load failure and test-worker file race are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:45561", + "accept": false, + "reason": "Beit legacy-field loading issue and xdist race on captured_info.txt are unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque underflow are different problems." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Missing TensorFlow backend dependency and nonexistent image_utils import are different issues." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "clean_up_tokenization_spaces behavior change and missing BOS/EOS with add_special_tokens are distinct tokenizer regressions." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Sentence-Transformers loading hang and embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version compatibility breakage and image_utils import error are not the same bug." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config default initialization bug and Whisper timestamp decoding crash are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config defaulting bug and ModelOutput key assignment bug are different code paths." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The core duplicate call on issue:42673 / issue:43704 looks plausible from the identical title, but the writeup overstates the evidence in a few places. The packet does not show their full descriptions, so phrases like \u201cexact same \u2026 in title and description\u201d are too strong. Also, the justification about having \u201cmore discussion/inbound references\u201d is only partly grounded: 42673 clearly has more discussion, but both items show 0 inbound references in the packet. The soft-edge verdicts themselves are mostly conservative, but the overall summary should be tightened to stay within the visible evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 37, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:22355", + "issue:28282", + "issue:30333", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:36331", + "issue:41628", + "issue:42757", + "issue:42831", + "issue:42898", + "issue:43116", + "issue:43232", + "issue:43278", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43874", + "issue:44060", + "issue:44188", + "issue:44190", + "issue:44206", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44448", + "issue:44483", + "issue:44484", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44933", + "issue:44977", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45137", + "issue:45310", + "issue:45356", + "issue:45405", + "issue:45406", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:45834" + ], + "soft_pairs": [ + "issue:44188|issue:44373", + "issue:42898|issue:45812", + "issue:43278|issue:45137", + "issue:44734|issue:45103", + "issue:30333|issue:33290", + "issue:43874|issue:45092", + "issue:43723|issue:44568", + "issue:42898|issue:45356", + "issue:44060|issue:45125", + "issue:42898|issue:44779", + "issue:45405|issue:45712", + "issue:42898|issue:44448", + "issue:44295|issue:45656", + "issue:44933|issue:45712", + "issue:44222|issue:44297", + "issue:44297|issue:45656", + "issue:43116|issue:44190", + "issue:43479|issue:43644", + "issue:44222|issue:44623", + "issue:42831|issue:44623", + "issue:44206|issue:45103", + "issue:41628|issue:42757", + "issue:22355|issue:28282", + "issue:30333|issue:36331", + "issue:30333|issue:34689", + "issue:44297|issue:45310", + "issue:44222|issue:44484", + "issue:43232|issue:45834", + "issue:43723|issue:44977", + "issue:28282|issue:30333", + "issue:44263|issue:45092", + "issue:30333|issue:35141", + "issue:44483|issue:45406" + ], + "prepared_review_unit_hash": "ed00c924f7aec4a08ed9223bb721939528da7fc92f02b078a0c2bc444d7d6a83", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37555, + "estimated_input_tokens": 9389, + "estimated_eval_tokens": 19034 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are a heterogeneous grab bag of unrelated bugs/regressions (tokenizers, generation, deepspeed, serving, save/load, docs). None of the soft pairs look like the same underlying change or defect, so all are rejected.", + "confidence": 0.99, + "canonical_issue_reason": "No true canonical duplicate emerges from this heterogeneous set; if one issue must anchor the cluster, issue 45812 is the best-documented open regression with broad impact.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45812 is the strongest standalone report: open, detailed, reproducible, and impacts all Granite models.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to a docstring-only position_ids report." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Tokenization-space cleanup behavior change is unrelated to Granite token ID regression." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift during eval is unrelated to a DeepSpeed ZeRO3 deque IndexError." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "KV-cache continuation crash in serve is unrelated to an auto_docstring AttributeError under future annotations." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow not reporting failed jobs is unrelated to an Adafactor/DeepSpeed OOM." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Missing image-processor method for GLM46V is unrelated to remote-code checkpoint/meta-init incompatibility." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Tokenizer loading failure and missing BOS/EOS insertion are different tokenizer regressions." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "General v5 tokenization-space behavior change is unrelated to Kimi-K2.5 codec handling regression." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Incorrect tied-weights warning is unrelated to missing tensor-parallel plan metadata." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Tokenization cleanup behavior change is unrelated to Deepseek tokenizer output corruption." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "PEFT version bump issue is unrelated to leftover dummy-class cleanup." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Tokenization cleanup behavior change is unrelated to Pegasus output regression." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Position_ids buffer access error is unrelated to a double optimizer-step DeepSpeed bug." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Missing image_utils import is unrelated to dummy-class leakage into dir(transformers)." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "FP8 save_pretrained issue is unrelated to tokenizer class metadata mismatch on save." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata mismatch is unrelated to duplicate optimizer steps." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Empty multi-label classification results are unrelated to local dataset loading failure." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Multimodal config default-init bug is unrelated to non-persistent buffer junk values." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "FP8 MoE save_pretrained behavior is unrelated to processor.save_pretrained missing files." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue is unrelated to processor save_pretrained file omission." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "Unsupported center arg in feature extractor is unrelated to auto_docstring crashing on annotations." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "AutoImageProcessor import failure is unrelated to is_offline_mode import failure from huggingface_hub." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Missing transformers.onnx module is unrelated to AutoModel requiring absent PyTorch." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "MLflow reporting on failed jobs is unrelated to CustomTrainer.compute_loss signature mismatch." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow failure handling is unrelated to Llama 3.2 Vision model-loading regression." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save metadata mismatch is unrelated to Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 MoE save behavior is unrelated to the default max_shard_size question." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs sync_gpus bug is unrelated to Kosmos2.5 OCR long-input index error." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading issue is unrelated to Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "Missing PyTorch import error is unrelated to failed-job MLflow reporting." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "GlmMoeDsaIndexer torch.split issue is unrelated to InternVL2 meta-init incompatibility." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "Training-job MLflow reporting is unrelated to embedding reinitialization after resize." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat-completions request rejection is unrelated to Gemma4Processor missing _tokenizer crash." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 38, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:30990", + "issue:33357", + "issue:38175", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:42886", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43232", + "issue:43479", + "issue:43576", + "issue:43723", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:44038", + "issue:44060", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44360", + "issue:44373", + "issue:44442", + "issue:44485", + "issue:44514", + "issue:44521", + "issue:44568", + "issue:44623", + "issue:44704", + "issue:44743", + "issue:44871", + "issue:44913", + "issue:44991", + "issue:45092", + "issue:45103", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45593", + "issue:45753" + ], + "soft_pairs": [ + "issue:44263|issue:44360", + "issue:44373|issue:44485", + "issue:44295|issue:44623", + "issue:44514|issue:45103", + "issue:44261|issue:45092", + "issue:45092|issue:45357", + "issue:43723|issue:44871", + "issue:43723|issue:44991", + "issue:30990|issue:38175", + "issue:43723|issue:44442", + "issue:41628|issue:43824", + "issue:43479|issue:44913", + "issue:42831|issue:45310", + "issue:44263|issue:44373", + "issue:39401|issue:44568", + "issue:42831|issue:44038", + "issue:42757|issue:45278", + "issue:42886|issue:42994", + "issue:43232|issue:45593", + "issue:43756|issue:44060", + "issue:43576|issue:43784", + "issue:42831|issue:45753", + "issue:42886|issue:43066", + "issue:43122|issue:44521", + "issue:42371|issue:44263", + "issue:22355|issue:28218", + "issue:44117|issue:44704", + "issue:30990|issue:33357", + "issue:44295|issue:44297", + "issue:43232|issue:44743" + ], + "prepared_review_unit_hash": "17ae752da0a7d4660a9a1a654bea36c8a277f1e0b1d11f2941c8f61ca2994646", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36309, + "estimated_input_tokens": 9078, + "estimated_eval_tokens": 18412 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is a mixed bag of tokenizer, import, config, and model-specific regression reports rather than a single duplicate thread. None of the soft pairs look like the same underlying bug or change; the broad tokenizer-loading regression issue is the best umbrella node.", + "confidence": 0.88, + "canonical_issue_reason": "issue:43723 is the broadest, most umbrella-like tokenizer regression here and best represents the tokenizer-loading side of this heterogeneous cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43723 is the most general issue in the set and can anchor the cluster better than the highly specific model/config tickets.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Both mention DSA/indexer internals, but one is about torch.split return handling and the other about a missing ReLU; different bugs." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "A docstring mismatch is not the same as a RoPE implementation discussion; no shared concrete failure." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading a buffer-backed position_ids tensor and missing processor save files are unrelated save/load problems." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "One is a Qwen2_5_VL chat-template batching crash; the other is an auto-docstring annotation crash. Different subsystems." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "MLA layernorm epsilon handling and InternVL2 meta-init compatibility are distinct multimodal issues." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Both are multimodal-model regressions, but one is meta initialization for remote-code checkpoints and the other is incorrect visual encoder keys on save." + }, + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Both involve tokenizer/model loading, but one is a generic AutoTokenizer load regression and the other is an eos_token_id config mismatch." + }, + { + "left": "issue:43723", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer load failures, but the affected model and likely root cause are different; too broad to merge as duplicates." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Sentence-Transformers loading hang and zero-probability outputs are unrelated symptoms and code paths." + }, + { + "left": "issue:43723", + "right": "issue:44442", + "accept": false, + "reason": "FastSpeech2ConformerTokenizer load failure is a model-specific tokenizer registration problem, not the same as the v5 AutoTokenizer regression." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols from different model families." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Default multimodal config initialization and GPTNeoX rotary_pct reload behavior are separate config bugs." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy and Qwen3.5 Moe from_pretrained errors are unrelated regressions." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "Indexer split-return behavior and a docstring issue are not the same functional bug." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Offset mapping correctness and add_special_tokens BOS/EOS behavior are different tokenizer behaviors." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "An FP8 accuracy issue and a Qwen3-VL-Moe bug are unrelated." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "A specific missing import from huggingface_hub is not the same as a broad set of import errors after upgrade." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline cache loading for tokenizers and quantized model saving are different storage-path failures." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation-state handling after sync_gpus and D-FINE auxiliary-loss logic are unrelated model behaviors." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Smollm3 RoPE-layer dropping and Qwen3-Next tied-weights warnings are different architecture-specific issues." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Broken transformers env CLI and a sentence-transformers import NameError are separate failures." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FP8 accuracy regression and Qwen3_5 infinite-loop behavior are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline tokenizer cache loading and wrong decoder type in v5 are different tokenizer/config problems." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Tokenization changes across versions and multimodal assistant_mask handling are different bugs." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 API guidance and DSA indexer split behavior do not share a concrete code-path problem." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Missing transformers.onnx module and an extra-space tokenizer bug are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "A tokenizer-mapping None bug and passing kwargs to cached_file in AutoProcessor are separate load-path issues." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "A loading hang and a MacOS bus error are different runtime failures." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Buffer access for position_ids and tokenizer_config class mismatch are unrelated serialization bugs." + }, + { + "left": "issue:43232", + "right": "issue:44743", + "accept": false, + "reason": "Both involve generation/cache behavior, but one resets recurrent states and the other is a separate modular_qwen3_5 cache 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": 17, + "review_unit_index": 39, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:29127", + "issue:30990", + "issue:36296", + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42757", + "issue:42907", + "issue:43066", + "issue:43116", + "issue:43122", + "issue:43278", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43493", + "issue:43784", + "issue:43824", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44222", + "issue:44261", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44373", + "issue:44485", + "issue:44488", + "issue:44514", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:44977", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45446", + "issue:45478", + "issue:45600", + "issue:45753" + ], + "soft_pairs": [ + "issue:44704|issue:45290", + "issue:42907|issue:44945", + "issue:43122|issue:44514", + "issue:41628|issue:45278", + "issue:43976|issue:44945", + "issue:43784|issue:45600", + "issue:39401|issue:44488", + "issue:44261|issue:44373", + "issue:43408|issue:43475", + "issue:42907|issue:43937", + "issue:44360|issue:44485", + "issue:44625|issue:44829", + "issue:39401|issue:44521", + "issue:43116|issue:45200", + "issue:44038|issue:44297", + "issue:29127|issue:30990", + "issue:42907|issue:44964", + "issue:43122|issue:44779", + "issue:44038|issue:44295", + "issue:36296|issue:43116", + "issue:42907|issue:43452", + "issue:43824|issue:44351", + "issue:44038|issue:44222", + "issue:45216|issue:45478", + "issue:42757|issue:44351", + "issue:43278|issue:44945", + "issue:43493|issue:45753", + "issue:42907|issue:44222", + "issue:43122|issue:44977", + "issue:43992|issue:44704", + "issue:43784|issue:45446", + "issue:43066|issue:44779", + "issue:38617|issue:45278" + ], + "prepared_review_unit_hash": "2552135a8a8c9ea209aa57914601d3c7c03b1a822959f2133beb4abd25a3c167", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37179, + "estimated_input_tokens": 9295, + "estimated_eval_tokens": 18846 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: most items are distinct model-, tokenizer-, import-, or training-path bugs, not duplicates. I did not find any soft pair that looks like the same underlying issue or change.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44779 is the most self-contained, actionable regression report in the set, with a clear version boundary and concrete tokenizer misbehavior.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44779 is the best representative issue for triage because it is specific, reproducible, and clearly scoped.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "Different code paths: cached_file kwarg forwarding vs chat-template tokenization crash with tool-call messages." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Unrelated failures: dequantized model saving vs incorrect outputs under pipeline parallelism." + }, + { + "left": "issue:43122", + "right": "issue:44514", + "accept": false, + "reason": "Both are tokenizer/chat-template complaints, but one is a tokenization regression and the other is a batched processor padding crash." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "One is a single missing export; the other is a broad upgrade-time import breakage across many symbols." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python-version compatibility problem vs pipeline-parallel inference correctness are not the same bug." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import NameError in sentence-transformers integration vs stale auto_mappings references to removed configs." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Tokenizer offset-mapping bug for Qwen3 is unrelated to loading a specific model repository." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "Runtime precision/config issue vs a docstring mismatch; not the same underlying defect." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "A warning about mismatched model types is different from an attribute error in SAM3 vision encoder output." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Different subsystems: save/dequantization failure vs invalid generation config handling." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated ML internals: DSA indexer activation vs GLM-5 RoPE implementation." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Config propagation bug in Qwen3.5 is unrelated to flash_attention_3 causing degenerate training." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization-like behavior, but offset mapping and assistant mask generation are different defects." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Example-script multi-label output bug vs Gemma 4 multimodal token-type defaulting are not the same path." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe loading bug vs tokenizer save metadata mismatch; different operations and symptoms." + }, + { + "left": "issue:29127", + "right": "issue:30990", + "accept": false, + "reason": "LayoutLMv3 error-message clarity vs Sentence Transformers loading hang are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Dequantized save failure vs loading Phi-4 multimodal instruct are separate model-loading/saving issues." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenization regressions, but they affect different models and different tokenization behaviors." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe bug vs position_ids buffer read error; no shared code path is evident." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training bug is unrelated to the multi-label classification example script returning empty results." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Saving a dequantized model and gguf_file loading failures are different serialization/loading problems." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Two import errors, but for different symbols and likely different missing-export causes." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Qwen3-VL-Moe loading issue vs FP8 save_pretrained failure are distinct save/load paths." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3.5 and newer versions, but one is save_pretrained checkpoint corruption and the other is from_pretrained loading failure." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Different packages and symbols: huggingface_hub import failure vs transformers import failure." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Training/evaluate dtype drift is unrelated to incorrect outputs under pipeline parallelism." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and Qwen3_5 image-specific infinite loop are different model-specific bugs." + }, + { + "left": "issue:42907", + "right": "issue:44222", + "accept": false, + "reason": "Saving dequantized models vs FP8 moe save_pretrained are similar only at a very broad level, not the same bug." + }, + { + "left": "issue:43122", + "right": "issue:44977", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but they concern different models and likely different fixes." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Missing embed_tokens.weight in UMT5Encoder loading is unrelated to AutoProcessor kwarg forwarding." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "Import NameError from latest transformers vs a PyTorch version check bug in flex_attention are unrelated." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Wrong tokenizer decoder type and incorrect Deepseek tokenization are both tokenizer-related, but not the same concrete defect." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "A single missing import symbol and a broad post-upgrade import breakage are not sufficiently the same root cause." + } + ] + }, + "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": 17, + "review_unit_index": 40, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:33453", + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:43064", + "issue:43116", + "issue:43421", + "issue:43441", + "issue:43450", + "issue:43493", + "issue:43673", + "issue:43825", + "issue:43827", + "issue:43874", + "issue:43906", + "issue:43937", + "issue:43976", + "issue:43986", + "issue:44038", + "issue:44075", + "issue:44117", + "issue:44291", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44779", + "issue:44829", + "issue:44936", + "issue:44977", + "issue:45071", + "issue:45072", + "issue:45081", + "issue:45083", + "issue:45216", + "issue:45310", + "issue:45341", + "issue:45357", + "issue:45405", + "issue:45446", + "issue:45478", + "issue:45561", + "issue:45563", + "issue:45588" + ], + "soft_pairs": [ + "issue:42673|issue:45072", + "issue:43937|issue:44485", + "issue:22355|issue:29127", + "issue:42371|issue:43874", + "issue:43976|issue:45446", + "issue:43064|issue:43673", + "issue:43116|issue:43493", + "issue:42222|issue:43441", + "issue:44075|issue:44936", + "issue:43116|issue:43825", + "issue:45357|issue:45478", + "issue:44038|issue:45478", + "issue:45083|issue:45446", + "issue:43937|issue:45563", + "issue:44117|issue:45588", + "issue:33453|issue:39401", + "issue:39401|issue:44779", + "issue:43827|issue:45446", + "issue:43937|issue:44977", + "issue:42673|issue:43906", + "issue:44291|issue:45071", + "issue:43450|issue:43986", + "issue:45405|issue:45446", + "issue:45341|issue:45561", + "issue:44117|issue:45081", + "issue:38617|issue:41628", + "issue:39401|issue:43421", + "issue:44410|issue:45310", + "issue:44829|issue:45216", + "issue:44484|issue:45310" + ], + "prepared_review_unit_hash": "8328568362319c77935459c38a2f7d7aadf6c1159643df8afd754179cd15a817", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36606, + "estimated_input_tokens": 9152, + "estimated_eval_tokens": 18560 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous; the only noticeable subtheme is a small set of Qwen3.5 save/load regressions, but the listed soft pairs are not strong duplicates. I would not merge any of the soft-edge candidates.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45310 is the closest representative of the small Qwen3.5 serialization/from_pretrained subtheme, but the cluster overall is too mixed to treat it as a true duplicate bucket.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the best single representative by breadth and relevance to the only recurring model-specific regression theme in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "Different failures: VRAM leak in dataloader-threaded generation vs dtype mismatch in bfloat16 inference." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Both touch GLM-5, but one is a GenerationConfig validation error and the other is a RoPE implementation discussion." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Unrelated issues: missing transformers.onnx module vs clearer layoutlmv3 box error messaging." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 API warning and missing image-patch method are different code paths and different bugs." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python-version compatibility regression is unrelated to the flex_attention AuxRequest version check." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Wrong optimizer state assignment in FSDP2/PEFT is not the same as a missing GenerationMixin cache in chunked prefill." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Example-script multi-label output bug and SigLIP2 implementation discrepancy are separate model/task issues." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "VitPose model breakage is unrelated to Ministral FlashAttention failures." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "SGD argument handling and trainer evaluate-after-train failure are different trainer bugs." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Empty multi-label results in an example script is not the same as a pipeline() error-message/doc-support issue." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but one is a save_pretrained key regression and the other is a from_pretrained load error; too different to treat as one bug." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Qwen3-VL-Moe/Transformers 5.0 issue and Qwen3.5 MoE from_pretrained error are distinct regressions." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Unexpected output-length helper behavior is unrelated to the PyTorch version check in flex_attention." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "GenerationConfig invalidity and a stale num_return_sequences warning are different symptoms and different fixes." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer mapping None and flash_attention s_aux handling are unrelated subsystems." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Tokenizer-loading regression and Qwen3 offset_mapping bug are not the same concrete failure." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Different models and different tokenizer bugs: offset_mapping vs incorrect Deepseek results in v5." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() is unrelated to the AuxRequest version check." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "GLM-5 generation config validation and Qwen3.5 flash-attention generation breakage are different bugs." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "VRAM leak in threaded dataloader use is unrelated to an isolated reproduction of another issue." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig type-checking breakage are different initialization/type issues." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video-processor shape bug is different from missing-torchvision crash on AutoProcessor load." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "PEFT minimum-version bump and PyTorch version check are unrelated release/compatibility issues." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "A vague testing_utils bug and a specific xdist race on captured_info.txt are not clearly the same defect." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer mapping fallback and Mistral regex patch crashing on backend_tokenizer are separate tokenizer-load failures." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different code paths." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Wrong offset_mapping and runtime post-processor updates are distinct tokenizer behaviors." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing qwen3next projections and Qwen3.5 from_pretrained errors are unrelated bugs." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Degenerate training with flash_attention_3 is unrelated to incorrect Qwen3.5 save_pretrained checkpoints." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "save_pretrained shard-size question and Qwen3.5 from_pretrained 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": 17, + "review_unit_index": 41, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:38617", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42490", + "issue:42617", + "issue:42673", + "issue:42831", + "issue:43012", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43335", + "issue:43408", + "issue:43441", + "issue:43452", + "issue:43577", + "issue:43582", + "issue:43673", + "issue:43742", + "issue:43874", + "issue:43931", + "issue:43937", + "issue:43986", + "issue:44038", + "issue:44075", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44496", + "issue:44829", + "issue:44933", + "issue:45003", + "issue:45071", + "issue:45278", + "issue:45341", + "issue:45459", + "issue:45464", + "issue:45478", + "issue:45542", + "issue:45656" + ], + "soft_pairs": [ + "issue:42371|issue:44261", + "issue:42673|issue:43582", + "issue:43408|issue:43742", + "issue:42617|issue:45341", + "issue:44933|issue:45278", + "issue:42222|issue:45542", + "issue:42490|issue:42673", + "issue:43937|issue:44496", + "issue:42175|issue:42222", + "issue:38617|issue:44220", + "issue:44483|issue:45464", + "issue:42371|issue:42831", + "issue:43066|issue:43122", + "issue:44075|issue:45656", + "issue:43673|issue:45341", + "issue:43012|issue:43577", + "issue:43874|issue:44261", + "issue:44484|issue:45478", + "issue:44410|issue:45478", + "issue:43097|issue:44493", + "issue:44263|issue:44485", + "issue:41628|issue:45071", + "issue:43012|issue:43408", + "issue:42222|issue:45459", + "issue:43441|issue:44829", + "issue:45071|issue:45542", + "issue:43931|issue:44410", + "issue:43335|issue:45071", + "issue:43931|issue:44038", + "issue:45003|issue:45341", + "issue:43097|issue:43452", + "issue:43986|issue:45341" + ], + "prepared_review_unit_hash": "51d59883150a9aeae12a1b118910608d697f7ab75456d02927eadb72d2aac40f", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "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 mostly a loose collection of unrelated Transformers issues, with only weak thematic overlap around v5 import/API regressions. None of the soft-pair candidates look like the same concrete bug, so I would not merge any of them as duplicates.", + "confidence": 0.44, + "canonical_issue_reason": "issue:45278 is the broadest umbrella for the import-regression subset and best represents the recurring v5 breakage reports in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most general, user-facing report here and the best anchor for duplicate triage among the issue candidates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API guidance vs MLA layernorm precision bug; different subsystems and failure modes." + }, + { + "left": "issue:42673", + "right": "issue:43582", + "accept": false, + "reason": "VRAM leak in multi-threaded Qwen3 loading vs Apple Silicon warmup TypeError; unrelated causes." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "SAM3 tracker config warning vs MobileLLM key error; not the same code path." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "3d_parallel.py runtime issue vs a testing_utils bug; no overlap in concrete failure." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-related, but one is a missing symbol and the other is a broad upgrade regression; not the same bug." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "vitpose model breakage vs TensorFlow backend/package-selection error; different areas." + }, + { + "left": "issue:42490", + "right": "issue:42673", + "accept": false, + "reason": "Local-path import warning vs VRAM leak under dataloader threads; unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GenerationConfig validation failure vs unknown-model loading error; distinct config/loading problems." + }, + { + "left": "issue:42175", + "right": "issue:42222", + "accept": false, + "reason": "Backend packaging issue vs vitpose model failure; not the same underlying defect." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "Missing import symbol vs audio feature extraction issue; different code paths." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Both touch chat APIs, but one is request acceptance and the other streaming inference; not a single concrete bug." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 settings warning vs FineGrainedFP8 accuracy regression; unrelated precision features." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Tokenizer decoder type mismatch vs tokenization change across versions; related domain, but not the same bug." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args ignored vs deepspeed stepping twice; different optimizer bugs." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Generation cache regression vs testing utility bug; no shared failure path." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "PyTorch bfloat16 compile warning vs Blip2 dtype staying float32; different symptoms and causes." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Missing multimodal patch-count method vs q_a_layernorm epsilon precision bug; unrelated." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "save_pretrained shard-size question vs Qwen3.5-Moe loading error; not the same issue." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Missing projections in qwen3next vs Qwen3.5-Moe from_pretrained error; different model families and failures." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Removed config option vs unexpected key warnings in many models; both version-related, but not the same concrete regression." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "torch.split return handling bug vs GLM-5 RoPE implementation discussion; not the same path." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "AutoImageProcessor import error vs PreTrainedConfig type checking regression; different API surfaces." + }, + { + "left": "issue:43012", + "right": "issue:43408", + "accept": false, + "reason": "bfloat16 compile warning vs SAM3 model-type warning; unrelated." + }, + { + "left": "issue:42222", + "right": "issue:45459", + "accept": false, + "reason": "vitpose model breakage vs protobuf-related tokenizer error handling; different layers." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "FlashAttention inference failure vs FlashAttention-3 training degeneration; same broad feature, but not one concrete bug." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "Type-checking regression vs undefined TensorFlow backend error; unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch vs qwen3next missing projections; distinct model-loading defects." + }, + { + "left": "issue:43335", + "right": "issue:45071", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug vs config type-checking regression; unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both involve Qwen3-VL/Transformers 5.0, but one is a specific checkpoint shape mismatch and the other is a generic bug report." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "sys.modules access bug vs testing_utils bug; no shared underlying code path." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Removed tie_embeddings option vs gguf_file from_pretrained breakage; different APIs." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "AutoProcessor video-model crash vs testing_utils bug; 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": 17, + "review_unit_index": 42, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:38617", + "issue:42757", + "issue:42886", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43381", + "issue:43408", + "issue:43502", + "issue:43531", + "issue:43550", + "issue:43824", + "issue:43825", + "issue:43873", + "issue:43874", + "issue:43901", + "issue:43906", + "issue:43976", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44062", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44315", + "issue:44351", + "issue:44368", + "issue:44380", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44655", + "issue:44704", + "issue:44829", + "issue:44987", + "issue:45071", + "issue:45103", + "issue:45245", + "issue:45341", + "issue:45372", + "issue:45375", + "issue:45588", + "issue:45753" + ], + "soft_pairs": [ + "issue:44117|issue:44987", + "issue:43986|issue:44484", + "issue:43408|issue:44368", + "issue:43874|issue:44263", + "issue:43976|issue:45071", + "issue:43986|issue:45372", + "issue:44829|issue:45245", + "issue:43531|issue:45341", + "issue:38617|issue:42757", + "issue:44261|issue:44485", + "issue:43064|issue:44315", + "issue:38617|issue:44351", + "issue:44410|issue:45375", + "issue:44062|issue:44117", + "issue:43097|issue:45588", + "issue:44038|issue:44410", + "issue:44704|issue:45103", + "issue:43502|issue:43992", + "issue:43986|issue:44117", + "issue:43097|issue:43873", + "issue:43824|issue:43986", + "issue:43066|issue:43906", + "issue:43097|issue:43901", + "issue:43825|issue:44655", + "issue:43992|issue:44484", + "issue:43208|issue:43550", + "issue:43012|issue:44368", + "issue:43012|issue:43550", + "issue:42886|issue:43381", + "issue:43352|issue:44380", + "issue:42907|issue:45753", + "issue:42907|issue:44655" + ], + "prepared_review_unit_hash": "f53b3a577ca6ffaf1b094da4a506afbd44ae2159029b07cbc6c0e4cfe942f66f", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37187, + "estimated_input_tokens": 9297, + "estimated_eval_tokens": 18850 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are mostly unrelated Transformers bug reports spanning imports, model configs, training, and docs. None of the soft pairs look like the same underlying bug/change, so all should be rejected. The most suitable anchor issue is the active quantization/offloading report.", + "confidence": 0.93, + "canonical_issue_reason": "issue:43873 is the most active and broadly scoped report in the set, with the highest discussion/inbound reference signal, so it is the best canonical anchor.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 stands out as the most representative and triage-worthy issue here because it is open, has substantial discussion, and concerns a broadly impactful runtime behavior.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Both are tokenizer/loading failures, but one is a tokenizer mapping assumption and the other is a specific model-loading regression; not the same bug." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "Video AutoProcessor/torchvision crash vs save_pretrained shard-size question; unrelated code paths." + }, + { + "left": "issue:43408", + "right": "issue:44368", + "accept": false, + "reason": "A model-type mismatch warning and a tie_word_embeddings warning are different config-validation issues." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Missing image-patch helper in GLM46V and torch.split return handling in GlmMoeDsaIndexer are separate model bugs." + }, + { + "left": "issue:43976", + "right": "issue:45071", + "accept": false, + "reason": "Both mention version breakage, but one is Python compatibility and the other is PretrainedConfig type checking; different failures." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Different crashes in different loading stacks: missing torchvision vs missing ReasoningEffort import from mistral_common." + }, + { + "left": "issue:44829", + "right": "issue:45245", + "accept": false, + "reason": "FlashAttention-3 degenerate training and category-count overflow are unrelated problems." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "A Qwen3-MoE sliding_window bug and a testing_utils bug do not share the same underlying cause." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Two separate ImportErrors from different symbols/modules; same symptom type, but not the same bug." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "RMS norm epsilon precision/config issue vs GLM-5 RoPE implementation issue; different model internals." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Optimizer-state corruption under FSDP2/PEFT is unrelated to Liger Kernel application with model_init." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports: configuration_utils.layer_type_validation vs transformers.HybridCache." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "One is missing attention projections in qwen3next; the other is a vision config field being dropped by @strict." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "AddedToken keyword duplication and TOKENIZER_MAPPING_NAMES returning None are different tokenizer-path failures." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "A removed config flag and a flash-attention sink-less crash are unrelated issues." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Both concern Qwen variants, but one is a broad v5 compatibility issue and the other is a specific missing-projection bug." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "Passing kwargs to cached_file and auto_docstring crashing on future annotations are separate utility-layer bugs." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "local_files_only still making API requests is unrelated to UMT5Encoder missing embed_tokens.weight." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Video model loading without torchvision is unrelated to tokenizer mapping returning None." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "A removed embedding-tie config flag and quantization offloading behavior are different problems." + }, + { + "left": "issue:43824", + "right": "issue:43986", + "accept": false, + "reason": "Different loading failures: missing Qwen2_5_VL export vs missing torchvision for a video model." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Wrong tokenizer decoder type and an isolated reproduction of another issue are not enough to establish the same bug." + }, + { + "left": "issue:43097", + "right": "issue:43901", + "accept": false, + "reason": "Config flag removal and outdated pipeline docs about return_all_scores are unrelated." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Pipeline task-support messaging and save_pretrained on Pipeline objects are different features and failure modes." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing UMT5Encoder weights and the max_shard_size default question do not share an underlying code-path issue." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM training blockers and Bamba torch.compile/SDPA failure are unrelated model-specific bugs." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "A bfloat16 compile warning and a tie_word_embeddings warning are different warning sources and behaviors." + }, + { + "left": "issue:42886", + "right": "issue:43381", + "accept": false, + "reason": "Offline cache loading failure and eval-mode gradient checkpointing incompatibility are separate runtime issues." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Unsupported Flash Attention 2 for Nemotron and GPT2 attention-scaling being ignored are distinct backend/model bugs." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Saving dequantized models and Qwen3_5 infinite-image-loop behavior are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both mention saving, but one is dequantized model persistence and the other is Pipeline.save_pretrained; different artifacts and code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 43, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:38617", + "issue:42490", + "issue:42757", + "issue:42898", + "issue:42994", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43208", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43525", + "issue:43526", + "issue:43582", + "issue:43611", + "issue:43618", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43867", + "issue:43901", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44077", + "issue:44222", + "issue:44230", + "issue:44242", + "issue:44263", + "issue:44380", + "issue:44483", + "issue:44485", + "issue:44514", + "issue:44655", + "issue:44683", + "issue:44938", + "issue:45216", + "issue:45290", + "issue:45381", + "issue:45440", + "issue:45478", + "issue:45479", + "issue:45593", + "issue:45753" + ], + "soft_pairs": [ + "issue:44380|issue:45479", + "issue:43352|issue:43976", + "issue:43064|issue:44380", + "issue:44483|issue:44514", + "issue:36683|issue:43582", + "issue:43208|issue:43526", + "issue:44222|issue:45478", + "issue:42490|issue:43012", + "issue:43012|issue:44683", + "issue:43618|issue:43761", + "issue:42757|issue:43824", + "issue:43618|issue:44380", + "issue:38617|issue:43502", + "issue:44483|issue:45290", + "issue:44038|issue:44230", + "issue:43299|issue:43867", + "issue:43756|issue:44485", + "issue:44483|issue:45381", + "issue:43901|issue:44655", + "issue:43299|issue:45753", + "issue:36683|issue:43475", + "issue:42994|issue:44655", + "issue:43352|issue:43937", + "issue:44077|issue:44380", + "issue:43867|issue:45216", + "issue:36010|issue:42757", + "issue:42898|issue:43066", + "issue:36683|issue:44077", + "issue:43611|issue:44655", + "issue:43976|issue:44938", + "issue:36683|issue:43525", + "issue:43352|issue:45593", + "issue:44263|issue:45440", + "issue:43618|issue:45216", + "issue:36683|issue:44242" + ], + "prepared_review_unit_hash": "9cb4dc6e8e88f0eba9c6a4ddc7da89b95297253b4625bbed65bc727b9c6871ec", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 37832, + "estimated_input_tokens": 9458, + "estimated_eval_tokens": 19172 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated Transformers bug reports and regressions. I did not find any soft pair that convincingly describes the same underlying issue or code change, so no merges are warranted from the provided evidence.", + "confidence": 0.34, + "canonical_issue_reason": "issue:43611 is the broadest, most central regression report in the set: a Transformers 5.0 loading breakage affecting models with `base_model_prefix`, which is more representative of the cluster\u2019s general v5-regression theme than the narrower model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best single issue to represent this cluster because it is a clear, high-impact version-regression report with wide applicability and strong triage value.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Different failures: GPT-2 attention scaling under SDPA/FlashAttention vs. degenerate zero loss in sequence classification with `num_labels=1`." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "One is a model/backend support limitation, the other is a Python version compatibility failure; no shared code-path bug." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Trainer/FSDP2/PEFT optimizer-state corruption is unrelated to GPT-2 attention scaling behavior." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "An API request validation issue is unrelated to a batched `apply_chat_template` crash in a VL processor." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing `vocab_size` and Apple Silicon allocator warmup TypeError are distinct bugs in different areas." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training blockers and BeitImageProcessorFast label reduction are unrelated model/processor bugs." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 `save_pretrained` on MoE and Qwen3.5 MoE `from_pretrained` load failure are different directions of failure and not the same concrete bug." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Local-path import warning and a bfloat16 compilation precision warning are unrelated." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "A PyTorch warning during bfloat16 compilation is not the same as flex_attention failing on torch >= 2.9." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both involve CLIP, but one is missing attentions and the other missing hidden_states; these look like separate output-field regressions." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Importing `is_offline_mode` from huggingface_hub and importing `Qwen2_5_VLForConditionalGeneration` from transformers are different import-path issues." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "CLIP output field assignment and GPT-2 attention scaling under specific backends are unrelated." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "A missing internal import and unwanted network requests despite `local_files_only=True` are different failure modes." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "A chat-completions API rejection is unrelated to tool-call assistant message handling in `apply_chat_template`." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Qwen3-VL-Moe loading breakage and FP8 inference support are not the same bug." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Qwen3VL MoE loading regression and a state-dict sorting load error are distinct loading problems." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE-layer mismatch and GLM-5 RoPE implementation discussion are not the same underlying issue." + }, + { + "left": "issue:44483", + "right": "issue:45381", + "accept": false, + "reason": "API request acceptance and Qwen2.5-VL video position-id correctness are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Documentation drift for `return_all_scores` and inability to save Pipeline objects are different issues." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "A Qwen3VL MoE model-loading regression is not the same as an infinite-loop generation bug on a specific image." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Gemma3Config missing `vocab_size` and SAM 3 video missing `fpn_position_embeddings` are separate missing-attribute bugs in different models." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure and saving Pipeline objects are different save-path bugs." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "FlashAttention 2 unsupported for Nemotron and GLM-5 generation config invalid are unrelated model-specific issues." + }, + { + "left": "issue:44077", + "right": "issue:44380", + "accept": false, + "reason": "`post_init` allowance in patchtsmixer is unrelated to GPT-2 attention scaling being ignored." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "State-dict sorting load error and Qwen3.5 checkpoint saving regression are different persistence bugs." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols and different upstream modules; not the same missing export." + }, + { + "left": "issue:42898", + "right": "issue:43066", + "accept": false, + "reason": "Tokenizer decoder-type behavior change and wrong tokenizer decoder type are related only superficially; the reported problems and likely fixes differ." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "A config attribute missing on Gemma3 and a `post_init` constraint issue in patchtsmixer are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Model loading via `base_model_prefix` and Pipeline serialization are separate APIs with different breakages." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Python version support failure in Transformers 5.1.0 and a Python 3.14 load failure are not the same issue." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3Config missing `vocab_size` and Llama4Config missing `pad_token_id` are different missing-config-field regressions." + }, + { + "left": "issue:43352", + "right": "issue:45593", + "accept": false, + "reason": "Unsupported FlashAttention 2 for Nemotron and D-FINE auxiliary-loss omission are unrelated model bugs." + }, + { + "left": "issue:44263", + "right": "issue:45440", + "accept": false, + "reason": "A `torch.split()` return-value issue in GlmMoeDsaIndexer and DeepseekV3MoE divergence from remote implementation do not describe the same bug." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "Missing CLIP attentions and incorrect Qwen3.5 checkpoint saving are unrelated regressions." + }, + { + "left": "issue:36683", + "right": "issue:44242", + "accept": false, + "reason": "A missing config attribute and missing load-balancing loss when `output_router_logits=False` are different defects." + } + ] + }, + "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": 17, + "review_unit_index": 44, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:28282", + "issue:33453", + "issue:36296", + "issue:36683", + "issue:41669", + "issue:42490", + "issue:42890", + "issue:43010", + "issue:43064", + "issue:43352", + "issue:43502", + "issue:43575", + "issue:43618", + "issue:43723", + "issue:43867", + "issue:43931", + "issue:43976", + "issue:43992", + "issue:44062", + "issue:44075", + "issue:44077", + "issue:44112", + "issue:44188", + "issue:44246", + "issue:44355", + "issue:44360", + "issue:44380", + "issue:44484", + "issue:44492", + "issue:44556", + "issue:44855", + "issue:44857", + "issue:44938", + "issue:44964", + "issue:44977", + "issue:45071", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45440", + "issue:45478", + "issue:45561", + "issue:45600", + "issue:45850" + ], + "soft_pairs": [ + "issue:44188|issue:44380", + "issue:43064|issue:44492", + "issue:43352|issue:43931", + "issue:43992|issue:45310", + "issue:43010|issue:45278", + "issue:43867|issue:45375", + "issue:44355|issue:44855", + "issue:41669|issue:44246", + "issue:36683|issue:43064", + "issue:42890|issue:45561", + "issue:44938|issue:45071", + "issue:43867|issue:45357", + "issue:43010|issue:45850", + "issue:43992|issue:45478", + "issue:43010|issue:43976", + "issue:44938|issue:45381", + "issue:43618|issue:44077", + "issue:42490|issue:44075", + "issue:36296|issue:43575", + "issue:44360|issue:45440", + "issue:28282|issue:33453", + "issue:43867|issue:44484", + "issue:43618|issue:44857", + "issue:43618|issue:44977", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:43502|issue:44062", + "issue:43723|issue:44556", + "issue:42890|issue:45341" + ], + "prepared_review_unit_hash": "7f99eedff8b3aa07e8bc0f6313560b6bb3f3c152b31f3d0a5e700c78358ac974", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35639, + "estimated_input_tokens": 8910, + "estimated_eval_tokens": 18076 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous. One soft duplicate is clear: 45310 and 45478 are the same Qwen3.5 MoE from_pretrained regression. Most other pairs only share broad themes like loading/import failures or training issues, but not the same concrete bug.", + "confidence": 0.78, + "canonical_issue_reason": "issue:45310 is the cleanest anchor for the only clear duplicate pair in this set: it is the earlier, slightly more specific report of the Qwen3.5 MoE from_pretrained failure, and issue:45478 is essentially the same title later on.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the best overall issue to keep as the representative duplicate target because it is concrete, reproducible, and has an exact duplicate; the rest of the cluster is mostly unrelated or too broad to serve as a better canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44188", + "right": "issue:44380", + "accept": false, + "reason": "Both are attention/backend bugs, but they describe different code paths and failures (compile branching vs SDPA/FlashAttention scaling)." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "One is a distributed training/optimizer-state bug; the other is a cache-strategy typo. No shared underlying fix." + }, + { + "left": "issue:43352", + "right": "issue:43931", + "accept": false, + "reason": "Different model-specific load failures: unsupported Flash Attention vs weight-shape mismatch." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "UMT5 encoder missing embed_tokens.weight is unrelated to the Qwen3.5 MoE from_pretrained regression." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "Both mention imports/regressions, but one is about decorating cache update with no_grad and the other is broad import errors after upgrade; not the same bug." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "State-dict sorting/load error is not the same as a missing config field silently dropped by strict config handling." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Both are import-time problems, but the root causes differ (compiled Python files vs torch.jit.script parsing on Python 3.13)." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Both concern import performance, but 41669 has a specific import-star root cause while 44246 is only a vague slowdown report." + }, + { + "left": "issue:36683", + "right": "issue:43064", + "accept": false, + "reason": "Gemma3Config vocab_size missing and distributed optimizer-state corruption are unrelated." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "A missing set_seed in integration tests is not the same as an xdist race on captured_info.txt." + }, + { + "left": "issue:44938", + "right": "issue:45071", + "accept": false, + "reason": "Python 3.14 import failure and PretrainedConfig type-checking breakage are different compatibility issues." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Load failure with sorted state_dict is not the same as save_pretrained writing incorrect visual encoder keys." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "One is a specific no_grad decoration bug; the other is a vague complaint about many broken models." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "The UMT5 encoder weight-missing issue is unrelated to the Qwen3.5 MoE from_pretrained regression." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache update no_grad and Python-version compatibility for Transformers 5.1.0 are unrelated." + }, + { + "left": "issue:44938", + "right": "issue:45381", + "accept": false, + "reason": "General Python 3.14 loading failure and qwen2.5-vl video position-id bug are different problems." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIPOutput attentions disappearing and patchtsmixer post_init constraints are unrelated model/API issues." + }, + { + "left": "issue:42490", + "right": "issue:44075", + "accept": false, + "reason": "Local-files-only path warning and SGD optimizer args being ignored are unrelated." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Tensor-parallel training bug and Qwen2-57B load OOM are not the same failure mode." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "A missing ReLU in the DSA indexer is unrelated to DeepseekV3MoE divergence from remote implementation." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "PyTorch-missing AutoModel ImportError and tokenizer-loading regression are different issues." + }, + { + "left": "issue:43867", + "right": "issue:44484", + "accept": false, + "reason": "State-dict sorting/load error is unrelated to max_shard_size being 50GB." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "CLIPOutput attentions and LwDetrImageLoss AMP/CUDA crash are unrelated." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "CLIPOutput attentions issue is unrelated to Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "A flaky integration test and a model-loading regression are not the same bug." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "A test seeding issue is unrelated to a stale device override CI failure." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "A flaky test case and an auto_mappings config reference breakage are unrelated." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "local_files_only network leakage and AddedToken keyword duplication are unrelated loader/tokenizer bugs." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Tokenizer loading in v5 and checkpoint reload failures after upgrade are too broad to be the same concrete bug." + }, + { + "left": "issue:42890", + "right": "issue:45341", + "accept": false, + "reason": "A test seeding problem and a testing_utils bug are different CI/test issues." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "These are the same Qwen3.5 MoE from_pretrained error with nearly identical titles; 45478 is a later duplicate of the same concrete bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 45, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:42548", + "issue:42757", + "issue:42831", + "issue:43010", + "issue:43065", + "issue:43232", + "issue:43299", + "issue:43502", + "issue:43526", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43827", + "issue:43874", + "issue:43901", + "issue:44222", + "issue:44230", + "issue:44261", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44655", + "issue:44749", + "issue:44908", + "issue:44938", + "issue:44961", + "issue:45071", + "issue:45125", + "issue:45230", + "issue:45362", + "issue:45375", + "issue:45468", + "issue:45478", + "issue:45850" + ], + "soft_pairs": [ + "issue:44222|issue:45125", + "issue:44387|issue:44556", + "issue:42548|issue:44351", + "issue:43010|issue:43065", + "issue:43502|issue:44485", + "issue:44355|issue:44749", + "issue:45125|issue:45478", + "issue:30990|issue:31795", + "issue:43010|issue:43232", + "issue:30333|issue:33453", + "issue:45125|issue:45375", + "issue:43299|issue:44556", + "issue:43502|issue:44222", + "issue:42831|issue:43874", + "issue:44315|issue:44360", + "issue:43502|issue:44961", + "issue:44355|issue:45071", + "issue:30333|issue:36296", + "issue:44355|issue:44655", + "issue:43827|issue:44961", + "issue:44222|issue:44230", + "issue:33453|issue:34689", + "issue:42757|issue:43576", + "issue:33453|issue:36296", + "issue:43576|issue:45071", + "issue:43576|issue:43824", + "issue:31795|issue:38617", + "issue:44961|issue:45468", + "issue:41628|issue:42548", + "issue:38617|issue:43576", + "issue:43576|issue:44938", + "issue:43673|issue:44492", + "issue:44908|issue:44961", + "issue:44485|issue:44961", + "issue:36296|issue:43901", + "issue:41628|issue:44355", + "issue:36296|issue:43526", + "issue:45362|issue:45850", + "issue:44261|issue:44360", + "issue:44623|issue:45230" + ], + "prepared_review_unit_hash": "ba0b530ff6a1f979552383221100b55dece16a967e5aa249decadb3a1ac55bba", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 40, + "serialized_chars": 38925, + "estimated_input_tokens": 9732, + "estimated_eval_tokens": 19720 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is a loose set of Transformers/HF regressions and docs/import issues, not a near-duplicate set. No soft edge looks like the same underlying bug/change, so all candidate pairs should be rejected.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 44556 is the best broad representative of the cluster\u2019s main theme: version-upgrade regressions that break model reload/loading across Transformers v5.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44556 is the clearest, most generally representative bug report in the set; it captures the repeated v5 loading regression pattern better than the more model-specific or documentation-only issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "Both are FP8/tensor-parallel themed, but one is about FP8 save_pretrained while the other is about a missing tensor-parallel plan for Qwen3.5-MoE; different failure modes." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "One reports CUDA reserved-memory OOM under int4 quantization, the other is a reload failure after upgrading versions; not the same bug." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and likely different missing exports; too broad to treat as duplicates." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "One is about no_grad decoration for cache/layer update methods, the other is a dummy Conv2d in Sam3PixelDecoder; unrelated code paths." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "local_files_only still making API requests is unrelated to a GLM-5 RoPE implementation discussion." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Both mention compiled/v5 behavior, but one is about compiled Python files failing and the other about a Chinese-described slowdown after upgrade; no clear same bug." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is tensor-parallel metadata and the other is a from_pretrained loading error in v5.4.0; not enough evidence of one underlying defect." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers loading hangs versus confusing model.forward documentation; one is a runtime bug, the other is docs." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Cache update no_grad issue versus generation kwargs after sync_gpus; different generation internals and symptoms." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLFlow reporting on training failure is unrelated to tokenizer loading regression." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Both mention Qwen3.5 MoE, but one is tensor-parallel planning and the other is a missing config field silently dropped by strict validation." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Qwen3VL MoE loading breakage in v5.0.0.dev0 is model-specific and separate from the general v4.57-to-v5.3 reload regression." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "API requests despite local_files_only=True is unrelated to FP8 save_pretrained behavior." + }, + { + "left": "issue:42831", + "right": "issue:43874", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue and missing get_number_of_image_patches are different FP8/multimodal problems." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel not applied with model_init is unrelated to the DSA indexer lacking a ReLU." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "local_files_only network leakage and a nonsense-titled issue are unrelated." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled file errors are not the same as PretrainedConfig type-checking regressions." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLFlow failure reporting and tensor-parallel training bugs affect different subsystems and failure modes." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled Python file failures are unrelated to inability to save Pipeline objects." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs still referencing pipeline() are unrelated to the unrelated placeholder issue." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "FP8 save_pretrained and FP8 inference support are related by theme but not the same concrete bug." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer loading regression and Llama 3.2 90B Vision model loading breakage are different model-loading failures." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing is_offline_mode import and broken transformers env command are distinct CLI/API regressions." + }, + { + "left": "issue:33453", + "right": "issue:36296", + "accept": false, + "reason": "Tokenizer loading regression and tensor-parallel training bug are different areas and symptoms." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken env command is unrelated to PretrainedConfig type checking." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Broken env command and missing Qwen2_5_VLForConditionalGeneration import are separate issues." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Docs confusion for model.forward is unrelated to the missing layer_type_validation import error." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "Placeholder/nonsense issue versus Gemma-4 audio positional encoding bug; no duplicate signal." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "AutoImageProcessor import failure and PreTrainedModel import failure are both imports, but not the same missing symbol or path." + }, + { + "left": "issue:38617", + "right": "issue:43576", + "accept": false, + "reason": "Missing import from configuration_utils and a broken env CLI command are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:44938", + "accept": false, + "reason": "Broken env command and Python 3.14 load failure are different compatibility problems." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Generation cache missing in chunked_prefill and a typo in cache strategies are unrelated." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs being ignored is unrelated to the placeholder issue." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training bug and docs mentioning return_all_scores are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:44355", + "accept": false, + "reason": "Importing AutoImageProcessor fails versus compiled Python files erroring; different problems." + }, + { + "left": "issue:36296", + "right": "issue:43526", + "accept": false, + "reason": "Tensor-parallel training bug and BeitImageProcessorFast reduce_labels bug are different subsystems." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Qwen3.5-35B chat crash and the broad 'too many models break' complaint are not a single concrete bug." + }, + { + "left": "issue:44261", + "right": "issue:44360", + "accept": false, + "reason": "Layernorm config precision error and DSA indexer ReLU issue are unrelated." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Missing files on processor.save_pretrained and a generic bug report 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": 17, + "review_unit_index": 46, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43704", + "issue:43827", + "issue:43874", + "issue:43976", + "issue:44038", + "issue:44077", + "issue:44246", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44655", + "issue:44704", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:44945", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45125", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45542", + "issue:45753", + "issue:45850" + ], + "soft_pairs": [ + "issue:44273|issue:45542", + "issue:44273|issue:45003", + "issue:44393|issue:45753", + "issue:44038|issue:45125", + "issue:44393|issue:45071", + "issue:43576|issue:44556", + "issue:43525|issue:45850", + "issue:41669|issue:43673", + "issue:44556|issue:45092", + "issue:44492|issue:45850", + "issue:43572|issue:45850", + "issue:42371|issue:44492", + "issue:44704|issue:44908", + "issue:42890|issue:44908", + "issue:45278|issue:45850", + "issue:44246|issue:44273", + "issue:43408|issue:44315", + "issue:43976|issue:44393", + "issue:43352|issue:44393", + "issue:44297|issue:45230", + "issue:43576|issue:44908", + "issue:44295|issue:45230", + "issue:44908|issue:45071", + "issue:41669|issue:44351", + "issue:41669|issue:45341", + "issue:44492|issue:44945", + "issue:43874|issue:45850", + "issue:44393|issue:45478", + "issue:44077|issue:44861", + "issue:43827|issue:44908", + "issue:31795|issue:33453", + "issue:41669|issue:44908", + "issue:44315|issue:44829", + "issue:43576|issue:44273", + "issue:44655|issue:44861", + "issue:44360|issue:45468", + "issue:45230|issue:45310", + "issue:44273|issue:44908", + "issue:43519|issue:44485", + "issue:43519|issue:44908", + "issue:41669|issue:43704" + ], + "prepared_review_unit_hash": "d98dfe4a6ac6e40dea166701353f7e49b023fdea0bbce10aff582cf635e5fbea", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 41, + "serialized_chars": 39338, + "estimated_input_tokens": 9835, + "estimated_eval_tokens": 19926 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly false-positive similarity: these issues span unrelated bugs, v5 regression complaints, model-specific errors, docs, and performance/import topics. No soft edge is safe to merge as the same underlying bug/change.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:44273 is the closest broad anchor because it is open and about loading/lazy-loading behavior, which overlaps the import-time regressions in the cluster. It is still only an umbrella, not a true duplicate canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Both mention loading/import behavior, but one is lazy loading and the other is a TensorFlow backend detection problem; not the same bug." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Import slowdown and unsafe sys.modules access are related themes, but they are distinct failure modes with different fixes." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Both are Qwen3-VL/Qwen3_5 user-facing failures, but one is 2D box output corruption and the other is an infinite-loop/image-specific hang." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Both involve Qwen3 MoE models, but one is a generic transformers 5.0 bug report and the other is a missing tensor-parallel plan attribute." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "A Qwen3-VL output bug and a PretrainedConfig type-checking regression are unrelated code paths." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Both are v5 upgrade regressions, but one breaks the env command and the other breaks checkpoint reload; different issues." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "A missing pad_token_id on Llama4Config is a concrete config bug; the other is only a generic complaint about many models breaking." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-star removal and missing GenerationMixin cache in chunked prefill are different regressions." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": false, + "reason": "Both concern v5 compatibility, but one is checkpoint reload after upgrade and the other is remote-code/meta-init incompatibility." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A cache-strategy typo is not the same underlying problem as a broad v5 breakage complaint." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLmConfig missing pad_token_idx is a specific config regression, not the same as a generic 'too many models break' report." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 API guidance and a cache-strategy typo are unrelated." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "Processor kwargs forwarding and scheduler kwargs handling affect different subsystems and code paths." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Missing set_seed in tests and inverse_sqrt scheduler kwargs are unrelated." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "A specific import-error report after upgrade is not the same underlying bug as a vague general complaint." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Both involve import behavior, but one is intermittent import slowness and the other is lazy-loading malfunction; not the same concrete defect." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "Model-type warning for sam3_tracker and Liger Kernel not applied with model_init are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python-version compatibility failure and Qwen3-VL hallucination/output error are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2 support missing for NemotronH vs a Qwen3-VL output bug are different model-specific problems." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer save metadata mismatch and a generic bug report are not the same issue." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "An env CLI breakage and a scheduler kwarg bug are different components." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Position_ids buffer access and a generic bug report do not match." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs and PretrainedConfig type checking are unrelated." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star cleanup and missing HybridCache export are different import-surface issues." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "A model import-performance issue and a testing_utils bug are unrelated." + }, + { + "left": "issue:44492", + "right": "issue:44945", + "accept": false, + "reason": "A cache-strategy typo and incorrect pipeline-parallel output are different bugs." + }, + { + "left": "issue:43874", + "right": "issue:45850", + "accept": false, + "reason": "A missing multimodal-token helper and a generic v5 complaint are not the same change or bug." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Two Qwen3.5/Qwen3-VL reports, but one is bbox output hallucination and the other is a from_pretrained error for Qwen3.5 MoE." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Optional post_init handling and tied-weight-key AttributeError are different implementation bugs." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs still referencing pipeline() and scheduler kwargs handling are unrelated." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Documentation confusion about forward args and tokenizer-loading regression are not the same issue." + }, + { + "left": "issue:41669", + "right": "issue:44908", + "accept": false, + "reason": "Import-time model loading changes and scheduler kwargs are unrelated." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel application with model_init and flash_attention_3 training degeneration are different problems." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "v5 env CLI breakage and lazy-loading malfunction are separate code paths." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Pipeline save_pretrained failure and tied-weight-key AttributeError are unrelated." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "DSA indexer/ReLU discussion and Gemma4AudioRelPositionalEncoding are completely different areas." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Both are generic bug reports around Qwen3.5 Moe, but the descriptions point to different failure modes and no shared concrete fix path." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy-loading/import behavior and scheduler kwargs are different subsystems." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are separate model math bugs." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "A processor timestamp bug and a scheduler kwargs bug are unrelated." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-star removal/performance and VRAM leakage in multi-threaded dataloaders are different issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 47, + "review_unit_count": 47, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-402", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:37428", + "issue:38617", + "issue:42371", + "issue:42913", + "issue:43066", + "issue:43298", + "issue:43502", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43606", + "issue:43618", + "issue:43704", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44315", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44861", + "issue:44908", + "issue:44998", + "issue:45092", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45850" + ], + "soft_pairs": [ + "issue:31795|issue:34689", + "issue:31795|issue:37428", + "issue:44315|issue:45092", + "issue:44485|issue:45468", + "issue:43704|issue:44485", + "issue:42371|issue:43704", + "issue:43502|issue:43519", + "issue:43519|issue:44623", + "issue:43519|issue:43704", + "issue:44263|issue:45468", + "issue:43606|issue:45468", + "issue:44261|issue:45468", + "issue:42913|issue:43066", + "issue:44908|issue:44998", + "issue:44861|issue:45290", + "issue:44492|issue:44998", + "issue:44998|issue:45850", + "issue:43618|issue:44998", + "issue:44998|issue:45341", + "issue:43525|issue:44861", + "issue:43572|issue:44861", + "issue:31515|issue:38617", + "issue:44230|issue:44623", + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230" + ], + "prepared_review_unit_hash": "9d58d72e9dd4d0bea4f322d9024f6ba304d633bf070d0057703d98c5c7bed327", + "component_budget": { + "node_count": 402, + "item_count": 402, + "soft_pair_count": 1504, + "serialized_chars": 775931, + "estimated_input_tokens": 193983, + "estimated_eval_tokens": 388222 + }, + "budget": { + "node_count": 35, + "item_count": 35, + "soft_pair_count": 29, + "serialized_chars": 28339, + "estimated_input_tokens": 7085, + "estimated_eval_tokens": 14426 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose collection of Transformers issues around loading, tokenizer/model regressions, and v5 compatibility. The soft links look like weak topical similarity rather than true duplicates; I would not merge any of the candidate pairs.", + "confidence": 0.78, + "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set and best matches the cluster\u2019s general theme of v5 regressions breaking many models, though it is too vague to serve as a precise duplicate target.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the most generally representative issue for this cluster because it describes widespread breakage, not a single narrow symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Both involve model/loading docs or regressions, but one is a documentation confusion issue and the other is a specific Llama 3.2 model-loading failure." + }, + { + "left": "issue:31795", + "right": "issue:37428", + "accept": false, + "reason": "Different problems: documentation ambiguity vs an import error for a flash-attention helper." + }, + { + "left": "issue:44315", + "right": "issue:45092", + "accept": false, + "reason": "Both touch model initialization, but one is about Liger Kernel not being applied and the other is remote-code checkpoint incompatibility with v5 meta init." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Both are architecture-specific math/positioning bugs, but they affect different models and different implementations." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leakage in multi-threaded inference is unrelated to RoPE implementation issues." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 API guidance and VRAM leakage are unrelated issues." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "One is unwanted network access during local loading; the other is a timestamp calculation bug in a processor." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Processor timestamp math and missing saved files are different code paths and symptoms." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "A tokenizer/processor timestamp bug is not the same as a dataloader-thread VRAM leak." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "Different models and different failure modes: torch.split return handling vs audio positional encoding." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to Gemma audio positional encoding." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "Precision error from missing rms_norm_eps is not the same as an audio positional encoding bug." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both concern tokenizer behavior in v5, but one is a broad behavior change and the other is a specific wrong decoder type issue." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs handling is unrelated to the unrelated 'Unemployment' issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "A tied-weights helper crash and a chat template crash are distinct failures in different utilities." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "A cache-strategy typo is unrelated to the unrelated 'Unemployment' issue." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "The second is a broad v5 breakage report; the first is not a technical bug report and is clearly unrelated." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression is unrelated to the unrelated 'Unemployment' issue." + }, + { + "left": "issue:44998", + "right": "issue:45341", + "accept": false, + "reason": "A testing_utils bug and an unrelated non-technical issue are not duplicates." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config and a tied-weights list crash are different config/utility bugs." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to _get_tied_weight_keys crashing on list input." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Both are import/loading-related, but one is slow checkpoint loading and the other is a missing import symbol." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 support for Qwen models and missing processor save files are different feature/bug areas." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow reporting on failed training jobs is unrelated to slow checkpoint loading." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Slow checkpoint loading and tensor-parallel training bugs are different problems." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Loading slowness is not the same as confusing forward-argument documentation." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen and v5-era errors, but one is FP8 inference support and the other is a from_pretrained failure." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Slow from_pretrained checkpoint loading and tokenizer-loading regression are related only at a high level, not the same bug." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching support for audio models is a different capability gap from Qwen FP8 inference support." + } + ] + }, + "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": 17, + "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": "8a6c846e734d045b44600385112d43b3f53d7b57a0c7532082b0b6f99aff1343", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2479, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2479, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These three issues do not look like duplicates: one is about Tapas tokenization mismatching TensorFlow, one is about an unhelpful auth error when downloading a private model, and one is about failing to load bert-base-uncased. They share only broad Transformers/model-loading context.", + "confidence": 0.89, + "canonical_issue_reason": "issue #16618 is the most concrete and user-facing of the set, with a specific model-load failure tied to bert-base-uncased.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the best representative issue because it has the clearest reproducible failure and a specific affected model.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Different symptoms and layers: Tapas tokenization parity vs. a model loading failure for bert-base-uncased." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Auth/download error for private models is unrelated to a load failure for a public model." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 3, + "component_count": 17, + "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": "13f17d72a3aed6f0289d0cadad0118b386fab32c1f56fe21683d43065ddf5517", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are superficially similar \u201csupport for more architectures\u201d requests, but they target different features: Flash Attention 2 vs GGUF. They should not be treated as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a duplicate of the other; they request support for different capabilities and should remain separate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Same broad framing, but different underlying work: Flash Attention 2 support is unrelated to GGUF support. They would not be merged into one issue." + } + ] + }, + "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": 17, + "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-27088-5", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:43834", + "issue:44365", + "issue:44370" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:28936", + "issue:28936|issue:43834", + "issue:28936|issue:44365", + "issue:28936|issue:44370", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:44365|issue:44370" + ], + "prepared_review_unit_hash": "a6c85be2455c5216a65b3503c14ecc827bf8e64f8ae98a75d392a2a78ee8f3e2", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 10, + "serialized_chars": 6443, + "estimated_input_tokens": 1611, + "estimated_eval_tokens": 3478 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 10, + "serialized_chars": 6443, + "estimated_input_tokens": 1611, + "estimated_eval_tokens": 3478 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are all documentation-translation requests, but they point to different target languages or separate placeholder instances, so they should not be deduplicated into one underlying issue.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43834 is the cleanest representative of the generic i18n translation template, but the cluster does not contain a true duplicate set because the requests appear language-specific.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Turkish translation request vs a separate placeholder translation issue; different underlying language target." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Turkish translation request vs a separate placeholder translation issue; not the same language or change." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Specific Turkish request and a generic placeholder translation issue are not the same underlying request." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Turkish vs Spanish docs translation; same broad category but clearly different underlying requests." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish request vs a separate generic placeholder translation issue; different language target." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Spanish request vs a separate placeholder translation issue; not the same concrete translation task." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Spanish request vs a separate placeholder translation issue; different underlying language target." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "Both are template-like i18n issues, but there is no evidence they refer to the same language or exact request." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "Near-identical template text is not enough; likely distinct language-specific translation requests." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "Two placeholder translation issues, but not enough evidence they represent the same concrete request." + } + ] + }, + "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": 17, + "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-36817-15", + "nodes": [ + "issue:36817", + "issue:36879", + "issue:37015", + "issue:41115", + "issue:41211", + "issue:42503", + "issue:42738", + "issue:43671", + "issue:43837", + "issue:43909", + "issue:44121", + "issue:44205", + "issue:45027", + "issue:45177", + "issue:45322" + ], + "soft_pairs": [ + "issue:36817|issue:41211", + "issue:36817|issue:43671", + "issue:36817|issue:36879", + "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:42738", + "issue:42738|issue:45027", + "issue:43671|issue:43837", + "issue:43671|issue:45027", + "issue:43671|issue:45177", + "issue:43671|issue:44121", + "issue:43671|issue:43909", + "issue:43837|issue:43909", + "issue:43837|issue:45322", + "issue:43837|issue:44121", + "issue:43909|issue:44205", + "issue:44121|issue:45322", + "issue:43837|issue:45027", + "issue:37015|issue:45027", + "issue:43837|issue:45177" + ], + "prepared_review_unit_hash": "80b5173ee9ceb62227c9dd5007d0f25731782ad32b1cdd73cca3d95126edbf61", + "component_budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 35, + "serialized_chars": 20708, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 + }, + "budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 35, + "serialized_chars": 20708, + "estimated_input_tokens": 5177, + "estimated_eval_tokens": 10610 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "A loose cluster of model-request issues spanning unrelated models and modalities (text, vision, audio, OCR). They are thematically similar but do not look like duplicate reports of the same underlying request.", + "confidence": 0.94, + "canonical_issue_reason": "issue:43671 is the most active representative of the cluster, but it is still a distinct Qwen3-TTS support request rather than a duplicate of the neighboring ASR/audio/model requests.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43671 is the best single issue to anchor this loose cluster because it is well-discussed and representative of the broader audio-model request subgroup, even though the items are not duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different requested models: EuroBert vs DEIMv2; same pattern of model request, not the same change." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert text model request vs Qwen3-TTS audio support; different code paths and artifacts." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "EuroBert config addition is unrelated to RF-DETR model support." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are distinct model additions with different architectures." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "RF-DETR detection model vs LFM2.5 Audio support; unrelated requests." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "RF-DETR and the weight-sparse transformer request are different concrete models." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR vision model request is not the same as Qwen3-ASR audio support." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR and Qwen3-TTS target different model families and functionality." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are unrelated model requests." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT text-model support vs a sparse-transformer request; not the same bug/change." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT and LFM2.5 Audio are different models/modalities." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are different model additions." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are distinct model requests, not duplicates." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture support and Qwen3-ASR support are different concrete requests." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 and the OpenAI sparse transformer request are unrelated." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 and ModernVBERT are separate model additions." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 and Voxtral TTS support are different model families and modalities." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 vision request and Qwen3-TTS audio request are unrelated." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT and Qwen3-ASR are different support requests." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT is a BERT variant request; Qwen3-TTS is an audio-model request." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "ModernVBERT and the generic RoPE BERT-like request are related in theme but not the same underlying issue." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "BERT-like models with RoPE vs Voxtral TTS; different model classes and code paths." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Qwen3-TTS and Qwen3-ASR are related family requests but distinct tasks and implementations." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are different audio model requests." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-TTS support and the sparse-transformer request are unrelated." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-TTS and Voxtral TTS are separate model integrations." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS support and DeepSeek-OCR2 are different modalities and requests." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are distinct audio model support requests." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-ASR and the sparse-transformer request are unrelated." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are different audio tasks/models." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are not the same underlying change." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR and EUPE are distinct model requests." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "LFM2.5 Audio 1.5B and SAM3-LiteText are different model additions." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Sparse transformer request vs EUPE request; no concrete overlap in the underlying model change." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are separate audio-model support requests." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral TTS are unrelated model requests." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model families and tasks." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 17, + "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-37934-6", + "nodes": [ + "issue:37934", + "issue:42936", + "issue:43240", + "issue:43668", + "issue:43813", + "issue:44031" + ], + "soft_pairs": [ + "issue:37934|issue:43668", + "issue:42936|issue:43240", + "issue:42936|issue:44031", + "issue:42936|issue:43813", + "issue:42936|issue:43668", + "issue:43240|issue:43668", + "issue:43668|issue:44031", + "issue:43668|issue:43813" + ], + "prepared_review_unit_hash": "c731ec4ecbc851041bc7f48f15f2dae9639e74c4fdb30083f95937f2f3ac51ad", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 6293, + "estimated_input_tokens": 1574, + "estimated_eval_tokens": 3404 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 6293, + "estimated_input_tokens": 1574, + "estimated_eval_tokens": 3404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues are only loosely text-similar; they span unrelated areas (normalization behavior, Mask2Former loss plumbing, config typing, quantization typos, and tokenizer warnings) and do not look like duplicates of the same underlying bug.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:44031 is the broadest user-facing regression in the set (tokenizers warning everywhere after a version bump), so it is the best representative issue, though it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Both mention normalization, but one is about Llama4TextL2Norm semantics and the other is a ModernBERTConfig type hint; different code paths and fixes." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Mask2Former ignore_value usage and kwargs not being forwarded to loss are different bugs in different parts of the loss pipeline." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2Former loss handling and tokenizer regex warnings are unrelated subsystems with no shared concrete bug." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "A Mask2Former loss/configuration issue is not the same as a quantization typo in a different feature area." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "These concern unrelated models and defect types: unused ignore_value versus an incorrect type hint." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Passing kwargs to a loss function is a different issue from a config annotation problem." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "ModernBERT config typing and tokenizer regex warnings are unrelated; no shared underlying change or bug." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Incorrect type hint in a config class and a specific typo in a quantization field are distinct defects." + } + ] + }, + "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": 17, + "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": "25defb6e84f9477312afafff52ed668cd4994c028a143d65f45ad21450aa6ce2", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are not duplicates: one requests block decoding behavior, the other requests DeepSeek v3.2 model support. They describe different user-facing changes and different code paths.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good canonical representative for the other; they are unrelated. If forced, 43037 is more specific, while 41084 is too vague to serve as a cluster anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different underlying requests: block decoding configuration vs adding support for a specific model version. No shared bug or change target." + } + ] + }, + "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": 17, + "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": "5290df8cbd033c3775db66ee7ffa8f9bc4be36475fed9a3c237d30ada5cf0904", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two closed issues are related only at the broad Trainer subsystem level, but they describe different problems: one is about `remove_unused_column` behavior/customizability, the other about FLOPS estimation for custom trainers. They should not be deduplicated.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #43598 is the more concrete, narrowly scoped Trainer behavior change; #43600 is a separate FLOPS-accounting concern and not the same underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "#43598 is the better representative issue because it has a specific actionable target (`remove_unused_column`) and a clearer expected fix path.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Different code paths and outcomes: one concerns column pruning customization in Trainer, the other concerns FLOPS estimation for custom trainers. They share the Trainer area but not the same underlying bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 9, + "component_count": 17, + "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": "3066966d528d6f060e5d7a3f64cdc0801355f436c79c94fc5d92707c4eca7a61", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1611, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1611, + "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 Transformers/model-registration level, but they are not the same issue: one asks for Qwen3-Omni auto-class registration, the other asks for sequence-classification support for Qwen3.5.", + "confidence": 0.95, + "canonical_issue_reason": "No canonical duplicate: issue 43684 is a broad AutoModel registration request for Qwen3-Omni, while 44405 is a separate task-specific request for AutoModelForSequenceClassification support on Qwen3.5.", + "canonical_pr_reason": null, + "best_issue_reason": "43684 is the broader, more central issue if a single issue must represent the topic, but it should not subsume 44405 because the model family and requested capability differ.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Reject: different underlying requests. 43684 is about adding Qwen3-Omni registration to auto classes; 44405 is about enabling sequence classification for Qwen3.5. They target different model/config paths and are not the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 10, + "component_count": 17, + "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": "6cd681caaebece3f205458182afd319b3f6f08aeacde9d178eeed33d0e704de9", + "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": "This cluster contains one clearly different issue ('Fett') and three issues with the same 'huggingface' title. The available evidence is limited to title similarity, timestamps, and body-length differences, so the huggingface items are best treated as related-looking but not proven duplicates. I\u2019d keep 43845 as the reference issue only because it is the earliest of that trio.", + "confidence": 0.61, + "canonical_issue_reason": "issue:43845 is the earliest of the three huggingface-titled issues and is the most reasonable anchor for that small cluster, but the evidence here is not strong enough to prove the others are duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43845 is the best overall issue candidate because it predates the other huggingface reports and appears to be the most representative of that subgroup.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles and no body text or other evidence tying them to the same underlying issue." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Different titles and no concrete evidence beyond a weak soft similarity signal." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "Different titles; the packet does not show enough to conclude duplication." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": false, + "reason": "Same title and close timestamps, but the packet only shows similarity signals and body-length differences, not enough to confirm the same issue." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "Same title, but only weak observable evidence is available here; body text is missing, so duplication is not established." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "Same title alone is insufficient; the packet does not provide enough detail to accept this as a duplicate." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but the soft-edge decisions are too strong for the packet. The data only shows title similarity, close timestamps for 43845/43846, and body-length differences; it does not show the actual body text or enough evidence to conclude the three 'huggingface' issues are duplicates with high confidence. In particular, the claims about 'placeholder content' and 'same vague subject matter' overstate what is observable here. A more conservative verdict would keep these as unaccepted soft matches." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 11, + "component_count": 17, + "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-44371-5", + "nodes": [ + "issue:44371", + "issue:44418", + "issue:44572", + "issue:44573", + "issue:44937" + ], + "soft_pairs": [ + "issue:44371|issue:44572", + "issue:44371|issue:44573", + "issue:44371|issue:44418", + "issue:44371|issue:44937", + "issue:44418|issue:44572", + "issue:44418|issue:44573", + "issue:44572|issue:44573", + "issue:44572|issue:44937", + "issue:44573|issue:44937" + ], + "prepared_review_unit_hash": "70723412500245f4618c7719cc67f9f9cffd1ca94d517cf1dbf068c1ce5b36a9", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5794, + "estimated_input_tokens": 1449, + "estimated_eval_tokens": 3154 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5794, + "estimated_input_tokens": 1449, + "estimated_eval_tokens": 3154 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is mostly low-signal spam-like issues plus one generic documentation suggestion; there is no clear underlying bug/change shared across them.", + "confidence": 0.96, + "canonical_issue_reason": "No canonical issue: the items are unrelated and too vague/spammy to represent one duplicated problem.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44418 is the least spam-like and most issue-like title, but it is still far too generic to be a strong canonical target.", + "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." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Both are just '' placeholders; no concrete shared bug or change." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "Spam placeholder versus generic documentation request; not the same underlying issue." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus another spam-like title; no shared concrete problem." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Generic docs suggestion versus spam placeholder; not a duplicate pair." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Generic docs suggestion versus spam placeholder; not a duplicate pair." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Identical spam-like titles, but there is no real underlying bug/change to unify." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus unrelated spam-like title; no concrete shared issue." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder versus unrelated spam-like title; no concrete shared issue." + } + ] + }, + "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": 17, + "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": "5258bfbf88b33c7d05662c4a9169cb35009b0e5a59469f8e7b8b242bbb938e4e", + "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 closed issues have an exact title match and very similar body sizes, so they are likely the same request, but the bodies themselves are not visible here, so this should be treated as a probable duplicate rather than a certain one.", + "confidence": 0.88, + "canonical_issue_reason": "issue #45145 is the better canonical thread because it was created first and already has an inbound reference, making it the more established issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue #45145 is the best representative issue because it is earlier, has slightly more reference activity, and appears to be the original thread for the same topic.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": true, + "reason": "Exact title match plus very similar body length strongly suggests the same underlying issue, but without the bodies this should be read as a likely/probable duplicate rather than a definite one." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The packet supports strong similarity and an exact title match, but it does not expose the issue bodies, so claiming \u201cstrongly overlapping content\u201d or a definite duplicate is a bit too strong. The soft-edge verdict should stay conservative (e.g. likely duplicate / probable same issue), and the confidence should be lower than 0.99 given the limited evidence." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 13, + "component_count": 17, + "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": 14, + "component_count": 17, + "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-45672-2", + "nodes": [ + "issue:45672", + "issue:45674" + ], + "soft_pairs": [ + "issue:45672|issue:45674" + ], + "prepared_review_unit_hash": "5d2a879d76386fd537b9b0f047977585d1167b218b6f1c41c429fa3fad4f1674", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1667, + "estimated_input_tokens": 417, + "estimated_eval_tokens": 1090 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1667, + "estimated_input_tokens": 417, + "estimated_eval_tokens": 1090 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate 4-bit inference bugs: one in Gemma4/audio modules causing a uint8 `torch.finfo()` TypeError, and one in `BitsAndBytesConfig` where `llm_int8_skip_modules` drops the default `lm_head` exclusion and triggers an AssertionError. They share only a broad theme, not the same underlying failure.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 45674 is the better canonical issue for the cluster because it states a clearer, more general regression in 4-bit inference configuration and has a concrete, actionable failure mode. Issue 45672 is a separate Gemma4/audio-module-specific type error.", + "canonical_pr_reason": null, + "best_issue_reason": "45674 is the strongest representative issue: it is broader, more self-contained, and describes a configuration bug with an explicit causal chain and reproducible assertion failure.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Both are 4-bit inference problems, but the code paths and failures differ: Gemma4/audio `torch.finfo()` on uint8 weights versus `BitsAndBytesConfig` skip-module handling and `lm_head` exclusion. 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": 15, + "component_count": 17, + "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": "ecea8af619346c989f5b900b9772e6679e44788b4c0501647c8dfaca0dd8fb7b", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues appear unrelated and spam-like, each containing only a different person\u2019s name and date of birth. They do not describe the same bug or request, so they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the items are unrelated and lack any shared problem statement or technical context.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a good canonical choice; both look like standalone, non-technical/spam entries rather than duplicate reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "Different names and dates of birth with no shared bug, feature, or code context; 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": 16, + "component_count": 17, + "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": "9e00e8d85a689e610ece8633b33399b2de1d6ef9148503043f8a5ef9a91619c7", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are body-less and their titles are just different personal-name/date strings; they do not appear to describe the same bug or request, so this is not a duplicate cluster.", + "confidence": 0.98, + "canonical_issue_reason": "No true canonical issue: both entries are content-free and unrelated, so there is no shared underlying problem to canonicalize.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45764 is only marginally cleaner as a representative because its title is slightly more formatted, but it is not a meaningful duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "Reject: the titles are different personal-name/date strings with no body, code path, or problem statement indicating 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": 17, + "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": "b3add61107fbb46a35dc68775a64dd282d1d15f7503d0c9760d7ed0c2c855311", + "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": "The set contains three unrelated issue topics: one dataset listing, one duplicate example request pair, and one duplicate scraping-service request pair. Only the exact-topic pairs should be treated as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits the whole set because the items split into separate topics; if forced to pick a representative, issue 45872 best captures the concrete feature request and has an exact duplicate in 45873.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45872 is the strongest representative issue overall: it is specific, actionable, and its duplicate 45873 is clearly the same request. The other issues are either a different topic or a less distinctive example request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Different topics: a dataset bundle listing versus an example request; they are not the same underlying issue." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Different topics: dataset listing versus example request; no evidence they describe the same bug or change." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Same title and apparent same example-request content; these look like exact duplicates." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Both mention financial time series, but one is an example request and the other adds a scraping service; different underlying requests." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Example-request issue versus scraping-service request; not the same concrete change." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Same as above: example request and scraping-service request are distinct issues." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Same as above: not the same underlying request." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Same title and same feature request wording; these look like exact duplicates." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The packet supports two exact duplicate pairs (45870/45871 and 45872/45873) plus the distinct dataset-listing issue 45869. The soft-edge verdicts are appropriately cautious, and the summary does not overclaim beyond the titles/body metadata provided." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-39895-14", + "nodes": [ + "pull_request:39895", + "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:45795", + "pull_request:45922" + ], + "soft_pairs": [ + "pull_request:39895|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:41855|pull_request:43424", + "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:45922", + "pull_request:43823|pull_request:43995", + "pull_request:43995|pull_request:44827", + "pull_request:43995|pull_request:45213", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:45795", + "pull_request:44827|pull_request:45655", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:43823|pull_request:45795", + "pull_request:44827|pull_request:45922", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "01eef6c2206db6eadceab27e75fd7f54e2654f196a3ced41e1047993a61e9a64", + "component_budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 21, + "serialized_chars": 39196, + "estimated_input_tokens": 9799, + "estimated_eval_tokens": 19854 + }, + "budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 21, + "serialized_chars": 39196, + "estimated_input_tokens": 9799, + "estimated_eval_tokens": 19854 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Most PRs are unrelated model/docs/tooling changes. The only accepted duplicate pair is the two `fixed_cross_entropy` patches, which both target the same `loss_utils.py` kwargs-support bug and issue 43240.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43251 is a solid canonical representative for the `fixed_cross_entropy` duplicate pair: it updates `loss_utils.py` to forward supported `cross_entropy` kwargs and explicitly targets issue 43240.", + "best_issue_reason": null, + "best_pr_reason": "PR 43251 is the strongest overall representative in this packet because it is the clearest example of the accepted `fixed_cross_entropy` fix cluster and has the more developed patch/review trail of the two near-duplicates.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "VideoPrism model addition vs Executorch dynamic-shape export test; unrelated changes." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Mistral tokenizer helper methods vs adding a new MobileLLM model; different workstreams." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Tokenizer API fix vs Executorch export test; not the same underlying bug." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor robustness changes vs new MobileLLM model implementation; distinct scopes." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy helper bugfix vs new MobileLLM model addition; unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "`fixed_cross_entropy` kwargs fix and Mistral4 model/test changes address different code paths." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch export testing is unrelated to Mistral4 fixes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Dynamic-shape export test vs new model addition; no common defect." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "MobileLLM model code vs config-resolution ordering fix; different problems." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45922", + "accept": false, + "reason": "New model addition vs vision-model lru memory leak fix; unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "MobileLLM model addition and Falcon output-interface refactor are different changes." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-collection refactor vs Mistral4 fixes; not the same bug or patch." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor is unrelated to the model-creation skill PR." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test/model fixes are unrelated to the `layrnorm` -> `layernorm` rename cleanup." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45795", + "accept": false, + "reason": "Mistral4 changes and audio/video processor docs do not address the same problem." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 fixes are unrelated to the config-resolution ordering change." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 test/model work does not match the model-creation skill PR." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45922", + "accept": false, + "reason": "Mistral4 fixes and BEiT lru/memory-leak fix are unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs patch `loss_utils.py` for the same `fixed_cross_entropy` kwargs-support bug and both target issue 43240." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "New MobileLLM model implementation is unrelated to the model-creation skill repository changes." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45795", + "accept": false, + "reason": "MobileLLM model code vs docs for audio/video processors; not the same underlying change." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45922", + "accept": false, + "reason": "Mistral4 test/model fixes and BEiT cache/memory-leak fix are unrelated issues." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "Reject: the output overstates evidence. `pull_request:43251` is not the only item with an explicit issue link; `pull_request:43254` also targets issue 43240. The duplicate-pair judgment for 43251/43254 is fine and conservative, but the representative/canonical rationale should avoid that claim and avoid implying validation/completeness beyond the packet." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 2, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40546-2", + "nodes": [ + "pull_request:40546", + "pull_request:42781" + ], + "soft_pairs": [ + "pull_request:40546|pull_request:42781" + ], + "prepared_review_unit_hash": "bf294f34623077bc9928368be50e94ba6f523a33509ac9fc336b370e643f949f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7225, + "estimated_input_tokens": 1807, + "estimated_eval_tokens": 3870 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7225, + "estimated_input_tokens": 1807, + "estimated_eval_tokens": 3870 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs introduce VibeVoice support, but they are not duplicates: #40546 is the original VibeVoice implementation, while #42781 adds a realtime variant plus the acoustic tokenizer and related generation changes. They overlap on shared model plumbing, but they do not fix the same concrete code-path problem.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "#40546 is the better canonical PR for the cluster because it is the core VibeVoice implementation and already has substantial review activity; #42781 is a later, draft extension focused on realtime support rather than the same base change.", + "best_issue_reason": null, + "best_pr_reason": "#40546 is the best representative PR for VibeVoice because it contains the primary model bring-up, docs, auto-registry wiring, and conversion/modeling pieces that define the feature.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "Shared filenames indicate common VibeVoice infrastructure, but #42781 is not the same change: it adds realtime and acoustic-tokenizer work on top of the base implementation instead of duplicating the exact code-path fix in #40546." + } + ] + }, + "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": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-40755-2", + "nodes": [ + "pull_request:40755", + "pull_request:41224" + ], + "soft_pairs": [ + "pull_request:40755|pull_request:41224" + ], + "prepared_review_unit_hash": "d39afd3876776a5ba5206f21e5011ec5ba54b47e91f9af9d9e18d6bbc62bd975", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two PRs are unrelated feature additions in different subsystems: TimesFM covariate forecasting vs. DINOv3 image classification support. They should not be clustered as duplicates or merged into one change.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: these PRs address different models and different code paths, so there is no shared underlying change to deduplicate around.", + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a good global representative for the cluster because the cluster is mixed and the PRs are about unrelated features rather than the same fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Reject: TimesFM covariate forecasting and DINOv3 image classification support are unrelated model features, with different files, APIs, and code 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": 4, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-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": 5, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-41797-3", + "nodes": [ + "pull_request:41797", + "pull_request:45186", + "pull_request:45630" + ], + "soft_pairs": [ + "pull_request:41797|pull_request:45186", + "pull_request:45186|pull_request:45630" + ], + "prepared_review_unit_hash": "4ab6116155ca22b10c5467ed2536705a1cc299cfec60c7bdba04f5a145669b32", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10407, + "estimated_input_tokens": 2602, + "estimated_eval_tokens": 5460 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10407, + "estimated_input_tokens": 2602, + "estimated_eval_tokens": 5460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three items are separate model-addition PRs (DeepSeek OCR, Isaac, Kimi2-6) that share Transformers scaffolding but target different models and code paths. They are not duplicates of one another.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:41797", + "right": "pull_request:45186", + "accept": false, + "reason": "Both add different models and implementations; overlap is only in shared registration/docs infrastructure, not the same underlying feature or bug fix." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "These are distinct new-model integrations for different architectures. Shared files are generic auto-mapping/documentation hooks, which is not enough to treat them as one change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 6, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42975-2", + "nodes": [ + "pull_request:42975", + "pull_request:42976" + ], + "soft_pairs": [ + "pull_request:42975|pull_request:42976" + ], + "prepared_review_unit_hash": "5e5eef70db3b291541671002a8a63c7f3a40a1be97d264052b272a3f1ecbfe06", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5935, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5935, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related maintenance PRs for GitHub Actions upgrades, but they target different actions and different concrete workflow changes, so they are not duplicates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #42975 is the better representative because it is a broader, clearly scoped workflow-action compatibility upgrade centered on Node 24 support.", + "best_issue_reason": null, + "best_pr_reason": "PR #42975 is the strongest cluster representative: it has the clearest overarching purpose and the wider set of workflow action updates, while #42976 is a separate action-version bump for Docker build steps.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Both are GitHub Actions dependency bumps, but they update different action families and fix different compatibility paths (checkout/cache/upload-artifact vs docker/build-push-action). They are related maintenance changes, not the same underlying bug or mergeable single 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": 7, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-42978-7", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:43973", + "pull_request:44601", + "pull_request:45490", + "pull_request:45838" + ], + "soft_pairs": [ + "pull_request:42978|pull_request:43448", + "pull_request:42978|pull_request:43451", + "pull_request:42978|pull_request:45490", + "pull_request:43448|pull_request:43451", + "pull_request:43448|pull_request:45490", + "pull_request:43451|pull_request:45490", + "pull_request:43973|pull_request:45838", + "pull_request:44601|pull_request:45838", + "pull_request:45490|pull_request:45838" + ], + "prepared_review_unit_hash": "84d19cbe1bd8ce14c0de1c125c10059fd4052729f4a20a02b66f571241a74ea9", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 9, + "serialized_chars": 23536, + "estimated_input_tokens": 5884, + "estimated_eval_tokens": 12024 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 9, + "serialized_chars": 23536, + "estimated_input_tokens": 5884, + "estimated_eval_tokens": 12024 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "No true duplicate cluster here: the PRs are separate model additions or one distributed-infra change, and the shared files are mostly generic docs/auto-registry plumbing. All soft edges should be rejected.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different models (ViT NEPA vs Molmo). Similarity is only from shared docs and auto-registry edits, not the same code path." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "Different model families (ViT NEPA vs Molmo2). Shared auto/model registry churn is superficial." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "ViT NEPA model add vs CTSM model add; only broad Transformers registry/docs overlap." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo vs Molmo2 are related names but distinct model integrations with different files, configs, and processors." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are unrelated model additions; overlap is limited to generic model-registration files." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 vs CTSM are different model implementations and docs, not the same underlying change." + }, + { + "left": "pull_request:43973", + "right": "pull_request:45838", + "accept": false, + "reason": "Lfm2 audio model add vs PP-OCRv6 text-recognition support; different modalities and code paths." + }, + { + "left": "pull_request:44601", + "right": "pull_request:45838", + "accept": false, + "reason": "Pipeline-parallel loading infrastructure is not the same as OCR model support; no concrete shared bug/change." + }, + { + "left": "pull_request:45490", + "right": "pull_request:45838", + "accept": false, + "reason": "CTSM model add and PP-OCRv6 model support are separate model integrations with only generic registry/docs overlap." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43291-3", + "nodes": [ + "pull_request:43291", + "pull_request:43665", + "pull_request:44815" + ], + "soft_pairs": [ + "pull_request:43291|pull_request:43665", + "pull_request:43291|pull_request:44815", + "pull_request:43665|pull_request:44815" + ], + "prepared_review_unit_hash": "8e8f13cc6741964f0695d9d77355e69b4b3f063f1463c46f9011e61b9e5f0238", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7091, + "estimated_input_tokens": 1773, + "estimated_eval_tokens": 3802 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7091, + "estimated_input_tokens": 1773, + "estimated_eval_tokens": 3802 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The three PRs are not duplicates: one fixes Whisper tokenizer decoding, one is a CLIP/ViT test-only change, and one is an FP8/dequantization fix across loading and quantization code. The shared similarity is only superficial/title-level.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44815 is the strongest representative because it makes a concrete runtime bugfix in loading/FP8 code, whereas 43291 is Whisper-specific and 43665 is just test-file churn with no shared underlying change.", + "best_issue_reason": null, + "best_pr_reason": "PR 44815 is the best standalone PR in the set: it has the clearest technical scope and the most substantive code-path fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Whisper tokenizer decoding fix vs CLIP/ViT test edits; different models, different code paths, no shared underlying bug." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper timestamp/batch decode changes are unrelated to FP8 dequantization/loading fixes." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "Test-file edits for vision models are unrelated to model loading and FP8 quantization 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": 9, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43488-2", + "nodes": [ + "pull_request:43488", + "pull_request:43532" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:43532" + ], + "prepared_review_unit_hash": "a7fd27b2d7792d89eba54cd608c6c695574b148dada4ef3eef33af1a260b3f1a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4562, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4562, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated synthetic PRs: one checks repo-bot/version formatting in setup.py, the other alters CI/workflow and test utilities. They only share a Vit test filename, not the same underlying change or bug.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43532 is the more substantial and representative change in the cluster, with multiple workflow and test utility files touched; #43488 is a narrower bot/formatting check.", + "best_issue_reason": null, + "best_pr_reason": "PR #43532 is the best cluster representative, but it is still a non-mergeable synthetic PR rather than a real duplicate target.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "Different intents and code paths: #43488 is a setup/version-format repo-bot check, while #43532 rewires CI/workflow logic and test helpers. Shared Vit test file is incidental." + } + ] + }, + "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": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43612-2", + "nodes": [ + "pull_request:43612", + "pull_request:43613" + ], + "soft_pairs": [ + "pull_request:43612|pull_request:43613" + ], + "prepared_review_unit_hash": "a2aa1081108b6c8030fc2aa455ae0a498969ab14120c7c5646c39ff0e7f037ff", + "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 two similar-looking but distinct PRs: one adds a Promptable Concept Segmentation pipeline, the other adds a Promptable Visual Segmentation pipeline. They share scaffolding and docs updates, but they target different pipelines/models and are not duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a duplicate of the other; both are standalone feature additions for different segmentation tasks. If forced to pick a representative, 43613 is slightly broader, but it should not be treated as canonical for this cluster.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Same general segmentation theme, but different concrete features: concept segmentation vs visual segmentation, with different pipeline names, task docs, and model/test coverage. They are not the same code-path problem 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": 11, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43651-2", + "nodes": [ + "pull_request:43651", + "pull_request:43663" + ], + "soft_pairs": [ + "pull_request:43651|pull_request:43663" + ], + "prepared_review_unit_hash": "127267880010f0dd1d488cbb18d2bb69cd34a7e5dfeefb1f80aa9c6dc54b282e", + "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": "Two open PRs touch the same file but implement unrelated Trainer extensibility changes: one adjusts gradient-accumulation loss scaling, the other refactors signature-column filtering for custom trainers. They are not duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different code paths and goals: 43651 adds _loss_is_scaled_for_ga to control loss scaling during training, while 43663 adds _get_signature_columns to customize input-column filtering. Sharing trainer.py and a broad subsystem is not enough to treat them as the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 12, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43656-3", + "nodes": [ + "pull_request:43656", + "pull_request:43836", + "pull_request:43842" + ], + "soft_pairs": [ + "pull_request:43656|pull_request:43842", + "pull_request:43836|pull_request:43842" + ], + "prepared_review_unit_hash": "a5657bf8bd8971f77abacf83534a8b51201261eae1e5a051307ddd631435c9c1", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5992, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5992, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster centers on `src/transformers/cli/serve.py` and TypeAdapter-related NameErrors, but the PRs are not close enough to treat as the same mergeable fix. PR 43842 is the most focused patch; 43656 is a related but different typing/runtime cleanup, and 43836 is a broader Serve refactor with the same area touched.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43842 is the cleanest and most targeted fix for the pydantic-absent TypeAdapter NameError, with the smallest patch and a direct annotation-only change in `serve.py`.", + "best_issue_reason": null, + "best_pr_reason": "PR 43842 best represents the cluster because it addresses the concrete runtime NameError with a narrowly scoped change, unlike the other PRs that mix in broader typing or refactor work.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "They both touch `serve.py` and mention TypeAdapter, but the packet does not show the same concrete bugfix path or a patch that is clearly mergeable into one PR; the explicit issue targets also differ." + }, + { + "left": "pull_request:43836", + "right": "pull_request:43842", + "accept": false, + "reason": "They share the same file and issue target, but 43836 is a broader Serve refactor and not clearly the same focused TypeAdapter NameError fix as 43842." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is only partly grounded, but one soft-edge verdict is too strong. For 43656 vs 43842, the packet supports that both touch `src/transformers/cli/serve.py` and both reference a TypeAdapter-related NameError, but it does not clearly show the same concrete bugfix path or that the changes are straightforwardly mergeable into one PR. The explicit issue targets also differ. The 43836 vs 43842 rejection is conservative and consistent." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-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": 14, + "component_count": 27, + "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:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44066|pull_request:44071", + "pull_request:44018|pull_request:44068", + "pull_request:44066|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44116", + "pull_request:44018|pull_request:44086", + "pull_request:44018|pull_request:44085", + "pull_request:44071|pull_request:44085", + "pull_request:44068|pull_request:44086", + "pull_request:44059|pull_request:44086", + "pull_request:44085|pull_request:44086", + "pull_request:44098|pull_request:44101", + "pull_request:44059|pull_request:44722", + "pull_request:44018|pull_request:44071", + "pull_request:44017|pull_request:44722", + "pull_request:44017|pull_request:44161", + "pull_request:44017|pull_request:44073", + "pull_request:44068|pull_request:44071", + "pull_request:44071|pull_request:44086", + "pull_request:43997|pull_request:44073", + "pull_request:44017|pull_request:44071", + "pull_request:44073|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44068", + "pull_request:44073|pull_request:44161", + "pull_request:44018|pull_request:44059", + "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:44018|pull_request:44161", + "pull_request:43997|pull_request:44071", + "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:43997|pull_request:44129", + "pull_request:44071|pull_request:44129", + "pull_request:44010|pull_request:44018", + "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" + ], + "prepared_review_unit_hash": "4b1bd05d6acc17c27375dcea736d70030435ad997b5afcb04c0461dacd366197", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84885, + "estimated_input_tokens": 21222, + "estimated_eval_tokens": 42700 + }, + "budget": { + "node_count": 27, + "item_count": 27, + "soft_pair_count": 72, + "serialized_chars": 75084, + "estimated_input_tokens": 18771, + "estimated_eval_tokens": 37798 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of model-specific output-tracing refactors, not true duplicates across different architectures. The only clear duplicate set is the GPT-J/CodeGen trio, which all target the same standardized output-capturing change in `modeling_gptj.py` (and, for two of them, `modeling_codegen.py`).", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "44066 is the best anchor: it has the explicit tracking issue, covers the shared GPT-J/CodeGen output-tracing change, and is the most complete representative of the near-duplicate GPT-J follow-ups.", + "best_issue_reason": null, + "best_pr_reason": "44066 is the strongest representative PR for the cluster\u2019s true duplicate set: same concrete GPT-J output-tracing refactor, explicit issue target, and broader scope than the GPT-J-only follow-up.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different models and code paths (CVT/FNet vs GPT-J); same refactor theme is not enough." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Different architectures/models; both touch output tracing, but not the same underlying bug/change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor in `modeling_gptj.py`; one PR is just a narrower slice of the same change." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "Different vision models/code paths (RegNet/ResNet vs EfficientNet); not the same fix." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet vs GPT-J/CodeGen; only the high-level refactor pattern overlaps." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen vs EfficientNet are unrelated concrete code paths." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 vs DeBERTa V2; same standardized-tracing theme, different fixes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen vs MGP-STR are different model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen vs MPT; not the same code-path problem." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-Neo and GPT-J are related but distinct model code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen vs GPT-Neo; shared refactor style, but different changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen are separate model ports." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet vs GPT-J/CodeGen; unrelated concrete fixes." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR vs GPT-J/CodeGen are different model paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet vs GPT-J; only the broad output-tracing theme overlaps." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo vs ResNet are unrelated implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet vs MPT; not the same underlying issue." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo vs Flaubert/XLM are different model families and code paths." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo is distinct from GPT-J/CodeGen; similar refactor style is not enough." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT vs GPT-J/CodeGen; different concrete changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo vs Flaubert/XLM are unrelated model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo vs MGP-STR are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo vs GPT-J; related architecture, but not the same change." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT vs MGP-STR are different models and fixes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo vs MGP-STR are unrelated code paths." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT2 vs MGP-STR; same refactor motif, different model bug." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J vs MGP-STR are different concrete changes." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT vs Flaubert/XLM are different model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT2 vs GPT-J/CodeGen are unrelated model paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo vs MPT; only the decorator migration pattern matches." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer vs GPT-J/CodeGen are unrelated implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer vs LongT5 are different code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet vs SegFormer are distinct model refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet vs MPNet; same output-tracing theme, different bugs." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT vs ImageGPT are different model implementations." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet vs Flaubert/XLM are unrelated code paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet vs LongT5 are different concrete changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet vs GPT-Neo are separate model refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet vs LongT5; not the same underlying bug/change." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44085", + "accept": false, + "reason": "MPT vs GPT-J are different model paths." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBERT vs MPT are unrelated implementations." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet vs LongT5 are different models and fixes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo vs GPT2 are distinct model code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44074", + "accept": false, + "reason": "RegNet vs TextNet; same tracing pattern, different concrete changes." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44018", + "accept": false, + "reason": "SqueezeBERT vs GPT-Neo are unrelated model implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa V2 vs TextNet are different code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor; both are plausibly one combined PR." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J output-tracing change in the same file; the title mismatch does not change the code-path match." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J vs SpeechT5 are unrelated model paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "CVT/FNet vs RegNet/ResNet/RT-DETR ResNet are different fixes." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet vs SpeechT5 are unrelated model implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa V2 vs SpeechT5 are different concrete changes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa V2 vs Vits; same broad theme, different code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen vs MGP-STR are unrelated model fixes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen vs MPT; only the standardized-output refactor is shared." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44161", + "accept": false, + "reason": "GPT-Neo vs LongT5 are different model families and bugs." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", + "accept": false, + "reason": "CVT/FNet vs SegFormer are unrelated model refactors." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBERT vs DeBERTa V2 are distinct code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44101", + "accept": false, + "reason": "DeBERTa V2 vs Flaubert/XLM are different models." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The summary stays conservative about cross-architecture similarity, and the GPT-J/CodeGen duplicate set is supported by the packet: 44066 and 44722 clearly share the GPT-J/CodeGen output-tracing refactor, while 44085 is a narrower GPT-J-only slice of the same file-level change. I don\u2019t see an overclaim that would require rejection." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 14, + "component_count": 27, + "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:44024", + "pull_request:44056", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44129" + ], + "soft_pairs": [ + "pull_request:44024|pull_request:44076", + "pull_request:44073|pull_request:44074", + "pull_request:44056|pull_request:44073", + "pull_request:44056|pull_request:44074", + "pull_request:43997|pull_request:44010", + "pull_request:44010|pull_request:44129" + ], + "prepared_review_unit_hash": "6ea512a0d315d20d314bae5e2db51ffff5e7c9eeb8995bdf8bf2479785e51921", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84885, + "estimated_input_tokens": 21222, + "estimated_eval_tokens": 42700 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 6, + "serialized_chars": 17122, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are not duplicates; they are separate model-specific PRs applying the same output-tracing refactor pattern across different code paths. The cluster is best treated as a thematic series, not one mergeable change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44056 is a clean representative of the shared refactor pattern: it adds standardized output tracing plumbing (`capture_outputs`, `_can_record_outputs`) directly in a single model implementation.", + "best_issue_reason": null, + "best_pr_reason": "44056 is the most representative single PR in the cluster because it targets the core standardized output-capture mechanism with a straightforward, self-contained model change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "Both are output-tracing refactors, but they affect different models (FocalNet vs ImageGPT) and different forward paths, so they are not the same change." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Same refactor theme, but VisualBert and TextNet are distinct model implementations with separate code paths; not mergeable as one PR." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "Both standardize output capture, but MPNet and VisualBert are different model-specific implementations, so this is not a duplicate bug/change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "They share the same refactor style, but MPNet and TextNet are separate model paths and could not plausibly be merged into one PR." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are unrelated model files; this is the same migration pattern, not the same underlying change." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBert and SpeechT5 modify different model code paths; both are output-capture refactors but not the same concrete fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 15, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-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": "0cd4f8d8924d08edb362cab52fb23f143e1e0c103698f82a398d045350cfcd65", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40174, + "estimated_input_tokens": 10044, + "estimated_eval_tokens": 20344 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40174, + "estimated_input_tokens": 10044, + "estimated_eval_tokens": 20344 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a set of similar-but-separate model-specific refactor PRs for output tracing / can_return_tuple cleanup. They share a common theme, but they do not look like duplicate fixes to the same underlying code path.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44003 is the strongest single representative: it shows the most complete output-tracing plumbing change across closely related Mamba/FalconMamba files. The rest are separate model-specific refactors, not duplicates of one another.", + "best_issue_reason": null, + "best_pr_reason": "44003 is the best global representative for the cluster\u2019s theme because it is broad and concrete, but it still represents only one model family\u2019s tracing refactor rather than a duplicate of the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different architectures and files (timm_backbone vs upernet); same refactor theme only." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different model families and code paths; both are output-tracing cleanup, not the same change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "timm_backbone and univnet are unrelated implementations; no shared concrete bug/change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and filenames; similar tracing refactor but not one duplicate PR." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "timm_backbone vs rwkv are unrelated code paths; only the meta-pattern matches." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "MobileNetV1 and UnivNet are unrelated models; same tracing refactor theme, not the same underlying fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families/files; no concrete shared bug or merged-able change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "MobileNetV1 vs Depth Anything are different code paths; overlap is only in output-tracing cleanup." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Different architectures and files; same refactor motif only." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "MobileNetV1 and SpeechEncoderDecoder are separate model implementations; not the same PR change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "VisionTextDualEncoder and UnivNet are unrelated model paths; similar tracing refactor theme only." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families/files; no evidence of the same underlying bug." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "VisionTextDualEncoder vs Mamba/FalconMamba are different code paths; not a duplicate fix." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Different architectures and filenames; only the output-tracing refactor motif is shared." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "UnivNet and UperNet are unrelated implementations; same style of refactor, not the same change." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families/files; no shared concrete code-path problem." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "UnivNet vs Depth Anything are separate model paths; not mergeable as one duplicate PR." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are unrelated code paths; the similarity is only in tracing cleanup." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and files; not the same underlying bug/change." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Mamba/FalconMamba vs SuperPoint are different code paths; similar refactor theme only." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Different implementations and filenames; both mention tracing, but they fix different model code." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "timm_backbone and codegen are unrelated models; no duplicate underlying change." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet and CodeGen are separate code paths; only the meta-refactor overlaps." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are unrelated; not the same fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Depth Anything and VisionEncoderDecoder are different model families/files; no concrete duplicate change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "Depth Anything and SpeechEncoderDecoder are unrelated code paths; same theme only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Different models and filenames; not the same underlying bug/change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Depth Anything vs RWKV are unrelated implementations; same refactor pattern only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model families/files; no shared concrete code-path problem." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "VisionEncoderDecoder and RWKV are unrelated model paths; not a duplicate PR." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are distinct code paths; only the tracing refactor theme matches." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families/files; same meta-change, not same bug or change." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "VisionEncoderDecoder and DPR are different implementations; no duplicate underlying fix." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "SpeechEncoderDecoder and SuperPoint are unrelated models; not the same change." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different code paths and filenames; only the output-tracing cleanup overlaps." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "SpeechEncoderDecoder and DPR are separate model implementations; not mergeable as one duplicate PR." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV are unrelated code paths; similar refactor style only." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "SuperPoint and DPR are different model families/files; no shared concrete fix." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are unrelated implementations; same tracing-refactor motif only." + } + ] + }, + "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": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44662-3", + "nodes": [ + "pull_request:44662", + "pull_request:45075", + "pull_request:45401" + ], + "soft_pairs": [ + "pull_request:44662|pull_request:45401", + "pull_request:44662|pull_request:45075", + "pull_request:45075|pull_request:45401" + ], + "prepared_review_unit_hash": "ef6ffd227a8203a03fcd335e0c38df0e82cbeea4df50463454cf4de9b082ee5d", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate model-implementation PRs (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They overlap only on shared registration/docs/auto-mapping boilerplate, not on the same underlying code change, so this cluster should not be deduplicated.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single PR is canonical here: each one adds a different model family with distinct configs, modeling code, and tests.", + "best_issue_reason": null, + "best_pr_reason": "No PR is a global best duplicate candidate; they address unrelated model additions rather than one shared bug or feature.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different models and code paths (PenguinVL vs Voxtral TTS); the overlap is only generic auto/docs wiring." + }, + { + "left": "pull_request:44662", + "right": "pull_request:45075", + "accept": false, + "reason": "PenguinVL and DeepSeek-OCR-2 are distinct model implementations; shared files are boilerplate registration changes only." + }, + { + "left": "pull_request:45075", + "right": "pull_request:45401", + "accept": false, + "reason": "These add different model stacks (DeepSeek-OCR-2 vs Voxtral TTS) and do not look like 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": 17, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-44979-2", + "nodes": [ + "pull_request:44979", + "pull_request:45363" + ], + "soft_pairs": [ + "pull_request:44979|pull_request:45363" + ], + "prepared_review_unit_hash": "33cb670797c5d4e0aee911ee4459ba26f554c099a59bd5badc4f3f568a9039db", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs are related in the broad area of model fusion, but they are not the same change: one introduces a general Module Fusion API, while the other wires n-to-1 kernel fusion through KernelConfig and related integration points. They are likely complementary rather than duplicates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44979 is the more foundational change: it introduces the standalone Module Fusion API and test coverage, which can serve as the generic abstraction layer for later fusion integrations.", + "best_issue_reason": null, + "best_pr_reason": "PR #45363 is the more concrete user-facing feature for kernel-based n-to-1 fusion, but it still depends on broader fusion infrastructure and is not a duplicate of the API-only PR.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Not the same concrete bug/change. #44979 adds a generic Module Fusion API, while #45363 adds KernelConfig-based n-to-1 fusion plumbing and an example. They touch related code, but they 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": 18, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45086-2", + "nodes": [ + "pull_request:45086", + "pull_request:45317" + ], + "soft_pairs": [ + "pull_request:45086|pull_request:45317" + ], + "prepared_review_unit_hash": "23673118accf8423a4094233d191412a13b3b42d0ee6202779152e7f16d74bc1", + "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 target the same Mistral regex patch in the same file and fix the same AttributeError by switching from backend_tokenizer.pre_tokenizer to tokenizer.pre_tokenizer; PR 45317 is the more complete version because it adds tests.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45317 is the more complete representative of the same fix: same code change as 45086 plus test coverage and a clearer problem statement.", + "best_issue_reason": null, + "best_pr_reason": "PR 45317 best captures the cluster because it includes the same bug fix with additional tests and discussion, making it the most complete candidate.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying AttributeError fix in the same code path and same file; the second PR is essentially the first fix plus tests, so they are mergeable as one change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 19, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45115-2", + "nodes": [ + "pull_request:45115", + "pull_request:45569" + ], + "soft_pairs": [ + "pull_request:45115|pull_request:45569" + ], + "prepared_review_unit_hash": "b3db35f9213f4365eeae472e69949ddc23e3efcbc373887c406757c3f7c0bb9d", + "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 NemotronH-related PRs, but they target different changes: one is a refactor/inheritance and conversion-mapping update, the other is a broader model split into dense/sparse variants with docs and tests. They are related but not duplicates.", + "confidence": 0.83, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45569 is the broader, more complete implementation of the NemotronH model split and associated docs/tests, so it best represents the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #45569 is the strongest standalone candidate because it covers the user-facing model reorganization and supporting documentation/testing, not just internal refactoring.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "Shared NemotronH files, but the concrete changes differ: #45115 focuses on inheritance/conversion mapping to GraniteMoeHybrid, while #45569 introduces dense/sparse model variants plus docs/tests. They are related but not the same underlying fix 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": 20, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45332-2", + "nodes": [ + "pull_request:45332", + "pull_request:45333" + ], + "soft_pairs": [ + "pull_request:45332|pull_request:45333" + ], + "prepared_review_unit_hash": "2536324f84c89e604d3a69bb2933610a9a7e84d756d6e180dfce42f98d9af336", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5398, + "estimated_input_tokens": 1350, + "estimated_eval_tokens": 2956 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5398, + "estimated_input_tokens": 1350, + "estimated_eval_tokens": 2956 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related heterogeneity feature PRs, but they are not duplicates: one adds per-layer config plumbing, the other adds downstream model/cache/masking support on top of it.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45332 is the broader umbrella change: it includes the model/runtime changes that actually consume heterogeneous configs, so it best represents the full feature cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR 45332 is the most representative end-to-end implementation and covers the larger concrete code-path change across modeling, cache, and masking logic.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Related but distinct work: 45333 adds heterogeneous config support and helpers, while 45332 adds model/cache/masking behavior that consumes those configs. They are prerequisite/follow-on changes, not the same underlying bug or a single 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": 21, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45415-2", + "nodes": [ + "pull_request:45415", + "pull_request:45425" + ], + "soft_pairs": [ + "pull_request:45415|pull_request:45425" + ], + "prepared_review_unit_hash": "11a56c8a04125bf52179caa118edf5248285ac1b20ce85d897ca590fdfe0c357", + "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": "Two open PRs in the typing area, but they are not duplicates: one is a broad type-checking sweep across many Transformers modules, while the other adds typing protocols/helpers and related modeling-utils plumbing. The overlap in a few shared files looks incidental rather than the same underlying change.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45415 is the broader, more representative typing cleanup across `src/transformers/*py` and best matches the cluster\u2019s overall surface area.", + "best_issue_reason": null, + "best_pr_reason": "PR #45415 is the best single representative for the cluster because it is the larger, more central typing sweep; PR #45425 is a narrower typing-infrastructure follow-up with different intent.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Both are typing-related, but they address different changes: #45415 is a wide type-checking pass, while #45425 introduces specific typing protocols and helpers. Shared files alone do not make them the same underlying bug/change, and they would not naturally 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": 22, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45487-6", + "nodes": [ + "pull_request:45487", + "pull_request:45817", + "pull_request:45843", + "pull_request:45858", + "pull_request:45890", + "pull_request:45926" + ], + "soft_pairs": [ + "pull_request:45487|pull_request:45843", + "pull_request:45487|pull_request:45926", + "pull_request:45817|pull_request:45843", + "pull_request:45858|pull_request:45890", + "pull_request:45858|pull_request:45926", + "pull_request:45890|pull_request:45926" + ], + "prepared_review_unit_hash": "cb0d30bcff04a3339453f2c9788f25361f9f06b820999652dbbd590388a918cd", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 6, + "serialized_chars": 14708, + "estimated_input_tokens": 3677, + "estimated_eval_tokens": 7610 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 6, + "serialized_chars": 14708, + "estimated_input_tokens": 3677, + "estimated_eval_tokens": 7610 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are all about accelerator/model-parallel or XPU-related behavior, but they target different models and different fixes. I don\u2019t see any pair that is the same underlying bug/change, so none of the soft edges should be merged.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45817 is the strongest representative because it makes a real code-path fix for Gemma4 model-parallel behavior, not just test expectation updates, and it has the clearest end-to-end change set.", + "best_issue_reason": null, + "best_pr_reason": "PR #45817 is the best standalone PR in the set: it addresses an actual model-parallel bug in model code and includes corresponding test coverage, making it the most substantial and actionable change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45487", + "right": "pull_request:45843", + "accept": false, + "reason": "Both are model-parallel fixes, but for different architectures (AltCLIP/ChineseCLIP vs Deimv2). The code paths and bugs are unrelated, so they should not be deduplicated." + }, + { + "left": "pull_request:45487", + "right": "pull_request:45926", + "accept": false, + "reason": "These touch different models and different accelerator concerns; one is a model-parallel fix, the other is XPU test expectations. Same broad theme, not the same bug." + }, + { + "left": "pull_request:45817", + "right": "pull_request:45843", + "accept": false, + "reason": "Gemma4 and Deimv2 are unrelated model implementations; the shared phrase 'model parallel bugs' is too broad to treat as one duplicate change." + }, + { + "left": "pull_request:45858", + "right": "pull_request:45890", + "accept": false, + "reason": "Both only add XPU expectations, but for different model test suites. This is test maintenance for separate models, not one underlying defect." + }, + { + "left": "pull_request:45858", + "right": "pull_request:45926", + "accept": false, + "reason": "Same accelerator class of change, but different models and different expected outputs. No evidence they fix the same concrete code-path problem." + }, + { + "left": "pull_request:45890", + "right": "pull_request:45926", + "accept": false, + "reason": "These are separate XPU expectation updates for distinct model tests; they do not appear to be 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": "pull_request", + "component_index": 23, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45548-2", + "nodes": [ + "pull_request:45548", + "pull_request:45662" + ], + "soft_pairs": [ + "pull_request:45548|pull_request:45662" + ], + "prepared_review_unit_hash": "8b9991acf91473b763654dc3318c8e370ac1dfd033614cae3daa1f6d456d81da", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are in the EP/distributed-loading area, but they fix different bugs: #45548 focuses on preserving distributed config / EP state during accelerate+ZeRO-3 loading, while #45662 fixes expert tensors being overwritten by rank-0 broadcast in the FSDP2 loading path. Shared filename and theme are not enough to treat them as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45662 is the better representative of the cluster theme because it targets a concrete EP loading corruption bug, whereas #45548 is a narrower serialization/config-preservation fix for a different code path.", + "best_issue_reason": null, + "best_pr_reason": "PR #45662 is the strongest standalone candidate: it addresses a specific, user-visible failure mode in the distributed loading path and has the more direct fix for the underlying bug.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Different underlying problems and code paths: #45548 removes distributed_config from serialization and adds has_ep for accelerate/ZeRO-3 loading, while #45662 prevents expert parameters from being overwritten during FSDP2/post-shard loading. Related area, but not the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 24, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45639-2", + "nodes": [ + "pull_request:45639", + "pull_request:45645" + ], + "soft_pairs": [ + "pull_request:45639|pull_request:45645" + ], + "prepared_review_unit_hash": "e683d343d857337f7772df82e698ee948f634161cd5c0b36c062835f44855393", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4589, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4589, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target the same xdist collision around `captured_info` debug artifacts; one is a narrower code-only fix, while the other bundles the same fix with CI log collection updates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45645 is the broader superset: it contains the xdist-safe `captured_info` handling in `testing_utils.py` plus the CI workflow update to read all `captured_info*.txt` files, so it best represents the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #45645 is the best representative because it addresses the same concrete bug as #45639 and also updates the downstream consumer of those files, making it the more complete end-to-end fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Both PRs fix the same underlying xdist collision for `captured_info` artifacts in the testing debug path; #45645 simply adds the CI-side aggregation needed for the same change and could plausibly be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 25, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45683-2", + "nodes": [ + "pull_request:45683", + "pull_request:45734" + ], + "soft_pairs": [ + "pull_request:45683|pull_request:45734" + ], + "prepared_review_unit_hash": "eba4ca0f26b2d3be0c00e316270a9341d31af2208b44dbab31b1d3a8e81e7538", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4469, + "estimated_input_tokens": 1118, + "estimated_eval_tokens": 2492 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4469, + "estimated_input_tokens": 1118, + "estimated_eval_tokens": 2492 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs touch the same quantization helper, but they fix different bugs: one special-cases audio submodules, the other changes how default skip lists are combined with user-supplied skips. They should not be clustered as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "45734 is the better representative PR because it fixes the core skip-module composition logic in `get_modules_to_not_convert`, which is broader and more central than the narrow audio-module special case in 45683.", + "best_issue_reason": null, + "best_pr_reason": "45734 is the strongest standalone PR in the set: it addresses a concrete quantization failure mode in the main code path and its fix is directly about the helper\u2019s behavior rather than a model-specific workaround.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45683", + "right": "pull_request:45734", + "accept": false, + "reason": "Both touch quantization skip logic, but 45683 adds audio-tower/embed_audio exclusions while 45734 changes additive handling of `skip_modules`; these are different underlying bugs and would not naturally 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": 26, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-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": 27, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45975-2", + "nodes": [ + "pull_request:45975", + "pull_request:45977" + ], + "soft_pairs": [ + "pull_request:45975|pull_request:45977" + ], + "prepared_review_unit_hash": "3f07da4de38f5f80d631618c18d5634fa68244a564f82c75c1cbb5e278edbb4c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4523, + "estimated_input_tokens": 1131, + "estimated_eval_tokens": 2518 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4523, + "estimated_input_tokens": 1131, + "estimated_eval_tokens": 2518 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs target GGUF on Apple Silicon and kernels-community integration, but they address different layers: one adds the core GGUF matmul/linear implementation, while the other adds an optional Metal dequant fast path plus CLI/model-manager wiring. Related topic, not the same underlying change.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45977 is the core functional change: it introduces the new GgufLinear inference path and related quantizer/modeling support. PR 45975 is narrower configuration/docs plumbing around selecting the kernels-community path.", + "best_issue_reason": null, + "best_pr_reason": "PR 45977 is the best representative for the cluster because it contains the substantive implementation that likely motivated the broader GGUF-on-Apple-Silicon work.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45975", + "right": "pull_request:45977", + "accept": false, + "reason": "They share the GGUF/Metal subsystem, but 45977 implements the main GGUF linear matmul path while 45975 only adjusts optional dequant/serve integration and help text. They are not the same concrete bug/change and would not naturally collapse into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/manifest.json b/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..28584983a5ff9d3268cd5d5e77e28e5ee0c23d33 --- /dev/null +++ b/snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260515t180026z", + "artifacts": { + "hybrid": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-05-15T18:10:41Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260515T180026Z", + "variant": "hybrid" +} diff --git a/snapshots/20260515T180026Z/manifest.json b/snapshots/20260515T180026Z/manifest.json index 7d0920817c3621ddf63de7d7a65a27f7f3c41ce1..878bbb24a82d4c3e79b2ca8b561979ca952429e0 100644 --- a/snapshots/20260515T180026Z/manifest.json +++ b/snapshots/20260515T180026Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 925 }, "extracted_at": "2026-05-15T18:00:26Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260515t180026z", + "runs": { + "hybrid-model-20260515t180026z": { + "analysis_id": "hybrid-model-20260515t180026z", + "artifacts": { + "hybrid": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260515T180026Z/analysis-runs/hybrid-model-20260515t180026z/manifest.json", + "model": null, + "published_at": "2026-05-15T18:10:41Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260515T180026Z", "watermark": {